Aircraft performance evaluation method based on multi-expert system reasoning

Abstract

The application discloses a kind of based on the inference of multi-expert system aircraft performance evaluation method, belong to fault diagnosis technical field, comprising: the multiple-source monitoring data collected is input aircraft performance evaluation model, based on multiple-source monitoring data constructs multiple different confidence rule base expert module;The rule in the multiple different confidence rule base expert module is fused by evidence reasoning algorithm, obtains nonlinear model, and according to nonlinear model obtains the module output result of each confidence rule base expert module, then according to the pre-constructed gate network, the module output result of each confidence rule base expert module is weighted fusion, obtains aircraft performance evaluation result.The application can fully combine multiple-source monitoring data and expert knowledge, and then the performance of aircraft is comprehensively evaluated, the evaluation accuracy of evaluation model is guaranteed, while the model structure is simplified, so as to facilitate intelligent decision-making for the performance of aircraft, with good engineering application value.

Description

Technical Field

[0001] This invention belongs to the field of fault diagnosis technology, specifically relating to an aircraft performance evaluation method based on multi-expert system reasoning. Background Technology

[0002] Because aircraft are subjected to multiple operating conditions such as strong vibration, thermal stress, impact load and combustion environment during service, their structural components are prone to degradation and failure problems such as fatigue, cracks and loose connections, resulting in high uncertainty of performance degradation. Therefore, it is of great significance to conduct a comprehensive and accurate evaluation of the performance status of aircraft in the early stage of service.

[0003] Existing aircraft performance evaluation methods primarily rely on statistical modeling and signal analysis driven by multi-source monitoring data. However, in practical engineering applications, multi-source monitoring data for aircraft often exhibits the following characteristics: 1) Limited multi-source monitoring data. Due to space, mass, and energy consumption constraints on the number of sensors deployed, the dimensions of acquired multi-source monitoring data are limited, and some key components are difficult to measure directly; 2) Uncertainty and noise interference. The environment during aircraft flight is extreme, and sensor data is often affected by high noise, missing data, and drift. Relying solely on data-driven methods can easily lead to evaluation bias; 3) The importance of expert experience. During the design, testing, and service of aircraft, a wealth of expert knowledge and empirical understanding of structural failure mechanisms has been accumulated. If only limited multi-source monitoring data is relied upon while ignoring expert knowledge, the evaluation model will struggle to fully reflect the potential risks of the structure. To address these issues, the Belief Rule Base (BRB), due to its ability to integrate numerical multi-source monitoring data with fuzzy expert knowledge, has become an effective performance evaluation modeling tool. The confidence rule base uses a "rule-weight-confidence" format to express complex uncertain reasoning relationships. It can describe experts' cognitive judgments of aircraft performance status and can be updated and corrected using multi-source monitoring data, thus achieving an organic integration of "data and knowledge." On the other hand, aircraft performance involves multi-source monitoring information, such as vibration signals, temperature and strain, thrust loads, and combustion chamber pressure. These information are coupled and jointly determine its performance status. Relying on only a single information source often fails to fully reveal performance failure symptoms. Therefore, integrating multi-source monitoring data is a necessary means to achieve reliable evaluation of aircraft performance. However, traditional single-confidence rule base models are prone to rule explosion when faced with multi-source monitoring data, i.e., the number of rules increases dramatically with the input dimension, leading to excessively large model size, decreased inference efficiency, and hindering practical engineering applications. Summary of the Invention

[0004] To address the aforementioned problems in the existing technology, this invention provides a method for evaluating aircraft performance based on multi-expert system reasoning.

[0005] The technical problem to be solved by this invention is achieved through the following technical solution: In a first aspect, the present invention provides a method for evaluating aircraft performance based on multi-expert system reasoning, the method comprising: Collect multi-source monitoring data on aircraft performance; The multi-source monitoring data is input into the aircraft performance evaluation model to obtain the aircraft performance evaluation result; wherein, the step of inputting the multi-source monitoring data into the aircraft performance evaluation model to obtain the aircraft performance evaluation result includes: Based on the multi-source monitoring data, several different confidence rule base expert modules are constructed; The rules in the multiple different confidence rule base expert modules are fused through the evidence reasoning algorithm to obtain a nonlinear model, and the module output results of each confidence rule base expert module are obtained based on the nonlinear model. The output results of each confidence rule base expert module are weighted and fused according to the pre-built gating network to obtain the aircraft performance evaluation result.

