CN114282570A - Fault tolerance method and fault tolerance system for sensor of aircraft engine - Google Patents
Fault tolerance method and fault tolerance system for sensor of aircraft engine Download PDFInfo
- Publication number
- CN114282570A CN114282570A CN202010986491.7A CN202010986491A CN114282570A CN 114282570 A CN114282570 A CN 114282570A CN 202010986491 A CN202010986491 A CN 202010986491A CN 114282570 A CN114282570 A CN 114282570A
- Authority
- CN
- China
- Prior art keywords
- sensor signal
- signal
- type sensor
- reconstruction system
- signal reconstruction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Feedback Control In General (AREA)
Abstract
The invention relates to a fault tolerance method and a fault tolerance system for an aircraft engine sensor. The fault-tolerant method comprises the following steps: step S1, establishing a first type sensor signal reconstruction system and a second type sensor signal reconstruction system; step S2, establishing a reliability evaluation mechanism for the first type sensor signal reconstruction system and the second type sensor signal reconstruction system, wherein the reliability evaluation mechanism selects to output the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system or the weight value of the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system according to the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system and the residual error of the output value of the second type sensor signal reconstruction system. According to the scheme provided by the invention, the reconstruction precision of the sensor signal of the control system of the aircraft engine can be improved on the premise of meeting the basic robustness safety of the control system.
Description
Technical Field
The invention relates to the technical field of fault tolerance of aero-engines, in particular to a fault tolerance method and a fault tolerance system of an aero-engine sensor.
Background
The aircraft engine is a control object with a complex structure and strong nonlinearity. The control system is required to have high reliability according to specific working conditions of the aircraft engine, and the requirement is more urgent than simply improving the performance of the system. The system may incur serious damage once it fails. The aeroengine sensor works in a high-temperature and strong-vibration environment and is one of the most unreliable control elements in the system, so that the technology for improving the fault tolerance of the aeroengine airborne sensor is an important operation means for improving the reliability of an engine control system and ensuring the flight safety.
At present, the airborne sensor mainly uses two sensor redundancy technologies, namely hardware redundancy and resolution redundancy. The hardware redundancy measures the same engine parameter by using a plurality of same sensors, and then a voter is used for detecting faults; the analytical redundancy engine model and its estimation technique, etc. provide an estimated value of the measured parameter as redundancy information, and the deviation between the estimated value and the measured value of the sensor is used to detect and isolate the fault. The sensor hardware redundancy is adopted to play a backup role, fault isolation can be carried out on the sensor hardware redundancy, the control system reliability is higher as the hardware redundancy is more, but the hardware redundancy can increase the weight, the size and the development cost of the control system and influence the performance and the economical efficiency of an engine, and therefore the analysis redundancy is also an important fault processing means. A full-authority digital electronic control (FADEC) system is widely adopted in the current engine, the analytic redundancy of key sensor input signals in the FADEC system needs to be constructed, the purpose is to use reconstructed sensor signals as an important basis for judging sensor faults in real time, and when the hardware redundancy of the sensor is completely failed, the reconstructed sensor signals can be used as backup input.
As for the research on constructing the signal analysis redundancy of the engine sensor, a method based on signal synthesis of other sensors is commonly adopted in domestic and foreign engines, and the service experience of the engines shows that the method is simple and reliable and can reduce the fault influence of an engine control system; in recent years, relatively hot researches are made on sensor reconstruction methods based on an airborne engine model or data driving, including adaptive or intelligent methods such as kalman estimation (virtual sensor technology), neural networks, support vector machines and the like, and compared with sensor signal synthesis methods, the methods can more accurately diagnose and reconstruct a fault sensor in transient state or performance deviation, so that great attention is paid to people.
The linear Kalman algorithm is represented in a recursion form, the algorithm is researched by learners more thoroughly, but the phenomenon of nonlinearity is quite common in practical application, for example, the nonlinearity problem caused by factors such as external interference on a physical model, existence of nonlinearity and a sick variance matrix and the like causes the traditional linear Kalman filtering algorithm not to be applicable any more, the Kalman filtering technology suitable for a nonlinear system is urgently needed to be obtained by improving the Kalman filtering algorithm, and the continuous nonlinear equation is firstly subjected to linearization and discretization treatment on the assumption that all transformations are quasi-linear, and a first-order Taylor expansion formula is used for approximating the nonlinear model to obtain an Extended Kalman Filtering (EKF) algorithm corresponding to the nonlinear system. And (3) the EKF develops the nonlinear state equation and observation equation into Taylor series around the estimation value of the last step, and a first-order approximation is taken to obtain a linearized model, so that a standard KF recursion system is used continuously. When the system is weak non-linear, the EKF filtering precision is high, but when the system is strong non-linear, the EKF filtering precision is greatly reduced and even the filtering divergence can be caused. Because the engine belongs to a strong nonlinear system, the use and robustness of the Kalman estimator have limitations, and in the aspect of a control system, the reliability and safety requirements of the FADEC control system determine that all control algorithms have enough robustness, so the use of the new methods in the aspect of control system signal reconstruction needs to be carefully considered.
In the aspect of sensor analysis redundancy design, two main methods are provided, one is a method based on signal synthesis of other sensors, and the other is (a sensor reconstruction method) based on adaptive Kalman estimation, a neural network, a support vector machine and other adaptive or intelligent methods, wherein a Kalman filtering estimation method is one of the technologies which have the most potential to be used for aircraft engine control at present.
The sensor signal synthesis method mainly focuses on sensor signal reconstruction in a steady state and a quasi-steady state, and the accuracy of a reconstructed signal can deviate from the original design effect in the transient process or under the conditions that an engine is subjected to individual difference, performance degradation, gas circuit component failure and the like.
