SYSTEM AND METHOD FOR REGULATING A PHYSICAL PARAMETER OF A REAL TURBOMACHINE SYSTEM FROM A PHYSICAL PARAMETER SET POINT

Abstract

A system REG for regulating a physical parameter of a real turbomachine system F(p) from a physical setpoint parameter, the regulation system REG comprising a system OPTK for optimising the parameterisation gain K during the regulation, the optimisation system OPTK comprising a stability correction module determining a first gain component K1, a response time correction module determining a second gain component K2, the stability correction module being designed to inhibit the response time correction module when an instability is detected during the regulation of the physical parameter, and a determination module configured to determine the parameterisation gain K as a function of the previously determined first gain component K1 and second gain component K2.

Claims

1-11. (canceled)

12. A regulation system for regulating a physical parameter of a real system F(p) of a turbomachine from a physical parameter set point, the regulation system having a response time and comprising: a corrector comprising a correction function C1(p) and a parameterization gain K, a theoretical inverse transfer function F−1(p) of the real system F(p), and a system for optimizing the parameterization gain K during regulation, the optimization system comprising: i. a stability correction module configured to determine a first gain constant K1, with a positive value, upon detecting an instability of the regulation system upon regulating the physical parameter, ii. a module for correcting the response time of the regulation system, configured to determine a second gain constant K2, with a negative value, upon detecting a delay upon regulating the physical parameter, the stability correction module being configured to inhibit the response time correction module upon detecting an instability upon regulating the physical parameter, and iii. a determination module configured to determine the parameterization gain K as a function of the first gain component K1 and second gain component K2 previously determined.

13. The regulation system according to claim 12, wherein, with a deviation being defined between the physical parameter and the physical parameter set point, the stability correction module comprising a stability detection module configured to compare the deviation with a high deviation threshold and a low deviation threshold, the stability detection module is configured to detect an instability if the deviation is successively higher than the high deviation threshold and then lower than the low deviation threshold.

14. The regulation system according to claim 13, wherein, with the deviation oscillating during an instability, the stability detection module is configured to count oscillations following detection of an instability and to determine a stability correction parameter as a function of the number of oscillations counted, the first gain component K1 depending on the stability correction parameter.

15. The regulation system REG according to claim 14, wherein, the stability correction module is configured to reset the number of oscillations counted to zero in case of detection of a transient phase generated by a significant variation in the physical parameter set point.

16. The regulation system according to claim 12, wherein the stability correction module comprises an overshoot detection module configured to determine an overshoot parameter, the first gain component K1 depending on the overshoot parameter.

17. The regulation system according to claim 16, wherein, with a deviation being defined between the physical parameter and the physical parameter set point, the overshoot detection module is configured to start a monitoring period of the deviation following a significant increasing variation in the parameter set point, the overshoot detection module being configured to compare the deviation to at least one overshoot threshold during the monitoring period, the overshoot detection module being configured to detect an overshoot if the deviation is greater than the overshoot threshold.

18. The regulation system according to claim 12, wherein the response time correction module is configured to determine a tolerance range around the physical parameter set point and to determine a second gain constant K2 if the physical parameter does not fall within the tolerance range.

19. A method for regulating a physical parameter by implementing the regulation system according to claim 12, the regulation method comprising: a step of monitoring the stability upon regulating the physical parameter, a step of determining a first gain constant K1, with a positive value, upon detecting an instability upon regulating the physical parameter, a step of monitoring the response time of the regulation system REG upon regulating the physical parameter in the absence of instability, a step of determining a second gain constant K2, with a negative value, upon detecting a delay upon regulating the physical parameter, a step of determining the parameterization gain K of the corrector C(p) from the first gain constant K1 and the second gain constant K2 so as to ensure stability of the regulation while optimizing the response time.

20. A computer program comprising instructions for executing the steps of the regulation method according to claim 19 when said program is executed by a computer.

21. An electronic control unit for a turbomachine comprising a memory comprising instructions of the computer program according to claim 20.