[0006] Optionally, the step of fusing the rules from the multiple different confidence rule base expert modules using an evidence reasoning algorithm to obtain a nonlinear model includes: The multi-source monitoring data is mapped to the corresponding reference state set according to a pre-constructed matching function; A first matching degree between the multi-source monitoring data and the reference state set is obtained based on the matching function, the multi-source monitoring data, and the reference state set; A second matching degree is obtained between the multi-source monitoring data and the rules in the multiple different confidence rule base expert modules based on the first matching degree. The activation weights of the rules in the multiple different confidence rule base expert modules for the multi-source monitoring data are obtained based on the second matching degree. Based on the evidence reasoning algorithm and the activation weight, the rules in the multiple different confidence rule base expert modules are aggregated to obtain the total confidence of the evaluation results corresponding to each rule; The nonlinear model is obtained based on the evaluation results corresponding to each rule and the total confidence level of the evaluation results corresponding to each rule.

[0007] Optionally, the pre-constructed matching function is represented as follows: ; in, This refers to the pre-built matching function. Indicates the first The first confidence rule base expert module Multi-source monitoring data acquired by multiple sensors Indicates the first The first confidence rule base expert module The multi-source monitoring data acquired by the sensor corresponds to the first... A reference state, This indicates the rule number in the confidence rule base expert module. This indicates the number of rules in the confidence rule base expert module.

[0008] Optionally, the activation weights are represented as follows: ; in, Indicates the first The first in the confidence rule base expert module The rules apply to the activation weights of the multi-source monitoring data. Indicates the first The first confidence rule base expert module The weight of each rule, Indicates the first The first confidence rule base expert module The second degree of matching between the rule and the multi-source monitoring data.

[0009] Optionally, the total confidence score of the evaluation results corresponding to each activated rule is represented as follows: ; in, Indicates the first The overall confidence level of the evaluation results corresponding to the rules in the expert module of the confidence rule base. Indicates intermediate variables. Indicates the first In the expert module of the confidence rule base, the first The first rule The confidence level of each evaluation result , Indicates the number of evaluation results. Indicates the first In the expert module of the confidence rule base, the first The first rule The confidence level of each evaluation result; The intermediate variable is represented as follows: .

[0010] Optionally, the pre-built gating network is represented as follows: ; in, Indicates the first The individual confidence rule base expert module at time Gated output, Denotes the first gating matrix. Indicates the first offset. This represents the first intermediate parameter.

[0011] Optionally, the first intermediate parameter is represented as follows: ; in, It is the hyperbolic tangent function. This indicates the second intermediate parameter.

[0012] Optionally, the second intermediate parameter is represented as follows: ; in, Represents the second gating matrix. This indicates the second offset. Indicates the first The individual confidence rule base expert module at time The module output results.

[0013] Optionally, the weights used to weight and fuse the outputs of the various confidence rule base expert modules are represented as follows: ; in, Indicates the first The individual confidence rule base expert module at time The weights corresponding to the module output results, Represents an exponential function. This indicates the number of expert modules in the confidence rule base.

[0014] Optionally, the performance evaluation results of the aircraft are expressed as follows: ; in, Indicates at time The results of the aircraft performance evaluation.

[0015] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: In the above technical solution, multi-source monitoring data is modeled and reasoned separately by different confidence rule base expert modules. Each module performs independent reasoning and then the data is fused through an evidence reasoning algorithm to obtain the output results of each confidence rule base expert module. Furthermore, a gating network is used for weighted fusion to obtain the aircraft performance evaluation result. This invention effectively alleviates the rule explosion problem of a single confidence rule base and improves the scalability and computational efficiency of the model by utilizing multi-expert parallel reasoning. Simultaneously, this invention can fully combine multi-source monitoring data and expert knowledge to achieve a multi-dimensional and comprehensive evaluation of aircraft performance. While ensuring the evaluation accuracy of the model, it also has a simple model structure and good engineering application value.