The Kalman estimation method is a technology which is most potential to be used in the control of an aircraft engine at present, a Kalman estimator is usually used for carrying out sensor signal reconstruction, and the method can consider the influences of factors such as individual difference of the engine, performance degradation, gas path component faults and the like through a state quantity augmentation mode, so that the reconstructed signals are more accurate. In practical engineering, the method is used for control, measurement parameters used in the Kalman estimator are more than parameters used in a signal synthesis method, the estimation precision of the Kalman estimator can be guaranteed, but the reliability is reduced, and errors can be brought into a reconstructed signal due to the fault of any measurement sensor. In addition, because the engine belongs to a strong nonlinear system, the robustness of the extended kalman estimator is limited, and the reliability and safety requirements of the FADEC control system determine that the control algorithm has enough robustness, the use of the method in the aspect of control system signal reconstruction needs to be controlled in a compromise manner.
Disclosure of Invention
The invention aims to provide a fault tolerance method and a fault tolerance system for an aircraft engine sensor, which can improve the reconstruction accuracy of a sensor signal of an aircraft engine control system on the premise of meeting the basic robustness and safety of the control system.
In order to solve the technical problem, the invention provides a fault tolerance method for an aircraft engine sensor, which comprises the following steps:
step S1, establishing a first type sensor signal reconstruction system and a second type sensor signal reconstruction system;
step S2, establishing a reliability evaluation mechanism for the first type sensor signal reconstruction system and the second type sensor signal reconstruction system, where the reliability evaluation mechanism selects to output the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system, and the residual error of the output value of the second type sensor signal reconstruction system, or selects a weighted value of the output values of the first type sensor signal reconstruction system, the second type sensor signal reconstruction system, or both.
According to one embodiment of the invention, the first type of sensor signal reconstruction system adopts a signal synthesis method, and the second type of sensor signal reconstruction system adopts a Kalman estimation method.
According to one embodiment of the invention, the first type of sensor signal reconstruction system has a first priority, the second type of sensor signal reconstruction system has a second priority, and the second priority is an alternative fault-tolerant manner with respect to the first priority based on the credibility assessment mechanism.
According to one embodiment of the invention, the signal synthesis method is based on sensor signal data, an engine non-linear model, an engine component level model or channel sensor values.
According to one embodiment of the present invention, the step of evaluating the credibility evaluation mechanism comprises:
step S21, calculating the residual error of the output value of the second type sensor signal reconstruction system;
step S22, acquiring a first reconstruction signal and a second reconstruction signal, where the first reconstruction signal is a sensor signal of the first type of sensor signal reconstruction system, and the second reconstruction signal is a sensor signal of the second type of sensor signal reconstruction system;
step S23, obtaining an absolute value of a difference between the first reconstructed signal and the second reconstructed signal;
step S24, dividing the absolute value of the difference value and the residual error to obtain a deviation value, and sorting the results of the deviation value according to the size;
and step S25, selecting the range of the deviation value in sections according to the sorting result, and selecting and outputting the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system or selecting the weight value of the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system according to the range.
According to an embodiment of the present invention, in step S21, the residual error S of the output value of the second type sensor signal reconstruction system is:
wherein y isiIs a sensor measurement of the engine and,
the values calculated for the Kalman estimator state space model, n is the number of sensor measurements.
According to an embodiment of the invention, in step S25, at least three values T1, T2 and T3 are selected within the range of the deviation value, wherein 0< T1< T2< T3;
when the deviation value is smaller than T1 or not smaller than T3, selecting a weight value for outputting the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system;
when the deviation value is not less than T1 and less than T2, selecting a weight value for outputting the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system;
and when the deviation value is not less than T2 and less than T3, selecting the weight value of the output value of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system.
The invention also provides an aircraft engine sensor fault tolerance system, which is used for executing the aircraft engine sensor fault tolerance method, the fault tolerance system comprises a numerical control channel A and a numerical control channel B which are completely the same, data exchange is carried out between the numerical control channel A and the numerical control channel B through a data bus CCDL, and the numerical control channel A and the numerical control channel B respectively comprise:
a signal detection confirmation module to determine that the sensor signal is valid;
each signal reconstruction module comprises a first type sensor signal reconstruction module, a second type sensor signal reconstruction module and a calculation module, and the calculation module is respectively connected with the first type sensor signal reconstruction module and the second type sensor signal reconstruction module;
and the control law module is used for controlling and outputting the analog quantity signal and the switching value signal according to the reliability evaluation mechanism so as to control the state of the engine.
According to an embodiment of the invention, the numerical control channel a and the numerical control channel B further include a signal state information module, respectively, and the signal state information module receives the analog signal of the signal detection confirmation module, converts the analog signal into a digital signal, and sends the digital signal to the first type sensor signal reconstruction module and the second type sensor signal reconstruction module.
According to an embodiment of the present invention, the nc a channel and the nc B channel further include a signal reconfiguration switch respectively, which is communicated between the calculation module and the control law module, and is configured to selectively transmit the result of the calculation module to the control law module.
The invention provides a fault tolerance method and a fault tolerance system for an aircraft engine sensor, which combine the advantages of a signal synthesis method and a Kalman estimation method by adopting a fusion strategy to form a set of fault tolerance method and a fault tolerance system for the sensor meeting the configuration of the conventional FADEC control system, and can improve the reconstruction precision of the sensor signal of the aircraft engine control system on the premise of meeting the basic robustness and safety of the control system.
It is to be understood that both the foregoing general description and the following detailed description of the present invention are exemplary and explanatory and are intended to provide further explanation of the invention as claimed.
