Dual Mode Control of a Variable Displacement Refrigerant Compressor

This invention relates to a control method for an electronically controlled variable displacement refrigerant compressor that is mechanically driven by a motor vehicle engine, where the displacement is controlled in a normal mode or a high speed mode.

Variable displacement refrigerant compressors have been utilized in automotive air conditioning systems, with the displacement regulated in accordance with cooling demand via either a hydraulic control valve or solenoid control valve. In a typical arrangement, the compressor includes one or more pistons coupled to a variable angle wobble plate or swash plate, and the control valve adjusts a differential pressure acting on a wobble plate control mechanism to vary the wobble plate angle, and hence the compressor displacement. In the control of such a compressor, it has been found that the compressor durability can be improved without significant loss of cooling during high speed operation of the engine by overriding the normal displacement control and de-stroking the compressor to reduce its displacement to a lower value. By way of example, a hydraulic de-stroking implementation is disclosed in the U.S. Pat. No. 5,071,321, and electronic de-stoking implementations are disclosed in the U.S. Pat. Nos. 4,606,705 and 5,145,326.

A drawback with known controls of the type described above is that the transition between normal and high-speed displacement control modes is abrupt and tends to produce disturbances in the compressor and the engine powertrain. These disturbances are undesired for reasons of both durability and driveability. Accordingly, what is desired is a dual mode control that minimizes disturbances to the compressor and engine powertrain when transitioning from one control mode to the other.

The present invention is directed to an improved control method for an electronically controlled variable displacement refrigerant compressor, where the control has normal and high speed displacement control modes, with smooth transitions between such control modes.

In a preferred embodiment, the displacement control mechanism is in the form of a solenoid valve, and the valve is pulse-width-modulated (PWM) at a variable duty cycle to regulate a displacement control pressure within the compressor. When the compressor is operating in a normal speed range, a first control determines a normal PWM duty cycle based on cooling demand; when the compressor is operating in a high speed range, a second control determines an alternate PWM duty cycle based on compressor speed. The solenoid valve is operated in accordance with the duty cycle corresponding to the lower displacement.

According to the invention, a smooth transition from normal control to high speed control is achieved by scheduling the alternate duty cycle as a function of compressor speed. Once in the high speed control mode, the parameters of the first control are adjusted so that the normal duty cycle matches the alternate duty cycle. This ensures a smooth transition from high speed operation to normal operation when the compressor speed falls back into the normal speed range.

FIG. 1 is a schematic diagram of an automotive air conditioning system including an electronically controlled variable displacement compressor and an electronic controller therefor.

FIG. 2 is a block diagram of the control carried out by the electronic controller of FIG. 1 according to this invention.

FIG. 3 is a main flow diagram representative of a computer program executed by the controller of FIG. 1 in carrying out the control of this invention.

FIG. 4 is a flow diagram detailing a portion of the main flow diagram of FIG. 3 in regard to a high speed logic routine for selecting an alternate solenoid duty cycle.

FIG. 5 is a flow diagram detailing a portion of the main flow diagram of FIG. 3 in regard to a PID evaporator outlet air temperature control.

FIG. 6 is a flow diagram detailing a portion of the main flow diagram of FIG. 3 in regard to a condenser outlet pressure control.

FIG. 7 is a flow diagram detailing a portion of the main flow diagram of FIG. 3 in regard to a PID post processing routine executed during the high speed mode of operation.

FIG. 8 is a flow diagram detailing a portion of the flow diagram of FIG. 7 in regard to the adjustment of PID data.

Referring to the drawings, and particularly to FIG. 1, the reference numeral 10 generally designates an automotive air conditioning (AC) system including an electronically controlled multi-cylinder variable displacement refrigerant compressor 12 of the variable angle wobble plate type. The other elements of the system 10 are conventional, and include condenser 13, orifice tube 14, evaporator 16 and accumulator 18 arranged in order between the compressor discharge cavity 20 and suction cavity 22.

