Motor parameter estimation method and apparatus

[0001] This application is a divisional of U.S. patent application Ser. No. 10/675,104, entitled “Motor Parameter Estimation Method and Apparatus,” filed Sep. 30, 2003, which is incorporated by reference herein.

[0002] The present technique relates generally to the field of electric motors. More particularly, the invention relates to a novel technique for estimating unknown parameters of an induction motor based on motor data obtained at one or more operating points or a no-load operating point.

[0003] A wide variety of induction motors are available and are currently in use throughout a range of industrial applications. In general, such motors include a stator provided in a motor housing and a rotor surrounded at least partially by the stator and supported for rotation within the housing. The stator and rotor may be mechanically and electrically configured in a variety of manners depending upon a number of factors, including: the application, the power available to drive the motor, and so forth. In general, however, electric power is applied to the stator to produce a rotating magnetic field to drive the rotor in rotation. Mechanical power is transmitted from the motor via an output shaft coupled to the rotor.

[0004] Motor operating parameters, such as output torque or efficiency, may only be determined with the motor in operation. Knowledge of these motor operating parameters can be important for a number of reasons. However, the devices used to measure motor operating parameters may interfere with the operation of the motor or may be relatively expensive. In addition, it may be difficult to measure the operating parameter. For example, it may be desirable to maintain the temperature of the rotor below a specific temperature. However, it is extremely difficult to measure the rotor temperature. In addition, it may be desirable to establish the torque and/or efficiency of a given motor to ensure that the proper motor is used in a given application. However, a typical torque measuring device requires the motor to be disconnected from its load each time the torque measurement is desired, interfering significantly with the operation of the motor. Previous attempts to develop a device to estimate motor operating parameters, such as torque and efficiency, have relied on motor nameplate data. However, these attempts have not yielded accurate results. Alternatively, a customer may not have the values of the motor electrical parameters that might be used to develop an estimate of various motor operating parameters.

[0005] A need exists for a technique for obtaining electric motor operating parameter data that is less expensive than conventional methods and which minimizes the disruption to the operation of a system incorporating the electric motor.

[0006] The present technique provides a novel system and method for establishing estimated values of electrical parameters of a motor. The electrical parameters may be established from motor databases, measured input electrical data, measured output data, and various estimations to account for unknown motor parameters. Compensations also may be provided for stator resistance, cable resistance, and other motor parameters. Based on the foregoing data, the system and method also may be used to estimate motor operating parameters, such as torque, efficiency, output power, output speed, and other performance criteria of the motor. The system and method also may establish energy and monetary comparison data between the motor and at least one alternative motor.

[0007] The foregoing and other advantages and features of the invention will become apparent upon reading the following detailed description and upon reference to the drawings in which:

[0008] FIG. 1 is a perspective view of an electric motor illustrating the various functional components of the motor including a rotor and a stator, in accordance with certain aspects of the invention;

[0009] FIG. 2 is the single-phase steady state equivalent schematic circuit of an induction motor, according to an exemplary embodiment of the present technique;

[0010] FIG. 3 is a system for providing estimated values of various motor operating parameters, according to an exemplary embodiment of the present technique;

[0011] FIG. 4 is a process for providing estimated values of various motor operating parameters based on data obtained at two load conditions of the motor, according to an exemplary embodiment of the present technique;

[0012] FIG. 5 is an alternative equivalent schematic circuit of a steady state induction motor, according to an exemplary embodiment of the present technique;

[0013] FIG. 6 is an alternative process for providing estimated values of various motor operating parameters based on data obtained with no-load on the motor, according to an exemplary embodiment of the present technique;

[0014] FIG. 7 is another alternative process for providing estimated values of various motor operating parameters based on data obtained at a single load on the motor, according to an exemplary embodiment of the present technique;

[0015] FIG. 8 is further alternative process for providing estimated values of various motor operating parameters based on data obtained at first, second, and third loads on the motor, according to an exemplary embodiment of the present technique;

[0016] FIG. 9 is another alternative process for providing estimated values of various motor operating parameters based on baseline motor parameters and data obtained at a desired operating load on the motor, according to an exemplary embodiment of the present technique; and

[0017] FIG. 10 is a system for providing estimated values of various motor operating parameters, according to an exemplary embodiment of the present technique.

[0018] Turning now to the drawings, and referring first to FIG. 1, an electric motor is shown and designated generally by the reference numeral 20. In the embodiment illustrated in FIG. 1, motor 20 is an induction motor housed in a conventional NEMA enclosure. Accordingly, motor 20 includes a frame 22 open at front and rear ends and capped by a front end cap 24 and a rear end cap 26. The frame 22, front end cap 24, and rear end cap 26 form a protective shell, or housing, for a stator assembly 28 and a rotor assembly 30. Stator windings are electrically interconnected to form groups, and the groups are, in turn, interconnected. The windings are further coupled to terminal leads 32. The terminal leads 32 are used to electrically connect the stator windings to an external power cable (not shown) coupled to a source of electrical power. Energizing the stator windings produces a magnetic field that induces rotation of the rotor assembly 30. The electrical connection between the terminal leads and the power cable is housed within a conduit box 34.

[0019] In the embodiment illustrated, rotor assembly 30 comprises a cast rotor 36 supported on a rotary shaft 38. As will be appreciated by those skilled in the art, shaft 38 is configured for coupling to a driven machine element (not shown), for transmitting torque to the machine element. Rotor 36 and shaft 38 are supported for rotation within frame 22 by a front bearing set 40 and a rear bearing set 42 carried by front end cap 24 and rear end cap 26, respectively. In the illustrated embodiment of electric motor 20, a cooling fan 44 is supported for rotation on shaft 38 to promote convective heat transfer through the frame 22. The frame 22 generally includes features permitting it to be mounted in a desired application, such as integral mounting feet 46. As will be appreciated by those skilled in the art, however, a wide variety of rotor configurations may be envisaged in motors that may employ the techniques outlined herein, including wound rotors of the type shown, and so forth. Similarly, the present technique may be applied to a variety of motor types having different frame designs, mounting and cooling styles, and so forth.

[0020] Referring generally to FIG. 2, an equivalent circuit for steady state operation of the induction motor of FIG. 1 is shown and designated generally by the reference numeral 50. The induction motor is powered by an AC power source, designated by reference numeral 52, having a voltage amplitude V₁ and a frequency ω. The stator of the motor has an electrical resistance R₁, as represented by reference numeral 54, and a leakage inductance L₁, as represented by reference numeral 56. The motor also has core loss resistance R_c due to core losses in the stator and rotor, designated by the reference numeral 58. The motor also has a magnetizing inductance L_m, designated by reference numeral 60. The rotor also has an electrical resistance R₂, designated by reference numeral 62. As illustrated, the rotor resistance R₂ is modified by dividing the rotor resistance R₂ by the slip s of the rotor. Finally, the rotor also has a leakage inductance L₂, as represented by reference numeral 64. Electric current flows through the stator to produce the magnetic field. The electric current I₁ through the stator is represented by arrow 66. In addition, the magnetic field induces an electric current I₂ in the rotor, as represented by arrow 68. Finally, electric current I₃ flowing through the core loss resistance R_c and the magnetizing inductance L_m is represented by arrow 70.

[0021] In a typical AC circuit, the voltage and current vary over time. In an inductive circuit, such as an induction motor, the voltage leads the current by an angle, known as the phase angle φ. In addition, some power is alternately stored and released by the inductance of the circuit. This power is known as the “reactive power.” In addition, the resistance of the circuit dissipates power as heat and the load utilizes a portion of the input power, this power is known as the “real power.” The “apparent power” is the product of the total voltage and the total current in the AC circuit. The ratio between the real power and the apparent power of a load in an AC circuit is known as the “power factor” of the load. The cosine of the phase angle is the power factor.

[0022] Referring generally to FIG. 3, a system for providing estimated values of various motor electrical parameters and motor operating parameters is shown and designated generally by reference numeral 80. The system 80 comprises a data processing module 82 that is electrically coupleable to the motor 20. The data processing module 82 is operable to utilize data obtained at two load conditions of the motor 20 to establish values of various electrical parameters of the motor, such as the electrical resistance of the rotor and the leakage inductance of the stator and rotor. The data processing module then uses the values of the estimated motor electrical parameters to estimate motor operating parameters, such as the temperature of the rotor, the torque of the motor, and the efficiency of the motor. The data processing module 82 may be provided as a stand-alone device, as part of a motor, or in a kit form to be added to an existing installed motor.

[0023] In the illustrated embodiment, the data processing module 82 has a processor module 84. Preferably, the processor module 84 utilizes a processor (not shown) and operates in accordance with programming instructions to produce estimates of various motor operating parameters. The processor module 84 may have analog-to-digital converters for converting analog data into digital data. In this embodiment, the processor module 84 is electrically coupled to each phase 86 of the input power to the motor 28 to enable the module to receive electrical input data, such as the input voltage, current, frequency, and power. However, the data also may be entered into the system manually. The input voltage data may be the line-to-line voltage or the phase voltage. The average phase voltage for a connection may be established by averaging the three line-to-line voltages and dividing by the √{square root over (3)}. The average line current is the phase current. Input power data also may be obtained directly or calculated from the stator voltage, current, and resistance data. A speed sensor 88 also is electrically coupled to the processor module 84. The speed sensor 88 may be integral with the motor or a separate device coupled to the processor module 84. The speed sensor 88 may measure the speed of the shaft 38 coupled to-the rotor 36 in order to establish rotor speed. Alternatively, the speed sensor 88 may measure the speed of the rotor 36 directly.

[0024] In the illustrated embodiment, the system 80 is operable to output motor electrical parameter data and motor operating parameter data to a control module 90. Preferably, the control module 90 has a visual display 92 to provide visual indications of the various parameters. Preferably, the control module 90 has a keypad or keyboard 94 to enable data, such as the electrical input data, rotor speed data, and any known motor electrical parameters, to be inputted into the processor module 84. In addition, in the illustrated embodiment the processor module 84 and the control module 90 are coupled to a network 96 to enable data to be transferred to or from remote terminals 98. The remote terminals 98 may be personal computers, or other digital communication devices.

