Universal excitation and control system for aircraft multi-subsystem integrated test

Abstract

The application belongs to the technical field of aircraft testing, and provides a general excitation and measurement and control system for comprehensive testing of multiple subsystems of an aircraft, which comprises obtaining state data of each subsystem before testing, and generating a state sorting sequence of the subsystems; an excitation action optimization model is constructed according to Pearson correlation coefficients of excitation actions and states of the subsystems and testing consumption time length, and an optimal excitation action is selected for a current to-be-tested subsystem; meanwhile, an influence curve is established based on historical excitation amplitudes and state response data, and a testing risk score is generated; then, an optimization model is established with action amplitude as an independent variable and the risk score as a dependent variable, and an optimal action amplitude is solved; and the optimal excitation action and amplitude are tested for each subsystem in turn according to the sorting sequence; the application can dynamically integrate real-time state data of multiple subsystems of the aircraft, optimize selection of the excitation action and setting of the action amplitude, thereby improving testing efficiency, safety and reliability, and reducing testing risk and cost.

Description

Technical Field

[0001] This invention belongs to the field of aircraft testing technology, and in particular relates to a general excitation and control system for integrated testing of multiple subsystems of aircraft. Background Technology

[0002] Integrated testing of multiple subsystems of an aircraft is a crucial step in ensuring flight safety and performance reliability, involving the coordinated verification of complex subsystems such as navigation, propulsion, and avionics. The testing process requires applying diverse excitation signals in a simulated flight environment to comprehensively evaluate the functional response of the subsystems.

[0003] However, existing testing practices have significant limitations. Test sequences are generally based on pre-defined rules or engineers' subjective experience, failing to dynamically integrate real-time subsystem status data. This leads to the potential delay in testing high-risk subsystems; for example, if a propulsion subsystem exhibiting abnormal conditions is not verified first, it can easily trigger a chain reaction of failures in subsequent tests, increasing the overall test failure probability. Regarding stimulus selection, existing systems lack mechanisms for in-depth analysis of historical test data, making it impossible to quantify the correlation between stimulus actions and subsystem state changes. This results in blind spots or redundancy in test coverage, making it difficult to customize efficient testing strategies for specific subsystem states. Furthermore, stimulus amplitude settings rely excessively on uniform thresholds or equipment manual recommendations, ignoring the differences in subsystem health states. For instance, applying excessive amplitude to sensitive subsystems may cause sensor overload damage, while using insufficient amplitude to robust subsystems may fail to adequately expose potential defects. This static approach fails to balance test adequacy and safety, and is ill-suited to adapting to risk changes in dynamic testing environments.

[0004] Furthermore, the testing process lacks a comprehensive consideration of the duration of stimulus actions and their correlation with the state, resulting in low testing efficiency and unreasonable resource allocation. These shortcomings not only prolong the testing cycle and increase costs, but also harbor risks of equipment damage and distorted test results, severely hindering the intelligent and universal development of integrated aircraft testing. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a universal excitation and control system for integrated testing of multiple subsystems of aircraft, thus solving the aforementioned problems.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a universal excitation and control system for integrated testing of multiple subsystems of an aircraft, the system specifically comprising: The data acquisition unit is used to acquire the state data of the aircraft's multi-subsystem, the set of excitation actions during the integrated test of the aircraft's multi-subsystem, the Pearson correlation coefficient between the excitation actions and the subsystem state, and the test duration of the excitation actions; among which, the state data of the aircraft's multi-subsystem refers to the state data of the aircraft's multi-subsystem before the integrated test. The sorting unit is used to generate a subsystem state sorting sequence based on the state data of the multi-subsystem of the aircraft. The stimulus action analysis unit is used to establish an stimulus action optimization model based on the Pearson correlation coefficient between the stimulus action and the subsystem state and the stimulus action test consumption time, and to generate the optimal stimulus action; whereby the stimulus action test consumption time refers to the average consumption time of stimulus action tests in historical data; The historical data analysis unit is used to acquire historical data on the changes in the amplitude of the stimulus action and the state of the subsystem, and to establish an impact curve on the state of the subsystem based on the historical data on the changes in the amplitude of the stimulus action and the state of the subsystem. The historical data on the changes in the amplitude of the stimulus action and the state of the subsystem refers to the changes in the state of the subsystem when the amplitude of the stimulus action changes in the historical data. The risk analysis unit is used to generate a test risk score for the aircraft subsystem based on the impact curve of the excitation action amplitude on the subsystem state and the action amplitude of the optimal excitation action. The motion amplitude analysis unit is used to establish a motion amplitude analysis model with the motion amplitude of the optimal excitation motion as the independent variable and the test risk score of the aircraft subsystem as the dependent variable, and to generate the optimal motion amplitude of the optimal excitation motion. The excitation and control unit is used to excite and control the subsystems in the subsystem state sorting sequence in sequence according to the optimal excitation action and the optimal action amplitude of the optimal excitation action.

[0007] Based on the above technical solutions, the present invention also provides the following optional technical solutions: Further technical solution: The sorting unit specifically includes: The status assessment module is used to generate subsystem status assessment values ​​based on the status data of the aircraft's multiple subsystems. The sequence output module is used to generate a sorted sequence of subsystem states based on the subsystem state evaluation values.

[0008] Further technical solutions: The specific methods for generating the subsystem state evaluation values ​​include: Through the formula: ; Generate subsystem state evaluation values ; In the formula, This represents the state assessment value of the i-th subsystem of the aircraft. This represents the normalized value of the j-th state data of the i-th subsystem of the aircraft. represents the weight coefficient of the j-th state data of the i-th subsystem of the aircraft, and m represents the number of state data of the i-th subsystem of the aircraft.

[0009] Further technical solution: The excitation action analysis unit specifically includes: The correlation analysis module is used to generate correlation scores based on the Pearson correlation coefficient between the excitation action and the subsystem state; The time consumption analysis module is used to generate a time score for the incentive action based on the time consumed in the incentive action test; The optimization result output module is used to establish an optimization model for incentive actions based on the correlation score and the incentive action time score, and generate the optimal incentive action.

[0010] Further technical solution: The specific method for generating the correlation score includes: Through the formula: ; Generate relevance scores ; In the formula, This represents the correlation score between the stimulus action k and the state of subsystem i. This represents the absolute value of the Pearson correlation coefficient between the excitation action k and the state of subsystem i. This represents the minimum absolute value of the Pearson correlation coefficient between all excitation actions and the state of subsystem i. This represents the maximum absolute value of the Pearson correlation coefficient between all excitation actions and the state of subsystem i.

[0011] Further technical solution: The method for generating the excitation action time score specifically includes: Through the formula: ; Generate motivational action time score ; In the formula, This represents the time score of the excitation action k used for subsystem i. This represents the test duration for the excitation action k used in subsystem i. This represents the threshold for test duration.

[0012] Further technical solution: The expression of the incentive action optimization model is specifically as follows: ; In the expression, This represents the optimal excitation action for subsystem i. This represents the correlation score between the stimulus action k and the state of subsystem i. G represents the time score of the excitation action k used for subsystem i, and G represents the set of excitation actions. , All are weighting coefficients, and .