A variable speed engine drive shaft (not shown) is coupled to a compressor pulley 48 via drive belt 50, and the pulley 48 is coupled to a compressor drive shaft 46 by an electromagnetic clutch 44. A number of pistons (only one of which is shown in FIG. 1) are mounted in the compressor crankcase 27 so as to be reciprocally driven by the shaft 46 through a tiltable wobble plate mechanism, generally designated by the reference numeral 25. The stroke of the pistons 24, and hence the displacement of the compressor 12, is determined by the operating angle of the wobble plate 26. The wobble plate operating angle is regulated by pulse-width-modulating (PWM) a solenoid actuated control valve 28 to control the pressure in crankcase 27. The control valve 28 includes two valves mechanically coupled to an armature 29: a normally closed ball poppet valve 30 coupling the crankcase 27 to a compressor discharge cavity 20 and a normally open flat poppet valve 31 coupling the crankcase 27 to a compressor suction cavity 22. When the solenoid coil 33 is de-energized, gas pressure in the crankcase 27 bleeds off into suction cavity 22 through poppet valve 31; when coil 33 is energized, high pressure gas enters crankcase 27 from discharge cavity 20 through poppet valve 30. In general, increasing the PWM duty cycle (i.e., the on/off energization ratio of solenoid coil 33) increases the crankcase pressure to decrease the operating angle of the wobble plate 26, and hence the compressor displacement, whereas decreasing the PWM duty cycle decreases the crankcase pressure, thereby increasing the operating angle of wobble plate 26, and hence the compressor displacement.

As described in detail below, the solenoid coil 33 is controllably energized by the electronic controller 54 based on four inputs: a target evaporator outlet air temperature EOAT.sub.tar, the drive speed CS of compressor 12, the actual outlet pressure COP.sub.act of condenser 13, and the actual outlet air temperature EOAT.sub.act of evaporator 18. The actual pressure and temperature values COP.sub.act and EOAT.sub.act may be obtained with suitable conventional transducers 13a, 16a, as indicated. The controller 54 may also be used to control the activation of the compressor clutch 44, as indicated by line 55.

FIG. 2 depicts the electronic controller 54 and air conditioning system 10 in block diagram form. A demand device (not shown) sets the target evaporator outlet air temperature EOAT.sub.tar, and the controller 54 outputs a PWM duty cycle DC.sub.out to the AC system 10. As indicated in FIG. 1, transducers in the AC system 10 provide signals indicative of the compressor speed CS, the actual evaporator outlet air temperature EOAT.sub.act and the actual condenser outlet pressure COP.sub.act.

In the preferred embodiment, the normal control comprises two closed loop PID controllers: Evaporator Outlet Air Temperature (EOAT) Controller 70, and Compressor Outlet Pressure (COP) Controller 72. That is, each Controller 70, 72 develops proportional, integral and differential control terms that contribute to the respective output. The input EOAT.sub.tar is compared to EOAT.sub.act at summing junction 74 to provide an error input EOAT.sub.err to EOAT Controller 70, which produces a target compressor outlet pressure COP.sub.tar. The COP.sub.tar signal, in turn, is compared to COP.sub.act at summing junction 76 to provide an error input COP.sub.err to COP Controller 72, which produces a PWM duty cycle PID.sub.-- DC. Thus, the EOAT Controller 70 develops a condenser outlet pressure COP.sub.tar for eliminating the evaporator outlet temperature error EOAT.sub.err, and the COP Controller 72 develops a solenoid duty cycle PID.sub.-- DC (referred to herein as the normal mode duty cycle) for eliminating the compressor pressure error COP.sub.err. In the preferred digital implementation, both control loops are executed many times per second so that COP.sub.tar may be a highly dynamic quantity based upon changes in vehicle speed, system voltage, ambient temperature, or any other disturbance to the AC system 10.

The compressor speed CS is applied as an input to the High Speed Destroke Logic block 80, which controls the status of a high speed indicator, referred to herein as the HIGH SPD flag. When the HIGH SPD flag is set to indicate a high speed condition, the block 82 determines an alternate PWM duty cycle DC.sub.min for solenoid valve 56, both the HIGH SPD flag and DC.sub.min being applied as inputs to the PID Post Processor 78. The PID Post Processor 78 sets the output duty cycle DC.sub.out equal to the higher of normal mode duty cycle PID.sub.-- DC and alternate duty cycle DC.sub.min when the HIGH SPD flag is set so that under high speed operation, the solenoid valve 56 is operated at a low duty cycle to suitably de-stroke, or reduce the displacement of, compressor 12.