[0025] The electrical input data may also be measured at the motor controller, rather than at the motor itself. However, in certain applications the motor controller may be quite remote from the motor. To facilitate the measurement of data at the motor, such as the rotor speed, and at other locations, such as at a motor controller, a time log of the measured voltages, currents, power and frequency may be used to record data. The voltages, currents, power and frequency corresponding to the time of the speed measurement are retrieved from the time log and matched to the speed of the rotor at that time. An adjustment also can be provided for the effect on the electrical input data caused by taking the measurement at the motor controller. First, the length of the cable between the motor and the starter may be measured. In addition, the ambient temperature is measured and the gauge of the cable identified. The diameter of the conductor may be calculated from the gauge of the cable. The resistance of the cable may be estimated based on the operating temperature, the length and diameter of the cable. The cable resistance is then subtracted from the total measured resistance to establish the stator resistance. Furthermore, the power loss in the cable may be established from the measured current and estimated cable resistance. The cable power is then subtracted from the measured power to obtain the power delivered to the motor.

[0026] Referring generally to FIG. 4, a process for establishing values of various motor electrical parameters and various motor operating parameters using the system of FIG. 3 is shown and designated generally by reference numeral 100. The process comprises obtaining the resistance of the stator, as represented by block 102. The process also comprises obtaining data at a first operating load point and providing the data to the processor module 84, as represented by block 104. In a presently contemplated embodiment, the data obtained at the first load point comprises: input voltage data, input current data, input power data, and shaft speed data. It should be noted that the input power can either be measured or calculated from the other input data. In addition, the process may measure motor frequency and temperature. Some data may be provided to the system 80 using the control module 90 or may be provided from a remote station 98 via the network 96. Preferably, the motor has a low load at the first operating point.

[0027] The process also comprises obtaining data from the motor at a second load point and providing the data to the processor module 84, as represented by block 106. The stator resistance R₁ data need only be obtained once if the stator temperature is obtained at each load point. Preferably, the motor has a full load at the second load point.

[0028] The data processing module 82 may then be operated to establish estimated values of various motor parameters, as represented by block 108. The programming instructions provided to the data processing module 82 are adapted to utilize a novel technique for establishing the values of the various motor parameters. The equivalent circuit of FIG. 2 provides a starting point to illustrate the development of the technique for estimating various motor parameters. Referring generally to FIG. 5, an equivalent circuit, designated generally by reference numeral 110, to the circuit of FIG. 2 is illustrated. In FIG. 5, each inductance illustrated in FIG. 2 is converted into an inductive reactance to facilitate solving for the unknown motor parameters. In addition, some of the reactances are combined to simplify the circuit 110. The stator leakage reactance X₁, designated by reference numeral 112, is a function of the electrical frequency ω of the power source and the stator leakage inductance L₁. The equivalent reactance X_e, designated by reference numeral 114, is a function of the magnetizing reactance X_m, the rotor resistance R₂, the slip s and the rotor leakage reactance X₂. The magnetizing reactance X_m, in turn, is a function of the electrical frequency ω and the magnetizing inductance L_m. The rotor leakage reactance X₂ is a function of the electrical frequency ω and the rotor leakage inductance L₂. The equivalent resistance R_e, designated by reference numeral 116, is a function of the rotor resistance R₂, the leakage reactance X₂, the slip s, and the core loss resistance R_c. Of the parameters illustrated in FIGS. 2 and 5, the stator resistance R₁ and the motor slip s can be measured relatively easily. This leaves the values of five parameters to be established: X₁, R₂, X₂, R_c, and X_m. These parameters are more difficult to measure than the stator resistance R₁ and the motor slip s.

[0029] Several assumptions and an approximation are made to simplify the process of developing a technique for estimating X₁, R₂, X₂, R_c, and X_m. Namely, it is assumed that the frequency of the power is constant, that the speed of the rotor does not change during the gathering of the load point data, and that the reading of the data is done quickly so that the rotor temperature is constant during the gathering of the data. Additionally, it has been established experimentally that excellent results are obtained by estimating the stator leakage reactance X₁ to be 5% of the magnetizing reactance X_m, or:
X₁=0.05X_m.   (1)
However, this factor may range from 0.02 to 0.07. By making this approximation the number of unknowns is reduced to four. Thus, only four equations are needed to solve for the values of the remaining unknown motor parameters. However, the equations relating these unknowns are highly nonlinear and an expression for the remaining unknowns by using measurements obtained at two load points is nontrivial. In the present technique, this process is facilitated by obtaining an actual value for the stator leakage reactance X₁. This value is then used in finding the values of the remaining unknowns.

[0030] In addition, the rotor leakage inductance L₂ and magnetizing inductance L_m are converted into reactances in FIG. 5 to assist in solving the various unknown motor parameters. Reactance is a function of the inductance and the frequency ω of the circuit. The reactances were combined with the rotor resistance R₂ and the core loss resistance R_c to form an equivalent reactance X_e and a total resistance R_t. At a first load point, the total resistance R_t1 is given by the following equation: $\begin{array}{cc}\frac{1}{R_{t1}}=\frac{1}{R_c}+\frac{1}{\left(\frac{R_2}{s_1}+\frac{s_1{X}_2^2}{R_2}\right)}.& \left(2\right)\end{array}$

[0031] The first term on the right side of the equation is the reciprocal of the core loss resistance R_c and the second term is the reciprocal of the new modified rotor resistance as a result of factoring the rotor leakage reactance X₂. At the second load point, the total resistance R_t2 is given by the following equation: $\begin{array}{cc}\frac{1}{R_{t2}}=\frac{1}{R_c}+\frac{1}{\left(\frac{R_2}{s_2}+\frac{s_2{X}_2^2}{R_2}\right)}.& \left(3\right)\end{array}$

[0032] Similarly, the equivalent reactances at the two motor load points X_e1 and X_e2 are given by the following equations: $\begin{array}{cc}\frac{1}{X_{e1}}=\frac{1}{X_m}+\frac{X_2}{\left(\frac{R_2^2}{s_1^2}+X_2^2\right)};\mathrm{and}& \left(4\right)\\ \frac{1}{X_{e2}}=\frac{1}{X_m}+\frac{X_2}{\left(\frac{R_2^2}{s_2^2}+X_2^2\right)}.& \left(5\right)\end{array}$
The right hand sides of equations (4) and (5) also have two terms, one resulting from the magnetizing reactance X_m and the other resulting from factoring the rotor leakage reactance X₂.

[0033] The following equations for equivalent reactance X_e and equivalent resistance R_e may be developed using FIG. 5 and data obtained at the two load points of the motor. The equation for the equivalent reactance X_e is given as follows: $\begin{array}{cc}{X}_e=\frac{-B}{2A}+\frac{\sqrt{B^2-4AC}}{2A},& \left(6\right)\end{array}$
where A, B, and C are given by:
A=1.05*0.05*sI₁²;   (7)
B=−1.1I₁V_1is; and   (8)
C=V_1i²s+(sR₁I₁−sV_1R)(I₁R₁−V_1R)   (9)
V_1i is the imaginary portion of the voltage and is a function of the amplitude of the power source voltage V₁ and the sine of the power factor angle. V_1R is the real portion of the voltage and is a function of the amplitude of the power source voltage V₁ and the cosine of the phase angle. In addition, the equivalent resistance R_e is given by the following equation: $\begin{array}{cc}{R}_e=\frac{{\mathrm{sX}}_e\left(V_{1j}-\mathrm{.05}{I}_1{X}_e\right)}{\left(V_{1R}-I_1{R}_1\right)}.& \left(10\right)\end{array}$

[0034] As discussed above, it was assumed that the stator leakage reactance is 5%, or 0.05 of the magnetizing reactance X_m. With no load on the motor, the rotor section of the circuit is considered open and the value for the slip s is considered to be zero. The total reactance of the circuit is made of the sum of the stator leakage reactance X₁ and the magnetizing reactance X_m. Since X₁ can be expressed as equal to 0.05 X_m, then the total no-load reactance can be written as 1.05 X_m. The value of X_e at the two load points is used to extrapolate the value at no-load to yield X_m. The value of X_e at zero-load is the magnetizing reactance X_m. In addition, the slip s is used as a measure of the load. Through experimentation using different load points and different motors, it has been found that the following equation yields a very close value for the magnetizing reactance X_mi to be used for estimating the stator leakage reactance X₁: $\begin{array}{cc}{X}_{mi}=X_{e1}+\frac{\left(X_{e2}-X_{e1}\right)s_1^{\frac{1}{4}}}{{\left(s_1-s_2\right)}^{\frac{1}{4}}}.& \left(11\right)\end{array}$

[0035] In equation (11) above, s₁ is the slip at a high load and s₂ is the slip at a low load, noting that s₁ is greater than s₂. The value of X_mi may then be used to establish the value of X₁, in accordance with equation (1) provided above.