[0013] Further technical solutions: The specific methods for generating the test risk score of the aircraft subsystem include: Through the formula: ; Generate test risk scores for aircraft subsystems ; In the formula, This represents the optimal incentive action. The test risk score for aircraft subsystem i when the amplitude of the action is h. This represents the state evaluation value of subsystem i. This represents the change in the subsystem state evaluation value of aircraft subsystem i when the optimal excitation amplitude is h, obtained based on the influence curve of the excitation amplitude on the subsystem state. This represents the state assessment threshold; where, when obtaining the change value of the subsystem state assessment value of aircraft subsystem i based on the influence curve of the excitation action amplitude on the subsystem state, the change in the action amplitude of the optimal excitation action is the difference between the action amplitude h and the initial action amplitude.

[0014] A further technical solution: The expression for the motion amplitude analysis model is specifically as follows: ; In the expression, This represents the optimal incentive action. The optimal range of motion, This represents the optimal incentive action. The test risk score for aircraft subsystem i when the amplitude of the action is h. This represents the optimal incentive action. The test completion score is given when the range of motion is h, and F represents the optimal motivational action. The set of range of motion, , All are weighting coefficients, and .

[0015] Further technical solution: The method for generating the test completion score specifically includes: Through the formula: ; Generate test completion score ; In the formula, This represents the optimal incentive action. The range of motion setting value, This refers to the optimal excitation action during the integrated testing of multiple subsystems of an aircraft. The minimum range of motion required.

[0016] This invention provides a universal excitation and control system for integrated testing of multiple subsystems of aircraft, which has the following advantages compared with the prior art: This invention solves the problems of relying on preset rules, lack of real-time data integration, and low efficiency in testing by dynamically integrating real-time status data and quantifying and optimizing excitation parameters. It has the advantages of being able to dynamically integrate real-time status data of multiple subsystems of an aircraft, optimize the selection of excitation actions and the setting of action amplitude, thereby improving testing efficiency, safety and reliability, and reducing testing risks and costs. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the general excitation and control system for integrated testing of multiple subsystems of an aircraft provided by the present invention; Figure 2 This is a schematic diagram of the structure of the sorting unit provided by the present invention; Figure 3 This is a schematic diagram of the excitation action analysis unit provided by the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0019] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0020] Please see Figure 1 This invention provides a universal excitation and control system for integrated testing of multiple subsystems of an aircraft, comprising, in one embodiment: The data acquisition unit 10 is used to acquire the state data of the multi-subsystem of the aircraft, the set of excitation actions during the integrated test of the multi-subsystem of the aircraft, the Pearson correlation coefficient between the excitation actions and the state of the subsystem, and the test time of the excitation actions; wherein, the state data of the multi-subsystem of the aircraft refers to the state data of the multi-subsystem of the aircraft before the integrated test is carried out. The sorting unit 20 is used to generate a subsystem state sorting sequence based on the state data of the multi-subsystem of the aircraft. The excitation action analysis unit 30 is used to establish an excitation action optimization model based on the Pearson correlation coefficient between the excitation action and the subsystem state and the excitation action test consumption time, and generate the optimal excitation action; wherein, the excitation action test consumption time refers to the average consumption time of excitation action tests in historical data; The historical data analysis unit 40 is used to acquire historical data on the changes in the amplitude of the stimulus action and the state of the subsystem, and to establish an influence curve on the state of the subsystem based on the historical data on the changes in the amplitude of the stimulus action and the state of the subsystem; wherein, the historical data on the changes in the amplitude of the stimulus action and the state of the subsystem refers to the data on the changes in the state of the subsystem when the amplitude of the stimulus action changes in the historical data. Risk analysis unit 50 is used to generate a test risk score for the aircraft subsystem based on the impact curve of the excitation action amplitude on the subsystem state and the action amplitude of the optimal excitation action. The motion amplitude analysis unit 60 is used to establish a motion amplitude analysis model with the motion amplitude of the optimal excitation motion as the independent variable and the test risk score of the aircraft subsystem as the dependent variable, and to generate the optimal motion amplitude of the optimal excitation motion. The excitation and control unit 70 is used to excite and control the subsystems in the subsystem state sorting sequence according to the optimal excitation action and the optimal action amplitude of the optimal excitation action. Among them, the excitation action refers to the preset or controllable external inputs or operations that can be applied to each subsystem during the integrated testing of multiple subsystems of an aircraft; the excitation action aims to simulate the actual flight environment or specific fault conditions in order to observe the response of the subsystem. The Pearson correlation coefficient is a statistic that measures the degree of linear correlation between two variables. In this embodiment, it is used to quantify the strength of the linear association between the stimulus action and the change in the state of the subsystem. The larger the absolute value of the coefficient, the more significant the impact of the stimulus action on the state of the subsystem. The influence curve of the excitation amplitude on the subsystem state refers to the functional relationship or graph established by analyzing historical data to describe how different amplitude changes of the excitation affect the changes of various state parameters of the subsystem; this curve helps to predict the subsystem response under different excitation intensities.

[0021] Specifically, firstly, the data acquisition unit 10 needs to acquire the status data of the aircraft's multi-subsystems, the set of excitation actions during the integrated testing of the aircraft's multi-subsystems, the Pearson correlation coefficient between the excitation actions and the subsystem status, and the test duration of the excitation actions. The status data of the aircraft's multi-subsystems can be acquired in real time through sensors or extracted from various operating parameters from the aircraft's health monitoring system. The set of excitation actions can be pre-stored in a database, containing all available test commands and parameter ranges. The Pearson correlation coefficient between the excitation actions and the subsystem status, as well as the test duration of the excitation actions, can be obtained through statistical analysis based on historical test data. For example, this data can be manually entered by test engineers or extracted from scattered test log files, and then manually calculated and organized.

[0022] Furthermore, in the sorting unit 20, a subsystem state sorting sequence is generated based on the state data of the aircraft's multiple subsystems. This step aims to determine the priority of each subsystem during testing. For example, based on the raw values ​​of each subsystem's state data, subsystems with state data exceeding a threshold can be ranked higher, while subsystems with normal state data can be ranked lower, through simple comparison or preset threshold rules. This sorting method can initially reflect the health status of the subsystems, but it may not be able to accurately assess their potential risks.

[0023] Subsequently, in the stimulus action analysis unit 30, an stimulus action optimization model is established based on the Pearson correlation coefficient between the stimulus action and the subsystem state and the test duration of the stimulus action, generating the optimal stimulus action. The goal of this model is to select the stimulus action most suitable for the current subsystem state from the set of stimulus actions. For example, one could simply select the stimulus action with the highest correlation to the current subsystem state, or select the stimulus action with the shortest test duration. This single-index optimization method may not fully consider both the effectiveness and efficiency of the test.

[0024] Next, in the historical data analysis unit 40, data on the changes in historical excitation amplitude and subsystem state are acquired, and an influence curve on the subsystem state is established based on this data. This curve describes the subsystem state response under different excitation intensities. For example, this influence can be approximated by linear interpolation over a limited number of historical test points or by constructing a simple piecewise function.

[0025] Based on this, in the risk analysis unit 50, a test risk score for the aircraft subsystem is generated according to the impact curve of the excitation amplitude on the subsystem state and the amplitude of the optimal excitation action. This score is used to quantify the risks that may be introduced during the test. For example, a fixed risk threshold can be set; when the predicted change in the subsystem state exceeds this threshold, it is directly judged as high risk, otherwise it is low risk. This qualitative or coarse risk assessment method may not provide refined risk management capabilities.