22. A turbomachine comprising the electronic unit according to claim 21.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The invention will be better understood upon reading the following description, given solely by way of example, and referring to the appended drawings given as non-limiting examples, in which identical references are given to similar objects and in which:

[0037] FIG. 1 is a schematic representation of a system for correcting a real system by an inverse model according to prior art;

[0038] FIG. 2 is a schematic representation of a regulation system according to one embodiment of the invention for a correction system of a real system by an inverse model;

[0039] FIG. 3 is a schematic representation of one embodiment of a stability correction module;

[0040] FIG. 4 is a schematic representation of one embodiment of an instability detection module;

[0041] FIG. 5 is a time course curve of the deviation ε for the detection of instabilities;

[0042] FIG. 6 is a schematic representation of one embodiment of an overshoot detection module;

[0043] FIG. 7 is a time course curve of the physical parameter following an increasing variation in the parameter set point, that is an increasing transient phase;

[0044] FIG. 8 is a set of curves representing the time course, during a transient phase, of the physical parameter and of the deviation to determine an overshoot correction value;

[0045] FIG. 9 is a schematic representation of one embodiment of a response time correction module;

[0046] FIG. 10 is a time course curve of the physical parameter y with respect to the physical parameter set point yc in stationary phase;

[0047] FIG. 11 is a set of curves representing the time course of the physical parameter following a periodic delay affecting the real system for a regulation system according to prior art;

[0048] FIG. 12 is a set of curves representing the time course of the physical parameter following a periodic delay affecting the real system for a regulation system according to the invention.

[0049] It should be noted that the figures set out the invention in a detailed manner for implementing the invention, said figures can of course be used to better define the invention if necessary.

DETAILED DESCRIPTION

[0050] With reference to FIG. 1, there is represented a correction system SC configured to determine a control parameter y as a function of a parameter set point yc. In this example, the correction system SC implements an inverse model corrector. In other words, the correction system SC successively comprises a corrector C(p), an inverse model F−1(p) and a real system F(p). The real system F(p) corresponds to the real system of the turbomachine which acts on the control parameter y. In practice, the real system F(p) implements several transfer functions. By nature, the real system F(p) changes over time (drift, wear, etc.) and does not respond identically to a parameter set point yc.

[0051] In a correction system SC implementing an inverse model corrector, it is assumed that the real system F(p) is mathematically invertible so as to define the inverse model F.sup.−1 (p). According to this hypothesis, the correction system SC then depends essentially on the corrector C(p) since F.sup.−1(p)*F(p)=1. In other words, the response time and stability of the correction system can then be directly determined by the corrector C(p).

[0052] In a known manner, the corrector C(p) comprises a transfer function Ci(p) and a tuning gain K known to those skilled in the art so as to obtain the following formula: C(p)=(p)*1/K.

[0053] According to the present invention, the tuning gain K is modified over time so as to adjust the response time and the stability of the corrector C(p). Thus, if a pure delay occurs in the real system F(p) or if the statistical gain of the real system F(p) is changed, the tuning gain K can be modified to maintain optimal performance.

[0054] Hereafter, the set comprising the corrector C(p) and the inverse model F.sup.−1(p) is referred to as regulator REG and provides a preliminary physical parameter ya to the real system F(p). With reference to FIG. 2, a regulator REG according to one embodiment of the invention is represented. The regulator REG has a response time.

[0055] With reference to FIG. 2, the corrector C(p) comprises a transfer function C.sub.1 (p) and an optimization system OPTK of the parameterization gain K.

[0056] The optimization system OPTK comprises a stability correction module 2 configured to determine a first gain component K1, a response time correction module 3 configured to determine a second gain component K2, and a determination module 4 configured to determine the parameterization gain K as a function of the previously determined gain components K1, K2. In the following, a deviation ε is defined which corresponds to the difference between the physical parameter set point yc and the physical parameter y (ε=yc−y).

[0057] The different modules will now be set forth in detail.

[0058] With reference to FIG. 2, the stability correction module 2 is configured to determine the first gain component K1 as well as an instability TopIS as a function of the control parameter y and the parameter set point yc.

[0059] Stability Correction Module 2 (FIG. 3)

[0060] The stability correction module 2 is schematically represented in FIG. 3. It comprises an instability detection module 21 configured to detect an instability TopIS and determine a stability correction parameter TopCS from the deviation ε and the parameter set point yc. In practice, as will be set forth later, the instability detection module 21 allows detection of an instability around the parameter set point yc.