[0036] Once the value of X₁ is obtained, new values for R_t and X_e may be obtained. These new values of R_t and X_e are based on a fixed known value of the stator reactance X₁, and may be determined in accordance with the following equations: $\begin{array}{cc}{\alpha }_1=\frac{1}{R_{t1}}-\frac{1}{R_{t2}};\mathrm{and}& \left(12\right)\\ {\alpha }_2=\frac{1}{X_{e1}}-\frac{1}{X_{e2}}.& \left(13\right)\end{array}$

[0037] There now are four equations and four unknowns. The unknowns are R₂, X₂, R_c, and X_m. To eliminate R_c, equation (3) is subtracted from equation (2) to yield the following equation: $\begin{array}{cc}{\alpha }_1=\frac{\frac{R_2}{s_1}\left(\frac{R_2^2}{s_2^2}+X_2^2\right)-\frac{R_2}{s_2}\left(\frac{R_2^2}{s_1^2}+X_2^2\right)}{\left(\frac{R_2^2}{s_1^2}+X_2^2\right)\left(\frac{R_2^2}{s_2^2}+X_2^2\right)}.& \left(14\right)\end{array}$
To eliminate X_m, equation (5) is subtracted from equation (4) yielding the following equation: $\begin{array}{cc}{\alpha }_2=\frac{X_2\left(\frac{R_2^2}{s_2^2}+X_2^2\right)-X_2\left(\frac{R_2^2}{s_1^2}+X_2^2\right)}{\left(\frac{R_2^2}{s_1^2}+X_2^2\right)\left(\frac{R_2^2}{s_2^2}+X_2^2\right)}.& \left(15\right)\end{array}$

[0038] From the equations provided above, equations may now be established for R₂, X₂, R_c, and X_m. By dividing equation (14) by equation (1 5), the following relationship for the X₂ and R₂ can be established:
X₂=γR₂.   (16)
where γ is given by the following equation: $\begin{array}{cc}\gamma =\frac{-{\alpha }_1\left(s_1+s_2\right)}{2{\alpha }_2{s}_1{s}_2}+\frac{\sqrt{{\left(\frac{{\alpha }_1}{{\alpha }_2}\right)}^2{\left(s_1+s_2\right)}^2+4s_1{s}_2}}{2s_1{s}_2}.& \left(17\right)\end{array}$
The rotor resistance R₂ may be established by substituting γR₂ for X₂ in equation (15) and using algebraic manipulation to produce the following equation: $\begin{array}{cc}{R}_2=\frac{\frac{\gamma }{{\alpha }_2}}{\left(\frac{1}{s_1^2+{\gamma }^2}\right)}-\frac{\frac{\gamma }{{\alpha }_2}}{\left(\frac{1}{s_2^2+{\gamma }^2}\right)}.& \left(18\right)\end{array}$
In addition, the core loss resistance R_c may be established in terms of R₂ and X₂ by manipulating equation (2) to produce the following equation: $\begin{array}{cc}{R}_c=\frac{1}{\left(\frac{1}{R_{t1}}-\frac{\frac{R_2}{s_1}}{\left(\frac{R_2^2}{s_1^2}+X_2^2\right)}\right)}& \left(19\right)\end{array}$
Finally, the magnetizing reactance X_m may be established in terms of R₂ and X₂ by manipulating equation (4) to produce the following equation: $\begin{array}{cc}{X}_m=\frac{1}{\left(\frac{1}{X_{e1}}-\frac{X_2}{\left(\frac{R_2^2}{s_1^2}+X_2^2\right)}\right)}.& \left(20\right)\end{array}$

[0039] The data processing module 82 is programmed to use the above-described equations and methodology to establish estimated values of rotor resistance R₂, leakage reactance X₂, core loss resistance R_c, and magnetizing reactance X_m. Voltage and current input data are obtained at the two load points and provided to the processor module 84. Input power data also may be obtained at the same two points or calculated from the voltage, current, and/or resistance data. In addition, motor speed data also is provided to the data processing module 82. The motor speed data may be the RPM of the motor or the slip. Ideally, the measurements at the two load points are made simultaneously to avoid potential change due to a change in the operating condition of the motor. In addition, in the illustrated embodiment the line-to-line electrical resistance of the stator is provided to the processor. The phase resistance is established by averaging the line-to-line resistance and dividing by 2. The data processing module 82 is operable to establish the value of the equivalent reactances X_e1 and X_e2 using equations (6) through (10) provided above at each load point. The processor also is operable to establish the initial magnetizing reactance X_mi using equation (11) provided above. In addition, the processor is operable to establish the value of the phase leakage reactance X₁ from the magnetizing reactance X_mi. Using the value of X₁, the processor is operable to find new values for the equivalent resistances R_t1, R_t2, X_e1, and X_e2, where: $\begin{array}{cc}{R}_{t1}=\frac{R_{e1}}{s_1};\mathrm{and}& \left(21\right)\\ R_{t2}=\frac{R_{e2}}{s_2}.& \left(22\right)\end{array}$

[0040] The system may also be operated to estimate motor operating parameters based on the values of X₁, R₂, X₂, R_c, and X_m, as represented by block 118. For example, the system may be adapted to establish the values of the rotor torque T, the rotor temperature, and the motor efficiency based on the values of R₂, X₂, R_c, and X_m, electrical input data and rotor speed data. The rotor current I₂ may be established using the following equation: $\begin{array}{cc}{I}_2=\left(I_1-\frac{\left(V_{1R}-I_1{R}_1\right)}{R_c}-\frac{\left(V_{1i}-I_1{X}_1\right)}{X_m}\right)+j\left(\frac{\left(V_{1R}-I_1{R}_1\right)}{X_m}-\frac{\left(V_{1i}-I_1{X}_1\right)}{R_c}\right).& \left(23\right)\end{array}$
The shaft torque may be obtained from the rotor resistance R₂ and the rotor current I₂, as follows: $\begin{array}{cc}T\left(N\text{-}m\right)=\frac{3I_{2\mathrm{rms}}^2{R}_2}{{\omega }_{s}s}.& \left(24\right)\end{array}$
In the above equation, I_2rms is the RMS value of the rotor current I₂, and ω_s is the mechanical synchronous speed in rad/second given by: $\begin{array}{cc}{\omega }_s=\frac{4\pi f}{p}.& \left(25\right)\end{array}$
In this equation, f is the alternating current frequency in Hz and p is the number of poles of the motor.

[0041] The shaft torque may be converted to foot-pounds by multiplying the torque in Newton-meters by 0.738. In addition, the shaft torque is modified by subtracting the friction and windage loss R_F&W and the stray load loss using published values and IEEE standards, as shown in the following table:

	Motor Power	SLL % of output power
	1-125 HP	1.8
	126-500 HP	1.5
	501-2499 HP	1.2

[0042] The motor efficiency is established by dividing the estimated output mechanical power by the input electrical power. $\begin{array}{cc}\eta =\frac{P_{\mathrm{out}}}{P_{\mathrm{in}}}.& \left(26\right)\end{array}$
The estimated output mechanical power P_out may be established from the torque T and the rotor speed data.

[0043] The above-described technique was used to estimate the efficiency of a 10 HP motor and a 600 HP motor using data from a motor design program and test data. The following are the results obtained for a 10 HP motor and the discussion of these results.

Motor Data:	HP: 10	Elec. Des.: E9893A A
	RPM: 1175	Frame: 0256T
	Enclosure: TEFC	Design: B
	Volts: 575	LR Code: G
	Amp: 10.1	Rotor: 418138071HE
	Duty: Cont.	Stator: 418126002AJ
	INS/AMB/S.F.: F/40/1.15	FAN: 702675001A
	TYP/PH/HZ: P/3/60

[0044] Using data from the program at full load and at ¼ load, the parameters of the motor were identified using the new method. The following is a summary of the results.

	Estimated Efficiency	Program Efficiency	% Error
Full Load	91.315	91.097	0.239%
¾ Load	92.154	91.850	0.330%
½ Load	92.101	91.661	0.479%
¼ Load	89.005	88.186	0.928%

[0045] From the above results it can be seen that the error in the estimated efficiency is less than 1% of the efficiency obtained from the program results. It can also be observed that the error increases as the load decreases. By examining the calculated losses it was noticed that the calculated core loss is less than the program value by 19 watts. This fixed error becomes a larger percentage of the total loss at low loads and as a result the percentage error in efficiency increases as the load decreases.

[0046] The estimated efficiency was also compared to laboratory test data. The following is a summary of the results for the 10 HP motor.

	Estimated Efficiency	Actual Efficiency	% Error
Full Load	89.98	90.310	−0.36%
¼ Load	86.18	86.530	−0.41%

The estimated core loss in this case was more than the measured value leading to a lower estimated efficiency than the measured efficiency.

[0047] The procedure was repeated for a 600 HP motor. The following are the results obtained for a 600 HP motor and the discussion of these results.

	Motor Data:	HP: 600	Elec. Des.:
		RPM: 1195	Frame: 35C5012Z
		Enclosure: TEFC	Design: 139481
		Volts: 575	LR Code:
		Amp: 532	Rotor: 710623-2-S
		Duty: Cont.	Stator: 710622-2-T
		INS/AMB/S.F.: F/ /1.15

[0048] Comparing the design program data to the estimated values from the above-described process, the following results were obtained:

	Estimated Efficiency	Program Efficiency	% Error
Full Load	95.794	95.791	0.003%
¾ Load	95.843	95.855	−0.013%
½ Load	95.318	95.352	−0.035%
¼ Load	92.655	92.710	−0.059%

The difference between the design program data and estimated value data is less 0.04%. Initially, the resolution selected for use with the design program data for the speed of the motor was one decimal point. The results obtained using one decimal point resolution on speed lead to higher error in estimation. The results provided above were obtained using a higher resolution on speed. In addition, this particular motor has a very low slip. The slip in RPM at full load is less than 5 RPM so that any error in the speed measurement will lead to a large error in estimation. The following are the results obtained using four decimal points resolution, three decimal points resolution, two decimal points and one decimal point resolution to illustrate the effect of resolution on the efficiency estimation.

[0049] Four Decimal Points Resolution:

	Estimated Efficiency	Program Efficiency	% Error
Full Load	95.795	95.791	0.0036%
¾ Load	95.844	95.855	−0.0122%
½ Load	95.320	95.352	−0.0338%
¼ Load	92.658	92.710	−0.0550%

[0050] Three Decimal Points Resolution:

	Estimated Efficiency	Program Efficiency	% Error
Full Load	95.797	95.791	0.0065%
¾ Load	95.848	95.855	−0.008%
½ Load	95.325	95.352	−0.028%
¼ Load	92.669	92.710	−0.044%

[0051] Two Decimal Points Resolution:

	Estimated Efficiency	Program Efficiency	% Error
Full Load	95.887	95.791	−0.0143%
¾ Load	95.969	95.855	−0.0364%
½ Load	95.509	95.352	−.0705%
¼ Load	93.031	92.710	−0.1297%

[0052] One Decimal Point Resolution:

	Estimated Efficiency	Program Efficiency	% Error
Full Load	95.008	95.791	−0.817%
¾ Load	94.776	95.855	−0.840%
½ Load	93.708	95.352	−1.200%
¼ Load	89.494	92.710	−3.486%

[0053] From the above results it can be concluded that to provide a good estimation of efficiency for low slip motors using this method it is preferable to have a resolution on speed to at least two decimal points. The reason for this is that if the resolution is less than two decimal points the error in slip causes an error in the estimation of the core loss, yielding a higher overall error.

