Robust control method for aero-engine starting control plan

Abstract

A robust control method of an aero-engine starting control plan, comprising: obtaining an acceleration fuel flow target at a current time j according to a control procedure of a first redundancy fuel control plan and a fusion output of the acceleration fuel flow at a previous time j-1 during an engine starting process of an aero-engine; obtaining a fusion coefficient at the current time by weighted calculation, and then obtaining a weight value according to a fuel deviation of the fuel control plan and an acceleration deviation of a second conversion acceleration control plan; inputting into a proportional-integral control to output the fusion output of the acceleration fuel flow at the current time j, and obtaining a steady-state fuel flow target at the current time j according to a control procedure of a third steady-state fuel control plan, and taking a minimum value of the two to obtain a final fuel flow target of the starting control plan.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aero-engine control, in particular to a robust control method of aero-engine starting control plan. BACKGROUND

[0002] Surge is an abnormal working state that must be avoided in the normal operation of an aero-engine, but in the starting process, it is often necessary to approach the surge boundary of the compressor in order to complete the starting faster. This contradiction makes the design of the starting control plan particularly strict. In order to prevent the starting surge caused by control interference factors such as starting load and starter power recession, the starting control plan designed often reserves a certain surge margin between the starting control plan and the surge boundary to ensure the safety of the starting process. In order to further improve the transient performance of advanced engines, reducing the design surge margin becomes a key means, which makes it necessary to strengthen the robustness of the starting control plan to sensor, actuator and power extraction and other control interference factors. A single starting control plan cannot avoid the starting surge caused by single or combined interference factors, and the mixed use of multiple plans becomes the main measure to improve the robustness of the starting control.

[0003] The currently used robust control method of multiple starting control plans is either too simple, with weak robustness or poor performance retention, or can only tolerate single or a few control interference factors, and cannot adapt to the complex control interference conditions in the real engine starting process.

[0004] Therefore, in view of the simplicity and universality of the robust control method of multiple starting control plans, a robust control method of aero-engine starting control plan with strong application ability is needed. SUMMARY

[0005] To solve the above problems, the purpose of the present application is to provide a robust control method of aero-engine starting control plan, which can adapt to the starting control plan of different engines and multiple control interference factors, avoid starting surge caused by single or combined interference factors, and maintain certain acceleration performance.

[0006] To achieve the above purpose, the present application adopts the following technical solutions.

[0007] A robust control method of aero-engine starting control plan, comprising the following steps:

[0008] S100: In the engine starting process of an aero-engine, the acceleration fuel flow target (W f,cmd ) obtained at the current time j and the acceleration fuel flow fusion output (W f,acc ) obtained at the last time j-1 are fused to obtain the acceleration fuel flow target (W j ) at the current time j. j-1, get the fuel deviation ; and process the fuel deviation to obtain the normalized fuel deviation ;

[0009] S200: Based on normalized fuel deviation The penalty coefficient is calculated with the asymmetric Gaussian distribution function , according to the high pressure shaft conversion speed at the current time j The switching coefficient is calculated with the flexible switching function , and according to and Calculate the fusion coefficient ;

[0010] S300: Calculate the acceleration target value of the high-pressure shaft according to the current time j , and the actual engine high-pressure shaft conversion acceleration target value obtained at the previous moment j-1 , get the converted acceleration deviation , according to the acceleration deviation , fuel deviation and fusion coefficient Calculate the weighted sum of the deviation of the converted rotor acceleration and the fuel flow deviation , output the acceleration fuel flow fusion output (W f,acc ) j ;

[0011] S400: Based on the steady-state fuel flow target (W f,steady ) j , and the acceleration fuel flow fusion output at the current moment j (W f,acc ) j , and obtain the final fuel flow target (W f,final ) j .

[0012] Optionally, the normalization formula for the fuel flow deviation of the redundancy fuel control plan in step S100 is:

[0013] .