[0061] The stability correction module 2 further comprises an overshoot detection module 22, configured to detect an overshoot TopOS from epsilon ε, the control parameter y and the parameter set point yc. In other words, during a quick increasing variation in the parameter set point yc, the physical parameter y can overshoot or undershoot the parameter set point yc and create an instability related to the transient phase. In what follows, the shorthand “a transient” will also be used to refer to a transient phase.

[0062] Similarly, the stability correction module 2 further comprises an undershoot detection module 23, configured to detect an undershoot TopUS from epsilon ε, the control parameter y and the parameter set point yc.

[0063] The stability correction module 2 finally comprises a stability correction module 24 configured to determine the first gain component K1 as a function of the overshoot detection parameter TopOS, the undershoot detection parameter TopUS and the stability correction parameter TopCS obtained by the other modules 21, 22, 23 of the stability correction module 2.

[0064] Stability Detection Module 21

[0065] The stability detection module 21 is schematically represented in FIG. 4. The stability detection module 21 comprises a first module 211 to detect an instability TopIS if the epsilon ε value is greater than a high threshold SH-CS or lower than a low threshold SB-CS. An instability TopIS is detected when a high threshold SH-CS and a low threshold SB-CS are successively overshot or undershot.

[0066] As an example, the first module 211 is configured to compare, on the one hand, the deviation ε to the high threshold SH-CS and, on the other hand, the deviation ε to the low threshold SB-CS. If the deviation ε is greater than the high threshold SH-CS, the over- or under-shoot is stored in a memory. Similarly, if the deviation ε is lower than the high threshold SB-CS, the over- or under-shoot is stored in a memory. When two over- or under-shoots of different natures are successively detected, an instability TopIS is detected as illustrated in FIG. 5. In the absence of threshold SB-CS, SH-CS crossing, no instability TopIS is detected.

[0067] As illustrated in FIG. 4, the stability detection module 21 further comprises a second module 212 to count oscillations NB-osc of the physical parameter y when an instability TopIS is detected. The number of oscillations NB-osc can be conveniently obtained by counting successive over- or under-shoots and storing them in a memory during a stationary phase.

[0068] The second module 212 is also configured to receive a zero reset command to reset the number of oscillations NB-osc to zero. To this end, as illustrated in FIG. 4, the stability detection module 21 comprises a third module 213 and a fourth module 214 to determine a zero reset RAZ.

[0069] As illustrated in FIG. 4, the stability detection module 21 comprises a fifth module 215 configured to determine the correction parameter TopCS from a number of oscillations NB-osc. In this embodiment, if the number of oscillations NB-osc is greater than 3, a correction is performed and a correction parameter TopCS is transmitted by the stability detection module 21. As illustrated in FIG. 4, during a correction, a zero reset information RAZ is also transmitted to avoid repeating the same correction.

[0070] Still referring to FIG. 4, the stability detection module 21 comprises a third module 213 to detect a transient, that is, a significant variation in the parameter set point yc, and to reset the number of oscillations NB-osc to zero following a transient detection. To this end, the third module 213 monitors whether the deviation ε is maintained between a second high threshold SH2 and a second low threshold SB2. Preferably, the second high threshold SH2 is larger than the high threshold SH-CS and the second low threshold SB2 is smaller than the low threshold SB-CS which were used previously so as to maintain a margin with respect to the detection of instability TopIS. If the second thresholds SB2, SH2 are exceeded, the presence of a transient is detected, which resets the number of oscillations NB-osc to zero and stops the stability correction. Indeed, transient-related corrections are processed by the overshoot detection module 22 and the undershoot detection module 23.

[0071] As illustrated in FIG. 4, the stability detection module 21 comprises a fourth timing module 214 configured to detect whether a correction has been made and command a zero reset RAZ in case of detection. In other words, the fourth module 4 enables a new correction to be inhibited when a previous correction has not yet taken effect. In this example, the fourth module 214 is in the form of a timing which is a function of the response time of the real system F(p).