[0054] The system was then operated using lab test data for the 600 HP motor. The resolution of the speed that was used was 1 RPM. This resolution is less than the minimum recommended for obtaining good results. The results using this coarse resolution are shown below.

	Estimated Efficiency	Program Efficiency	% Error
Full Load	96.59	96.65	−0.052%
¾ Load	95.87	96.68	−0.840%
½ Load	95.02	96.17	−1.20%
¼ Load	95.95	93.62	2.480%

From these results, it can be concluded that the method yields excellent results for regular slip motors. However, for low slip motors the resolution on the RPM of the motor is preferably at least two decimal points so as to get a good estimate of the motor efficiency in the field. One way of obtaining excellent resolution of the motor speed is by using accelerometers to measure the motor vibration and find its spectrum.

[0055] A comparison between the losses seen in the design program and the estimated losses using the above-described method is provided below.

	Design Program	New Method
	Rotor Loss:
	Full Load	1.79 KW	1.785 KW
	¾ Load	.980 KW	.979 KW
	½ Load	.430 KW	.429 KW
	¼ Load	.107 KW	.107 KW
	Core Loss:
	Full Load	5.77 KW	5.756 KW
	¾ Load	5.81 KW	5.852 KW
	½ Load	5.85 KW	5.924 KW
	¼ Load	5.9 KW	5.975 KW

The results illustrate general agreement between the design program results and the new method of estimating motor parameters describe above.

[0056] Referring generally to FIG. 6, an alternative process for establishing estimated values of various motor electrical parameters using data obtained at a single operating point with no load on the motor is shown and designated generally by reference numeral 120. In addition, the estimated values of the motor electrical parameters may be used to establish estimated values of various motor operating parameters. The process comprises obtaining stator resistance RI data, as represented by block 122. The line-to-line input resistance may be measured, averaged, and divided by 2 to determine the phase resistance R₁. The process also comprises obtaining electrical input data with no load on the motor and providing the data to the processor module 84, as represented by block 124. To achieve the no-load condition, the motor is disconnected from its load. The electrical input data obtained at the first load points comprises: input voltage data, input current data. Some data may be provided to the system 80 using the control module 90 or may be provided from a remote station 98 via the network 96. The current with no-load I_n1 may be measured for each phase and averaged. The three line voltages may be measured, averaged, and divide by √{square root over (3)} to determine the phase voltage V₁.

[0057] The data processing module 82 may then be operated to establish estimated values of various motor parameters, as represented by block 126. The programming instructions are provided to the data processing module 82 are adapted to utilize a novel technique for establishing the values of the various motor parameters using data obtained with no-load on the motor. With no load on the motor, the rotor portion of the circuit will effectively be an open circuit and is assumed to be an open circuit for these purposes. The current I₂ will be sufficiently small to handle the windage and friction load of the rotor. With no load on the motor, the stator current I₁ will be the no-load current I_n1. The stator leakage inductance L₁, the magnetizing inductance L_m and the core loss resistance R_c may be established using the following equations. First, the total resistance R_t may be obtained by the following equation: $\begin{array}{cc}{R}_t=\frac{P}{I_{\mathrm{nl}}^2}.& \left(27\right)\end{array}$

[0058] The total impedance Z may be found by dividing the input voltage V₁ by the no-load current I_n1, as follows: $\begin{array}{cc}Z=\frac{V_1}{I_{\mathrm{nl}}}.& \left(28\right)\end{array}$

[0059] The total reactance X₁+X_m may be found from the total impedance Z and the total resistance R_t, as follows:
X₁+X_m=√{square root over (Z²−R_t²)}.   (29)

[0060] The individual values for the stator reactance X₁ and the magnetizing reactance X_m may be found from the assumed relationship of X₁=0.05 X_m, as follows:
X₁+X_m=1.05X_m.   (30)

[0061] Next, the motor friction and windage power P_F&W may be estimated based on the motor size and construction, if known. If not, the motor friction and windage power P_F&W is combined with the core loss. The equivalent resistance R_W&F due to motor friction and windage power P_F&W may be estimated as follows: $\begin{array}{cc}{R}_{W\&F}=\frac{P_{W\&F}}{I_{\mathrm{nl}}^2}.& \left(31\right)\end{array}$

[0062] The series core loss resistance R_m, may be established as follows:
R_m=R_t−R₁−R_W&F.   (32)

[0063] The parallel magnetizing inductance L_m, may be established as follows: $\begin{array}{cc}{L}_m=\frac{X_m^2+R_m^2}{X_m\omega }.& \left(33\right)\end{array}$

[0064] The parallel core resistance Rc, may be established as follows: $\begin{array}{cc}{R}_c=\frac{X_m^2+R_m^2}{R_m}.& \left(34\right)\end{array}$

[0065] The stator leakage inductance L₁, may be established as follows: $\begin{array}{cc}{L}_1=\frac{X_1}{\omega }.& \left(35\right)\end{array}$

[0066] As with the previous two load point method, the data processing module 82 may be used to estimate other motor parameters based on the estimated motor electrical parameter data obtained above, as represented by block 128. An expression of the rotor current I₂ may be obtained from the voltage across the rotor and the rotor impedance. Designating the voltage across the rotor as V_a and the rotor current as I₂, the following equation can be written: $\begin{array}{cc}{I}_2=\frac{V_a}{\frac{R_2}{S}+\mathrm{j\omega }{L}_2}.& \left(36\right)\end{array}$

[0067] The rotor current can also be expressed using the input current I₁, the current through the magnetizing inductance I_m, and the current through the core resistance I_c, as follows:
I₂=I₁−I_c−I_m.   (37)

[0068] The above currents can be expressed in terms of the voltage and the value of the motor parameters as follows: $\begin{array}{cc}{V}_a=V_1-I_1\left(R_1+j\omega L_1\right);& \left(38\right)\\ I_c=\frac{V_a}{R_c};\mathrm{and}& \left(39\right)\\ I_m=\frac{V_a}{\mathrm{j\omega }{L}_m}.& \left(40\right)\end{array}$

[0069] The following expression for I₂ may be obtained by manipulating the equations above and substituting the expressions for I₁, I_c, and I_m from equations (38)-(40) into equation (37): $\begin{array}{cc}{I}_2=I_1-\frac{\left(V_1-I_1\left(R_1+\mathrm{j\omega }{L}_1\right)\right)}{R_c}-\frac{(V_1-I_1\left(R_1+\mathrm{j\omega }{L}_1\right)}{\mathrm{j\omega }{L}_m}.& \left(41\right)\end{array}$

[0070] Equations (36) and (41) can now be equated to obtain an equation relating the input current, the input voltage, and the motor parameters. Because the resulting equation has a real part and imaginary part, this will yield two equations. The input current can be written as a complex quantity:
I₁=I_1R−jI_1i.   (42)

[0071] Two equations, one representing the real part and one representing the imaginary part, may be obtained using equations (34), (39) and (40). The real part is as follows: $\begin{array}{cc}\left(I_{1R}-\frac{V_1}{R_c}+\frac{I_{1R}{R}_1}{R_c}+\frac{I_{1i}\omega L_1}{R_c}+\frac{I_{1R}{L}_1}{L_m}-\frac{R_1{I}_{1i}}{\omega L_m}\right)\left(\frac{R_2^2}{s^2}+{\omega }^2{L}_2^2\right)=\frac{R_2}{S}\left(V_1-I_{1R}{R}_1-I_{1i}\omega L_1\right)-\omega L_2\left(\omega L_1{I}_{1R}-R_1{I}_{1i}\right).& \left(43\right)\end{array}$

[0072] The imaginary part will be given by: $\begin{array}{cc}\left(-I_{1i}+\frac{{\mathrm{\omega L}}_1{L}_{1R}}{R_c}-\frac{R_1{I}_{1i}}{R_c}+\frac{V_1}{\omega L_m}-\frac{I_{1R}{R}_1}{\omega L_m}-\frac{I_{1i}{L}_1}{L_m}\right)·\left(\frac{R_2^2}{s^2}+{\omega }^2{L}_2^2\right)=-\frac{R_2}{s}\left(\omega L_1{I}_{1R}-R_1{I}_{1i}\right)-\omega L_2\left(V_1-I_{1R}{R}_1-I_{1i}\omega L_1\right).& \left(44\right)\end{array}$

[0073] Equations 43 and 44 can be written as: $\begin{array}{cc}{\alpha }_1\left(\frac{R_2^2}{S^2}+{\omega }^2{L}_2^2\right)={\alpha }_2{R}_2+{\alpha }_3{L}_2;\mathrm{and}& \left(45\right)\\ {\beta }_1\left(\frac{R_2^2}{S^2}+{\omega }^2{L}_2^2\right)={\beta }_2{R}_2+{\beta }_3{L}_2;& \left(46\right)\end{array}$
where the different variables are given by: $\begin{array}{cc}{\alpha }_1=I_{1R}-\frac{V_1}{R_c}+\frac{I_{1R}{R}_1}{R_c}+\frac{I_{1i}\omega L_1}{R_c}+\frac{L_1{I}_{1R}}{L_m}-\frac{R_1{I}_{1i}}{\omega L_m};& \left(47\right)\\ {\alpha }_2=\frac{V_1-I_{1R}{R}_1-I_{1i}\omega L_1}{s};& \left(48\right)\\ {\alpha }_3=-\omega \left(\omega L_1{I}_{1R}-R_1{I}_{1i}\right);& \left(49\right)\\ {\beta }_1=-I_{1i}+\frac{\omega L_1{I}_{1R}}{R_c}-\frac{R_1{I}_{1i}}{R_c}+\frac{V_1}{\omega L_m}-\frac{I_{1R}{R}_1}{\omega L_m}-\frac{I_{1i}{L}_1}{L_m};& \left(50\right)\\ {\beta }_2=\frac{{\alpha }_{3}S}{\omega };\mathrm{and}& \left(51\right)\\ {\beta }_3=-{\alpha }_2\omega S.& \left(52\right)\end{array}$

[0074] Dividing equations (43) and (44) and solving for the rotor inductance in terms of the rotor resistance one gets: $\begin{array}{cc}{L}_2=\gamma R_2;& \left(53\right)\\ \mathrm{where}\text{:}\gamma =\frac{{\alpha }_1{\beta }_2-{\alpha }_2{\beta }_1}{{\alpha }_3{\beta }_1-{\alpha }_1{\beta }_3}.& \left(54\right)\end{array}$

[0075] Solving for the rotor resistance, the following relationship results: $\begin{array}{cc}{R}_2=\frac{\frac{{\alpha }_2}{{\alpha }_1}+\frac{{\alpha }_3\gamma }{{\alpha }_1}}{{\omega }^2{\gamma }^2+1/s^2}.& \left(55\right)\end{array}$

[0076] The following process may be used for calculating motor torque and motor efficiency. First, estimate the slip s from the shaft speed N and the synchronous speed N_s, as follows: $\begin{array}{cc}s=\frac{N_S-N}{N_S}.& \left(56\right)\end{array}$
The synchronous speed Ns may be obtained from the input frequency and the number of poles of the motor. The power factor may then be computed using the input current, input voltage, and input power.