[0014] Optionally, in step S100, the acceleration fuel flow target (W f,cmd ) j The calculation formula is:

[0015] ,

[0016] in, 、 、 and The fuel control plan corresponding to the different pressure sensors calculates the result at the current time j.

[0017] Optionally, in step S200, the penalty coefficient , the switching coefficient and the fusion coefficient The calculation formula is:

[0018]

[0019]

[0020] ;

[0021] Wherein, and are different scaling coefficients of the asymmetric Gaussian distribution function on the positive and negative sides, both greater than 0 and not more than 0.1, ; and are the scaling coefficient and the translation coefficient of the switching function respectively, greater than 0 and not more than 1.

[0022] Optionally, in step S300, the deviation weighted sum The calculation formula is:

[0023] ,

[0024] Wherein, adjusts the size of the scaling coefficient, adjusts the size of the scaling coefficient, both greater than 0 and not more than 10.

[0025] Optionally, in step S400, the final fuel flow target (W f,final ) j The calculation formula is: .

[0026] Compared with the prior art, the present application has the following advantages:

[0027] The present invention has few parameters for optimization settings, is simple to use and easy to implement algorithms. Except for a very small number of parameters that require prior knowledge to determine the initial values, the settings of other parameters are independent of the specific engine characteristics. It can be generally applied to the weighted fusion needs of two acceleration control plans without paying attention to the disturbance characteristics of the specific plan. It has the robust control capability of single fault tolerance or combined fault tolerance for at least six control interference factors such as fuel metering error, starter power degradation, compressor rear pressure sensor error, starting load increase, high-pressure compressor efficiency degradation and high-pressure turbine efficiency degradation. The starting process controlled by the disclosed method can not only avoid surge, but also maintain a certain acceleration performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a control schematic diagram of a robust control method in an aircraft engine starting controller provided by one embodiment of the present disclosure;

[0029] Figure 2 This is a flowchart of a robust control method for two starting control plans provided by an embodiment of the present disclosure;

[0030] Figure 3 is an asymmetric Gaussian distribution function image provided by one embodiment of the present disclosure;

[0031] Figure 4 is a flexible switching function image provided by an embodiment of the present disclosure;

[0032] Figure 5 This is a line chart showing the statistics of the start-up time of different control methods during the start-up process under the influence of combined interference factors of the sea level standard atmosphere provided by an embodiment of the present disclosure;

[0033] Figure 6 This is a line chart showing the statistics of minimum surge margins during startup using different control methods under the influence of combined interference factors in a sea level standard atmosphere, provided by an embodiment of the present disclosure;

[0034] Figure 7 It is a line chart showing the statistics of minimum surge margin after starter disengagement during the starting process under the influence of combined interference factors of the sea level standard atmosphere provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] The following will refer to the attached Figures 1 to 7 Specific embodiments of the present invention will now be described in detail. Although specific embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention may be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to facilitate a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0036] It should be noted that certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that technicians may use different nouns to refer to the same component. This specification and claims do not use the difference in nouns as a way to distinguish components, but use the difference in the functions of the components as the criterion for distinction. As mentioned throughout the specification and claims, "including" or "comprising" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment of the present invention, but the description is based on the general principles of the specification and is not intended to limit the scope of the invention. The scope of protection of the present invention shall be as defined in the attached claims.

[0037] To facilitate understanding of the embodiments of the present invention, further explanation will be given below using specific embodiments as examples in conjunction with the accompanying drawings, and the accompanying drawings do not constitute a limitation on the embodiments of the present invention.