During high speed operation, the PID Post Processor 78 additionally determines alternate control parameters for EOAT Controller 70 and COP Controller 72 so that the normal mode duty cycle PID.sub.-- DC matches the alternate duty cycle DC.sub.min. In the illustrated embodiment, the EOAT and COP Controllers 70, 72 are PID controllers, and the PID Post Processor 78 determines alternate integral terms for causing COP.sub.tar to equal COP.sub.act, and PID.sub.-- DC to equal DC.sub.min. When the compressor speed falls back into the normal speed range, the alternate control parameters are replaced with calculated parameters based on cumulative error, and the integral terms of the EOAT and COP Controllers 70, 72 transition toward values that provide a normal mode duty cycle PID.sub.-- DC that brings EOAT.sub.act into conformance with EOAT.sub.tar, as described above. This results in a smooth transition from the high speed control mode to the normal control mode, improving the quality of control of the AC system 10.

FIG. 3 is a flow diagram representative of a main loop computer program executed by the controller 54 in carrying out the control of this invention. Block 90 designates a series of initialization instructions for setting various parameters and variables to initial values at the onset of each period of operation of the AC system 10. For example, the cumulative error and integral control terms may be reset to zero. After initialization, the blocks 92-100 are repeatedly executed in sequence to read and process input signals such as CS, EOAT.sub.tar, EOAT.sub.act and COP.sub.act, to perform the High-Speed Logic, to update the EOAT and COP control loops, and to perform the PID Post Processing function.

The input signal processing designated by block 92 provides a background updating function for the EOAT and COP control loops, computing and storing parameters such as EOAT.sub.err and COP.sub.err, as well as differential error terms EOAT.sub.del and COP.sub.del. The differential error terms EOAT.sub.del, COP.sub.del represent the difference in error terms EOAT.sub.err, COP.sub.err, respectively, over a predetermined time interval or execution loop. The EOAT and COP control loops subsequently use these terms to update COP.sub.tar and PID.sub.-- DC, respectively.

The High-Speed Logic function of block 94 is described in further detail in the flow diagram of FIG. 4. Referring to FIG. 4, the blocks 110-118 are executed to determine the status of the HIGH SPD flag. The compressor speed CS is compared to an upper reference speed REF.sub.up at block 110, and to a lower reference speed REF.sub.lo at block 116. If CS exceeds REF.sub.up, block 112 is executed to set the HIGH SPD flag, indicating a high speed operating condition. Once the HIGH SPD flag is set, as determined at block 114, it remains set until CS falls below REF.sub.lo. When CS falls below REF.sub.lo, blocks 118-122 are executed to clear the HIGH SPD flag, to reset a HIGH SPD TIMER to zero, and to set the alternate duty cycle DC.sub.min to zero.

Once the HIGH-SPD flag is set, the blocks 124-126 increment the HIGH-SPD TIMER, one count per loop, until the timer count reaches a reference count REF corresponding to a predefined time interval. So long as the timer count is less than REF, the block 128 sets the alternate duty cycle DC.sub.min to a predetermined initial value, referred to as DC.sub.init. Once the timer count reaches REF, the block 130 sets the alternate duty cycle DC.sub.min to a value determined as a function of the compressor speed CS. The CS vs. DC.sub.min relationship may be determined based on field testing; in general, DC.sub.min increases with increasing compressor speed. Alternatively, the HIGH-SPD TIMER may be dispensed with, and DC.sub.min may be determined as a function of the compressor speed CS as soon as the HIGH-SPD flag is set.