Drawings
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below, wherein:
FIG. 1 shows a dual channel signal processing flow diagram of a prior art FADEC system;
FIG. 2 illustrates an aircraft engine sensor fault tolerance method in an embodiment of the invention;
FIG. 3 illustrates a method of evaluation of a trustworthiness evaluation mechanism;
FIG. 4 shows a diagram of the Ps3 sensor signal synthesis;
FIG. 5 shows a block diagram of a Kalman estimator that performs the reconstruction of the Ps3 signal;
FIG. 6 shows a comparison plot one of the engine, Kalman estimator, and Ps3 for signal synthesis;
FIG. 7 shows a comparison graph two of the engine, Kalman estimator, and signal synthesized Ps 3;
FIG. 8 shows a schematic diagram of the signal synthesis and output selection strategy of the Kalman estimator;
FIG. 9 shows a schematic diagram of signal synthesis and confidence evaluation of a Kalman estimator;
FIG. 10 illustrates a schematic structural diagram of an aircraft engine sensor fault tolerance system in an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways than those specifically described herein, and thus the present invention is not limited to the specific embodiments disclosed below.
As used in this application and the appended claims, the terms "a," "an," "the," and/or "the" are not intended to be inclusive in the singular, but rather are intended to be inclusive in the plural unless the context clearly dictates otherwise. In general, the terms "comprises" and "comprising" merely indicate that steps and elements are included which are explicitly identified, that the steps and elements do not form an exclusive list, and that a method or apparatus may include other steps or elements.
In describing the embodiments of the present application in detail, the cross-sectional views illustrating the structure of the device are not enlarged partially in a general scale for convenience of illustration, and the schematic drawings are only examples, which should not limit the scope of the present application. In addition, the three-dimensional dimensions of length, width and depth should be included in the actual fabrication.
For convenience in description, spatial relational terms such as "below," "beneath," "below," "under," "over," "upper," and the like may be used herein to describe one element or feature's relationship to another element or feature as illustrated in the figures. It will be understood that these terms of spatial relationship are intended to encompass other orientations of the device in use or operation in addition to the orientation depicted in the figures. For example, if the device in the figures is turned over, elements described as "below" or "beneath" other elements or features would then be oriented "above" the other elements or features. Thus, the exemplary words "below" and "beneath" can encompass both an orientation of up and down. The device may have other orientations (rotated 90 degrees or at other orientations) and the spatial relationship descriptors used herein should be interpreted accordingly. Further, it will also be understood that when a layer is referred to as being "between" two layers, it can be the only layer between the two layers, or one or more intervening layers may also be present.
In the context of this application, a structure described as having a first feature "on" a second feature may include embodiments in which the first and second features are formed in direct contact, and may also include embodiments in which additional features are formed in between the first and second features, such that the first and second features may not be in direct contact.
First, the basic aircraft engine control system sensor fault tolerance framework is described below. Fig. 1 shows a two-channel signal processing flow diagram of a prior art FADEC system. Taking a typical dual-channel full-authority engine digital electronic control system as an example, the electronic controller is composed of dual-redundancy numerical control channels with completely the same functions, and the dual-redundancy numerical control channels are divided into an a channel 101 (main channel) and a B channel 102 (backup channel). Any one channel can complete all control functions of the engine, and two channels with completely equivalent functions adopt similar redundancy design. The two channels of the controller are mutually hot backups, and data exchange is carried out between the two channels through an interchannel data bus CCDL 103. During operation, A, B two channels convert the engine status signal and the sensor signal of the numerical control system into corresponding digital signals through respective processing circuits, and the signals are detected and confirmed in the signal detection module 104, and the general detection and confirmation methods include extremum detection, slope detection, BIT self-check, and whether the signals are valid or not is confirmed through the signal detection module, and then the signals enter the parameter selection module 105. The parameter selection module 105 not only receives the value of the local channel, but also obtains the signal value of the backup channel through the CCDL data, and the selection logic in the parameter selection module 105 determines which signal is to be used for the control law calculation module 106 later. The common fault tolerance strategy for two-channel signals includes: if both signals are valid, the average value is adopted; when the cross channel signal is invalid, using the channel signal; when the channel signal fails, the cross channel signal is used; when the two channel signals do not coincide, the preference to use the better one or use the value of the model value module 107; when both channel signals are invalid, a model value or default value is used. The control signal determined by the selection logic is used as the input of the control law calculation module 106, and finally, the main control channel outputs the corresponding analog quantity signal and the corresponding switching value signal to the corresponding execution mechanism (the output control module 107), so that the state of the engine is controlled.
The parameter selection module 105 and model value module 107 in fig. 1 determine how the fault is fault tolerant. This will depend on the hardware configuration, the failure impact of the signal and the ability to reconstruct the signal, different signals having different fault tolerance strategies, where signal reconstruction is a very important part of the module, and most ideally a virtual sensor signal with a very high confidence level consistent with the real engine sensor can be reconstructed. The sensor reconstruction method of signal synthesis is used in aeroengines more frequently. The signal synthesis method is mainly based on the gas circuit coupling characteristic of the engine, and the analytical redundancy is constructed through the relation fitting of signal data of other sensors.
FIG. 2 illustrates an aircraft engine sensor fault tolerance method in an embodiment of the invention; fig. 3 illustrates an evaluation method of the credibility evaluation mechanism. As shown in the figure, the fault tolerance method for the sensor of the aircraft engine provided by the invention comprises the following steps:
step S1, establishing a first type sensor signal reconstruction system and a second type sensor signal reconstruction system;
step S2, establishing a reliability evaluation mechanism for the first type sensor signal reconstruction system and the second type sensor signal reconstruction system, wherein the reliability evaluation mechanism selects to output the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system or the weight value of the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system according to the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system and the residual error of the output value of the second type sensor signal reconstruction system.