[0038] In one embodiment, the present invention provides a robust control method for an aircraft engine start control plan, comprising the following steps:

[0039] S100: During the engine start-up process of the aircraft engine, according to the acceleration fuel flow target (W f,cmd ) j , and the accelerated fuel flow fusion output (W f,acc ) j-1 , get the fuel deviation ; and process the fuel deviation to obtain the normalized fuel deviation ;

[0040] S200: Based on normalized fuel deviation The penalty coefficient is calculated with the asymmetric Gaussian distribution function , according to the high pressure shaft conversion speed at the current time j The switching coefficient is calculated with the flexible switching function , and according to and Calculate the fusion coefficient ;

[0041] S300: Calculate the acceleration target value of the high-pressure shaft according to the current time j , and the actual engine high-pressure shaft conversion acceleration target value obtained at the previous moment j-1 , get the converted acceleration deviation , according to the acceleration deviation , fuel deviation and fusion coefficient Calculate the weighted sum of the deviation of the converted rotor acceleration and the fuel flow deviation , output the acceleration fuel flow fusion output (W f,acc ) j ;

[0042] S400: Based on the steady-state fuel flow target (W f,steady ) j , and the acceleration fuel flow fusion output at the current moment j (W f,acc ) j , and obtain the final fuel flow target (W f,final ) j .

[0043] Optionally, the normalization formula for the fuel flow deviation of the redundancy fuel control plan in step S100 is:

[0044] .

[0045] Optionally, in step S100, the acceleration fuel flow target (W f,cmd ) j The calculation formula is:

[0046] ,

[0047] in, 、 、 and Calculation results of the fuel control plan corresponding to different pressure sensors at the current time j.

[0048] Optionally, in step S200, the penalty coefficient , switching coefficient and fusion coefficient The calculation formula is:

[0049]

[0050]

[0051] ;

[0052] in, and is an asymmetric Gaussian distribution function with different scaling coefficients on the positive and negative sides, both greater than 0 and No more than 0.1, ; and They are the scaling coefficient and the translation coefficient of the switching function, Greater than 0 and No more than 1.

[0053] Optionally, in step S300, the deviation weighted sum The calculation formula is:

[0054] ,

[0055] in, Adjustment The scaling factor of the size, Adjustment The scaling factor for size, both greater than 0 and not exceeding 10.

[0056] Optionally, the final fuel flow target (W f,final ) j The calculation formula is: .

[0057] In one embodiment, Figure 2 As shown, the present invention provides a robust control method for an aircraft engine starting control plan, comprising the following steps:

[0058] S100: During the engine start-up process of the aircraft engine, the acceleration fuel flow target (W f,cmd ) j , which is achieved by replacing the original single conversion fuel flow control plan with P2 conversion fuel flow control plan, P 25 Converted fuel flow control plan and P 31 The fuel flow control plan is converted into three different fuel control plans, thereby obtaining three converted fuel flow results W f,p2 、W f,p25 and W f,p31 , and take the middle value of these three results to get (W f,cmd ) j Then, the accelerated fuel flow rate obtained at the previous moment j-1 is fused and output (W f,acc ) j-1 , subtract the front and back to get the fuel deviation , ; and divide the fuel deviation by the acceleration fuel flow target (W f,cmd ) j Get the normalized fuel deviation , ;

[0059] S200: Based on normalized fuel deviation With asymmetric Gaussian distribution function Calculate the penalty coefficient , according to the high pressure shaft conversion speed at the current time j With flexible switching function Calculate the switching coefficient , and according to and Calculate the fusion coefficient ;

[0060]

[0061]

[0062] in and is an asymmetric Gaussian distribution function with different scaling coefficients on the positive and negative sides, both greater than 0 and No more than 0.1, ; and They are the scaling coefficient and the translation coefficient of the switching function, Greater than 0 and No more than 1.

[0063] S300: The high-pressure shaft converted acceleration target value obtained at the current time j by the control program of the second commonly used conversion acceleration control plan in engineering , and the actual engine high pressure shaft conversion acceleration target value at the previous moment j-1 , subtract the front and back to get the converted acceleration deviation , ,according to 、 and Calculate the weighted sum of the deviation of the converted rotor acceleration and the fuel flow deviation , and Input to the proportional-integral control to output the acceleration fuel flow fusion output (W f,acc ) j ;

[0064] S400: The steady-state fuel flow rate target (W) obtained at the current time j according to the control program of the third steady-state fuel control plan commonly used in engineering f,steady ) j , and the acceleration fuel flow fusion output at the current moment j (W f,acc ) j The minimum value of the two is used to obtain the final fuel flow target of the starting control plan (W f,final ) j , .

[0065] ReferenceFigure 1 , a control diagram of a robust control method in an aero-engine starting controller, first, based on the parameters P2, P 25 and P 31 , and using the control program of the redundant fuel control plan to obtain the acceleration fuel flow target (W f,cmd ) j at the current time j, which is obtained by replacing the original single conversion fuel flow control plan with three different fuel control plans, i.e., the P2 conversion fuel flow control plan, the P 25 conversion fuel flow control plan and the P 31 conversion fuel flow control plan, so as to obtain three conversion fuel flow results W f,p2 , W f,p25 and W f,p31 , and taking the middle value of the three results to obtain (W f,cmd ) j , then subtracting the acceleration fuel flow fusion output (W f,acc ) j-1 obtained at the previous time j-1 to obtain the fuel deviation ; and dividing the fuel deviation by the acceleration fuel flow target (W f,cmd ) j to obtain the normalized fuel deviation ; thereafter, according to the normalized fuel deviation and the aforementioned asymmetric Gaussian distribution function, the penalty coefficient is calculated, according to the sensor measurement high-pressure shaft conversion speed at the current time j and the aforementioned flexible switching function, the switching coefficient is calculated, and according to and , the fusion coefficient is calculated; then, according to the high-pressure shaft conversion acceleration target value obtained at the current time j by the control program of the second commonly used conversion acceleration control plan in engineering, and the actual engine high-pressure shaft conversion acceleration target value at the previous time j-1, the conversion acceleration deviation is obtained by subtracting the former from the latter, according to , and , the weighted sum of the conversion rotor acceleration deviation and the fuel flow deviation is calculated, and is input into the Proportional-integral control to output the acceleration fuel flow fusion output (W f,acc ) jFinally, the steady-state fuel flow target (W f,steady ) j obtained by the control program of the third commonly used steady-state fuel control plan at the current time j, and the acceleration fuel flow fusion output (W f,acc ) j at the current time j, and the minimum value of the two is the final fuel flow target (W f,final ) j .

[0066] The above embodiment, wherein the number of the redundancy conversion fuel control plans can be any number greater than or equal to two, so as to enhance the ability of the starting control to cope with single sensor noise or even failure. The starting process controlled by the method of the present disclosure can avoid the significant decrease of surge margin caused by single control interference factors or combined control interference factors, avoid the risk of surge, and can also avoid the significant decrease of acceleration performance.

[0067] In another embodiment, the normalization method of the fuel flow deviation of the redundancy fuel control plan in step S100 is:

[0068]

[0069] The acceleration fuel flow target (W f,cmd ) j The calculation method is:

[0070]

[0071] Wherein , , and are the calculation results of the fuel control plans corresponding to different pressure sensors at the current time j.

[0072] In another embodiment, the image of the asymmetric Gaussian distribution in step S200 is as shown in Figure 3 , the image of the flexible switching function is as shown in Figure 4 , the penalty coefficient , the switching coefficient and the fusion coefficient The calculation method is:

[0073]

[0074]

[0075]

[0076] Wherein and are different scaling factors of the asymmetric Gaussian distribution function, both greater than 0 and not more than 10 not more than 0.1, ; and are scaling factor and translation factor of the switching function, respectively, greater than 0 and not more than 1.

[0077] In another embodiment, the deviation weighted sum in step S300 is The calculation method is:

[0078]

[0079] wherein, adjusts the scaling factor of the size, adjusts the scaling factor of the size, both greater than 0 and not more than 10.

[0080] In another embodiment, the final fuel flow target (W f,final ) j The calculation method is:

[0081] .

[0082] The control interference factors considered in this example include fuel metering error, starter power degradation, high pressure compressor efficiency degradation, high pressure turbine efficiency degradation, compressor after pressure sensor error, and normal range and abnormal range of starter load increase.

[0083] In one embodiment, a simulation is performed for a specific engine component level model to verify the effectiveness of the proposed robust starting control method. The method is compared with two conventional starting control plans, fuel flow plan converted from pt31 (high pressure compressor outlet total pressure) and N-dot control plan, to prove that the method can prevent compressor surge during the starting process better than the two conventional plans (fuel flow plan converted from pt31 and N-dot plan) under the influence of a single interference factor or combined interference factors, to verify the characteristics of the method described in the disclosure, the specific process is as follows:

[0084] Table 1

[0085]

[0086] wherein, HPC is high pressure compressor, and HPT is high pressure turbine.

[0087] First, according to the provisions of the control disturbance size described in Table 1, it is assumed that under the sea level standard atmospheric design conditions, for a specific engine component level model, no control disturbance factor is injected and a single control disturbance factor is injected respectively, the conventional fuel flow plan (CFCS) and N-dot plan (NDCS) converted by pt31, and the robust control method (IPCS) described in the present disclosure are respectively tested in the start-up process, and the excess degree of fuel control plan in IPCS refers to three kinds of fuel flow plans converted by fan inlet total pressure pt2, fan outlet total pressure pt25 and pt31, The start-up process test here refers to the acceleration process from engine static to engine speed reaching the slow speed minus 1.5% of the maximum speed.

[0088] The minimum surge margin Minimum SM and start-up time in each start-up process of each method are counted and compared, as shown in Table 2.

[0089] Table 2 shows that the IPCS method proposed in this paper in each simulation test, the start-up time and the minimum surge margin, two key parameters, are not always the best, but are within an acceptable range, and the performance is more balanced, which can take into account the start-up time and the minimum surge margin, while ensuring the safety and rapidity of the start-up process, so it can better meet the actual engineering needs.

[0090] Table 2

[0091]

[0092] Further, the present disclosure also simulates and compares the three control methods through more comprehensive combination control disturbance random simulation, and the specific process is as follows:

[0093] Table 3

[0094]

[0095] According to the provisions of the control disturbance size described in Table 3, 1000 kinds of engine combination disturbance working conditions are randomly selected, and each disturbance factor in the 1000 working conditions is randomly selected from the distribution range with uniform distribution. It is assumed that under the sea level standard atmospheric design conditions, 1000 kinds of combination disturbance working conditions are injected into the engine component level model respectively, and the start-up process is repeated, and the start-up time (start-up time) and Minimum SM in each acceleration process are counted and compared, as shown in Table 4. Figures 5 to 6 and Table 4.

[0096] Table 4 shows that among the three methods, in terms of start-up time, whether in terms of mean, worst-case, or variance, CFCS > IPCS > NDCS. IPCS and NDCS are similar, but significantly smaller than CFCS. This indicates that NDCS has the best start-up time, followed by IPCS, with CFCS having the worst start-up time. Regarding minimum surge margin, CFCS > IPCS > NDCS in terms of mean and worst-case values. IPCS and CFCS are similar, but significantly larger than NDCS. However, in terms of variance, IPCS is significantly smaller than both CFCS and NDCS. This indicates that IPCS and CFCS have similar minimum surge margins, both being optimal, while NDCS has the worst minimum surge margin. Therefore, considering both start-up time and minimum surge margin, IPCS is the optimal choice among the three methods.

[0097] Figure 5 The IPCS method has a shorter average startup time and a narrower distribution, indicating that the IPCS method's acceleration performance is more robust to these disturbances. The CFCS method has a significantly longer average startup time, and it can be seen that the CFCS method's startup time is significantly affected by the disturbances, with the longest startup time reaching 77.1 seconds.

[0098] Figure 6 The results show that the average minimum surge margin of the IPCS method is 4.9% higher than that of the NDCS method. Under all combined operating conditions, the minimum surge margin of the NDCS method is smaller than that of the IPCS method. Furthermore, in 299 test groups, the acceleration line during the startup process exceeded the compressor stall line, indicating that the engine had stalled or even surged under these test conditions.

[0099] In addition, in order to further compare the robustness of the IPCS method and the NDCS method, the minimum surge margin distribution diagrams of the proposed IPCS method and the NDCS method after the starter is disengaged are shown in Figure 2. Figure 7 shown. Figure 7 Overall, the IPCS method demonstrates a higher minimum surge margin, averaging 0.9% higher than the NDCS method, with a maximum difference of 2.3%. Overall, only the robust control method proposed in this disclosure can maintain control within the surge margin under the influence of multiple combined disturbances while also avoiding a significant degradation in acceleration performance. This demonstrates that this method is not only simple and easy to use, but also offers significant benefits.

[0100] Table 4

[0101]

[0102] Although the embodiments of the present application have been described above with reference to the accompanying drawings, the present application is not limited to the above-described specific embodiments and areas of application, and the above-described specific embodiments are merely illustrative and instructive, but are not restrictive. A person of ordinary skill in the art can make many forms under the guidance of the present specification and without departing from the scope protected by the claims of the present application, and these all belong to the present application.

Claims

1. A robust control method for an aeroengine start control schedule, characterized in that, The method comprises the following steps: S100: In the engine starting process of the aero-engine, the acceleration fuel flow target (W f,cmd ) j , and the acceleration fuel flow fusion output (W f,acc ) j-1 obtained at the last time j-1 are obtained to obtain the fuel deviation ; and the fuel bias is processed to obtain a normalized fuel bias ; S200: the penalty coefficient is calculated according to the normalized fuel deviation The penalty coefficient is calculated according to the asymmetric Gaussian distribution function The switching coefficient is calculated according to the high-pressure shaft conversion speed of the current time j The switching coefficient is calculated according to the flexible switching function The fusion coefficient is calculated according to and The fusion coefficient is calculated according to ; S300: Calculate the acceleration target value of the high-pressure shaft according to the current time j , and the actual engine high-pressure shaft conversion acceleration target value obtained at the previous moment j-1 , get the converted acceleration deviation , according to the acceleration deviation , fuel deviation and fusion coefficient Calculate the weighted sum of the deviation of the converted rotor acceleration and the fuel flow deviation , output the acceleration fuel flow fusion output (W f,acc ) j ; S400: Obtain the steady-state fuel flow target (W f,steady ) j , and the acceleration fuel flow fusion output (W f,acc ) j at the current time j, to obtain the start control plan final fuel flow target (W f,final ) j .

2. The method of claim 1, wherein, The normalization formula of the fuel flow deviation of the fuel control plan is as follows: 。 3. The method of claim 1, wherein, In step S100, the acceleration fuel flow target (W f,cmd ) j The calculation formula is: ， wherein, is the result of the fuel control plan for the respective pressure sensor at the current time j.

4. The method of claim 1, wherein, In step S200, the penalty coefficient , the switching coefficient and the fusion coefficient are calculated according to the following formulas: ； ； ； wherein, and are different scaling factors for the positive and negative sides of the asymmetric Gaussian distribution function, both greater than 0 and not more than 0.1, ; and are scaling and translation factors for the switching function, respectively, greater than 0 and not more than 1.

5. The method of claim 4, wherein, In step S300, the deviation weighted sum The calculation formula is: ， wherein, adjusting the size of the scaling factor, adjusting the size of the scaling factor, both being greater than 0 and not exceeding 10.

6. The method according to any one of claims 1 to 5, characterized in that, The final fuel flow target (W f,final ) j The calculation formula is: 。

Images