The EOAT Control of main flow diagram block 96 is described in further detail in the flow diagram of FIG. 5. Referring to FIG. 5, the blocks 190-192 are first executed to update the proportional and differential control terms EOAT.sub.prop and EOAT.sub.diff. The proportional control term EOAT.sub.prop is determined according to the product of EOAT.sub.err and a proportional gain term G.sub.eoat.sbsb.--.sub.prop, whereas the differential control term EOAT.sub.diff is determined according to the product of EOAT.sub.del and a differential gain term G.sub.eoat.sbsb.--.sub.diff. The blocks 194-198 then determine if the EOAT or COP control loops are already at their respective authority limits. Block 194 identifies a minimum compressor displacement condition wherein PID.sub.-- DC is at 100%, COP.sub.act is greater than COP.sub.tar, and EOAT.sub.act is less than EOAT.sub.tar. Block 196 identifies a maximum compressor displacement condition wherein PID.sub.-- DC is at 0%, COP.sub.act is less than COP.sub.tar, and EOAT.sub.act is greater than EOAT.sub.tar. Finally, block 200 identifies a COP Control loop limit condition where COP.sub.tar is at either a maximum allowable value COP.sub.tarmax or a minimum allowable value COP.sub.tarmin. If none of the limit conditions defined by blocks 194-198 are in effect, the block 200 updates the cumulative error term EOAT.sub.errsum. If any of the limit conditions defined by blocks 194-198 are in effect, block 200 is skipped, freezing the cumulative error term EOAT.sub.errsum at its previous value. The blocks 194-198 thus perform an anti-wind-up function that limits the value of the integral control term EOAT.sub.int in a limit condition where the EOAT or COP control loops are already at the limit of their respective authority. The block 202 then computes the integral control term EOAT.sub.int based on the cumulative error term EOAT.sub.errsum and an integral gain term G.sub.eoat.sbsb.--.sub.int. The block 204 computes the combined error term EOAT.sub.-- PID according to the sum of the proportional, differential and integral control terms, and block 206 computes COP.sub.tar as a function of EOAT.sub.-- PID, and the maximum and minimum allowable target condenser outlet pressure values COP.sub.tarmax and COP.sub.tarmin. In general, block 206 scales EOAT.sub.-- PID such that COP.sub.tar is midway between COP.sub.tarmax and COP.sub.tarmin when EOAT.sub.-- PID is equal to zero, and such that it increases toward COP.sub.tarmax when EOAT.sub.-- PID is positive, and decreases toward COP.sub.tarmin when EOAT.sub.-- PID is negative.

The COP Control of main flow diagram block 98 is described in further detail in the flow diagram of FIG. 6. Referring to FIG. 6, the blocks 210-212 are first executed to update the proportional and differential control terms COP.sub.prop and COP.sub.diff. The proportional control term COP.sub.prop is determined according to the product of COP.sub.err and a proportional gain term G.sub.cop.sbsb.--.sub.prop, whereas the differential control term COP.sub.diff is determined according to the product of COP.sub.del and a differential gain term G.sub.cop.sbsb.--.sub.diff. The block 214 then determines if the COP Control loop is in a limit condition where COP.sub.tar is at either the maximum or minimum allowable values COP.sub.tarmax or COP.sub.tarmin. If not, there is no wind-up condition, and the block 216 updates the cumulative error term COP.sub.errsum. If block 214 is answered in the affirmative, block 216 is skipped, freezing the cumulative error term COP.sub.errsum at its previous value. It will be recognized that this mirrors the function performed by blocks 194-198 of the EOAT control of FIG. 5. The block 218 then computes the integral control term COP.sub.int based on the cumulative error term COP.sub.errsum and an integral gain term G.sub.cop.sbsb.--.sub.int. The block 220 computes the combined error term COP.sub.-- PID according to the sum of the proportional, differential and integral control terms, and block 222 scales COP.sub.-- PID to form the normal duty cycle PID.sub.-- DC.

The PID Post Processing function of main flow diagram block 100 is described in further detail in the flow diagram of FIG. 7. Referring to FIG. 7, the block 140 initially sets the output duty cycle DC.sub.out to the normal duty cycle PID.sub.-- DC. Block 142 then determines the current EOAT State. The Off State signifies normal mode operation, the Low State signifies high-speed mode operation with EOAT.sub.act lower than EOAT.sub.tar, and the High State signifies high-speed mode operation with EOAT.sub.act higher than EOAT.sub.tar.

Initially, of course, the EOAT State is set equal to Off. If the HIGH-SPD flag is set during the Off State, and the alternate duty cycle DC.sub.min is greater than or equal to the normal duty cycle PID.sub.-- DC, as determined at blocks 142-146, the block 148 sets the output duty cycle DC.sub.out equal to the alternate duty cycle DC.sub.min, and the block 150 determines the polarity of the error term EOAT.sub.err. If EOAT.sub.err is negative, block 152 sets the EOAT State to High. If EOAT.sub.err is positive, blocks 154-156 set the EOAT State to Low and adjust the PID data of the EOAT and COP control loops to ensure a smooth transition back to the normal control mode. As explained above, the adjusted PID data serves to modify the EOAT and COP controls so that the normal mode duty cycle PID.sub.-- DC matches the alternate duty cycle DC.sub.min. A flow diagram detailing the adjustment of the PID data is described below in reference to FIG. 8.