[0077] Next, the real and imaginary components of the current I_1R & I_1i are established using equations (47-54). The rotor resistance may then be established using the following equation: $\begin{array}{cc}{R}_2=\frac{\left(\frac{{\alpha }_2}{{\alpha }_1}+\frac{{\alpha }_3}{{\alpha }_1}\gamma \right)}{\left({\omega }^2{\gamma }^2+\frac{1}{s^2}\right)}.& \left(57\right)\end{array}$

[0078] The rotor current and torque can be calculated using the following equations:
I₂=√{square root over (α₁²+β₁²)}.   (58)

[0079] The torque T may be estimated by: $\begin{array}{cc}T\left(\mathrm{in}\mathrm{Newton}\text{-}\mathrm{meters}\right)=\frac{3*I_{2\mathrm{rms}}^2*R_2}{{\omega }_2*S};& \left(59\right)\end{array}$
where ${\omega }_s=\frac{4\pi f}{p}$
is the synchronous speed and p is the number of poles. To convert the torque to ft-lbs multiply the T in Newton-meters by 0.738.

[0080] For the purpose of calculating motor efficiency the output power needs to be calculated. This can be obtained using the following equation: $\begin{array}{cc}\mathrm{Output}\mathrm{Power}{P}_{\mathrm{out}}=\frac{\mathrm{TN}}{5252}-P_{F\&W}-\mathrm{SLL};& \left(60\right)\end{array}$
where, T is shaft torque in ft-lb and SLL is the stray load power loss, which is typically a known percentage motor power depending on motor size and varies with the square of the torque. The IEEE standard specifies certain percentage of output power as SLL. This percentage changes as the motor power changes. For example, for 1 to 125 HP motors, the SLL is equal to 1.8% of maximum power. For 126 to 500 HP motors, the SLL is equal to 1.5% of maximum power. Finally, for 501 to 2499 HP motors, the SLL is equal to 1.2% of maximum power.

[0081] As mentioned above, if the friction and windage loss is not known, its value can be lumped with the core loss. The effect of lumping the friction and windage loss with core loss is to cause the rotor loss to be lower than the actual loss, thus raising the estimated efficiency, since the effect of lumping the friction and windage loss with the core loss is to reduce the power across the air gap by the friction and windage loss. In this circumstance, the rotor loss is the motor slip times the friction and windage loss. To obtain an estimate of the maximum error using this approximation, a value of slip equal to 0.025 and a maximum percentage of friction and windage loss of motor power equal to 3% may be used. This yields a maximum error in estimating the efficiency equal to 0.075%, which is within the measurement error. Tests conducted on different motors indicate the validity of the assumption. If the value of the friction and windage loss is known, then that value may be used. The motor efficiency may then be estimated using the ratio of the estimated output power to the input power. The above-described method was applied to experimental data and the results indicate an accuracy of over 99%.

[0082] It is important to note that the core loss is obtained at a constant frequency. If the motor used at a different frequency, then the core loss needs to be estimated at the new frequency. In general the core loss is proportional to the square of frequency and to the magnitude of the flux density. If the flux density is constant then a simple equation can be used to estimate the core loss at a different operating frequency.

Test Results:

[0083] The no-load data from three motors were used to test the accuracy of the above method. The following is a summary of the data obtained.

10 HP Motor:
Motor Data:	HP: 10	Elec. Des.: E9893A A
	RPM: 1175	Frame: 0256T
	Enclosure: TEFC
	Volts: 575	Design: B
	Amp: 10.1	LR Code: G
	Duty: Cont.	Rotor: 418138071HE
	INS/AMB/S.F.: F/40/1.15	Stator: 418126002AJ
	TYP/PH/HZ: P/3/60	FAN: 702675001A
	No load Current:	4.41 ampere
	No Load Voltage:	574.9 volts
	No Load Power:	261.73 watts
	Stator Resistance:	0.8765 ohm
	F&W power:	57 watts
	Stray Load Loss:	1.13% obtained from
		experimental data

The results obtained are as follows:

[0084] Actual Motor Efficiency at full load=90.2434%

[0085] Estimated Motor Efficiency=90.8452%

[0086] Estimation Error=0.6357%

150 HP Motor:
Motor Data:	HP: 150	Elec. Des.: W00868-A-A001
	RPM: 1180	Frame: EC360
	Enclosure: TENV
	Volts: 460
	Amp: 10.1
	Duty: 15 Min
	INS/AMB/S.F.: F//1.15
	No Load Current:	66.09 ampere
	No Load Voltage:	460 volts
	No Load Power:	2261 watts
	Stator resistance:	0.03509 ohm
	F & W power:	896 watts
	Stray Load Loss:	0.85% from test data

The results obtained are as follows:

[0087] Actual Motor Efficiency at full load=93.106%

[0088] Estimated Motor Efficiency=93.413%

[0089] Estimation Error=0.3303%

600 HP Motor:
Motor Data:	HP: 600	Elec. Des.:
	RPM: 1195	Frame: 35C5012Z
	Enclosure: TEFC
	Volts: 575	Design: 139481
	Amp: 532	LR Code:
	Duty: Cont.	Rotor: 710623-2-S
	INS/AMB/S.F.: F//1.15	Stator: 710622-2-T
	No Load Current	= 148.45 ampere
	No Load Voltage	= 575 volts
	No Load Power	= 6860 watts
	Stator resistance	= .0091 ohm
	F & W power	= 1725 watts
	Stray Load Loss	= 1.3% from Test data

The results obtained are as follows:

[0090] Actual Motor Efficiency at full load=96.025%

[0091] Estimated Motor Efficiency=95.976%

[0092] Estimation Error=−0.0500%

[0093] To make the estimation of the motor efficiency less sensitive to slight errors in measured frequency, the following process may be performed. First, the stator loss is calculated using the input current and the estimated stator resistance R₁. The friction and windage loss is estimated based on the motor size, type, and speed. The rotor loss may be estimated by subtracting the stator loss from the Input power P and multiplying the remainder by the slip. The stray load loss SLL is estimated based on the IEEE standard, as described above, with the exception that the core loss is neglected. The modified input power is then calculated at the two measurement points by subtracting the above losses from the input power P.

[0094] A plot of the modified input power versus measured speed may then be performed to determine the core loss. The core loss is the modified input power at the synchronous speed n_s. This can be determined mathematically using the following equation: $\begin{array}{cc}\mathrm{CoreLoss}=\left(P_1-n_1\left(\frac{P_2-P_1}{n_2-n_1}\right)\right)+\left(\frac{P_2-P_1}{n_2-n_1}\right)n_S;& \left(61\right)\end{array}$
where:

[0095] P₁ Modified Input power at point 1 “low load”

[0096] P₂ Modified Input power at point 2 “high load”

[0097] n₁ Motor speed at point 1

[0098] n₂ Motor speed at point 2

[0099] n_s Synchronous speed using the measured frequency at low load.

[0100] The rotor loss and the stray load loss SLL may then be recalculated using the new core loss value. The magnetizing inductance L_m, rotor resistance R₂, and rotor leakage inductance L₂ are calculated as provided previously. This method was found to be less sensitive to error in frequency measurements.

[0101] The temperature of the rotor during motor operation may be estimated using the estimated value of the rotor resistance R₂ and the following equation relating changes in electrical resistance of the rotor to changes in temperature:
R_{2 hot}=R_{2 cold}(1+α(T_hot−T_cold));   (62)
where: R_{2 cold} is the rotor resistance at a first temperature; R_{2 hot} is the rotor resistance at a second temperature; T_cold is the rotor temperature at a first temperature; T_hot is the rotor temperature at a second temperature; and α is the temperature coefficient of electrical resistance of the rotor in Ω/unit of temperature.

[0102] As an example, the above equation may be manipulated algebraically to obtain the following equation for an aluminum rotor: $\begin{array}{cc}{T}_{\mathrm{hot}}=\frac{R_{2\mathrm{hot}}}{R_{2\mathrm{cold}}}*\left(225+T_{\mathrm{cold}}\right)-225.& \left(63\right)\end{array}$

[0103] The value used for R_{2 hot} is the estimated value for the rotor resistance R₂ at the second temperature T_hot. The control module 90 may be used to input the rotor temperature at the first temperature T_cold and the rotor resistance at the first temperature R_{2 cold}. In addition, the data may be provided by the remote stations 98 via the network 96.

[0104] Referring generally to FIG. 7, a process for establishing values of various motor electrical parameters and various motor operating parameters using the system of FIG. 3 is shown and designated generally by reference numeral 130. The process comprises obtaining the resistance of the stator, as represented by block 132. The process also comprises obtaining data at a single operating load point and providing the data to the processor module 84, as represented by block 134. In a presently contemplated embodiment, the data obtained at the first load point comprises: input voltage data, input current data, input power data, shaft speed data, and stator temperature data. It should be noted that the input power can either be measured or calculated from the other input data. Some data may be provided to the system 80 using the control module 90 or may be provided from a remote station 98 via the network 96.

