Technical Field
This invention relates to a method for protecting turbocompressors from adverse surges and stalls, specifically by utilizing sets of coordinates which are invariant to inlet conditions. And it is concerned with measuring distance from a turbocompressor's operating point to the Surge Limit Interface.
Background Art
Unstable or oscillatory flow conditions within a turbocompressor, known as surge and stall, are detrimental to process machinery and to the overall process operation. The proximity of the compressor to these unfavorable conditions is detected by process monitoring apparatuses that interact with control algorithms which regulate compressor flow rates within a stable operating region, thus avoiding surge and stall.
Surge control is initiated by analog input signals emanating from various sources located throughout the compressor-process system. Although these signals are many, the set used must consist of relevant data to initiate control-algorithm response (by recirculating or blowing off some of the process gas) to any disturbance before the process flow rate reaches a surge condition.
Prior art surge control can be divided into two categories: surge parameters which are invariant to inlet conditions, and those parameters which are not. Invariant parameters in the prior art consist of different combinations of reduced flow and pressure ratio; or combinations of volumetric flow divided by rotational speed, and polytropic head divided by rotational speed squared. The calculation of these parameters requires knowledge of at least the pressures at the suction and discharge of the turbocompressor, and a flow measurement (.DELTA.p.sub.o). One advantage of the present invention is that it is not limited to this combination of transmitter signals. Control strategies can be implemented using, for instance, a power measurement, suction pressure, and discharge pressure. Furthermore, the concept of this invention can be applied to the detection of fault and fallback strategies, which will keep the turbocompressors running under adverse circumstances.
Of the second category of parameters (those not invariant), some are based on the same pressure and flow measurements as the first category, while others utilize a power or rotational speed measurement as a replacement for flow or discharge pressure measurement. Thus, a control scheme can be applied even if the turbocompressor lacks a flow or discharge pressure measurement. The advantage of the present invention over this prior art is that it does not require that corrections be made for changing inlet conditions.
Thus, there is a need for a method of surge control that provides the flexibility of having a multiplicity of control strategies, together with fault checking and fallback. There is also a need for a surge control system, invariant to inlet conditions, that can accommodate compressor-process systems which are not fully instrumented or have faulty transmitters. A typical turbocompressor performance map (FIG. 5) will depict a surge region (zone) and a stable operating region that are separated by a sharp interface referred to as the Surge Limit Line. Also shown on this map is a Surge Control Line, and the distance between this line and the Surge Limit Line is a safety margin. If the operating point crosses the Surge Control Line, into the safety margin, the antisurge controller calculates a finite error; this error is used in the PI loop. The output of the loop is used to activate an electromechanical sequence in which gas is recycled or blown off to reestablish and maintain a safe flow rate. Should this safety margin be excessive, the frequency and duration of flow recycling will increase, resulting in a reduction of energy efficiency of the compression process. Conversely, should the margin be too brief, the prospect of inadequate protection is amplified.
It is, therefore, obvious that considerable economic advantages can be derived from a narrow margin of safety that incorporates enhanced surge protection with a resultant lessening of process upset. Additional spin-off benefits would: better ensure efficient operation; extend the intervals between scheduled shutdowns; and increase annual monetary savings.
For the foregoing reasons, there is a need to easily and accurately calculate (using invariant coordinate systems) at what point instability occurs under all inlet conditions.
Disclosure of the Invention
The present invention is directed to a method that satisfies the need to protect turbocompressors from detrimental surges and stalls by the use of various combinations of coordinate systems which are invariant to inlet conditions.
To initiate surge detection and control, it is necessary to easily and accurately calculate a compressor's operating point and its distance from the interface (Surge Limit Interface) between the surge and stable regions using information from the transmitters at hand. Moreover, it is important to be able to calculate this relationship for all inlet conditions--pressure, temperature, molecular weight, compressibility, and specific heat ratio. To protect compressors under varying inlet conditions, one must either construct the Surge Limit Interface in a space which is invariant to these inlet conditions, or be able to correct for them.
Using dimensional analysis, three-dimensional (without inlet guide vanes) and four-dimensional (with inlet guide vanes) coordinate systems are constructed which are invariant to inlet conditions, under the assumption that the Reynolds number is of negligible effect.