[0026] Furthermore, in the motion amplitude analysis unit 60, a motion amplitude analysis model is established with the motion amplitude of the optimal excitation motion as the independent variable and the test risk score of the aircraft subsystem as the dependent variable, generating the optimal motion amplitude of the optimal excitation motion. This model aims to determine the optimal excitation intensity that controls risk within an acceptable range while ensuring test effectiveness. For example, several fixed motion amplitude values ​​can be preset, and their corresponding risk scores can be calculated for each, selecting the amplitude with the lowest risk score. This discrete amplitude selection method may not find a globally optimal continuous amplitude value.

[0027] Finally, in the excitation and control unit 70, the subsystems in the subsystem state sequence are excited and controlled sequentially based on the optimal excitation action and its optimal amplitude. Specifically, after the amplitude analysis unit 60 is executed, the system returns to the excitation action analysis unit 30 to excite and control the next subsystem in the subsystem state sequence. This process ensures that all subsystems requiring testing can be tested according to the optimized strategy. For example, subsystems could be tested in a preset fixed order without considering changes in their state during testing, or the entire testing process could end directly after testing one subsystem. Such a non-adaptive testing process may not be able to handle dynamic changes that occur during testing.

[0028] In practical implementation, the aircraft comprises three main subsystems: avionics, propulsion, and flight control. Before testing begins, a comprehensive functional and performance verification of these three subsystems is required. First, the test system acquired initial state data for each subsystem: the avionics system reported slight data drift in its navigation module, the power system reported slightly below-normal fuel pump pressure, and the flight control system showed all parameters were within normal range. Simultaneously, the system acquired all available excitation sets (such as "navigation module self-test," "fuel pump pressure cycle test," and "control surface response test"), as well as the historical Pearson correlation coefficients and average test durations of these excitations with the states of each subsystem; this data was centrally stored and managed.

[0029] Next, the system generates a subsystem status ranking sequence based on the acquired subsystem status data. Due to data drift in the avionics system's navigation module, its status evaluation value is calculated as the highest (worst), followed by the power system's fuel pump pressure anomaly, and the flight control system's status evaluation value is the lowest (best). Thus, a test ranking sequence of "avionics system > power system -> flight control system" is generated.

[0030] Subsequently, the system begins processing the first subsystem in the sorted sequence—the avionics system: based on the Pearson correlation coefficient between the excitation action and the avionics system state ("Navigation module self-test" has a high correlation with data drift) and the excitation action test duration ("Navigation module self-test" has a moderate duration), an excitation action optimization model is established. Through this model, the optimal excitation action for the avionics system is calculated to be "Navigation module self-test".

[0031] After determining the optimal excitation action, the impact data of different amplitudes (self-test frequency, signal strength) of the "navigation module self-test" action on the avionics system status (data drift degree) is obtained from historical data. Based on this historical data, an impact curve describing the relationship between the amplitude of the "navigation module self-test" action and the changes in avionics system data drift is established.

[0032] Based on this, and taking into account the impact curve and the current data drift status of the avionics system, a test risk score is generated for the "navigation module self-test" under different maneuver amplitudes.

[0033] Furthermore, using the range of motion for the "navigation module self-test" as the independent variable and the test risk score and test completion score (test sufficiency) of the avionics system as the dependent variables, a range of motion analysis model is established. This model is used to calculate the optimal range of motion for the "navigation module self-test." This optimal range of motion aims to minimize potential risks to the avionics system while ensuring sufficient testing.

[0034] Finally, based on the determined optimal excitation action "navigation module self-test" and its optimal action amplitude, the avionics system is excited and controlled. After the avionics system test is completed, the excitation action optimization operation is returned, and the above process of excitation action optimization operation to the optimal action amplitude operation is repeated for the next subsystem (power system) in the sorted sequence, until all subsystems have completed the test.

[0035] Through the above technical solution, the present invention dynamically integrates real-time status data and quantitatively optimizes excitation parameters, solving the problems of relying on preset rules, lack of real-time data integration, and low efficiency in testing. It can dynamically integrate real-time status data of multiple subsystems of the aircraft, optimize the selection of excitation actions and the setting of action amplitude, thereby improving testing efficiency, safety and reliability, and reducing testing risks and costs.

[0036] For preferred options, please refer to [link / reference]. Figure 2 The present invention further proposes that the sorting unit 20 specifically includes: The status assessment module 21 is used to generate subsystem status assessment values ​​based on the status data of the multi-subsystem of the aircraft. The sequence output module 22 is used to generate a subsystem state sorting sequence based on the subsystem state evaluation value; The subsystem status assessment value is a quantitative indicator used to comprehensively reflect the current health status, performance, or potential risks of a specific subsystem of an aircraft. The concept involves integrating multiple raw status data (such as temperature, pressure, voltage, vibration frequency, etc.) through a specific algorithm or model to generate a single, comparable value. This assessment value can be a dimensionless score or an index with specific physical meaning. Generating the subsystem status assessment value can be done in various ways, such as by using expert experience to set weights for each status data point and performing a weighted sum, or by using machine learning models (such as support vector machines or neural networks) to train historical data to predict the health status of the subsystem and output the assessment value. The subsystem status ranking sequence refers to an ordered list that arranges all subsystems under test according to a preset priority rule based on their status assessment values. The purpose of this sequence is to provide a clear execution order for subsequent excitation and control operations, ensuring that test resources are preferentially allocated to subsystems in poor condition or with higher risks. The generation of the subsystem state ranking sequence can be a simple sorting of the subsystem state assessment values ​​from high to low (or from low to high, depending on the definition of the assessment values), or it can be a process of dividing the subsystems into different risk levels based on the assessment values ​​and then sorting the subsystems within each level.

[0037] Specifically, the state assessment module 21 aims to transform raw, dispersed state data of multiple aircraft subsystems into unified and comparable assessment metrics. Its role is to refine and abstract complex and diverse state information, providing a reliable basis for subsequent ranking. Implementation methods may include: first, preprocessing the raw state data, such as data cleaning, missing value imputation, normalization, or standardization, to eliminate differences in dimensions and ranges; then, a weighted average method can be used, assigning different weights based on the importance of each state data point to the subsystem's health status, and then calculating the weighted average as the assessment value; alternatively, a rule-based expert system can be constructed to judge the state data and provide assessment scores according to preset logical rules.

[0038] The sequence output module 22 is used to arrange the evaluated subsystems in order of importance or urgency, thereby guiding the execution of the testing process. Implementation methods may include: sorting the state evaluation values ​​of all subsystems in descending order, with higher evaluation values ​​appearing earlier in the sequence, indicating a worse state and requiring priority testing; or, setting multiple evaluation value thresholds to classify subsystems into different levels such as "high risk," "medium risk," and "low risk," then sorting the high-risk subsystems first, followed by processing the medium-risk and low-risk subsystems in sequence.