Preferably, the first type of sensor signal reconstruction system adopts a signal synthesis method, and the second type of sensor signal reconstruction system adopts a Kalman estimation method. These two types of methods will be specifically described later.
Preferably, the signal synthesis method is based on sensor signal data, an engine nonlinear model, an engine component level model, or channel sensor values. Fig. 4 shows a diagram of the Ps3 sensor signal synthesis. As shown, in order to illustrate the features of the present invention, the high-pressure compressor outlet static pressure sensor Ps3 is taken as an example to illustrate the signal synthesis method. Wherein, the Ps3 sensor is a compressor outlet static pressure sensor, and mainly ensures the operability of the engine and the stability of the compressor. The maximum value of the high pressure compressor exit static pressure Ps3 is limited primarily by combustor case strength and fuel pump pressure. The minimum value of the high-pressure compressor outlet static pressure Ps3 is limited primarily by the combustor flameout boundary and the aircraft bleed air pressure, which are relatively important signals used in engine control. According to the following formulas (1) - (5), Ps3 is mainly calculated by total temperature, total pressure, flow and cross-sectional area, for any point on the common working line of the high-pressure compressor, if the input conditions T25 and P25 and N2R25 and PS3/P25 are not changed, the efficiency, the converted flow, the pressure ratio and the flow of the compressor are not changed, and the value of Ps3 obtained by the model is the same as the value of the true engine Ps 3. In order to ensure the calculation time, a relationship table of the sensors of N2R25 and P25 and Ps3 can be established, and the signal synthesis structure diagram is shown in FIG. 4.
Eff＝MAPHPC(N2R25, P3/P25) formula (1)
Wac＝MAPHPC(N2R25, P3/P25) formula (2)
Wa＝f1(Wac, P25, T25) formula (3)
T3＝f2(P3/P25, T25, Eff) equation (4)
Ps3＝f3(P3, T3, A, Wa) formula (5)
In the formula, MAPHPCReferring to the characteristics of the compressor, Eff is the efficiency of the compressor, Wac is the converted flow of the compressor, Wa is the flow of the compressor, P25 is the inlet pressure of the compressor, T25 is the inlet temperature of the compressor, T3 is the outlet temperature of the compressor, A is the cross-sectional area of the compressor, and Ps3 is the outlet static pressure of the compressor. Similarly, other sensor signals of the engine can be combined with an aeroengine aerodynamic coupling mode to establish a similar signal synthesis method.
Still taking the Ps3 sensor as an example, the sensor fault signal reconstruction method of the kalman estimation method is described. It will be readily appreciated that all of the teachings set forth herein are applicable to other sensor signals. From the point of view of signal synthesis, there is an assumption that the characteristics of the compressor must be constant. In practice, the characteristics of the compressor are changed due to manufacturing differences, performance degradation of the compressor in service, and engine-induced changes, and actually, the synthesized signal of the Ps3 deviates from the Ps3 signal of a real sensor. Therefore, the constructed Kalman estimation method comprises the influence of the characteristic change of the compressor, and the degree of the characteristic deviation of the compressor from a design value is indicated through the health parameters of the efficiency and the flow of the compressor. Firstly, a state space model is designed, and meanwhile, a compressor efficiency health parameter KE3D25 and a flow health parameter KW25R are used as augmentation state quantities, and the state space model listed in the text is shown as a formula 6. The state space model is not unique, and can be increased or decreased according to actual conditions regardless of input quantity, state quantity or output quantity.
With the state space model, an augmented linear kalman estimator can be designed, and a kalman estimation method is adopted, which belongs to a traditional control theory method and is not described in detail herein. The sensor signal reconstruction method adopting the Kalman estimation method is shown in FIG. 5, the structure adopts N1, N2, T25, T3 and P25 as measurement parameters of a Kalman estimator 501, health parameters of N1, N2, KE3D25 and KW25R are estimated by the Kalman estimation method, the health parameters and fuel quantity are used as output of a state space model 502, and a predicted Ps3 signal can be obtained, and FIG. 5 shows a structure diagram of the Kalman estimation method for reconstructing the Ps3 signal.
Compared with a signal synthesis method, the kalman estimation method additionally introduces a measurement signal of an N1 sensor, a measurement signal of a T3 sensor, and health parameters of the efficiency and the flow of the compressor, so that the fault-tolerant capability of the fault-tolerant method under transient state and Ps3 sensor fault is enhanced, but the reliability of the fault tolerance of the Ps3 signal is actually reduced due to the fact that the number of dependent sensors is increased, and assuming that the reliability of the sensors is the same, the more the number of dependent sensors is, the higher the probability of failure of the overall measurement system is, the higher the probability of Ps3 signal error is, and taking the above example as an example, the reconstructed results of the kalman estimation method and the signal synthesis method under the condition that the N1 sensor fails at the same time and the N1 sensor fails are shown in fig. 4. Fig. 6 shows a comparison graph one of the engine, kalman estimator, and signal synthesized Ps 3. As can be seen from fig. 6, the Ps3 signal reconstructed by the signal synthesis method is consistent with the Ps3 signal of the real engine, but the Ps3 signal reconstructed based on the kalman estimation method is affected because the error information of N1 is "likely" to be carried into the reconstruction of Ps3, and in this example, the failure of the N1 sensor causes a large deviation of the Ps3 signal reconstructed by the kalman estimator.