[0072] Advantageously, the stability detection module 21 enables a correction parameter TopCS to be determined as a function of the number of oscillations NB-osc measured following the detection of an instability TopIS i. Advantageously, any correction is inhibited in the case of a transient or a correction not yet taken by the real system F(p) into account. The correction parameter TopCS calculated in this way makes it possible to improve regulation stability, as will be set forth later.

[0073] Overshoot Detection Module 22

[0074] The overshoot detection module 22 is schematically represented in FIG. 3. It allows an upper correction value TopOS to be determined if the physical parameter y to be regulated makes an overshoot at the end of the transient. In other words, the overshoot detection module 22 enables an immediate correction to be made following an increasing variation in the parameter set point yc.

[0075] In practice, with reference to FIG. 6, the overshoot detection module 22 comprises a module for determining 221 the response of the real system F(p) in a closed loop Yb following the reception of a physical parameter set point yc. This allows a theoretical ideal response of the real system F(p) to be determined. In this example, the determination module 221 implements a “low-pass” type, in particular, first- or second-order filter.

[0076] The overshoot detection module 22 further comprises a transient detection module 222, that is, a variation in the deviation ε from the closed loop response yBF, in particular, of its derivative. In this way, it is determined whether the regulation is actually in an increasing transient phase, that is, an increasing variation in the control set point yc. If the deviation ε deviates from the closed loop response yBF, an acceleration is detected TopAccel. An example of a transient with an instability at the output of the transient is schematically illustrated in FIG. 7 (damping defect).

[0077] Still referring to FIG. 6, the overshoot detection module 22 further comprises a storage module 223 configured to start a monitoring period when an acceleration TopAccel is detected. Thus, the overshoot detection module 22 focuses on detecting an instability during a transient acceleration phase. Indeed, the overshoot detection module 22 does not aim at detecting an over- or under-shoot of the physical parameter y when the parameter set point yc is stationary, which corresponds to a regulation instability.

[0078] The overshoot detection module 22 further comprises a module for monitoring the deviation ε with respect to overshoot thresholds SD1, SD2. In this embodiment, the monitoring module 222 comprises two overshoot thresholds SD1, SD2, which in this example are hysteresis type thresholds.

[0079] As illustrated in FIG. 6, if an overshoot of an overshoot threshold SD1, SD2 is detected during the monitoring period, an overshoot correction value TopOS is output by the overshoot detection module 22.

[0080] Advantageously, the overshoot detection module 22 comprises modules to stop the monitoring period of the storage module 224 in case of detection of stabilization (module 225) or in case of deceleration set point (module 226). Indeed, it is necessary to avoid that an overshoot is detected by a deceleration of the physical parameter set point yc. This avoids making untimely corrections which are a source of instability.

[0081] An example of implementation of an overshoot detection is illustrated in FIG. 8. In this example, following a transient acceleration phase, the deviation ε overshoots the first overshoot threshold SD1, which activates an overshoot correction TopOS, and then the second overshoot threshold SD2, which again activates an overshoot correction TopOS. In other words, the correction is responsive and allows the regulation to be corrected as soon as an overshoot occurs at the end of the transient. Such a correction is possible because it is only carried out in the presence of a transient.

[0082] Undershoot Detection Module 23

[0083] The undershoot detection module 23 is schematically represented in FIG. 3. It will not be set forth in detail hereafter given that it is similar to the overshoot detection module 22 but aims at detecting an undershoot during a decreasing variation in a parameter set point yc, that is a deceleration of the physical parameter set point yc.

[0084] Similarly to the overshoot detection module 22, if a transient phase is detected and an exceedance of a lower threshold is detected, an undershoot correction value TopUS is output from the undershoot detection module 23.

[0085] Calculation Module 24

[0086] The calculation module 24 is schematically represented in FIG. 3 and has the function of determining the first correction component K1 from the correction values TopCS, TopOS and TopUS.

[0087] Response Time Correction Module 3 (FIG. 9)

[0088] The response time correction module 3 is schematically represented in FIG. 9. It comprises a stability detection module 31 configured to detect that the parameter set point yc is actually stable. In practice, the stability detection module 31 checks that the parameter set point yc does not vary much, that is, without transients. The stability detection module 31 produces a stability confirmation signal ConfS for a calculation module 33.