[0016] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0017] Figure 1 This invention provides a method for evaluating aircraft performance based on multi-expert system reasoning. Figure 2 This is a schematic diagram illustrating an example of multi-source monitoring data for an aircraft provided in an embodiment of the present invention; Figure 3 This is a schematic diagram illustrating another example of multi-source monitoring data for an aircraft provided in this embodiment of the invention; Figure 4 This is a comparison chart of aircraft performance evaluation results and actual aircraft performance values ​​provided by an embodiment of the present invention. Detailed Implementation

[0018] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0019] Figure 1 This invention provides a method for evaluating aircraft performance based on multi-expert system reasoning, such as... Figure 1 As shown, the method may include the following steps: S1. Collect multi-source monitoring data on aircraft performance.

[0020] It is understandable that multi-source monitoring data refers to different types of multi-source monitoring data of the aircraft. For example, it may include gas concentration (Gc, unit vol%), distance (Dis, unit mm), bolt loosening force (Bl, unit kN), temperature (Tem, unit °C), vibration velocity (Vib, unit mm / s), and settlement (Set, unit mm), etc. The monitoring data of each type in the multi-source monitoring data can be obtained from the corresponding type of sensor.

[0021] S2. Input the multi-source monitoring data into the aircraft performance evaluation model to obtain the aircraft performance evaluation results.

[0022] Specifically, the output aircraft performance evaluation results can be categorized into Excellent, Good, Average, and Poor levels. The final performance evaluation result is obtained by comparing the value of the evaluation result with the reference values ​​corresponding to different performance evaluation levels. For example, the reference value for Excellent is 1, for Good it is 0.8, for Average it is 0.6, and for Poor it is 0. A performance evaluation result of 1 indicates Excellent; a result greater than or equal to 0.8 and less than 1 indicates Good; a result greater than or equal to 0.6 and less than 0.8 indicates Average; and a result greater than or equal to 0 and less than 0.6 indicates Poor. It is worth noting that the method for determining the aircraft performance evaluation result can be adjusted according to actual circumstances; the above method is only one possible approach.

[0023] The process involves inputting multi-source monitoring data into the aircraft performance evaluation model to obtain the aircraft performance evaluation results, including: S21. Construct multiple expert modules for confidence rule bases based on multi-source monitoring data.

[0024] A confidence rule base is an intelligent modeling method based on evidence-based reasoning. In a confidence rule base, rules and premise attributes are hierarchically related: on the one hand, a single rule is composed of multiple premise attributes logically combined, with each premise attribute assigned a corresponding weight to reflect its relative importance; on the other hand, the conclusion of a rule consists of the evaluation result and its confidence distribution, where confidence is used to measure the reliability of the conclusion. A rule is only activated when all premise attributes are simultaneously satisfied. Its typical form is: IF [Premise] THEN [Conclusion] WITH [Confidence Distribution]. In this invention, the first... The first in the confidence rule base expert module The rule can be represented as: ; in, , This indicates the number of rules in the confidence rule base expert module. At time respectively No. The confidence rule base expert module received the first Multi-source monitoring data acquired by sensors with the same attribute: Since each type of monitoring data in the multi-source monitoring data has multiple sensors used to acquire that type of monitoring data, sensors that collect the same type of data in the multi-source monitoring data can be called sensors with the same attribute. The number of sensors with the same attribute is the number of prerequisite attributes. They are the first The prerequisite attribute at time Reference values ​​for multi-source monitoring data. Indicates "and". It is the first The first in the confidence rule base expert module The first rule corresponding to the One evaluation result, express The corresponding confidence level.