[0039] This application optimizes the subsystem state ranking sequence generation step in the general excitation and control method for integrated testing of multiple subsystems of aircraft by introducing subsystem state evaluation values ​​and a ranking mechanism based on these evaluation values. Specifically, after acquiring the state data of the multiple subsystems of the aircraft, the state evaluation module 21 is first executed to generate subsystem state evaluation values ​​for each subsystem based on these raw state data. This process integrates heterogeneous, multi-dimensional state data into a unified quantitative index, enabling objective and accurate comparison of the health status or risk level between different subsystems. Subsequently, in the sequence output module 22, a subsystem state ranking sequence is further generated based on these generated subsystem state evaluation values. In this way, the ranking sequence is no longer simply based on raw data, but on indicators that have been professionally evaluated and quantified, thus ensuring the scientific and rational nature of the ranking. This evaluation-ranking mechanism allows for priority processing of subsystems with higher state evaluation values ​​(i.e., poorer state or higher risk) during subsequent excitation and control, thereby concentrating limited testing resources on the areas requiring the most attention and significantly improving the efficiency and relevance of the entire integrated testing process.

[0040] In the specific implementation process, the propulsion system in the aircraft subsystem includes three key state data: temperature, pressure, and vibration frequency. In this method, these three state data are first normalized. The temperature data is normalized to the range of 0-1 (0 represents the lower limit of normal operating temperature, and 1 represents the upper limit of dangerous temperature; the pressure data is normalized in the same way). The vibration frequency data is also normalized to the range of 0-1.

[0041] Then, based on historical data analysis, the weights of these three normalized data points were set: temperature weight was 0.4, pressure weight was 0.3, and vibration frequency weight was 0.3.

[0042] The subsystem state evaluation value is obtained by multiplying the normalized data by the corresponding weights and summing the results.

[0043] Specifically, after normalization, the normalized temperature is 0.8, the normalized pressure is 0.7, the normalized vibration frequency is 0.9, and the subsystem state assessment value is 0.8. 0.4 + 0.7 0.3 + 0.9 0.3 = 0.32 + 0.21 + 0.27 = 0.8. Repeat this process for all subsystems to obtain their respective subsystem state evaluation values. Then, sort all subsystem state evaluation values ​​in descending order, with the subsystem having the highest evaluation value at the top, forming a subsystem state ranking sequence: the power system's evaluation value is 0.8, the avionics system's evaluation value is 0.65, and the flight control system's evaluation value is 0.9. The ranking sequence is: flight control system, avionics system, power system.

[0044] Through the above technical solution, this application can transform the complex and diverse state data of multiple subsystems of an aircraft into unified subsystem state evaluation values, thereby overcoming the problems of inconsistent dimensions and information redundancy that may exist when directly sorting based on raw data. The subsystem state ranking sequence generated based on this evaluation value can more accurately and scientifically reflect the true health status and testing priority of each subsystem. This enables priority testing of subsystems with poor conditions or high risks during subsequent excitation and control processes, avoiding blind testing or waste of resources, significantly improving the efficiency and effectiveness of integrated testing of multiple subsystems of an aircraft, ensuring that critical faults can be detected and handled in a timely manner, and thus improving the overall safety and reliability of the aircraft.

[0045] Preferably, the present invention further proposes a method for generating the subsystem state evaluation value, specifically including: Through the formula: ; Generate subsystem state evaluation values ; In the formula, This represents the state assessment value of the i-th subsystem of the aircraft. This represents the normalized value of the j-th state data of the i-th subsystem of the aircraft. This represents the weight coefficient of the j-th state data of the i-th subsystem of the aircraft, and m represents the number of state data of the i-th subsystem of the aircraft. The subsystem status assessment value is a quantitative representation of the current health status or performance level of the i-th subsystem of an aircraft. This value aims to comprehensively reflect the overall impact of various subsystem status data, facilitating a unified comparison and ranking of different subsystems. A higher value indicates a worse subsystem status, potentially more numerous or severe problems, thus requiring a higher testing priority. By introducing this assessment value, complex, multi-dimensional subsystem status information can be simplified into a single, comparable indicator.

[0046] The normalized value of the j-th state data of the i-th subsystem of an aircraft refers to the value obtained by performing a specific mathematical transformation on the original state data so that it falls within a preset range (e.g., 0 to 1 or -1 to 1). The purpose of normalization is to eliminate differences in units and orders of magnitude between different state data, ensuring that all data have the same weight or comparability during the calculation process, and preventing certain data with large numerical ranges from dominating the evaluation results. Common normalization methods include min-max normalization or Z-score normalization. For example, min-max normalization can be used to linearly map the original data to the [0,1] interval; or Z-score normalization can be used to convert the data into a distribution with a mean of 0 and a standard deviation of 1.

[0047] The weighting coefficient of the j-th state data of the i-th subsystem of an aircraft is used to characterize the relative importance or influence of the j-th state data of the i-th subsystem in assessing the overall state of the subsystem. Different state data may have different indicative meanings and degrees of influence on the health status of the subsystem. For example, a temperature anomaly of a critical component may reflect a potential failure of the subsystem more clearly than a slight fluctuation in a non-critical parameter. By assigning appropriate weighting coefficients to each state data, it can be ensured that the more important state data has a greater impact on the final assessment result when calculating the subsystem state assessment value. The weighting coefficients can be determined based on expert experience, historical failure data analysis, sensitivity analysis, or machine learning methods.

[0048] This application's solution effectively resolves the contradiction between the diversity of raw state data and the uniformity of evaluation by introducing a calculation mechanism for subsystem state evaluation values. Specifically, for the state data of the multi-subsystems of the aircraft acquired before the integrated testing of the multi-subsystems, the state data of each subsystem are first normalized to obtain the normalized value of the j-th state data of the i-th subsystem, thus ensuring the comparability of state data with different dimensions and numerical ranges in subsequent calculations and eliminating evaluation bias caused by differences in data scales. Subsequently, according to the different degrees of influence of each state data on the overall state of the subsystem, corresponding weight coefficients are assigned to each normalized value. These weight coefficients reflect the importance of each state data, making key parameters dominant in the evaluation. Finally, by multiplying and summing all the normalized state data of each subsystem with their corresponding weight coefficients, the subsystem state evaluation value of that subsystem is calculated. This evaluation value integrates all relevant state information of the subsystem and considers the importance of each piece of information, thus accurately and quantitatively reflecting the current state of the subsystem. The higher the subsystem state evaluation value, the worse the subsystem state. This provides a reliable basis for generating the subsystem state ranking sequence, enabling the ranking results to more realistically reflect the priority of each subsystem in the comprehensive test, thereby optimizing the excitation and control strategies and improving test efficiency and accuracy.

[0049] The aforementioned technical solution transforms the complex and multi-dimensional state data of multiple aircraft subsystems into a single, physically meaningful subsystem state assessment value. This quantitative assessment method overcomes the limitations of traditional methods, which struggle to uniformly measure different types of state data, resulting in a more objective and accurate assessment of the health status of each subsystem. The subsystem state ranking sequence generated based on this assessment value more accurately reflects the priority of each subsystem in integrated testing, ensuring that test resources are preferentially allocated to subsystems in poorer condition or at higher risk. This avoids wasting test resources or overlooking critical issues due to inaccurate assessments, significantly improving the relevance and efficiency of integrated testing of multiple aircraft subsystems.