[0105] As represented by block 136, the data processing module 82 then operates to establish estimated values of various motor parameters. As discussed above, these estimated motor parameters may comprise one or more of the circuit parameters in the motor equivalent circuits 50 and 110 illustrated in FIGS. 2 and 5. Accordingly, the various motor parameters may comprise the stator resistance R₁, the slip s, the stator leakage reactance X₁, the rotor resistance R₂, the rotor leakage reactance X₂, the core loss resistance R_c, and the magnetizing reactance X_m. The stator resistance R₁ and the motor slip s can be measured relatively easily, while the remaining parameters (i.e., X₁, R₂, X₂, R_c, and X_m) are estimated by the processor module 84 in accordance with unique aspects of the process 130 illustrated in FIG. 7.

[0106] As represented by block 138, the data processing module 82 then operates to establish estimated values of other unknown motor parameters based on the one or more parameters estimated in block 136. For example, the data processing module 82 may estimate output power, efficiency, torque, and other characteristics of the motor 20. Accordingly, in certain embodiments, the data processing module 82 operates in accordance with the process 130 to obtain various losses associated with the motor 20. For example, the losses may comprise stator loss, rotor loss, core loss, friction and windage, and stray load loss. The stator loss can be estimated accurately by measuring the stator resistance R₁ and the stator current I₁. The friction and windage loss can be estimated using simulated data on different motor sizes. For example, the data processing module 82 can access a database of motors to obtain the appropriate friction and windage loss. An exemplary motor database may list the motor frame size, number of poles, fan diameter, and the loss associated with the motor. The stray load loss can be estimated using the IEEE standard. Finally, data processing module 82 estimates the rotor loss and the core loss, as described in further detail below.

[0107] The rotor loss can be estimated approximately by multiplying the input power minus the stator loss by the slip, as follows:
RotorLoss=(P_in−3I₁²R₁)s   (64)
P_in is the input power in watts, I₁ is the input current, R₁ is the stator phase resistance, and s is the slip of the rotor. As discussed above with reference to FIG. 7, these parameters are obtained in blocks 132 and 134 of the process 130. Accordingly, the data processing module 82 readily estimates the rotor loss according to equation (64). The error in estimating the rotor loss using this method is equal to the slip s multiplied by the core loss. In view of the equivalent circuit 50 of FIG. 2, the core loss can be expressed as follows: $\begin{array}{cc}\mathrm{CoreLoss}=\frac{3V_a^2}{R_c}& \left(65\right)\end{array}$
V_a is the voltage across the rotor and R_c is the core loss resistance. Accordingly, the error associated with the rotor loss calculated above in equation (64) can be expressed as follows: $\begin{array}{cc}\mathrm{Rotor}\mathrm{Loss}\mathrm{Error}=\frac{3V_a^2}{R_c}s& \left(66\right)\end{array}$
The foregoing calculation provides an accurate estimation of rotor loss for motors having low to moderate core loss (e.g., less than 50% of the total losses). For example, if the core loss (65) is roughly 20% of the losses, then a motor having 85% efficiency will have a core loss of approximately the 3% of the input power. If the motor has four poles and a 40-rpm slip at full load, then the slip s will be approximately 0.0227. Applying these values to equation (66), the percentage error in rotor loss is equal to 0.068% of input power. Accordingly, the rotor loss error (66) has a negligible effect on the calculation of rotor loss (64) and motor efficiency, as discussed in further detail below.

[0108] The only loss left to be estimated is the core loss. For this estimated motor parameter, the data processing module 82 operates to calculate the various parameters of the equivalent circuits 50 and 110, as illustrated in FIGS. 2 and 5. In the illustrated embodiment of FIG. 7, the data processing module 82 operates to obtain or estimate the various parameters: R₁, s, X₁, R₂, X₂, R_c, and X_m. The calculation of the stator resistance R₁ and the motor slip s can be obtained relatively easily. However, the data processing module 82 estimates the remaining parameters (i.e., X₁, R₂, X₂, R_c, and X_m) using unique aspects of the process 130, as set forth below. Once all circuit parameters are obtained, the data processing module 82 estimates the core loss. In turn, the data processing module 82 can estimate other operating parameters of the motor, such as motor efficiency, torque, and so forth.

[0109] As discussed above, the process 130 comprises several assumptions and approximations to simplify the process of estimating X₁, R₂, X₂, R_c, and X_m. For example, the frequency of the power is assumed to be constant, the speed of the rotor is assumed to be constant during the gathering of the single load point data, and the rotor temperature is assumed to be constant during the gathering of the data. Additionally, it has been shown that the stator leakage reactance X₁ can be expressed as a fraction of the magnetizing reactance X_m using the following equation:
X₁=0.0325X_m   (67)
As discussed above, this factor may range from 0.02 to 0.07.

[0110] According to the IEEE standard, the rotor leakage reactance X₂ can be expressed as a function of the stator leakage reactance X₁ as follows:
For design A motors: X₂=X₁   (68)
For design B motors: X₂=1.492X₁   (69)
For design C motors: X₂=2.325X₁   (70)
For design D motors: X₂=X₁   (71)
Accordingly, the calculation of rotor leakage reactance X₂ provided by equations (68) through (71) depends on the calculation of stator leakage reactance X₁ provided by equation (67), which in turn depends on the calculation of magnetizing reactance X_m. The magnetizing reactance X_m is estimated by the data processing module 82, as set forth below.

[0111] In view of the simplified motor equivalent circuit 110 illustrated in FIG. 5, the data processing module 82 estimates an approximate value of equivalent resistance X_e using the following equations relating the measured input current, voltage and power: $\begin{array}{cc}\mathrm{Real}\mathrm{Part}\mathrm{of}\mathrm{Input}\mathrm{Impedance}{Z}_{\mathrm{inR}}=\frac{V_{\mathrm{inR}}}{I_1}& \left(72\right)\\ \mathrm{Imaginary}\mathrm{Part}\mathrm{of}\mathrm{input}\mathrm{Impedance}{Z}_{\mathrm{inI}}=\frac{V_{\mathrm{inI}}}{I_1}& \left(73\right)\\ X_e=\left(Z_{\mathrm{inI}}-X_1\right)+\frac{{\left(Z_{\mathrm{inR}}-R_1\right)}^2}{\left(Z_{\mathrm{inI}}-X_1\right)}& \left(74\right)\end{array}$
V_inR is the real portion of the input voltage, V_inI is the imaginary portion of the input voltage, I₁ is the electric current through the stator, R₁ is the stator resistance, and X₁ is the stator leakage reactance. The data processing module 82 also defines the parallel resistive element or total resistance R_t, as set forth in the following equation: $\begin{array}{cc}{R}_t=\left(Z_{\mathrm{inR}}-R_1\right)+\frac{{\left(Z_{\mathrm{inI}}-X_1\right)}^2}{\left(Z_{\mathrm{inR}}-R_1\right)}& \left(75\right)\end{array}$
In this exemplary embodiment, the data processing module 82 initially assumes the stator leakage reactance X₁ equal to zero to estimate a first approximation of the equivalent resistance X_e, as set forth below: $\begin{array}{cc}{X}_e\left(\mathrm{approximate}\right)=Z_{\mathrm{inI}}+\frac{{\left(Z_{\mathrm{inR}}-R_1\right)}^2}{Z_{\mathrm{inI}}}& \left(76\right)\end{array}$

[0112] In view of the relationships set forth above in equations (67) through (76), the data processing module 82 estimates an approximate value for the stator leakage reactance X₁ as a fraction of the first approximation of the equivalent resistance X_e, as follows:
X₁(approximate)=0.0325X_e(approximate)   (77)
Again, this factor may range from 0.02 to 0.07. After calculating an approximate value for the stator leakage reactance X₁ as set forth by equation (77), the data processing module 82 can estimate the rotor leakage reactance X₂ using the appropriate one of equations (68) through (71). Accordingly, only the rotor resistance R₂, the core loss resistance R_c, and the magnetizing reactance X_m remain to be estimated by the data processing module 82.

[0113] In the illustrated embodiment of FIG. 7, data processing module 82 estimates the rotor resistance R₂ using the following relationships: $\begin{array}{cc}{R}_2=\frac{3V_a^2{s}^2}{\mathrm{RotorLoss}}& \left(78\right)\end{array}$
Again, V_a is the voltage across the rotor, s is the slip of the rotor, and the rotor loss is estimated according to equation (64). The rotor voltage V_a can be calculated from the real and imaginary parts V_aR and V_aI of the rotor voltage V_a, as set forth in the following equations:
V_aR=V_1R−I₁R₁   (79)
V_aI=V_1l−I₁X₁   (80)
V_a=(V_aR²+V_aI²)^1/2   (81)
After calculating the rotor voltage V_a, the data processing module 82 proceeds to calculate the rotor resistance defined by equation (77). Accordingly, only the core loss resistance R_c and the magnetizing reactance X_m remain to be estimated by the data processing module 82.

[0114] The data processing module 82 can calculate the magnetizing reactance X_m and that core loss resistance R_c from the following relationships: $\begin{array}{cc}\frac{1}{X_e}=\frac{1}{X_m}+\frac{1}{X_2+\frac{R_2^2}{X_2}}& \left(82\right)\\ \frac{1}{R_t}=\frac{1}{R_c}+\frac{1}{\frac{R_2}{S}+\frac{X_2^2}{\frac{R_2}{S}}}& \left(83\right)\end{array}$
Finally, using the core loss resistance R_c calculated from equation (83), the processing module 82 can calculate the core loss defined by equation (65).

[0115] At this point, the data processing module 82 has estimated values for all of the motor parameters (e.g., X₁, R₂, X₂, R_c, and X_m) and all the motor losses (e.g., stator loss, rotor loss, core loss, friction and windage loss, and stray load loss). If desired, after calculating the magnetizing reactance X_m as set forth in equation (82), the data processing module 82 can recalculate the stator leakage reactance X₁ according to equation (67). In turn, the data processing module 82 can recalculate the other motor parameters (e.g., R₂, X₂, R_c, and X_m) and the core loss using the newly estimated value of stator leakage reactance X₁. Accordingly, the data processing module 82 can reiterate the calculations set forth in equations (67) through (83) any number of times to improve the accuracy of the estimated motor parameters.