[0025] S22. The rules from multiple different confidence rule base expert modules are fused using an evidence reasoning algorithm to obtain a nonlinear model, and the module output results of each confidence rule base expert module are obtained based on the nonlinear model.

[0026] Optionally, S22 may include: (1) Map the multi-source monitoring data to the corresponding reference state set according to the pre-built matching function.

[0027] (2) The first matching degree between the multi-source monitoring data and the reference state set is obtained based on the matching function, the multi-source monitoring data and the reference state set.

[0028] Specifically, in the In the expert module of the confidence rule base, a matching function is pre-built to match the input multi-source monitoring data. Mapped to its corresponding set of reference states This enables the conversion between multi-source monitoring data and a defined reference state: ; ; in, This represents a pre-built matching function. Indicates the first The first confidence rule base expert module Multi-source monitoring data acquired by multiple sensors Indicates the first The first confidence rule base expert module The multi-source monitoring data acquired by the sensor corresponds to the first... A reference state, This indicates the rule number in the confidence rule base expert module. This indicates the number of rules in the confidence rule base expert module. Indicates multi-source monitoring data Compared with reference state The first degree of matching between them , This represents the distribution of the first degree of match among all reference states. Indicates the number of reference states.

[0029] (3) Based on the first matching degree, the second matching degree between the multi-source monitoring data and the rules in multiple different confidence rule base expert modules is obtained.

[0030] Specifically, determine the input multi-source monitoring data and the first The first in the confidence rule base expert module The second degree of matching between the rules: ; in, Indicates the first Rule number 1 The first confidence rule base expert module The degree of matching of the reference state of the multi-source monitoring data acquired by each sensor. Indicates the first The first confidence rule base expert module The attribute weights of each sensor.

[0031] (4) Based on the second matching degree, obtain the activation weights of the rules in the expert modules of multiple different confidence rule bases for multi-source monitoring data.

[0032] Specifically, by integrating the previously determined rule weights, the activation weights of rules from multiple different confidence rule base expert modules for multi-source monitoring data are obtained based on the second matching degree: ; in, Indicates the first The first in the confidence rule base expert module The activation weight of each rule for multi-source monitoring data reflects the relative importance of that rule relative to all rules activated by the given input multi-source monitoring data. Indicates the first The first confidence rule base expert module The weight of each rule, Indicates the first The first confidence rule base expert module The second degree of matching between the rule and the multi-source monitoring data.

[0033] (5) Aggregate the rules in multiple different confidence rule base expert modules according to the evidence reasoning algorithm and activation weight to obtain the total confidence of the evaluation results corresponding to each rule.

[0034] Optionally, the overall confidence score of the evaluation results for each rule is expressed as follows: ; in, Indicates the first The overall confidence level of the evaluation results corresponding to the rules in the expert module of the confidence rule base. Indicates intermediate variables. Indicates the first In the expert module of the confidence rule base, the first The first rule The confidence level of each evaluation result , Indicates the number of evaluation results. Indicates the first In the expert module of the confidence rule base, the first The first rule The confidence level of each evaluation result.

[0035] intermediate variables It is expressed as follows: .

[0036] (6) Obtain the nonlinear model based on the evaluation results corresponding to each rule and the total confidence of the evaluation results corresponding to each rule.

[0037] It is understandable that the nonlinear model obtained by integrating the evaluation results of each rule and the total confidence of the evaluation results of each rule through modular inference can be expressed as: ; in, Represents a nonlinear model. It is the first n One evaluation result.

[0038] (7) Obtain the module output results of each confidence rule base expert module based on the nonlinear model.

[0039] Furthermore, based on this nonlinear model, the module outputs of each confidence rule base expert module can be represented as follows: ; in, Indicates the first The individual confidence rule base expert module at time The module output results, Indicates the expectation. Indicates at time The n One evaluation result.

[0040] S23. The module outputs of each confidence rule base expert module are weighted and fused according to the pre-built gating network to obtain the aircraft performance evaluation result.