The steady state operating point resides on a manifold which is one dimension less than the complete space in which it resides. Thus, for the purpose of control, the problem is reduced to two dimensions when inlet guide vanes are not used, and three dimensions when they are. These coordinate systems (fundamental coordinates), as shown below, yield several possibilities for control; however, linear or nonlinear combinations of the fundamental coordinates are also invariant and can be utilized.
Tables 1 and 2 of FIG. 6 contain three new parameters not found in the prior art: T.sub.r (reduced torque), P.sub.r (reduced power), and N.sub.e.sup.2 (equivalent speed), each is divided by k.sub.s. Not only are T.sub.r and P.sub.r paired with N.sub.e.sup.2, but all three are combined with one or two of the remaining coordinates (h.sub.r /k.sub.s, R.sub.c, q.sub.s.sup.2 /k.sub.s, .alpha.) to formulate a two-dimensional system for turbocompressors without guide vanes, or a three-dimensional system for units with guide vanes.
Since the ratio of specific heats (k.sub.s) is presently unmeasurable, it may be assumed to be constant in many instances, without loss of significant accuracy. It may also be calculated from known values when more accuracy is desired.
From these tables of FIG. 6, the coordinate system that provides the most accurate control can be easily chosen. Furthermore, accurate control can be accomplished when the installation lacks certain transmitters such as flow measurement, temperatures, or downstream pressure. Besides providing flexibility for the primary control strategy of a given installation, the above mentioned alternatives provide avenues for fallback strategies in the event of transmitter failure and for fault tolerance.
Additional advantages revealed are that no downstream information is required; accurate control or measurement is certain under varying inlet conditions; and any of the methods can be checked against other methods to improve control integrity.
The basic invariant coordinate systems are based on polytropic head, torque, and power as functions of flow, rotational speed, and inlet guide-vane position. Another coordinate system is presented using pressure ratio instead of polytropic head.
Since power and torque are independent of head and pressure ratio, combinations of power and head, power and pressure ratio, torque and head, or torque and pressure ratio, can be used for control.
Brief Description of the Drawings
These and other features, aspects, and advantages of the present invention will become better understood with regard to the following description, appended claims, and accompanying drawings where:
FIG. 1 shows a turbocompressor and its surge protection system (with measuring devices);
FIG. 2 shows a schematic diagram of a computing-module setup for turbocompressors without inlet guide vanes;
FIG. 3 shows a schematic diagram of a computing-module setup for turbocompressors with inlet guide vanes;
FIG. 4A shows a surge limit line for a turbocompressor without inlet guide vanes in (P.sub.r, R.sub.c) coordinates;
FIG. 4B shows a surge limit line for a turbocompressor with inlet guide vanes in (P.sub.r, R.sub.c, .alpha.) coordinates;
FIG. 5 shows a turbocompressor performance map depicting the different operating regimes; and
FIG. 6 shows two tables of fundamental coordinates: Table 1 shows viable combinations for turbocompressors without inlet guide vanes, and Table 2 for units with inlet guide vanes.
Best Modes for Carrying Out the Invention
To protect a turbocompressor from unstable or oscillatory flow conditions (surge or stall) it must be known at what point this instability occurs. There is an interface between a turbocompressor's stable operating region and the region in which it encounters surge or stall; and it is necessary to accurately calculate the operating point and its distance from this interface (Surge Limit Interface).
The operating conditions that are used to calculate the distance from surge or stall are detected by process monitoring (measuring) devices located throughout the compressor-process system.
FIG. 1 shows a surge protection system (with measuring devices) depicting a turbocompressor 101 pumping gas from a source 102 to an end user 106. Gas enters the compressor through an inlet line 103, into which is installed an orifice plate 104, and leaves by a discharge line 105. Flow is recycled to the source 102 via an antisurge valve 107.
FIG. 1 also illustrates the antisurge control setup and its connections to the compression process. This arrangement includes a rotational speed transmitter 108, a guide vane position transmitter 109, an inlet pressure transmitter 110, a discharge pressure transmitter 111, an inlet temperature transmitter 112, a discharge temperature transmitter 113, a flow rate transmitter 114, (which measures differential pressure across the flow measuring device 104), an antisurge valve position transducer 115, a torque transmitter 116, a driver 117, and a power transmitter 118.
The monitoring equipment of FIG. 1 interacts with those computing modules shown in FIG. 2 and FIG. 3 which, in turn, display schematic diagram setups for turbocompressors without and with inlet guide vanes, respectively. Both assume constant k.sub.s.