[0050] For preferred options, please refer to [link / reference]. Figure 3 The present invention further proposes that the excitation action analysis unit 30 specifically includes: The correlation analysis module 31 is used to generate a correlation score based on the Pearson correlation coefficient between the excitation action and the subsystem state; The time consumption analysis module 32 is used to generate a time score for the stimulus action based on the time consumed by the stimulus action test. The optimization result output module 33 is used to establish an incentive action optimization model based on the correlation score and the incentive action time score, and generate the optimal incentive action. The correlation analysis module 31 aims to quantify the strength of the association between the incentive action and the subsystem state. The Pearson correlation coefficient reflects the degree of linear correlation between two variables; converting it into a score allows for a more intuitive assessment of the incentive action's influence on the subsystem state. This score can be generated by normalizing the original Pearson correlation coefficient and mapping it to a preset scoring range, such as 0 to 1, enabling a fair comparison of the correlations of different incentive actions. Alternatively, it can be divided into different levels based on the absolute value of the correlation coefficient, combined with expert experience or a preset threshold, and assigned a corresponding score. Furthermore, the method for obtaining the Pearson correlation coefficient between the incentive action and the subsystem state is existing technology and will not be elaborated upon here.

[0051] The time consumption analysis module 32 is used to quantify the time cost required to execute a specific incentive action. The testing time of an incentive action is a crucial indicator of testing efficiency; converting it into a score helps to consider efficiency factors when selecting the optimal incentive action. This score can be generated by comparing the testing time of the incentive action with a preset baseline time or threshold to calculate a score reflecting time efficiency—for example, a shorter time results in a higher score. Alternatively, it can be converted into a standardized score based on the average consumption time statistically analyzed from historical data, using methods such as inverse proportional functions or piecewise functions.

[0052] The optimization result output module 33 combines the two quantitative indicators mentioned above to determine the most suitable incentive action. This optimization model can be a multi-objective optimization model, balancing the effectiveness (relevance score) and efficiency (time score) of the incentive action by setting different weight coefficients. For example, a weighted summation method can be used, multiplying the relevance score and the incentive action time score by their respective weight coefficients and then adding them together, selecting the incentive action with the highest total score as the optimal incentive action. Alternatively, a predictive model based on machine learning algorithms such as decision trees and support vector machines can be used to predict the optimal incentive action in the current situation by learning the relationship between the scores of incentive actions and the optimal choice in historical data.

[0053] Specifically, firstly, this application transforms the Pearson correlation coefficient between the incentive action and the subsystem state into a standardized correlation score, enabling the quantification and comparison of the impact of different incentive actions on the subsystem state. Next, the testing time of the incentive action is transformed into an incentive action time score, reflecting the time cost required to execute the incentive action. These two scores quantify the incentive action from both effectiveness and efficiency dimensions. Subsequently, these quantified scores are input into an incentive action optimization model. This model comprehensively considers the correlation between the incentive action and the subsystem state, as well as its testing time. By balancing these two factors, it systematically evaluates and selects the incentive action that is most effective for the current subsystem state and has relatively high testing efficiency, thus generating the optimal incentive action. This step-by-step quantification and model optimization approach ensures that the selection of the optimal incentive action is no longer a simple empirical judgment, but a data-driven, quantifiable decision-making process. This significantly improves the scientific rigor and accuracy of incentive action selection, avoiding the problems of low testing efficiency or poor testing results caused by insufficient consideration of a single indicator.

[0054] In the specific implementation process, for the propulsion system within the aircraft subsystem, there are two excitation actions: A (power adjustment) and B (cooling operation). First, the Pearson correlation coefficient between excitation action A and the subsystem state is calculated to be 0.8, and for excitation action B, it is 0.6. After normalization, the correlation coefficient range is mapped to 0-100 points, with excitation action A receiving a correlation score of 80 and excitation action B receiving 60. The average test duration for excitation action A is 10 seconds, and for excitation action B, it is 25 seconds. Setting a test duration threshold of 30 seconds, the time score for excitation action A is calculated as (30-10) / 30. 100 = 66.7 points. The time score for motivational action B is calculated as (30-25) / 30. 100 = 16.7 points. When establishing the incentive action optimization model, the weight of the relevance score is set to 0.7, and the weight of the time score is set to 0.3. The overall score for incentive action A is 80. 0.7 + 66.7 0.3 = 56 + 20.01 = 76.01 points. The overall score for motivational action B is 60. 0.7 + 16.7 0.3 = 42 + 5.01 = 47.01 points. Based on the comprehensive score comparison, motivational action A was selected as the optimal motivational action.

[0055] Through the above technical solution, this application can quantitatively evaluate the correlation between excitation actions and subsystem states, as well as the time cost required for testing, and establish an optimization model based on this. This enables the systematic and objective selection of the optimal excitation action in integrated testing of multiple subsystems of an aircraft, which effectively reflects changes in subsystem states while also considering testing efficiency. This method avoids the problems of inaccurate testing or low efficiency that may result from selection based solely on experience or a single indicator, significantly improving the scientific nature, accuracy, and overall efficiency of the testing process, thus providing a more reliable basis for subsequent excitation and control.

[0056] Preferably, the present invention further proposes a method for generating the relevance score that specifically includes: Through the formula: ; Generate relevance scores ; In the formula, This represents the correlation score between the stimulus action k and the state of subsystem i. This represents the absolute value of the Pearson correlation coefficient between the excitation action k and the state of subsystem i. This represents the minimum absolute value of the Pearson correlation coefficient between all excitation actions and the state of subsystem i. This represents the maximum absolute value of the Pearson correlation coefficients between all excitation actions and the state of subsystem i; The correlation score is a standardized indicator used to measure the influence of stimulus action k on the state of subsystem i. Its function is to normalize the absolute value of the original Pearson correlation coefficient to a value between 0 and 1, facilitating comparisons between different stimulus actions and subsequent weighted calculations.

[0057] The above formula maps the absolute value of the Pearson correlation coefficient between the stimulus action k and the state of subsystem i to the interval between 0 and 1. Its purpose is to ensure that all correlation scores are compared on a uniform scale, eliminating the influence of the original data's dimensions, and enabling the subsequent stimulus action optimization model to fairly evaluate the effectiveness of different stimulus actions. Implementing this formula first requires collecting the absolute values ​​of the Pearson correlation coefficients between all stimulus actions and the specific state of subsystem i. Then, the maximum and minimum values ​​are found through comparison operations. Finally, the absolute value of the Pearson correlation coefficient for each stimulus action is substituted into the formula for calculation.

[0058] The Pearson correlation coefficient measures the degree of linear correlation between two variables (in this case, the stimulus action k and the state of subsystem i). The absolute value is used to represent the strength of the correlation, regardless of its direction (positive or negative). As raw input data, it reflects the direct impact of the stimulus action on the state of the subsystem. The Pearson correlation coefficient between the execution of stimulus action k and the change in the state of subsystem i in historical data can be calculated using statistical analysis software (such as Python's SciPy library or R language) and then its absolute value can be taken. Alternatively, a dedicated data analysis module can be used to perform correlation analysis on the collected stimulus action data and subsystem state data in real time or offline to derive the coefficient.