The above description of the use of the word "may" is meant to be specific. Because in the signal reconstruction of the kalman estimation method, if there is a linear relationship between the measurement sensor signals, if one of the sensors fails, the kalman estimation method superimposes the influence on the state quantity, and the influence on the reconstructed sensor signals becomes small. Still using the above example, assuming simultaneous failures of the N2 sensors, the kalman estimator and signal synthesis method reconstruction results under the N2 sensor failure are shown in fig. 7. As can be seen from fig. 7, the Ps3 signal reconstructed by the kalman estimator is consistent with the Ps3 signal of the real engine, and in contrast, the Ps3 signal reconstructed by the signal synthesis method has a large deviation, which indicates that the added sensor signal in the kalman filter can increase the estimation accuracy of the Ps3 in some cases. The basic idea of the invention is to provide a method to fuse the two methods, namely the signal synthesis method and the Kalman estimation method, each of which takes the length and increases the accuracy of the reconstructed signal, thereby constructing a more accurate fault-tolerant system, wherein an important link is how to fuse the two signals.
Preferably, the first type of sensor signal reconstruction system has a first priority and the second type of sensor signal reconstruction system has a second priority. Based on a reliability evaluation mechanism, the second priority is an alternative fault-tolerant mode relative to the first priority, which means that the output value of the first-class sensor signal reconstruction system is a priority selection scheme under the condition that signal deviations are the same.
Preferably, referring to fig. 3, the step of evaluating the credibility evaluation mechanism comprises:
step S21, calculating the residual error of the output value of the second type sensor signal reconstruction system;
step S22, acquiring a first reconstruction signal and a second reconstruction signal, wherein the first reconstruction signal is a sensor signal of a first sensor signal reconstruction system, and the second reconstruction signal is a sensor signal of a second sensor signal reconstruction system;
step S23, obtaining an absolute value of a difference between the first reconstructed signal and the second reconstructed signal;
step S24, dividing the absolute value of the difference value and the residual error to obtain a deviation value, and sorting the results of the deviation value according to the size;
and step S25, selecting the range of the deviation value in sections according to the sorting result, and selecting and outputting the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system or selecting the weight value of the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system according to the range.
Preferably, in step S21, the residual error S of the output value of the second type sensor signal reconstruction system is:
wherein y isiIs a sensor measurement of the engine and,
the values calculated for the Kalman estimator state space model, n is the number of sensor measurements.
Preferably, in step S25, at least three values T1, T2 and T3 are selected within the range of the deviation value, wherein 0< T1< T2< T3;
when the deviation value is less than T1 or not less than T3, selecting to output the output value of the second type sensor signal reconstruction system;
when the deviation value is not less than T1 and less than T2, selecting to output an output value of the first type of sensor signal reconstruction system;
and when the deviation value is not less than T2 and less than T3, selecting the weight value of the output value of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system.
The process of fusing the signal synthesis method and the kalman estimation method will be specifically described below. The present invention uses the output of the Kalman calculation method and the output residual error of the measurement sensor, in this embodiment, the output residual error S is defined as shown in formula 7, where yiFor 5 sensor measurements of the engine, i.e. nEqual to 5, and is equal to,
values calculated for the kalman estimator state space model:
the results of the two Ps3 reconstruction methods under the combined sensor fault and engine degradation conditions are compared to obtain the output residual error of the kalman estimator, and the deviation between the Ps3 signal reconstructed by the kalman estimation method and the true engine Ps3 signal (table 1, column 5) and the deviation between the Ps3 signal reconstructed by the signal synthesis method and the true engine Ps3 signal (table 1, column 6) are obtained simultaneously. The information shown in table 1 can be obtained.
TABLE 1 statistical information of two methods to reconstruct the Ps3 signal
Table 1 is transformed according to the following principle, so that table 2 can be obtained.
(1) Since a model value needs to be used, the Ps3 signal of the control system channel is invalid and only depends on two reconstructed signals, so the difference between the Ps3 signal reconstructed by the kalman estimator and the Ps3 signal synthesized by the signal is obtained by subtracting the 5 th column and the 6 th column in table 1 and is recorded as SVM _ Syn.
(2) And obtaining the absolute value of the difference value between the Ps3 signal reconstructed by the Kalman estimation method and the Ps3 signal reconstructed by the signal synthesis method, and recording the absolute value as | SVM _ Syn |.
(3) The method has the advantages that the absolute value SVM _ Syn and the residual error S are divided, deviation values of two methods under the unit residual error are obtained and recorded as absolute value SVM _ Syn/S, the calculation brings two advantages, the first method can eliminate the influence of the fault amplitude value, and the second method can measure the reliability of the deviation between the Kalman estimator and the signal reconstruction.
(4) Sorting is performed according to the value of SVM _ Syn/S (Table 2, column 6).
TABLE 2 Ps3 signal comparison table reconstructed by two methods
The data in table 2 can be summarized into three parts. The serial numbers 1-3, 18-21 show that the Ps3 signal reconstructed by the Kalman estimation method is superior to or equivalent to the Ps3 signal reconstructed by the signal synthesis method, the corresponding range of the SVM _ Syn/S is defined as 0-0.2 and >5, and the data can be modified correspondingly; the sequence number of the line 4-12 part shows that the Ps3 signal reconstructed by the signal synthesis method is superior to the Ps3 signal reconstructed by the Kalman estimation method, and the corresponding range of | SVM _ Syn |/S is defined to be 0.2-1.0; the serial numbers 13-17 indicate that the Ps3 signals reconstructed by adopting the Kalman estimation method or the signal synthesis method have large errors, but considering that the deviation of the two methods is in the opposite direction, in this case, the Ps3 is obtained by adopting a weighted average method, so that the deviation of the reconstructed signals of the Ps3 can be reduced, and the corresponding range of | SVM _ Syn |/S is defined to be 1.0-5. From the above results, the reliability evaluation mechanism of the signal synthesis method and the kalman estimation method shown in fig. 8, or referred to as a fusion strategy, can be summarized, and the fusion structure is shown in fig. 9. As can be seen in fig. 8:
it will be readily appreciated that at least three values T1, T2, and T3 are selected within the range of deviation values herein. In practice, four or more values may be selected. And selecting the weight values of the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system according to the segmentation range formed by the selected values.