If the EOAT State is Low, and the HIGH-SPD flag is cleared, as determined at blocks 142 and 158, the blocks 160-162 are executed to set the EOAT State to Off, and to adjust the PID data of the EOAT and COP control loops. Similarly, if the EOAT State is High and the HIGH-SPD flag is cleared, as determined at blocks 142 and 164, the blocks 166-168 are executed to set the EOAT State to Off, and to adjust the PID data of the EOAT and COP control loops.

If the EOAT State is Low, and the HIGH-SPD flag remains set, block 170 is executed to determine if the error term EOAT.sub.err is more negative than a reference value such as -1.degree. C. If so, the block 172 sets the EOAT State to High, and the block 174 sets the output duty cycle DC.sub.out equal to the alternate duty cycle DC.sub.min. If the error term EOAT.sub.err is less negative than the reference value, and the alternate duty cycle DC.sub.min is greater than or equal to the normal duty cycle PID.sub.-- DC, as determined at blocks 170 and 176, the block 178 sets the output duty cycle DC.sub.out equal to the alternate duty cycle DC.sub.min, and the block 180 adjusts the PID data of the EOAT and COP control loops.

If the EOAT State is High, and the HIGH-SPD flag remains set, block 182 sets the output duty cycle DC.sub.out equal to the alternate duty cycle DC.sub.min, and the block 184 determines if the error term EOAT.sub.err is more positive than a reference value such as +1.degree. C. If so, the block 186 sets the EOAT State to Low, and block 188 adjusts the PID data of the EOAT and COP control loops.

In view of the above, it will be understood that the PID Post Processing routine signals adjustment of the PID data of the EOAT and COP control loops whenever the high speed control mode is in effect and EOAT.sub.act is lower than the target EOAT.sub.tar. When EOAT.sub.act is higher than the target EOAT.sub.tar, DC.sub.min will be greater than the control loop output PID.sub.-- DC, and adjustment of the PID data can be suspended. Finally, it will be seen that the PID Post Processing routine signals adjustment of the PID data whenever the HIGH-SPD flag transitions from set to clear.

Referring to FIG. 8, the adjustment of the PID data as performed by the blocks 156, 162, 180, 168 and 188 of the PID Post Processing flow diagram of FIG. 7 involves (1) setting COP.sub.tar equal to COP.sub.act (block 230), (2) computing and storing a cumulative error term EOAT.sub.errsum that will cause the output of the EOAT control loop, COP.sub.tar, to be equal to COP.sub.act (blocks 232-240), and (3) computing and storing a cumulative error term COP.sub.errsum that will cause the normal mode duty cycle PID.sub.-- DC to be equal to the alternate duty cycle DC.sub.min (blocks 242-248). The adjusted error term EOAT.sub.errsum is determined by computing the proportional and differential control terms EOAT.sub.prop and EOAT.sub.diff at blocks 232-234, computing a combined PID term EOAT.sub.-- PID that would cause COP.sub.tar to be equal to COP.sub.act at block 236, computing an integral control term EOAT.sub.int based on EOAT.sub.-- PED, EOAT.sub.prop and EOAT.sub.diff at block 238, and computing the cumulative error term EOAT.sub.errsum based on the EOAT.sub.int and G.sub.eoat.sbsb.--.sub.int at block 240. Similarly, the adjusted error term COP.sub.errsum is determined by computing the differential control term COP.sub.diff at block 242, computing a combined PID term COP.sub.-- PID that would cause PID.sub.-- DC to be equal to DC.sub.out at block 244, computing an integral control term COP.sub.int based on COP.sub.-- PID and COP.sub.diff at block 246 (COP.sub.prop is zero since COP.sub.tar is equal to COP.sub.act), and computing the cumulative error term COP.sub.errsum based on the COP.sub.int and G.sub.cop.sbsb.--.sub.int at block 248.

In summary, the present invention provides an improved dual control method for a variable displacement compressor where transitions between a normal control mode and a high speed control mode occur without causing unnecessary disturbances in the control and powertrain. While the invention has been described in reference to the illustrated embodiment, it is expected that various modifications in addition to those suggested above will occur to those skilled in the art. In this regard, it will be understood that the scope of this invention is not limited to the illustrated embodiment, and that controls incorporating such modifications may fall within the scope of this invention, which is defined by the appended claims.