[0059] Specifically, for any subsystem i in the aircraft, the absolute values ​​of the Pearson correlation coefficients between all possible excitation actions k and the state of subsystem i are first obtained. Then, the maximum and minimum values ​​are identified among these absolute values. Using these extreme values, the absolute value of the Pearson correlation coefficient for each excitation action k is substituted into the above formula for calculation, thereby generating a correlation score between excitation action k and the state of subsystem i. This score uniformly maps the correlation strength between all excitation actions and the specific subsystem state to the interval between 0 and 1, where 0 represents the weakest correlation and 1 represents the strongest correlation. This normalization process allows for a fair comparison of the correlation strength between different excitation actions on a uniform scale, providing standardized and comparable input parameters for the subsequent establishment of an excitation action optimization model. In this way, the excitation action optimization model can more accurately balance factors such as correlation and test duration, thereby more effectively identifying the excitation actions that have the greatest impact on the subsystem state and the optimal test efficiency, thus improving the accuracy and efficiency of multi-subsystem integrated testing of the aircraft.

[0060] In the specific implementation process, for the flight control system of the aircraft, there are three candidate excitation actions: pitch excitation (action k1), roll excitation (action k2), and yaw excitation (action k3). Through historical data analysis, the absolute values ​​of the Pearson correlation coefficients between these three excitation actions and the attitude control system state can be calculated as 0.75, 0.40, and 0.90, respectively. Based on these data, the minimum absolute value of the Pearson correlation coefficient for attitude control system i is determined to be 0.40, and the maximum value is 0.90. Then, the correlation score for each excitation action is calculated according to the above formula: for pitch excitation (action k1): 0.70; for roll excitation (action k2): 0; for yaw excitation (action k3): 1. Therefore, the yaw excitation is determined to be the action with the strongest correlation to the attitude control system state (score 1), while the roll excitation is determined to be the action with the weakest correlation (score 0), and the pitch excitation is in the middle.

[0061] Through the above technical solution, this application provides a standardized method for generating correlation scores. This method maps the absolute values ​​of the Pearson correlation coefficients between different excitation actions and subsystem states to a uniform range of 0 to 1, effectively eliminating the differences in dimensions and numerical ranges of the original correlation coefficients. This allows for a fair and accurate comparison of the impact of different excitation actions on the subsystem state on a unified scale. This significantly improves the accuracy and reliability of the excitation action optimization model in selecting the optimal excitation action, avoids evaluation bias caused by excessive differences in the original data, and thus optimizes the efficiency and effectiveness of integrated testing of multiple subsystems of aircraft.

[0062] Preferably, the present invention further proposes a method for generating the excitation action time score, specifically including: Through the formula: ; Generate motivational action time score ; In the formula, This represents the time score of the excitation action k used for subsystem i. This represents the test duration for the excitation action k used in subsystem i. This represents the threshold for test duration; The test duration refers to the average time required to complete an stimulus action k when it is applied to subsystem i. This duration is a key parameter for measuring the execution cost of the stimulus action. The test duration can be obtained in various ways. For example, it can be calculated by statistical analysis based on historical test data to determine the average execution time of a specific stimulus action in the past; or it can be estimated or simulated by conducting a detailed analysis of the execution process of the stimulus action, combined with equipment response time, operation steps, etc.

[0063] The test duration threshold is a preset, acceptable maximum test duration. This threshold serves as a benchmark for normalizing the test duration of different stimulus actions, thereby generating a uniform stimulus action time score. The test duration threshold can be determined based on the overall time budget of the actual testing task, the availability of testing resources, industry standards, or empirical values. For example, it can be set as the maximum test duration among all candidate stimulus actions, or a hard time limit can be set based on the urgency of the testing task.

[0064] This application's solution introduces a computational mechanism for excitation action time scoring, enabling the selection of optimal excitation actions during integrated testing of multiple subsystems of an aircraft to take into account the execution efficiency of the excitation actions. Specifically, when it is necessary to excite and control a subsystem in a subsystem state sequence, the set of excitation actions corresponding to that subsystem and the test duration of each excitation action are first obtained. Subsequently, using a preset test duration threshold and the test duration of excitation action k, the excitation action time score is calculated using the aforementioned formula. This formula is designed so that the smaller the test duration, the larger the excitation action time score, thus reflecting the time efficiency advantage of the excitation action. This score, along with the correlation score between excitation action k and the state of subsystem i, is used as input to the excitation action optimization model. In this way, the optimization model can not only identify excitation actions highly correlated with the subsystem state, but also prioritize those excitation actions with shorter execution times and higher efficiency, thereby optimizing the utilization of test resources and the management of the test cycle while ensuring test effectiveness.

[0065] In the specific implementation process, when conducting comprehensive testing of the propulsion system within the aircraft subsystem, it is necessary to evaluate two candidate stimulus actions, A and B. The historical test duration for stimulus action A is 12 minutes, and for stimulus action B, it is 8 minutes. A test duration threshold of 15 minutes is set, and the stimulus action time scores for both are calculated using the formula described above. For stimulus action A, the stimulus action time score is 0.2. For stimulus action B, the stimulus action time score is 0.47. By comparison, the stimulus action time score for stimulus action B is higher than that for stimulus action A, indicating that stimulus action B has a greater advantage in time efficiency.

[0066] Through the above technical solution, in the integrated testing of multiple subsystems of an aircraft, the test duration of excitation actions can be converted into a quantified excitation action time score. This allows for the generation of optimal excitation actions not only to consider the correlation between the excitation action and the subsystem state, but also to effectively assess and weigh the time cost required for their execution. Therefore, this solution avoids selecting excitation actions that, while highly relevant, have excessively long execution times, thus ensuring that the selected optimal excitation actions are both effective and efficient in actual operation. This significantly improves the overall efficiency and practicality of integrated testing of multiple subsystems of an aircraft, helps to shorten the testing cycle, and optimizes the allocation of testing resources.

[0067] Preferably, the present invention further proposes the following expression for the incentive action optimization model: ; In the expression, This represents the optimal excitation action for subsystem i. This represents the correlation score between the stimulus action k and the state of subsystem i. G represents the time score of the excitation action k used for subsystem i, and G represents the set of excitation actions. , All are weighting coefficients, and ; The excitation action optimization model is a mathematical optimization model designed to select an optimal excitation action for aircraft subsystem i from a given set of excitation actions G. The model achieves this selection by maximizing a weighted sum, which consists of a correlation score between excitation action k and the state of subsystem i, and an excitation action time score for excitation action k. The optimal excitation action, calculated by this model, is determined to be the most suitable excitation action for testing subsystem i.

[0068] The set of stimuli G is a complete list of all stimuli that can be used for testing. It can be a predefined static list, such as a list of stimuli stored in a database, or it can be a dynamically generated set based on the current aircraft state, mission requirements, or available resources.

[0069] Weighting coefficient , This setting adjusts the relative importance of relevance scores and incentive action timing scores during the optimization process. Their sum is 1, allowing testers to flexibly prioritize the effectiveness of incentive actions or testing efficiency based on actual needs. For example, in certain critical testing scenarios, this setting can be configured... For larger datasets, prioritize actions with high relevance; however, in time-constrained scenarios, you can set... Larger actions are prioritized based on their shorter execution time. These weighting coefficients can be preset by the system administrator or dynamically configured through the user interface.