[0089] Still referring to FIG. 9, the response time correction module 3 comprises a module for determining a tolerance range 32 around the physical parameter set point yc. In this example, a low template GabBTR and a high template GabHTR are determined beforehand and a tolerance range defined between yc−GabBTR and yc+GabHTR is derived therefrom. The tolerance range corresponds to the lag that is considered allowable between the physical parameter y and the physical parameter set point yc. Preferably, the templates GabBTR, GabHTR are determined from the response of the real system F(p) in a closed loop yBF following the reception of a physical parameter set point yc. Thus, the templates GabBTR, GabHTR are ideally determined to define a reference tolerance range.

[0090] The response time correction module 3 further comprises a calculation module 33 configured to determine a second gain constant K2 if the physical parameter y does not belong to the tolerance range during a stationary phase confirmed by the confirmation signal ConfS.

[0091] In practice, the response time correction module 3 can monitor any delay of the physical parameter y with respect to the physical parameter set point yc. Such a delay can, for example, be related to a parameterization gain K that has been increased too much, in particular, following a detection of an instability. A second gain constant K2 with a negative value makes it possible to improve the response time. Advantageously, several point corrections are thus performed.

[0092] As illustrated in FIG. 10, the physical parameter y deviates from the monitoring range in the zones P31 and P32, which reflects a delay in regulation. The response time correction module 3 determines the second gain constant K2 based on the number of deviations outside the monitoring range.

[0093] Module 4 for Determining the Parameterization Gain K (FIG. 2)

[0094] As illustrated in FIG. 2, the module 4 for determining the parameterization gain K is configured to sum the first gain constant K1, determined by the stability correction module 2, and the second gain constant K2, determined by the response time correction module 3. Preferably, a static gain constant is also added to determine the parameterization gain K. In fact, the parameterization gain K is modified in real time.

[0095] Given that the first gain constant K1 is positive and the second gain constant K2 is negative, the parameterization gain K is dynamically modified during the regulation in order to adapt to the changes over time and correct any drift.

[0096] Example of implementation of the regulation method with dynamic optimization of the tuning gain K

[0097] According to the regulation method according to the invention, the method comprises a step of monitoring the stability upon regulating the physical parameter, a step of determining a first gain constant K1, with a positive value, upon detecting an instability upon regulating the physical parameter y, a step of monitoring the response time upon regulating the physical parameter in the absence of instability, a step of determining a second gain constant K2, with a negative value, upon detecting a delay upon regulating the physical parameter, and a step of determining the parameterization gain K of the corrector C(p) from the first gain constant K1 and the second gain constant K2 so as to ensure stability of the regulation while optimizing the response time.

[0098] As an example, in order to set forth the advantages of the invention with respect to prior art, the time course of the physical parameter y as a function of the physical parameter set point yc (upper curves) when the real system F(p) undergoes a periodic delay RET (middle curve) for a static parameterization gain (lower curve) is represented in FIG. 11. In this example, the delay RET varies between 0.2 s and 3 s.

[0099] As illustrated in FIG. 11, when the delay RET becomes significant, significant instabilities of the physical parameter y appear, both in stabilized operation and during transients. Moreover, the quickness is not optimal. Such a regulation is not satisfactory.

[0100] With reference to FIG. 12, the time course of the physical parameter y as a function of the physical parameter set point yc (upper curves) when the real system F(p) undergoes a periodic delay RET (middle curve) for a dynamic parameterization gain (lower curve) as optimized by the regulation system REG according to the invention is represented.

[0101] As illustrated in FIG. 12, when the delay RET becomes significant, an instability of the physical parameter y appears which is responsively corrected by an increase in the parameterization gain K. This increase is related to the stability correction module 2 which has increased the first gain constant K1 following the detection of the instability TopIS. When the delay RET becomes smaller, the past increase of the parameterization gain K is detrimental to the response time and introduces a lag. This lag is responsively corrected by a decrease of the parameterization gain K. This decrease is related to the response time correction module 3 which has decreased the second gain constant K2 following the detection of a lag.

[0102] By virtue of the invention, the stability and response time of the regulation system REG are dynamically and responsively corrected over time. The performance of the regulation system REG is optimal due to its self-adaptation.