[0116] After obtaining final estimations of these motor parameters and losses, the data processing module 82 can proceed to estimate motor operating parameters, such as motor efficiency, torque, and so forth (block 138). For example, the system may be adapted to calculate the rotor torque, the rotor temperature, and the motor efficiency based on the values of R₂, X₂, R_c, and X_m, electrical input data, and rotor speed data. As discussed above, the shaft torque may be obtained from the rotor resistance R₂ and the rotor current I₂ as set forth in equation (24). In addition, the motor efficiency can be estimated from the following equation: $\begin{array}{cc}\eta =\frac{P_{\mathrm{out}}}{P_{in}}=\frac{P_{in}-\mathrm{SL}-\mathrm{RL}-\mathrm{CL}-\mathrm{FWL}-\mathrm{SLL}}{P_{in}}& \left(84\right)\end{array}$
SL is the stator loss, RL is the rotor loss estimated above in equation (64), CL is the core loss estimated above in equation (65), FWL is the friction and windage loss, and SLL is the stray load loss.

[0117] The process 130 described above with reference to FIG. 7 provides an exceptional estimation of the motor parameters, motor losses, and the motor efficiency. These estimations are particularly accurate if the core loss moderate to low, e.g., less than 50% of the total motor losses. The reason for this correlation between accuracy and core loss is due to the assumption of zero core loss in the estimation of rotor loss, rotor resistance R₂, and core loss resistance R_C. In the illustrated embodiment of FIG. 7, the magnitude of the core loss resistance R_C relates to the addition of the rotor leakage reactance X₂. Accordingly, the value of the rotor leakage reactance X₂ results in a different core loss resistance R_C.

[0118] As an example of the accuracy of the single load point techniques described above, the process 130 was used to estimate the efficiency of a 60 HP motor. The actual efficiency of the motor is 94.4%, whereas the process 130 estimated the motor efficiency as 94% using a single load point. In addition, a TENV 75 HP motor (design B) was evaluated using both the two load points and single load point techniques described with reference to FIGS. 4 and 7. The motor efficiency estimated according to the two load points technique was 90.6%, while the single load point technique yielded a motor efficiency of 91.6%. The single load point technique of FIG. 7 estimated motor efficiencies based on the high load point and different reactance ratios X₂/X₁, i.e., 1.5, 2, 2.5, and 3, as tabulated below:

Reactance Ratio	Core Resistance Ohm	Efficiency
1.5	316	91.6
2	158	90.7
2.5	90.6	89.26
3	56.1	87.25
Zero Core Loss Max Eff.	Infinity	92.6

At a reactance ratio of 1.5, the motor efficiency estimated by the single point technique of FIG. 7 is approximately 1% higher than the motor efficiency estimated by the two point technique of FIG. 4. Accordingly, it can be concluded that the single point technique of FIG. 7 estimates the motor efficiency at an accuracy of 98%, because the accuracy of the two point technique of FIG. 4 has been established at approximately 99%. As a result, the single point technique of FIG. 7 can provide a range of motor efficiencies by estimating the maximum efficiency based on zero core loss and a lower bound using a leakage reactance ratio of 2.5.

[0119] Referring generally to FIG. 8, a process for establishing values of various motor electrical parameters and various motor operating parameters using the system of FIG. 3 is shown and designated generally by reference numeral 140. The process comprises obtaining data at a first operating load point and providing the data to the processor module 84, as represented by block 142. The process also comprises obtaining data at a second operating load point and providing the data to the processor module 84, as represented by block 144. The process further comprises obtaining data at a third operating load point and providing the data to the processor module 84, as represented by block 146. In a presently contemplated embodiment, the data obtained at the first, second, and third load points comprises: input voltage data, input current data, input power data, shaft speed data, and frequency of the motor 20. It should be noted that the input power can either be measured or calculated from the other input data. In addition, the process may measure motor temperature. However, the data obtained at these three points generally does not include the stator resistance of the motor 20. Some data may be provided to the system 80 using the control module 90 or may be provided from a remote station 98 via the network 96.

[0120] As represented by block 148, the data processing module 82 then operates to establish estimated values of various motor parameters without the need to measure the stator resistance. As discussed above, these estimated motor parameters may comprise one or more of the circuit parameters in the motor equivalent circuits 50 and 110 illustrated in FIGS. 2 and 5. Accordingly, the various motor parameters may comprise the stator resistance R₁, the slip s, the stator leakage reactance X₁, the rotor resistance R₂, the rotor leakage reactance X₂, the core loss resistance R_c, and the magnetizing reactance X_m. These seven parameters determine the motor equivalent circuit 50. Accordingly, the motor equivalent circuit 50 can be fully analyzed by measuring and/or estimating these parameters. The motor slip s can be calculated based on the speed of the motor, which is obtained in blocks 142, 144, and 146. The remaining parameters (i.e., R₁, X₁, R₂, X₂, R_c, and X_m) are estimated by the processor module 84 in accordance with unique aspects of the process 140 illustrated in FIG. 8.

[0121] As represented by block 150, the data processing module 82 then operates to establish estimated values of other unknown motor parameters based on the one or more parameters estimated in block 148. For example, the data processing module 82 may estimate output power, efficiency, torque, and other characteristics of the motor 20. In addition, in certain embodiments, the data processing module 82 operates in accordance with the process 140 to obtain various losses associated with the motor 20. For example, the losses may comprise stator loss, rotor loss, core loss, friction and windage, and stray load loss. Based on these losses, the data processing module 82 can then estimate the motor efficiency.

[0122] To estimate the six unknown motor parameters, the process 140 of FIG. 8 proceeds to solve six equations relating to measurements of the input voltage, current, power, and output speed at the three load points. In view of the motor equivalent circuit 50 of FIG. 2, the input impedance at three load points may be defined by the following equations: $\begin{array}{cc}{Z}_{in1}=Z_s+\frac{Z_c{Z}_{r1}}{Z_c+Z_{r1}}& \left(85\right)\\ Z_{in2}=Z_s+\frac{Z_c{Z}_{r2}}{Z_c+Z_{r2}}& \left(86\right)\\ Z_{in3}=Z_s+\frac{Z_c{Z}_{r3}}{Z_c+Z_{r3}}& \left(87\right)\end{array}$
Z_s is the stator impedance, Z_c is of the core impedance, and Z_r is the rotor impedance. These three input impedance equations can be combined by subtracting equation (86) from equation (85), subtracting equation (87) from equation (85), and dividing the resulting two equations to obtain the following: $\begin{array}{cc}\frac{\left(Z_{in1}-Z_{in2}\right)\left(Z_c+Z_{r2}\right)}{\left(Z_{in1}-Z_{in3}\right)\left(Z_c+Z_{r3}\right)}=\frac{\left(Z_{r1}-Z_{r2}\right)}{\left(Z_{r1}-Z_{r3}\right)}& \left(88\right)\end{array}$
In addition, given that the rotor leakage inductance L₂ and the rotor resistance R₂ are the same at each of the three load points, the right hand side of equation (88) can be simplified to the following equation: $\begin{array}{cc}\lambda =\frac{\left(\frac{1}{S_1}-\frac{1}{S_2}\right)}{\left(\frac{1}{S_1}-\frac{1}{S_3}\right)}& \left(89\right)\end{array}$
S₁, S₂, and S₃ are the motor slips at the three load points. As discussed above, these motor slips S₁, S₂, and S₃ can be calculated from speed measurements obtained in blocks 142, 144, and 146 of the process 140. Denoting the quantity in equation (89) by λ, the foregoing equation (88) can be solved for core impedance Z_c in terms of the rotor impedance Z_r at load points 2 and 3 as set forth in the following equation: $\begin{array}{cc}{Z}_c=\frac{\left(Z_{in1}-Z_{in2}\right)Z_{r2}-\lambda \left(Z_{in1}-Z_{in3}\right)Z_{r3}}{\left[\lambda \left(Z_{in1}-Z_{in3}\right)-\left(Z_{in1}-Z_{in2}\right)\right]}& \left(90\right)\end{array}$

[0123] The core impedance Z_c defined by equation (90) can then be substituted into equation (88) to obtain an equation in the rotor input impedances Z_r1, Z_r2 and Z_r3 at the first, second, and third load points. These rotor input impedances Z_r1, Z_r2 and Z_r3 are functions of the rotor leakage reactance X₂ and the motor resistance R₂. The resulting equation can be decomposed into a real part and an imaginary part yielding two equations in the two rotor unknowns. The rotor impedance Z_r can be expressed as: $\begin{array}{cc}{Z}_r=\frac{R_2}{S}+j{X}_2& \left(91\right)\end{array}$
Using equation (91) at the three load points yields the rotor impedance Z_r at the three different slips S₁, S₂, and S₃. After the foregoing substitution of the core impedance Z_c decomposition into real and imaginary parts, equation (88) can be redefined as set forth below:
The real part is given by: $\begin{array}{cc}\left(\frac{A_{22}}{S_2^2}+\frac{A_{33}}{S_3^2}+\frac{A_{12}}{S_1{S}_2}+\frac{A_{13}}{S_1{S}_3}+\frac{A_{23}}{S_2{S}_3}\right)R_2^2-\left(A_{22}+A_{33}+A_{12}+A_{13}+A_{23}\right)X_2^2-\left[\frac{2B_{22}}{S_2}+\frac{2B_{33}}{S_3}+\left(\frac{1}{S_1}+\frac{1}{S_2}\right)B_{12}+\left(\frac{1}{S_1}+\frac{1}{S_3}\right)B_{13}-\left(\frac{1}{S_2}+\frac{1}{S_3}\right)B_{23}\right]R_2{X}_2=0& \left(92\right)\end{array}$
The imaginary part is given by: $\begin{array}{cc}\left(\frac{B_{22}}{S_2^2}+\frac{B_{33}}{S_3^2}+\frac{B_{12}}{S_1{S}_2}+\frac{B_{13}}{S_1{S}_3}+\frac{B_{23}}{S_2{S}_3}\right)R_2^2-\left(B_{22}+B_{33}+B_{12}+B_{13}+B_{23}\right)X_2^2+\left[\frac{2A_{22}}{S_2}+\frac{2A_{33}}{S_3}+\left(\frac{1}{S_1}+\frac{1}{S_2}\right)A_{12}+\left(\frac{1}{S_1}+\frac{1}{S_3}\right)A_{13}+\left(\frac{1}{S_2}+\frac{1}{S_3}\right)A_{23}\right]R_2{X}_2=0& \left(93\right)\end{array}$
The A's and the B's in equations (92) and (93) are functions of the measured input impedance at the three load points and, also, the rotor resistance R₂.