When the | SVM _ Syn |/S is smaller than T1, the index is smaller, the difference between the signal synthesis method and the Kalman filtering method is small, and a more accurate Kalman estimator method, namely a fault-tolerant method can be used for selecting the second type sensor signal reconstruction system output value (SVM). It is to be understood that here the output values of the sensor signal reconstruction systems of the first type are weighted 0, whereas the output values of the sensor signal reconstruction systems of the second type are weighted 1.0.
When the absolute value SVM _ Syn/S is in a range from T1 to T2, the difference between the signal synthesis method and the Kalman filtering method is in a residual error range, at the moment, the more accurate of the two signals cannot be judged, and the error possibility of the Kalman estimator is higher, so that a signal synthesis method with better robustness is selected, namely, the fault tolerance method selects a first type sensor signal reconstruction system output value (Syn) with higher priority level. It is to be understood that here the output values of the first type of sensor signal reconstruction system are weighted by 1.0, whereas the output values of the second type of sensor signal reconstruction system are weighted by 0.
When the absolute value SVM _ Syn/S is in a range from T2 to T3, the difference between a signal synthesis method and a Kalman estimation method is shown to be in a range of a plurality of residual errors, at the moment, the more accurate of the two signals still cannot be judged, but the larger deviation of the two signals can be judged, at the moment, the probability that both a Kalman estimator result and the signal synthesis method have faults is very high, the main reason is that a certain common input signal used by the two signals has a problem, and at the moment, the mean value (SVM + Syn) of the output value of a first type sensor signal reconstruction system and the output value of a second type sensor signal reconstruction system is selected. It is also possible as an alternative to select a signal synthesis method with a higher priority level at this stage, or it is also possible to select the first type of sensor signal reconstruction system output value (Syn). It is to be understood that here the output values of the first type of sensor signal reconstruction system are weighted 0.5, while the output values of the second type of sensor signal reconstruction system are weighted 0.5. Instead of taking the mean of the two, non-mean schemes may be used, such as 0.6 and 0.4, as long as the weights of the two are between 0 and 1.0.
When the absolute value SVM _ Syn/S is larger than T3, the S is small, the Kalman estimator method can well track the output of the sensor, and at the moment, a more accurate Kalman estimator method, namely a fault-tolerant method can be used for selecting the second type sensor signal reconstruction system output value (SVM). The output values of the sensor signal reconstruction systems of the first type are here weighted 0, whereas the output values of the sensor signal reconstruction systems of the second type are weighted 1.0.
FIG. 10 illustrates a schematic structural diagram of an aircraft engine sensor fault tolerance system in an embodiment of the present invention. As shown, the present invention also provides an aircraft engine sensor fault tolerance system 1000. The fault-tolerant system is used for executing the fault-tolerant method for the aero-engine sensor. The fault-tolerant system comprises a numerical control channel A and a numerical control channel B which are completely the same, and data exchange is carried out between the numerical control channel A and the numerical control channel B through a data bus CCDL. One of the numerical control a channel and the numerical control B channel is specifically illustrated in fig. 10. For example, taking numerical control a channel as an example, it includes:
the signal detection validation module 1001 determines that the sensor signal is valid.
Each signal reconstruction module 1002 comprises a first type sensor signal reconstruction module 1003, a second type sensor signal reconstruction module 1004 and a calculation module 1005, and the calculation module 1005 is connected to the first type sensor signal reconstruction module 1003 and the second type sensor signal reconstruction module 1004 respectively.
And the control law module 1006 controls and outputs the analog quantity signal and the switching value signal according to the reliability evaluation mechanism so as to control the state of the engine.
Preferably, the nc a channel and the nc B channel further include a signal status information module 1007, respectively, and the signal status information module 1007 converts the analog signal of the signal detection confirmation module 1001 into a digital signal after receiving the analog signal, and sends the digital signal to the first type sensor signal reconstruction module 1003 and the second type sensor signal reconstruction module 1004.
Preferably, the nc a channel and the nc B channel further include a signal reconfiguration switch 1008 respectively, which is connected between the calculation module 1005 and the control law module 1006, and is configured to selectively transmit the result of the calculation module 1005 to the control law module 1006.
The invention provides a fault tolerance method for an aircraft engine sensor, which is characterized in that a first type sensor signal reconstruction system and a second type sensor signal reconstruction system are arranged, and different priorities are set. The first type of sensor signal reconstruction system adopts a fault-tolerant method (such as a signal synthesis method, an engine nonlinear model, an engine component level model or other modeling methods, even a channel sensor value) with better robustness, the system has a first priority, and the second type of sensor signal reconstruction system adopts a Kalman estimation method, so that the system has better reconstruction performance and a second priority. Establishing a reliability evaluation mechanism aiming at the first type of sensor signal reconstruction system and the second type of sensor signal reconstruction system, determining the output of the reconstruction signal according to the reliability evaluation mechanism, and outputting an output value of the selectable first type of sensor signal reconstruction system, an output value of the second type of sensor signal reconstruction system or selecting the weight values of the output values of the two types of sensor signal reconstruction systems. The output value is used as a dual-channel voter of the fault tolerance system of the aircraft engine sensor and as a backup model control signal. The evaluation method of the credibility evaluation mechanism is determined based on the residual errors of the output values of the first type sensor signal reconstruction system, the output values of the second type sensor signal reconstruction system and the output values of the second type sensor signal reconstruction system.