[0041] Alternatively, the pre-built gated network can be represented as follows: ; in, Indicates the first The individual confidence rule base expert module at time Gated output, Denotes the first gating matrix. Indicates the first offset. This represents the first intermediate parameter.

[0042] The first intermediate parameter is represented as follows: ; in, It is the hyperbolic tangent function. This indicates the second intermediate parameter.

[0043] The second intermediate parameter is represented as follows: ; in, Represents the second gating matrix. This indicates the second offset. Indicates the first The individual confidence rule base expert module at time The module output results.

[0044] The weights used to weight and fuse the outputs of the various confidence rule base expert modules are represented as follows: ; in, Indicates the first The individual confidence rule base expert module at time The weights corresponding to the module output results, Represents an exponential function. This indicates the number of expert modules in the confidence rule base.

[0045] The performance evaluation results of the aircraft are as follows: ; in, Indicates at time The results of the aircraft performance evaluation.

[0046] In this invention, the weights of multiple different expert modules on the output results are adjusted by a gating network, thereby enabling collaborative reasoning among multiple different expert modules and a comprehensive evaluation of the aircraft's performance.

[0047] In one implementation, Figure 2 This is a schematic diagram illustrating an example of multi-source monitoring data for an aircraft provided in an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating another example of multi-source monitoring data for an aircraft provided in this embodiment of the invention, such as... Figure 2 and Figure 3 As shown, this embodiment uses multi-source monitoring data of an aircraft structure as an example. By building an aircraft performance monitoring platform, a total of 400 sets of test data were acquired. Each set of data contains six monitoring data: gas concentration (Gc, vol%), distance (Dis, mm), bolt loosening force (Bl, kN), temperature (Tem, °C), vibration velocity (Vib, mm / s), and settlement (Set, mm). Each monitoring data comes from two sensors, denoted as Sensor 1 and Sensor 2. Each monitoring data has a center point, indicating that the performance state of the aircraft structure at the center point under that monitoring data is the most ideal and the best. The center points of the six monitoring data are 29 vol% (Gc), 2000 mm (Dis), 662 kN (Bl), 25 °C (Tem), 0 mm (Set), and 0 mm / s (Vib). The reference levels are set based on the distance from the center point of the multi-source monitoring data, including "low", "medium", "high", and "very high", represented by "L", "M", "H", and "VH", respectively. Taking a vibration sensor as an example, the correspondence between reference values ​​and reference levels is shown in Table 1 below. The output performance evaluation results include four categories: "Excellent," "Good," "Medium," and "Poor," with the corresponding values ​​for each level shown in Table 2. Table 1

[0048] Table 2

[0049] Furthermore, a confidence rule base was constructed based on the aforementioned multi-source monitoring data. This confidence rule base includes 16 rules. The second column of Table 3 shows the rule weights, which are set to 1, indicating that all rules are equally important. The last column of Table 3 shows the confidence level of the evaluation results for each rule.

[0050] Table 3

[0051] Figure 4This is a comparison chart of aircraft performance evaluation results and actual aircraft performance values ​​provided by an embodiment of the present invention, such as... Figure 4 As shown, the aircraft performance evaluation model provided in this application can effectively and accurately evaluate the performance level of an aircraft.

[0052] This invention can fully integrate multi-source monitoring data and expert knowledge to comprehensively evaluate aircraft performance, ensuring the evaluation accuracy of the evaluation model while simplifying the model structure. This invention can be used to make intelligent decisions on aircraft performance and has good engineering application value.

[0053] It should be noted that the terms "first," "second," etc., are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention.

[0054] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, those skilled in the art can combine and integrate the different embodiments or examples described in this specification.

[0055] Although the invention has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings and the disclosure in carrying out the claimed invention. In the description of the invention, the word "comprising" does not exclude other components or steps, "a" or "an" does not exclude a plurality, and "a plurality" means two or more, unless otherwise explicitly specified. Furthermore, while different embodiments may describe certain measures, this does not mean that these measures cannot be combined to produce good results.