[0124] As set forth below in detail below, these two equations (92) and (93) can be simplified as:
(M₁+M₂R₂)R₂²+(N₁+N₂R₂)X₂²+(L₁+L₂R₂)R₂X₂=0   (94)
(U₁+U₂R₂)R₂²+(V₁+V₂R₂)X₂²+(W₁+W₂R₂)R₂X₂=0   (95)
Equation (95) essentially obtains the rotor leakage reactance X₂ in terms of the rotor resistance R₂. The A's and B's can be defined by the following equations in which β_R and β_j are the real and imaginary parts of the measured differential input impedance β₁ and β₂ and the subscripts 1, 2, and 3 refer to the first, second, and third load points obtained at blocks 142, 144, and 146 of the process 140 of FIG. 7. $\begin{array}{cc}{A}_{22}=\left({\beta }_{1R}^2-{\beta }_{1j}^2\right)\left(\left(\frac{1}{S_2}-\frac{1}{S_1}\right)R_2+{\mathrm{\lambda \beta }}_{2R}\right)-2{\beta }_{1R}{\beta }_{1j}{\beta }_{2j}\lambda & \left(96\right)\\ {\beta }_1=Z_{in1}-Z_{in2}& \left(97\right)\\ {\beta }_2=Z_{in1}-Z_{in3}& \left(98\right)\\ A_{33}={\lambda }^2\left[\left({\beta }_{2R}^2-{\beta }_{2j}^2\right)\left(\left(\frac{1}{S_2}-\frac{1}{S_1}\right)R_2+{\beta }_{1R}\right)-2{\beta }_{2R}{\beta }_{2j}{\beta }_{1j}\right]& \left(99\right)\\ A_{12}=\left[\begin{array}{c}\left({\beta }_{1R}{\beta }_{2R}-{\beta }_{1j}{\beta }_{2j}\right)\left({\lambda }^2{\beta }_{2R}-{\mathrm{\lambda \beta }}_{1R}\right)-\\ \left({\beta }_{1R}{\beta }_{2j}+{\beta }_{1j}{\beta }_{2R}\right)\left({\lambda }^2{\beta }_{2j}-{\mathrm{\lambda \beta }}_{1j}\right)\end{array}\right]& \left(100\right)\\ A_{13}=-A_{12}& \left(101\right)\\ A_{23}=-\left[\begin{array}{c}\left({\beta }_{1R}{\beta }_{2R}-{\beta }_{1j}{\beta }_{2j}\right)\left({\mathrm{\lambda \beta }}_{1R}+{\lambda }^2{\beta }_{2R}+2R_2\left(\frac{1}{S_1}-\frac{1}{S_2}\right)\right)-\\ \left({\beta }_{1R}{\beta }_{2j}+{\beta }_{1j}{\beta }_{2R}\right)\left({\mathrm{\lambda \beta }}_{1j}+{\lambda }^2{\beta }_{2j}\right)\end{array}\right]& \left(102\right)\\ B_{22}={\mathrm{\lambda \beta }}_{2j}\left({\beta }_{1R}^2-{\beta }_{1j}^2\right)+2{\beta }_{1R}{\beta }_{1j}\left(\left(\frac{1}{S_2}-\frac{1}{S_1}\right)R_2+{\mathrm{\lambda \beta }}_{2R}\right)& \left(103\right)\\ B_{33}={\lambda }^2\left[{\beta }_{1j}\left({\beta }_{2R}^2-{\beta }_{2j}^2\right)+2{\beta }_{2R}{\beta }_{2j}\left(\left(\frac{1}{S_2}-\frac{1}{S_1}\right)R_2+{\beta }_{1R}\right)\right]& \left(104\right)\\ B_{12}=\left[\begin{array}{c}\left({\beta }_{1R}{\beta }_{2j}+{\beta }_{1j}{\beta }_{2R}\right)\left({\lambda }^2{\beta }_{2R}-{\mathrm{\lambda \beta }}_{1R}\right)+\\ \left({\beta }_{1R}{\beta }_{2R}-{\beta }_{1j}{\beta }_{2j}\right)\left({\lambda }^2{\beta }_{2j}-{\mathrm{\lambda \beta }}_{1j}\right)\end{array}\right]& \left(105\right)\\ B_{13}=-B_{12}& \left(106\right)\\ B_{23}=-\left[\begin{array}{c}\left({\beta }_{1R}{\beta }_{2R}-{\beta }_{1j}{\beta }_{2j}\right)\left({\mathrm{\lambda \beta }}_{1j}+{\lambda }^2{\beta }_{2j}\right)+\\ \left({\beta }_{1R}{\beta }_{2j}+{\beta }_{1j}{\beta }_{2R}\right)\left({\mathrm{\lambda \beta }}_{1R}+{\lambda }^2{\beta }_{2R}+2R_2\left(\frac{1}{S_1}-\frac{1}{S_2}\right)\right)\end{array}\right]& \left(107\right)\end{array}$

[0125] In turn, an equation (108) can be achieved by dividing equations (94) and (95) by the square of the rotor leakage reactance X₂² and by defining α=R₂/X₂ (i.e., the ratio of rotor resistance R₂ to rotor leakage reactance X₂), as set forth below: $\begin{array}{cc}\alpha =\frac{\left(W_1+W_2{R}_2\right)\left(N_1+N_2{R}_2\right)-\left(L_1+L_2{R}_2\right)\left(V_1+V_2{R}_2\right)}{\left(M_1+M_2{R}_2\right)\left(V_1+V_2{R}_2\right)-\left(N_1+N_2{R}_2\right)\left(U_1+U_2{R}_2\right)}& \left(108\right)\end{array}$

[0126] In view of the equations set forth above, the rotor leakage reactance X₂ can be obtained in terms of the rotor resistance R₂ based on equation (95). The rotor leakage reactance X₂ from equation (108) can then be substituted into the equation (94). This substitution yields a cubic equation in the rotor resistance R₂. Using a spreadsheet, it was found that the foregoing equation reduces to a quadratic equation in the rotor resistance R₂. Accordingly, after solving for the rotor resistance R₂, the rotor leakage reactance X₂ can be obtained using equation (108). In turn, the core impedance Z_c can be obtained using equation (90). Moreover, the stator impedance can then be obtained using equation (85). If desired, the data processing module 82 can calculate other parameters based on the foregoing calculations. For example, the data processing module 82 uses these estimated parameters to estimate the efficiency of the motor 20.

[0127] The process 140 described above with reference to FIG. 8 was evaluated with data obtained at three load points on a real motor. The process 140 provided exceptionally accurate estimations of motor efficiency. Numerical analysis and the foregoing tests indicated an estimation error of approximately 1.5% as it pertains to the estimation of motor efficiency. A portion of this error can be attributed to inaccuracies encountered in the field due to instrumentation.

[0128] Referring generally to FIG. 9, a process for establishing values of various motor electrical parameters and various motor operating parameters, such as motor torque and speed, is shown and designated generally by reference numeral 160. The process 160 comprises obtaining baseline motor parameters and providing the data to the processor module 84, as represented by block 162. For example, the data processing module 82 may obtain the various parameters for the motor equivalent circuits 50 and 110 of FIGS. 2 and 5 at a particular baseline condition. In certain embodiments, as described below, the baseline condition may comprise a motor frequency (e.g., 60 Hz) for an inverter-driven motor. If desired, the process 160 may employ any one of the processes 100, 120, 130, or 140 described above with reference to FIGS. 4, 6, 7, and 8. The process 160 also comprises obtaining motor data at a desired operating load point or condition (e.g., a new motor frequency other than baseline) and providing the data to the processor module 84, as represented by block 164. In a presently contemplated embodiment, the data obtained at the desired operating load comprises: input voltage data, input current data, input power data, shaft temperature data, and frequency data of the motor 20. It should be noted that the input power can either be measured or calculated from the other input data. In addition, the data obtained at these three points generally does not include the speed and/or torque of the motor 20. Some data may be provided to the system 80 using the control module 90 or may be provided from a remote station 98 via the network 96.

[0129] As represented by block 166, the data processing module 82 then operates to establish estimated values of various motor parameters at the desired operating load based on the baseline motor parameters and the data obtained at the desired operation load. Again, this estimation step 166 may be performed without measurements of the speed and/or torque of the motor 20. As discussed above, these estimated motor parameters may comprise one or more of the circuit parameters in the motor equivalent circuits 50 and 110 illustrated in FIGS. 2 and 5. Accordingly, the various motor parameters may comprise the stator resistance R₁, the slip s, the stator leakage reactance X₁, the rotor resistance R₂, the rotor leakage reactance X₂, the core loss resistance R_c, and the magnetizing reactance X_m. These seven parameters determine the motor equivalent circuit 50. Accordingly, the motor equivalent circuit 50 can be fully analyzed by measuring and/or estimating these parameters. As discussed below, the data processing model 82 estimates these parameters in accordance with unique aspects of the process 160 illustrated in FIG. 9.

[0130] As represented by block 168, the data processing module 82 then operates to establish estimated values of other unknown motor parameters based on the one or more parameters estimated in block 166. For example, the data processing module 82 may estimate output power, speed, efficiency, torque, and other characteristics of the motor 20. In addition, in certain embodiments, the data processing module 82 operates in accordance with the process 160 to obtain various losses associated with the motor 20. For example, the losses may comprise stator loss, rotor loss, core loss, friction and windage, and stray load loss.