The fault tolerance method and the fault tolerance system for the sensor of the aircraft engine provided by the invention have the beneficial effects that:
(1) on the premise of keeping the original control fault-tolerant robustness, the fault-tolerant performance is improved. The sensor fault reconstruction system of the signal synthesis method is robust and safe as shown by the operation service experience of the existing aircraft engine, the robustness of the engine control fault tolerance can be ensured as long as the fault-tolerant signal is ensured within the range, the sensor fault reconstruction system established by the Kalman estimation method with constraint limitation is introduced, the more excellent fault-tolerant performance of the sensor is selected within the reasonable credibility range, the influence of factors such as individual difference, performance degradation and gas path component fault on the signal synthesis method is solved, the better transient fault-tolerant capability is realized, and the aim of still keeping the robustness and the safety under the condition of improving the reconstruction performance is fulfilled.
(2) By introducing a reliability evaluation mechanism, the signal synthesis method is always used as a preferred fault-tolerant mode for the fault-tolerant control of the engine, the special safety requirement of an aircraft engine control system is not damaged, and the original control logic and control structure of the control system are not greatly influenced. Before the invention is used, only the sensor fault reconstruction system of the signal synthesis method outputs the signals for double-channel signal voting or serves as a backup control signal.
(3) The reliability evaluation mechanism defined by the invention can be applied to all the first type sensor signal reconstruction systems and the second type sensor signal reconstruction systems established by the self-adaptive method. Therefore, the system framework can be suitable for fault-tolerant evaluation of all the first-class sensor signal reconstruction systems and the second-class sensor signal reconstruction systems established by the self-adaptive method, and the use of a new fault-tolerant technology in the field of slightly conservative aircraft engines is promoted.
Although the present invention has been described with reference to the present specific embodiments, it will be appreciated by those skilled in the art that the above embodiments are merely illustrative of the present invention, and various equivalent changes and substitutions may be made without departing from the spirit of the invention, and therefore, it is intended that all changes and modifications to the above embodiments within the spirit and scope of the present invention be covered by the appended claims.
Claims (10)
1. An aircraft engine sensor fault tolerance method, the fault tolerance method comprising:
step S1, establishing a first type sensor signal reconstruction system and a second type sensor signal reconstruction system;
step S2, establishing a reliability evaluation mechanism for the first type sensor signal reconstruction system and the second type sensor signal reconstruction system, where the reliability evaluation mechanism selects to output the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system, and the residual error of the output value of the second type sensor signal reconstruction system, or selects a weighted value of the output values of the first type sensor signal reconstruction system, the second type sensor signal reconstruction system, or both.
2. The fault tolerant method according to claim 1, characterized in that the reconstruction system of the first type of sensor signal employs a signal synthesis method and the reconstruction system of the second type of sensor signal employs a kalman estimation method.
3. The fault-tolerant method of claim 2, wherein the first type of sensor signal reconstruction system has a first priority and the second type of sensor signal reconstruction system has a second priority, the second priority being an alternative fault-tolerant manner with respect to the first priority based on the confidence-assessment mechanism.
4. A fault tolerant method as claimed in claim 3 wherein the signal synthesis method is based on sensor signal data, an engine non-linear model, an engine component level model or channel sensor values.
5. The fault-tolerant method of claim 1, wherein the step of evaluating the trustworthiness evaluation mechanism comprises:
step S21, calculating the residual error of the output value of the second type sensor signal reconstruction system;
step S22, acquiring a first reconstruction signal and a second reconstruction signal, where the first reconstruction signal is a sensor signal of the first type of sensor signal reconstruction system, and the second reconstruction signal is a sensor signal of the second type of sensor signal reconstruction system;
step S23, obtaining an absolute value of a difference between the first reconstructed signal and the second reconstructed signal;
step S24, dividing the absolute value of the difference value and the residual error to obtain a deviation value, and sorting the results of the deviation value according to the size;
and step S25, selecting the range of the deviation value in sections according to the sorting result, and selecting and outputting the output value of the first type sensor signal reconstruction system, the output value of the second type sensor signal reconstruction system or selecting the weight value of the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system according to the range.
6. The fault tolerant method according to claim 5 characterized in that in step S21, the residual S of the output values of the second type of sensor signal reconstruction system is:
wherein y isiIs a sensor measurement of the engine and,
the values calculated for the Kalman estimator state space model, n is the number of sensor measurements.
7. The fault tolerant method of claim 5 wherein in step S25, at least three values T1, T2 and T3 are selected within the range of deviation values, wherein 0< T1< T2< T3;
when the deviation value is smaller than T1 or not smaller than T3, selecting a weight value for outputting the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system;
when the deviation value is not less than T1 and less than T2, selecting a weight value for outputting the output values of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system;
and when the deviation value is not less than T2 and less than T3, selecting the weight value of the output value of the first type sensor signal reconstruction system and the second type sensor signal reconstruction system.