[0056] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A method for evaluating aircraft performance based on multi-expert system reasoning, characterized in that, The method includes: Collect multi-source monitoring data on aircraft performance; The multi-source monitoring data is input into the aircraft performance evaluation model to obtain the aircraft performance evaluation results; The step of inputting the multi-source monitoring data into the aircraft performance evaluation model to obtain the aircraft performance evaluation results includes: Based on the multi-source monitoring data, several different confidence rule base expert modules are constructed; The rules in the multiple different confidence rule base expert modules are fused through the evidence reasoning algorithm to obtain a nonlinear model, and the module output results of each confidence rule base expert module are obtained based on the nonlinear model. The output results of each confidence rule base expert module are weighted and fused according to the pre-built gating network to obtain the aircraft performance evaluation result.

2. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 1, characterized in that, The process of fusing rules from multiple different confidence rule base expert modules using an evidence reasoning algorithm to obtain a nonlinear model includes: The multi-source monitoring data is mapped to the corresponding reference state set according to a pre-constructed matching function; A first matching degree between the multi-source monitoring data and the reference state set is obtained based on the matching function, the multi-source monitoring data, and the reference state set; A second matching degree is obtained between the multi-source monitoring data and the rules in the multiple different confidence rule base expert modules based on the first matching degree. The activation weights of the rules in the multiple different confidence rule base expert modules for the multi-source monitoring data are obtained based on the second matching degree. Based on the evidence reasoning algorithm and the activation weight, the rules in the multiple different confidence rule base expert modules are aggregated to obtain the total confidence of the evaluation results corresponding to each rule; The nonlinear model is obtained based on the evaluation results corresponding to each rule and the total confidence level of the evaluation results corresponding to each rule.

3. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 2, characterized in that, The pre-constructed matching function is represented as follows: ； in, This refers to the pre-built matching function. Indicates the first The first confidence rule base expert module Multi-source monitoring data acquired by multiple sensors Indicates the first The first confidence rule base expert module The multi-source monitoring data acquired by the sensor corresponds to the first... A reference state, This indicates the rule number in the confidence rule base expert module. This indicates the number of rules in the confidence rule base expert module.

4. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 3, characterized in that, The activation weights are represented as follows: ； in, Indicates the first The first in the confidence rule base expert module The rules apply to the activation weights of the multi-source monitoring data. Indicates the first The first confidence rule base expert module The weight of each rule, Indicates the first The first confidence rule base expert module The second degree of matching between the rule and the multi-source monitoring data.

5. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 4, characterized in that, The overall confidence level of the evaluation results corresponding to each rule is expressed as follows: ； in, Indicates the first The overall confidence level of the evaluation results corresponding to the rules in the expert module of the confidence rule base. Indicates intermediate variables. Indicates the first In the expert module of the confidence rule base, the first The first rule The confidence level of each evaluation result , Indicates the number of evaluation results. Indicates the first In the expert module of the confidence rule base, the first The first rule The confidence level of each evaluation result; The intermediate variable is represented as follows: 。 6. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 1, characterized in that, The pre-built gating network is represented as follows: ； in, Indicates the first The individual confidence rule base expert module at time Gated output, Denotes the first gating matrix. Indicates the first offset. This represents the first intermediate parameter.

7. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 6, characterized in that, The first intermediate parameter is represented as follows: ； in, It is the hyperbolic tangent function. This indicates the second intermediate parameter.

8. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 7, characterized in that, The second intermediate parameter is represented as follows: ； in, Represents the second gating matrix. This indicates the second offset. Indicates the first The individual confidence rule base expert module at time The module output results.

9. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 8, characterized in that, The weights used to perform weighted fusion of the output results of the various confidence rule base expert modules are represented as follows: ； in, Indicates the first The individual confidence rule base expert module at time The weights corresponding to the module output results, Represents an exponential function. This indicates the number of expert modules in the confidence rule base.

10. The aircraft performance evaluation method based on multi-expert system reasoning according to claim 9, characterized in that, The performance evaluation results of the aircraft are as follows: ； in, Indicates at time The results of the aircraft performance evaluation.

Images