FIG. 2 illustrates an arrangement for turbocompressors without inlet guide vanes in (P.sub.r, R.sub.c) coordinates. The equipment includes a module 119 which calculates pressure ratio, as the ratio of discharge pressure to suction pressure; while a module 120 determines reduced power at the surge limit (as a function of pressure ratio). Another module 121 calculates the ratio of power to rotational speed (rpm), the division of this ratio with suction pressure is computed as reduced power by a module 122. And, finally, the relative slope is determined by a module 123, from the ratio of reduced power (at surge) to reduced power. The relative slope information then interacts with a control system to regulate turbocompressor flow rates.
FIG. 3 shows a computing-module arrangement for turbocompressors with inlet guide vanes in (P.sub.r, R.sub.c, .alpha.) coordinates. The equipment includes a module 119 which calculates pressure ratio as the ratio of discharge pressure to suction; while a module 124 determines reduced power at the surge limit (as a function of pressure ratio and inlet guide vane angle). Another module 121 calculates the ratio of power to rotational speed (rpm), the division of this ratio with suction pressure is computed as reduced power by a module 122. And, lastly, the relative slope is determined by a module 123, from the ratio of reduced power (at surge) to reduced power. This means that module 123 divides the values of reduced power (P.sub.r) into the value of reduced power at surge (P.sub.r,surge), to determine the relative slope (S.sub.rel). P.sub.r,surge and f(R.sub.c) are the same. ##EQU1## which is the ratio of reduced power at surge to reduced power. The relative slope information then interacts with a control system to regulate turbocompressor flow rates.
FIG. 4A depicts a surge limit line plot for a turbocompressor without inlet guide vanes, in the fundamental coordinates (Table 1) shown on FIG. 2. Likewise, FIG. 4B also depicts a surge limit line plot, but for a turbocompressor with inlet guide vanes, in the fundamental coordinates (Table 2) on FIG. 3.
FIG. 5 shows a turbocompressor performance map which depicts characteristic curves along with the surge limit and control lines that define regions (zones) of operation.
The fundamental coordinate systems (see FIG. 6) are invariant to inlet conditions, and are founded on the theory of dimensional analysis or similitude. Except for inlet guide vane position, this invention focuses exclusively on fixed-geometry compressors.
Tables 3 and 4 of FIG. 6 contain sets of fundamental coordinates for control with and without inlet guide vanes. The sets are combinations of the following:
T.sub.r =reduced torque
h.sub.r =reduced polytropic head
q.sub.s =reduced flow rate in suction
P.sub.r =reduced power
N.sub.e =equivalent speed
R.sub.c =pressure ratio
.alpha.=inlet guide vane position
k.sub.s =ratio of specific heats in suction
where; ##EQU2## and;
T=torque
.sigma.=exponent:
n=polytropic exponent
k=ratio of specific heats
P.sub.d =absolute pressure at discharge
P.sub.s =absolute pressure in suction
.DELTA.p.sub.o,s =A flow measurement signal in suction
P=power
P=pressure
N=rotational speed
Z=compressibility
R=gas constant: R.sub.u /MW
R=universal gas constant
MW=molecular weight
T=temperature
s=subscript: suction
r=reduced
c.sub.p =specific heat at constant pressure
c.sub.v =specific heat at constant volume
Although the present invention has been described in detail, and with reference to several possibilities for control, linear or nonlinear combinations of these fundamental coordinates are also invariant and can be used.
An infinite number of coordinate systems can be constructed based on the invariant coordinates presented in the previous sections. Many of these would be viable coordinates for control purposes. These combinations are considered part of the scope of this inventions.
By way of example, consider a compressor without inlet guide vanes. We can construct the compressor map in a coordinate system made up of nonlinear combinations of reduced polytropic head, reduced power, and reduced flow for control. In particular, the map may be constructed in the space: ##EQU3##
This combination may be attractive because it is equivalent to ##EQU4## which is made up of parameters which are completely invariant to initial conditions--including the ratio of specific heats, k.sub.3. The advantage is that, using the form of Equation 1, k.sub.3 need not be known at all.
The flow measurement has been referred to as located in suction. Flow measurement in discharge is also acceptable and may be substituted anywhere suction flow measurement appears.
It is therefore to be understood that, within the scope of the appended claims, the invention may be practiced otherwise than as specifically described.