8. An aircraft engine sensor fault tolerance system for performing the aircraft engine sensor fault tolerance method of any one of claims 1 to 7, the fault tolerance system comprising an identical numerical control channel a and a numerical control channel B for data exchange therebetween via a data bus CCDL, the numerical control channel a and the numerical control channel B each comprising:
a signal detection confirmation module to determine that the sensor signal is valid;
each signal reconstruction module comprises a first type sensor signal reconstruction module, a second type sensor signal reconstruction module and a calculation module, and the calculation module is respectively connected with the first type sensor signal reconstruction module and the second type sensor signal reconstruction module;
and the control law module is used for controlling and outputting the analog quantity signal and the switching value signal according to the reliability evaluation mechanism so as to control the state of the engine.
9. The fault tolerant system of claim 8 wherein said nc a channel and said nc B channel further comprise a signal status information module, respectively, said signal status information module receiving said analog signal from said signal detection confirmation module, converting said analog signal into a digital signal, and sending said digital signal to said first type sensor signal reconstruction module and said second type sensor signal reconstruction module.
10. The fault tolerant system of claim 8 wherein said nc a channel and said nc B channel each further comprise a signal reconfiguration switch, said signal reconfiguration switch being in communication between said calculation module and said control law module for selectively communicating the results of said calculation module to said control law module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010986491.7A CN114282570A (en) | 2020-09-18 | 2020-09-18 | Fault tolerance method and fault tolerance system for sensor of aircraft engine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010986491.7A CN114282570A (en) | 2020-09-18 | 2020-09-18 | Fault tolerance method and fault tolerance system for sensor of aircraft engine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114282570A true CN114282570A (en) | 2022-04-05 |
Family
ID=80867402
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010986491.7A Pending CN114282570A (en) | 2020-09-18 | 2020-09-18 | Fault tolerance method and fault tolerance system for sensor of aircraft engine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114282570A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113934153A (en) * | 2020-06-29 | 2022-01-14 | 中国航发商用航空发动机有限责任公司 | Multi-channel simulation method and system for aircraft engine control system |
| CN115576294A (en) * | 2022-09-08 | 2023-01-06 | 大连理工大学 | Fault-tolerant soft-hard hybrid control method for fault diagnosis of aircraft engine sensor |
| CN118170124A (en) * | 2024-05-14 | 2024-06-11 | 中车工业研究院(青岛)有限公司 | Fault analysis method, device, equipment and storage medium of instrument control system |
-
2020
- 2020-09-18 CN CN202010986491.7A patent/CN114282570A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113934153A (en) * | 2020-06-29 | 2022-01-14 | 中国航发商用航空发动机有限责任公司 | Multi-channel simulation method and system for aircraft engine control system |
| CN113934153B (en) * | 2020-06-29 | 2024-09-03 | 中国航发商用航空发动机有限责任公司 | Multichannel simulation method and system for aero-engine control system |
| CN115576294A (en) * | 2022-09-08 | 2023-01-06 | 大连理工大学 | Fault-tolerant soft-hard hybrid control method for fault diagnosis of aircraft engine sensor |
| CN118170124A (en) * | 2024-05-14 | 2024-06-11 | 中车工业研究院(青岛)有限公司 | Fault analysis method, device, equipment and storage medium of instrument control system |
| CN118170124B (en) * | 2024-05-14 | 2024-07-23 | 中车工业研究院(青岛)有限公司 | Fault analysis method, device, equipment and storage medium of instrument control system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114282570A (en) | Fault tolerance method and fault tolerance system for sensor of aircraft engine | |
| Kobayashi et al. | Application of a bank of Kalman filters for aircraft engine fault diagnostics | |
| CN110118128B (en) | Fault diagnosis and fault-tolerant control method for micro gas turbine sensor | |
| CA2354944C (en) | Sensor fault detection, isolation and accomodation | |
| Kobayashi et al. | Evaluation of an enhanced bank of Kalman filters for in-flight aircraft engine sensor fault diagnostics | |
| Jiang | Sensor Fault Detection and Isolation Using System Dynamics Identification Techniques. | |
| CN111474919B (en) | Aeroengine control system sensor fault diagnosis method based on AANN network group | |
| WO2017198528A1 (en) | Controlling a gas turbine considering a sensor failure | |
| US20020066054A1 (en) | Fault detection in a physical system | |
| Xin et al. | Fault diagnosis based on measurement reconstruction of HPT exit pressure for turbofan engine | |
| CN111506049B (en) | Multiple fault diagnosis method for aero-engine control system based on AANN network system | |
| Zedda et al. | Fault diagnosis of a turbofan engine using neural networks-A quantitative approach | |
| CN104428727B (en) | The method that the deterioration of the turbine is detected for by the performance of monitoring turbine | |
| Yang et al. | Multiple model-based detection and estimation scheme for gas turbine sensor and gas path fault simultaneous diagnosis | |
| Kobayashi et al. | Hybrid Kalman filter approach for aircraft engine in-flight diagnostics: Sensor fault detection case | |
| CN104698833B (en) | Redundancy control method and system | |
| Cruz-Manzo et al. | GAS turbine sensor fault diagnostic system in a real-time executable digital-twin | |
| Goebel et al. | Diagnostic information fusion: requirements flowdown and interface issues | |
| Li et al. | Multiple fault diagnosis of aeroengine control system based on autoassociative neural network | |
| Panov et al. | Model-based compensation of sensor failure in industrial gas turbine | |
| CN112749789A (en) | Aero-engine multiple fault diagnosis device based on self-association neural network | |
| Xue et al. | Application of Kalman filters for the fault diagnoses of aircraft engine | |
| Lu et al. | Fault diagnosis for sensors and components of aero-engine based on grey analysis | |
| Ren et al. | Multiple sensor validation for natural gas combined cycle power plants based on robust input training neural networks | |
| Gabor et al. | Redundancy method used in PLC related applications |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |