Fault tolerant control method for aero-engine acceleration control program

Abstract

A fault-tolerant control method for an aircraft engine acceleration control plan, wherein during the aircraft engine acceleration process, i A control program that accelerates the control plan at the current moment j The obtained acceleration fuel flow target and the last moment j The acceleration fuel flow fusion output of the 1-1 is weighted to obtain the overall variance at the current moment, and then the overall variance of each acceleration control plan is used to calculate the weight. The statistical variance of the fuel of multiple plans at the current moment is calculated. The new weighted average relative fuel amount is calculated by combining the posterior probability and semi-variance until the relative error of the weighted average relative fuel amount before and after the iteration is less than the set maximum error. Finally, the product of the terminated weighted average relative fuel amount and the acceleration fuel flow fusion output is used as the acceleration fuel flow fusion output at the current moment.

Description

Technical Field

[0001] The present invention relates to the technical field of aero-engine control, and in particular to a fault-tolerant control method for an aero-engine acceleration control plan. Background Art

[0002] Surge is an abnormal operating state that must be avoided for the normal operation of aircraft engines. However, during acceleration, the compressor surge limit must be approached in order to complete acceleration quickly. This contradiction makes the design of the acceleration control plan particularly rigorous. To prevent acceleration surge caused by uncertain factors such as intake distortion and engine degradation, the designed acceleration control plan often reserves a certain surge margin between the surge limit and the surge limit to ensure the safety of the acceleration process. In order to further improve the transient performance of advanced engines, reducing the designed surge margin has become a feasible key means. This makes it necessary to strengthen the acceleration control plan's adaptability to disturbance factors such as sensors, actuators, and power extraction. A single acceleration control plan is unlikely to avoid abnormal operation or failure of core components caused by a single factor failure. The mixed use of multiple plans has become the main response measure to improve fault tolerance.

[0003] The fault-tolerant control methods for the various acceleration control plans currently used are either too simple, with disadvantages such as weak fault tolerance or poor performance retention, or too complex in parameter setting and optimization adjustment, with poor adaptability to different engines. The corresponding calculation methods cannot adapt well to changes in the number of weighted fusion acceleration control plans.

[0004] Therefore, in order to solve the problems of simplicity and versatility of various fault-tolerant control methods for acceleration control plans, a fault-tolerant control method for aircraft engine acceleration control plans with strong application capabilities is needed.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0006] The present invention provides a fault-tolerant control method for an aircraft engine acceleration control plan, which can adapt to acceleration control plans of different engines and different fusion numbers, avoid acceleration surge caused by significant failure of a single disturbance factor and small-scale errors of other disturbance factors, and maintain a certain acceleration performance.

[0007] A fault-tolerant control method for an aircraft engine acceleration control plan includes:

[0008] S100: During the acceleration of the aircraft engine, i A control program that accelerates the control plan at the current moment j The obtained acceleration fuel flow target (W fi ) j , and the half variance of the last time j -1 W f,acc ) j-1 , and the normalized relative fuel quantity is obtained by dividing the front by the back x i,j， ;

[0009] S200: the instantaneous variance of the first acceleration control plan is calculated according to the current time j i d ti,j and the overall variance of the last time j -1 d i,j-1 The overall variance of the current time is calculated by weighting d i,j , and the initial value d i,1 is set according to prior knowledge, and then the weight is calculated according to the overall variance of each acceleration control plan d i,j ω i,j ;

[0010] S300: the weighted average relative fuel quantity μ j The initial value is set to the minimum value of the relative fuel quantity j x i,j at each time, and the statistical variance of the fuel of multiple plans at the current time is calculated σ j ;

[0011] S400: the half variance is obtained according to the size relationship between the relative fuel quantity x i,j and the weighted average relative fuel quantity μ j , and the statistical variance of the fuel of multiple plans at the current time σ j h i,j , the posterior probability is calculated according to the weight ω i,j and the asymmetric Gaussian distribution function γ i,j , and then the new weighted average relative fuel quantity is calculated according to the posterior probability γ i,j and the half variance h i,j μ ​​​​​​j S400 step is repeated until the relative fuel quantity of the weighted average before and after iteration μ j relative error is less than the set maximum error, and finally the relative fuel quantity of the terminated weighted average μ j and the acceleration fuel flow rate fusion output W f,acc ) j-1 product as the acceleration fuel flow rate fusion output at the current time W f,acc ) j .

[0012] The fault-tolerant control method of the turbofan engine acceleration control plan, in step S200, the weight ω i,j The calculation method is:

[0013] ,

[0014] ,

[0015] ,

[0016] Where ρ is a constant set, ρ is not more than 10 -4 .

[0017] The fault-tolerant control method of the turbofan engine acceleration control plan, the initial value d i,1 :

[0018]

[0019] Where ( W f,id ) k-1 and ( W f,id ) k is the real-time optimal fuel at the time of model simulation k -1 and k , m is the total number of statistical samples, not less than 3000.

[0020] The fault-tolerant control method of the turbofan engine acceleration control plan, in step S300, the variance σ j The calculation method is:

[0021] ,

[0022] ,

[0023] ,

[0024] Wherein n is the total number of acceleration control plans used in the weighted fusion, not less than 2.

[0025] In the fault-tolerant control method of the aero-engine acceleration control plan, in step S400, h i,j , γ i,j And the new μ j The specific calculation steps are as follows:

[0026] ,

[0027] ,

[0028] ,

[0029] ,

[0030] ,

[0031] Wherein β is a constant set, μ j ) new The new μ j , μ j ) final The final μ j .

[0032] In the fault-tolerant control method of the aero-engine acceleration control plan, in step S400, β is 0.618.

[0033] Compared with the prior art, the present application has the following advantages: the present application has fewer parameters to be optimized, is simple to use and easy to implement the algorithm, except that a few parameters need to be determined by prior knowledge to determine the initial value, the remaining parameter settings are irrelevant to the specific engine characteristics. It can be universally applied to the weighted fusion of not less than two acceleration control plans, without the need to pay attention to the disturbed characteristics of the specific plan. The acceleration process controlled by the disclosed method can still maintain a certain acceleration performance in addition to avoiding the surge caused by a single fault. BRIEF DESCRIPTION OF DRAWINGS

[0034] Various other advantages and benefits of the present application will become apparent to those of ordinary skill in the art, upon reading the following detailed description of the preferred embodiment. The accompanying drawings are intended to further assist in the understanding of the present application, and are not intended to limit the present application in any way. It is to be clearly understood that the drawings depicted are only a few of numerous possibilities for the present application, and that many other drawings can be derived from these drawings by one of ordinary skill in the art without paying creative effort. Throughout the drawings, like reference numerals will be used to refer to like components.

[0035] In the drawings:

[0036] Figure 1 is a control schematic diagram of a fault-tolerant control method in an aero-engine acceleration controller according to an embodiment of the present disclosure;

[0037] Figure 2 is a program block diagram of a fault-tolerant control method of various acceleration control plans according to an embodiment of the present disclosure;

[0038] Figure 3 is an asymmetric Gaussian distribution function image according to an embodiment of the present disclosure;

[0039] Figure 4 is a box plot of minimum surge margin statistics of different control methods in an acceleration process under the influence of a single factor fault and other small range deviations of sea level standard atmosphere according to an embodiment of the present disclosure;

[0040] Figure 5 is a box plot of acceleration time statistics of different control methods in an acceleration process under the influence of a single factor fault and other small range deviations of sea level standard atmosphere according to an embodiment of the present disclosure;

[0041] Figure 6 is a flight envelope range and simulation point according to an embodiment of the present disclosure;

[0042] Figure 7 is a box plot of minimum surge margin statistics of different control methods in an acceleration process under the influence of a single factor fault and other small range deviations within a flight envelope according to an embodiment of the present disclosure;

[0043] Figure 8 is a box plot of acceleration time statistics of different control methods in an acceleration process under the influence of a single factor fault and other small range deviations within a flight envelope according to an embodiment of the present disclosure.

[0044] The present application will be further explained in conjunction with the drawings and embodiments. DETAILED DESCRIPTION

[0045] Specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although specific embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0046] 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.

[0047] 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.

[0048] like Figures 1 to 8 As shown, the fault-tolerant control method of the aircraft engine acceleration control plan includes the following steps:

[0049] S100: During the acceleration of the aircraft engine, i A control program that accelerates the control plan at the current moment j The obtained acceleration fuel flow target ( W fi ) j , and the previous moment j -1 accelerated fuel flow fusion output ( W f,acc ) j-1 , divide the front and back to get the normalized relative fuel quantity x i,j， ;

[0050] S200: According to the current time j Calculated i Instantaneous variance of an accelerated control plan d ti,j and the previous moment j A population variance of -1 d i,j-1 Weighted calculation to get the overall variance at the current moment di,j its initial value d i,1 According to prior knowledge, then use each kind of acceleration control plan to set the overall variance d i,j Calculate the weight value ω i,j ;

[0051] S300: Weighted average relative fuel quantity μ j At each time j Need to set the initial value as the relative fuel quantity x i,j The minimum value in each time σ j ;

[0052] S400: According to the relative fuel quantity x i,j And the weighted average relative fuel quantity μ j The size relationship and the statistical variance of the multiple plan fuels at the current time σ j Get the half variance h i,j According to the weight value ω i,j And the asymmetric Gaussian distribution function to calculate the posterior probability γ i,j Then combine the posterior probability γ i,j And the half variance h i,j Calculate the new weighted average relative fuel quantity μ j Repeat the S400 step until the relative error of the weighted average relative fuel quantity μ j Before and after iteration is less than the set maximum error, and finally with the terminated weighted average relative fuel quantity μ j And the acceleration fuel flow rate fusion output W f,acc ) j-1 The product is the acceleration fuel flow rate fusion output W f,acc ) j .

[0053] The preferred embodiment of the fault-tolerant control method of the aircraft engine acceleration control plan, in step S200, the weight value ω i,j The calculation method is:

[0054] ,

[0055] ,

[0056] ,

[0057] Where ρ is a set constant, which does not exceed 10 -4 .

[0058] In a preferred embodiment of the fault-tolerant control method of the aircraft engine acceleration control plan, the initial value d i,1 for:

[0059]

[0060] in( W f,id ) k-1 and( W f,id ) k The model simulation results k -1 and k The real-time optimal fuel consumption at the moment, where m is the total number of statistical samples and must be no less than 3000.

[0061] In a preferred embodiment of the fault-tolerant control method for an aircraft engine acceleration control plan, in step S300, the variance σ j The calculation method is:

[0062] ,

[0063] ,

[0064] ,

[0065] Where n is the total number of accelerated control plans used in weighted fusion, which shall not be less than 2.

[0066] In a preferred embodiment of the fault-tolerant control method for an aircraft engine acceleration control plan, in step S400, h i,j , γ i,j and new μ j The specific calculation steps are as follows:

[0067] ,

[0068] ,

[0069] ,

[0070] ,

[0071] ,

[0072] where β is a constant set, μ j ) new is the new μ j , μ j ) final is the final μ j .

[0073] In the preferred embodiment of the fault-tolerant control method for the acceleration control program of an aero-engine, in step S400, β is 0.618.

[0074] In one embodiment, as shown in Figure 1 , a simple fault-tolerant control method for the acceleration control program of an aero-engine is applied to the acceleration controller to output the weighted fused final acceleration fuel flow, and the specific control program block diagram is as shown in Figure 2 , which includes the following steps:

[0075] S100: In the engine acceleration process of an aero-engine, the acceleration fuel flow target i j obtained at the current time according to the control program of the first acceleration control program, W fi ) j , and the acceleration fuel flow fused output j W f,acc j-1 at the previous time t-1 are divided to obtain the normalized relative fuel amount x i,j， ;

[0076] S200: The instantaneous variance j i of the first acceleration control program calculated at the current time t and the overall variance d ti,j at the previous time t-1 are weighted to obtain the overall variance j d i,j-1 at the current time t, and the initial value d i,j d i,1 ​​​​​​According to prior knowledge setting, then use each kind of acceleration control plan's overall variance d i,j The weight is calculated ω i,j ;

[0077] S300: the weighted average relative fuel quantity μ j At each time j Need to set the initial value as the relative fuel quantity x i,j The minimum value in each time, and calculate the statistical variance of the multiple plan fuels at the current time σ j ;

[0078] S400: according to the relative fuel quantity x i,j And the weighted average relative fuel quantity μ j The size relationship and the statistical variance of the multiple plan fuels at the current time σ j Get the half variance h i,j According to the weight ω i,j And the asymmetric Gaussian distribution function to calculate the posterior probability γ i,j Then combine the posterior probability γ i,j And the half variance h i,j Calculate the new weighted average relative fuel quantity μ j Repeat S400 step until the relative error of the weighted average relative fuel quantity μ j Before and after iteration is less than the set maximum error, finally with the terminated weighted average relative fuel quantity μ j And the acceleration fuel flow rate fusion output ( W f,acc ) j-1 The product as the acceleration fuel flow rate fusion output ( W f,acc ) j .

[0079] The above embodiments constitute the complete technical solution of the present disclosure, wherein the number of acceleration control plans can be any number not less than two. The acceleration process controlled by the method of the present disclosure can avoid the significant decrease in surge margin caused by single disturbance factor failure and other small-range errors of disturbance factors, and avoid the risk of surge; and can avoid significant decrease in acceleration performance when there is single failure influence.

[0080] In another embodiment, the normalization method of the fuel flow of the plurality of acceleration control plans in step S100 is:

[0081]

[0082] In another embodiment, in step S200, the ω i,j The calculation method is:

[0083]

[0084]

[0085] wherein p is a constant set according to experience, and is generally small, not more than 10 -4 .

[0086] The initial value is d i,1 According to prior knowledge, the average variance between the normalized fuel of the plurality of plans and the normalized optimal fuel under different failure conditions can be obtained through model simulation, as shown in the formula:

[0087]

[0088] wherein (t) and (t) are the real-time optimal fuel at time W f,id k-1 W f,id k and are the real-time optimal fuel at time k -1 and k obtained through model simulation, and m is the total number of statistical samples, generally not less than 3000.

[0089]

[0090] In another embodiment, in step S300, the variance σ j The calculation method is:

[0091]

[0092] ​​​

[0093]

[0094] wherein n is the total number of plans adopted for the weighted fusion, not less than 2.

[0095] In another embodiment, the image of the asymmetric Gaussian distribution in step S400 is as shown in Figure 3 h i,j , γ i,j and the new μ j The specific calculation steps are as follows:

[0096]

[0097]

[0098]

[0099]

[0100]

[0101] wherein β is a constant set, generally with a value of 0.618 having a better use effect, μ j new is the new μ j , μ j final is the final μ j .

[0102] The disturbance factors considered in the present example, in addition to the temperature sensor considering only normal error influence, also consider the wide range of fault errors of the pressure sensor, fuel metering errors and faults, and the normal range and abnormal range of power extraction of the engine itself, as shown in Table 1.

[0103] Table 1

[0104]

[0105] wherein, L ex is the power extraction size of the engine, p t2 is the total pressure of engine intake, p t25 is the total pressure of compressor inlet, p t31 is the total pressure of compressor outlet,​​​T t2 is the total temperature of the engine intake air.

[0106] This example simulates a specific engine component level model to verify the effectiveness of the proposed simple fault-tolerant control method. p t2 Converted fuel flow plan, p t31 The weighted fusion of the converted fuel flow plan and the N-dot plan is used to prove that the method is better than the two plans (for p t31 The converted fuel flow plan and N-dot plan) take the minimum value, the median of the three plans and directly use ω i,j The method of weighting the three planned fuels can better prevent compressor surge during the acceleration process to verify the characteristics of the method disclosed in the present disclosure. The specific process is as follows:

[0107] According to the deviation of disturbance factors and the provisions of fault impact as described in Table 1, assuming that under the design conditions of sea level standard atmosphere, a large-scale fault deviation of a random single disturbance factor and a small-scale deviation of other factors are injected into a specific engine component-level model, the minimum value (2ASs) of two plans, the median of three plans, and the direct ω i,j The three weighted fuel plans (WA) and the simplified fault-tolerant control method (EMAD) described in this disclosure were tested for full-range acceleration, and 3,000 different disturbance combinations were tested. The so-called full-range acceleration refers to the acceleration process from the engine idling state to the intermediate state. During this process, the throttle lever is quickly pushed to input the speed control step command, fully reflecting the effect of the acceleration control plan at different speeds. The minimum surge margin of each method and each acceleration process was calculated. SM min For comparison, such as Figure 4 As shown, SM id Represents the theoretical surge margin target for the design. Figure 4 It shows that the proposed EMAD method can effectively stay away from the surge boundary under fault conditions, the Median method has reached the surge boundary, and the WA method is also very close to the surge boundary. t acc Distribution box plot Figure 5 As shown in Figure 1, according to the requirements of JSSG-2007B, the full acceleration time is defined as the time to reach 95% of the target thrust change. Figure 5It is shown that the proposed EMAD method can ensure the stability of the acceleration performance compared with the method of 2ASs. In summary, the proposed simple fault-tolerant control method in the present disclosure can effectively avoid surge and avoid significant decline in acceleration performance at sea level.

[0108] Furthermore, the present disclosure also simulates and compares four fault-tolerant control methods through large sample random disturbance simulation within the full flight envelope. The specific process is as follows:

[0109] According to the deviation and fault influence size of the disturbance factors described in Table 1, the engine model is randomly injected with the same single disturbance factor fault error and small range deviation of other disturbance factors in each full acceleration within the full flight envelope. The flight envelope is shown in Figure 6 10000 groups of repeated acceleration processes under different disturbance conditions are carried out within the envelope, and the minimum surge margin in each acceleration process is counted. SM min The comparison is shown in Figure 7 wherein SM b represents the surge boundary. Figure 7 It is shown that compared with the Median and WA methods, the proposed EMAD method can ensure that the surge boundary is far away from the surge boundary under fault conditions within the full envelope. The distribution box plot of the full acceleration time difference between the proposed EMAD method and the 2ASs method is shown in Figure 8 The distribution shows that the acceleration time of the proposed EMAD method within the full flight envelope is generally smaller than that of the 2ASs method under disturbance, which shows that the EMAD method can also avoid significant decline in acceleration time within the full flight envelope. In general, only the proposed simple fault-tolerant control method in the present disclosure can ensure that the surge boundary is far away from the surge boundary while avoiding significant decline in acceleration performance within the full envelope. It can be seen that the method is not only simple and easy to use, but also has obvious benefits.

[0110] Although the embodiments of the present application are described above in combination with the drawings, the present application is not limited to the above specific embodiments and application fields, and the above specific embodiments are only illustrative and guiding, but not limiting. Those skilled in the art can make many forms under the guidance of the present disclosure and without departing from the scope protected by the claims of the present application, which are all included in the protection of the present application.

Claims

1. A fault-tolerant control method for an aircraft engine acceleration control schedule, characterized by, The method comprises the following steps: S100: Acceleration fuel flow target (W fi ) j obtained at the current time j according to the control program of the i-th acceleration control plan in the engine acceleration process of the aero-engine f,acc j-1 , and the acceleration fuel flow fusion output (W i,j， ​ S200: instantaneous variance d of the i-th acceleration control plan calculated according to the current time j ti,j and the overall variance d of the previous time j-1 i,j-1 The overall variance d of the current time is calculated by weighting i,j The initial value d i,1 is set according to prior knowledge, and then the overall variance d of each acceleration control plan is calculated i,j to obtain the weight ω i,j ; S300: Weighted average relative fuel quantity μ j At each time j, the initial value is set to the minimum value of the relative fuel quantity x i,j and the statistical variance σ j of the planned fuel at the current time is calculated S4 00: according to the relative fuel quantity x i,j with the weighted average relative fuel quantity μ j the size relationship and the statistical variance σ of the multiple plan fuel at the current time j get the half variance h i,j , according to the weight ω i,j and the asymmetric Gaussian distribution function to calculate the posterior probability γ i,j , then combine the posterior probability γ i,j and the half variance h i,j to calculate the new weighted average relative fuel quantity μ j , repeat the S400 step until the relative error of the weighted average relative fuel quantity μ j before and after iteration is less than the set maximum error, and finally output the terminated weighted average relative fuel quantity μ j and the acceleration fuel flow fusion (W f,acc ) j-1 f,acc j product as the acceleration fuel flow fusion output (W j ) at the current time 2. The fault-tolerant control method for an acceleration control schedule of a gas turbine engine according to claim 1, wherein, In step S200, the weight ω i,j The calculation method is: d i,j = (1 - p) - d i,j-1 + p - min(d ti,j , 3d i,j-1 ), where p is a constant set to no more than 10 -4 .

3. The fault-tolerant control method for an acceleration control schedule of a gas turbine engine according to claim 1, wherein, Initial value d i,1 is: where (W f,id ) k-1 and (W f,id ) k k-1 and k are the real-time optimal fuel at the k-1 and k moments obtained by model simulation, and m is the total number of statistical samples, not less than 3000.

4. The fault-tolerant control method for an acceleration control schedule of a gas turbine engine according to claim 1, wherein, In step S300, the variance σ j The calculation method is: Wherein n is the total number of the accelerated control plans adopted in the weighted fusion, and is not less than 2.

5. A fault-tolerant control method for an acceleration control schedule of a gas turbine engine according to claim 4, wherein, In step S400, h i,j , γ i,j and new μ j are calculated as follows: (W f,acc ) j = (μ j ) final · (W f,acc ) j-1 , where β is a constant set, (μ j ) new is the new μ j calculated for this iteration j ) final is the final μ j for which the iteration is terminated.

6. A fault-tolerant control method for an acceleration control schedule of a gas turbine engine according to claim 5, wherein, In step S400, β is 0.618.

Images