[0070] The proposed solution constructs an optimization model for the aforementioned stimulus actions, quantifying and integrating the effectiveness (reflected by correlation scores) and testing efficiency (reflected by stimulus action time scores) of stimulus actions in the integrated testing of multiple subsystems of an aircraft. After acquiring the state data of the multiple subsystems, the set of stimulus actions, the Pearson correlation coefficient between the stimulus actions and the subsystem states, and the testing time consumed by the stimulus actions, the system first generates and sorts the subsystem state evaluation values ​​based on this data, while simultaneously calculating the correlation score between each stimulus action and the subsystem state, and the stimulus action time score. Subsequently, for each subsystem in the sorted sequence, the model iterates through all stimulus actions k in the stimulus action set G and calculates the comprehensive score of each stimulus action according to preset weight coefficients. Finally, the stimulus action with the highest comprehensive score is selected as the optimal stimulus action for that subsystem. This mechanism ensures that the selected stimulus actions not only effectively influence the state of the target subsystem but also have high efficiency during the testing process, thereby achieving the best testing results within limited testing resources and time.

[0071] In the specific implementation process, the flight control subsystem within the aircraft subsystem needs to be tested, and there are three candidate excitation actions: pitch excitation (action k1), roll excitation (action k2), and yaw excitation (action k3). Through preliminary data analysis, the correlation scores between these excitation actions and the attitude control subsystem state, as well as the excitation action timing scores, have been obtained. The specific values ​​are as follows: Motivational action k1: Relevance score = 0.9, Motivational action time score = 0.8; Motivational action k2: Relevance score = 0.6, Motivational action time score = 0.9; Motivational action k3: Relevance score = 0.4, Motivational action time score = 0.95; The testing strategy focuses on relevance, with weighting coefficients a1 and a2 set to 0.6 and 0.4, respectively.

[0072] Calculations are performed based on the expression of the incentive action optimization model: The overall score for the motivational action k1 is 0.86; The overall score for the motivational action k2 is 0.72. The overall score for motivational action k3 is 0.62; By comparison, the overall score of the excitation action A is the highest. Therefore, for the flight control subsystem, the optimal excitation action is determined to be excitation action k1 (pitch excitation).

[0073] Through the above technical solution, this application provides a quantitative and configurable stimulus action selection mechanism that can systematically balance the impact of stimulus actions on the subsystem state with the time cost required for testing. This allows testers to flexibly adjust testing strategies according to actual testing needs and resource constraints, thereby selecting the most suitable stimulus actions for the current testing scenario. This effectively improves the efficiency and relevance of integrated testing of multiple subsystems of aircraft, and avoids the problems of wasted testing resources and poor testing results that may result from blindly or empirically selecting stimulus actions.

[0074] Preferably, the present invention further proposes a method for generating the test risk score of the aircraft subsystem, specifically including: Through the formula: ; Generate test risk scores for aircraft subsystems ; In the formula, This represents the optimal incentive action. The test risk score for aircraft subsystem i when the amplitude of the action is h. This represents the state evaluation value of subsystem i. This represents the change in the subsystem state evaluation value of aircraft subsystem i when the optimal excitation amplitude is h, obtained based on the influence curve of the excitation amplitude on the subsystem state. This represents the state assessment threshold; where, when obtaining the change value of the subsystem state assessment value of aircraft subsystem i based on the influence curve of the excitation action amplitude on the subsystem state, the change in the action amplitude of the optimal excitation action is the difference between the action amplitude h and the initial action amplitude. The test risk score for an aircraft subsystem aims to quantify the potential risk inherent in a test operation when an optimal excitation action of amplitude h is applied to aircraft subsystem i. A higher score indicates a greater test risk. This score allows testers to intuitively assess the safety of tests under different amplitudes of motion.

[0075] The state evaluation value of subsystem i reflects its health status or performance level before testing. This value can be calculated based on various state data of the subsystem through weighted summation or other methods. For example, it can comprehensively consider data from multiple dimensions such as temperature, pressure, vibration, and response time of the subsystem, and assign different weights to reflect their influence on the overall state of the subsystem.

[0076] The change in the subsystem state assessment value of aircraft subsystem i is obtained from the influence curve of the excitation amplitude on the subsystem state, when the amplitude of the optimal excitation is h. This change is predictive, quantifying the potential change in the state assessment value of subsystem i when the optimal excitation is executed with amplitude h. This change can be obtained by consulting or calculating a pre-established influence curve of the excitation amplitude on the subsystem state. For example, this influence curve can be constructed using historical test data, simulation models, or expert experience, describing the functional relationship between the excitation amplitude and the change in the subsystem state assessment value. When obtaining this change value, the change in the amplitude of the optimal excitation is defined as the difference between the current amplitude h and the initial amplitude, meaning that this change reflects the expected deviation relative to the initial state.

[0077] The state assessment threshold serves as a baseline value to normalize the potential state assessment values ​​of subsystem i. This threshold can be set based on aircraft design specifications, safety standards, or historical operational data, representing the upper limit of acceptable state assessment values ​​for the subsystem. For example, it can be set as the maximum state assessment value allowed for the subsystem under normal operation or safety testing conditions.

[0078] This application's solution effectively addresses the problem of quantifying test risks in integrated testing of multiple aircraft subsystems by introducing a test risk score for the aircraft subsystems. After acquiring the state data of the multiple aircraft subsystems and generating subsystem state assessment values, and after generating the optimal stimulus action based on the Pearson correlation coefficient between the stimulus action and the subsystem state and the test duration of the stimulus action, this solution further utilizes historical stimulus action amplitude and subsystem state change data to establish an influence curve on the subsystem state. Based on this influence curve, for any given amplitude h of the optimal stimulus action, the change in the state assessment value of subsystem i can be predicted. Subsequently, the current state assessment value of subsystem i is added to the predicted change to obtain a comprehensive value reflecting the potential state of the subsystem after testing. By comparing this comprehensive value with a preset state assessment threshold, a test risk score for the aircraft subsystem can be generated. This calculation method allows the test risk score to comprehensively consider the initial health status of the subsystem and the potential impact of a specific stimulus action amplitude, thus providing a quantitative and comprehensive risk assessment. Its ingenuity lies in combining static subsystem state assessment with dynamic stimulus effects, allowing testers to consider not only the effectiveness of the stimulus but also its safety when selecting the stimulus amplitude, thus avoiding the risks of blind or overtesting.

[0079] In the specific implementation process, a comprehensive test was conducted on the aircraft's propulsion system. First, based on the sensor data (temperature, vibration, etc.) of the actuator, its current state assessment value was calculated using a weighted summation formula, which was 0.6. Simultaneously, based on historical data analysis, "adjusting the power" was determined to be the optimal excitation action for the actuator. Next, based on historical test data, an influence curve describing the relationship between the amplitude of the "adjusting the power" action (power adjustment amount) and the change in the actuator's state assessment value was established. According to this curve, when the amplitude of the optimal excitation action is set to +5 km / h, by referring to or calculating the influence curve, it can be found that when the amplitude is 5 Hz, the change in the state assessment value of actuator subsystem i is 0.2. The preset state assessment threshold is 1.0. At this point, the test risk score for the aircraft's propulsion system is 0.8.

[0080] Through the above technical solution, this application provides a quantitative risk assessment mechanism for integrated testing of multiple subsystems of aircraft. This mechanism comprehensively considers the initial state of the subsystem and the potential state changes caused by specific excitation amplitudes, thereby generating an intuitive test risk score. This allows testers to have a basis for selecting the optimal excitation amplitude, avoiding the uncertainty that may arise from relying solely on experience. Specifically, by quantitatively assessing the test risk under different amplitudes, unnecessary damage to the aircraft subsystem due to excessively large test amplitudes can be effectively avoided, while also preventing low test efficiency or failure to fully expose potential problems due to overly conservative test amplitudes. Therefore, this solution helps improve the safety, effectiveness, and efficiency of integrated testing of multiple subsystems of aircraft, providing important quantitative basis for test decisions.

[0081] Preferably, the present invention further proposes the following expression for the motion amplitude analysis model: ; In the expression, This represents the optimal incentive action. The optimal range of motion, This represents the optimal incentive action. The test risk score for aircraft subsystem i when the amplitude of the action is h. This represents the optimal incentive action. The test completion score is given when the range of motion is h, and F represents the optimal motivational action. The set of range of motion, , All are weighting coefficients, and ; Among them, the motion amplitude analysis model aims to quantify the test effect under different motion amplitudes by comprehensively considering test risk and test completion, thereby determining the optimal incentive motion amplitude.

[0082] Test completion score measures the degree of completion or effectiveness of a test task at a specific action range h. Generally, a larger action range may indicate a higher test completion rate because it can more comprehensively expose problems in the subsystem. The test completion score can be calculated based on the relationship between a preset minimum action range requirement and the current action range setting.

[0083] The set of action amplitudes, F, refers to the set of all possible action amplitudes that can be adopted for the optimal excitation action. This set can be discrete, such as a few preset levels, or continuous, such as a numerical range. The determination of this set can be based on the design specifications of the aircraft subsystem, historical test data, or expert experience.

[0084] Weighting coefficient , This is used to adjust the relative importance of test risk score and test completion score in the action amplitude analysis model. For example, when more emphasis is placed on test safety, the importance can be appropriately increased. The value can be appropriately increased when more emphasis is placed on the thoroughness of the test. The sum of these two coefficients is usually set to 1 to maintain the normalization of the weights.

[0085] This application's solution addresses the problem of determining the specific execution intensity of the optimal excitation action by introducing an action amplitude analysis model. This model uses the action amplitude *h* of the optimal excitation action as the independent variable, comprehensively considering the test risk score and test completion score of aircraft subsystem *i* under that action amplitude. This is achieved by introducing weighting coefficients. , The model allows for flexible adjustment of the priority of risk and completion in the decision-making process based on actual testing needs. For example, in testing scenarios with extremely high safety requirements, a higher weight can be assigned to the risk score. This model minimizes testing risk while ensuring testing effectiveness by finding the action magnitude that minimizes the objective function (i.e., the weighted sum of the risk score and the completion score). This approach enables testers to intelligently select the most appropriate stimulus intensity based on the specific state of the aircraft subsystem and the testing objectives, avoiding the uncertainty and suboptimal results that may result from testing based on experience or fixed parameters.

[0086] In the specific implementation process, the optimal excitation action targets the aircraft's propulsion system, and its action amplitude can take values ​​from a preset set F, for example, F = {+2km / h, +5km / h, +8km / h, +10km / h, ..., +30km / h}. For each action amplitude in set F, the system first calculates the corresponding test risk score based on the generation method of the test risk score for the propulsion system when the optimal excitation action amplitude is [value missing]. Simultaneously, it calculates the corresponding test completion score based on the generation method of the test completion score when the optimal excitation action amplitude is +5km / h. Then, these scores are substituted into the expression of the action amplitude analysis model. In the current test, the focus on test risk is slightly higher than that on test completion, with weighting coefficients of 0.6 and 0.4 set. The system iterates through all action amplitudes in F, calculates the weighted sum corresponding to each action amplitude, and selects the action amplitude that minimizes the weighted sum as the optimal action amplitude.

[0087] Through the above technical solution, this application can intelligently determine the optimal amplitude of the optimal excitation action based on the specific state and testing requirements of the aircraft subsystem. This solution comprehensively considers testing risk and test completion, avoiding the risks of insufficient or excessive testing that may result from single-dimensional decision-making. By quantifying risk and completion and performing weighted optimization, the testing process becomes more accurate and efficient, effectively reducing the possibility of unnecessary damage to the aircraft subsystem during testing, while ensuring the comprehensiveness and reliability of test results, thereby improving the overall quality and safety of integrated testing of multiple aircraft subsystems.

[0088] Preferably, the present invention further proposes a method for generating the test completion score, specifically including: Through the formula: ; Generate test completion score ; In the formula, This represents the optimal incentive action. The range of motion setting value, This refers to the optimal excitation action during the integrated testing of multiple subsystems of an aircraft. Minimum required range of motion; The optimal stimulus amplitude setting value refers to the specific amplitude value set for the optimal stimulus action during integrated testing of multiple subsystems of an aircraft. This value is a variable to be optimized, and its magnitude directly affects the test risk and test completion rate.

[0089] In integrated testing of multiple subsystems of an aircraft, the minimum required amplitude of the optimal excitation action represents the minimum amplitude that the optimal excitation action must achieve to ensure the effectiveness and adequacy of the test. A amplitude below this requirement may lead to insufficient testing and failure to fully expose potential problems in the subsystems. This minimum requirement can be determined by design specifications, industry standards, historical testing experience, or expert knowledge. For example, for a valve action, the minimum requirement might be a full opening or closing stroke; for a motor, it might be reaching a certain minimum speed or output torque.

[0090] This application's solution improves the aforementioned optimal motion amplitude analysis model by introducing a specific calculation method for test completion score. When determining the optimal motion amplitude for the optimal stimulus action, this solution compares the setpoint of the optimal stimulus action's motion amplitude with the minimum requirement for the motion amplitude of the optimal stimulus action in the integrated testing of multiple subsystems of an aircraft. Specifically, the closer the motion amplitude setpoint is to or reaches the minimum requirement, the smaller the numerator value, resulting in a lower test completion score. In the optimal motion amplitude analysis model, since the objective is to minimize this weighted sum, it tends to select motion amplitudes that result in a smaller test completion score, i.e., it tends to select motion amplitudes that meet or exceed the minimum requirements. Simultaneously, the model also considers test risk score, thereby minimizing test risk while meeting test completion requirements. This combination makes the selection of the optimal motion amplitude more comprehensive and reasonable, avoiding the problem of only considering risk while ignoring test adequacy, and also avoiding blindly pursuing high amplitudes that increase unnecessary risks.

[0091] During the implementation process, when conducting comprehensive testing of the aircraft's propulsion system, the optimal excitation action was determined to be "adjusting power," and the minimum required amplitude of this action was +6 km / h. The set amplitude of the action was +5 km / h, and the test completion score was 0.167.

[0092] Through the above technical solution, the impact of the optimal excitation amplitude setting value on the test completion rate can be clearly quantified in the integrated testing of multiple subsystems of an aircraft. This solution ensures that the optimal amplitude analysis model, when selecting the optimal amplitude, not only considers test risks but also the sufficiency and effectiveness of the test, guaranteeing that the selected amplitude meets or exceeds the preset minimum test requirements. This helps avoid problems such as incomplete testing or inaccurate test results due to insufficient amplitude, thereby improving the quality and reliability of integrated testing, making the test results more instructive, and providing a more solid foundation for the safe operation of the aircraft.

[0093] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A universal excitation and control system for integrated testing of multiple subsystems of an aircraft, characterized in that, The system specifically includes: The data acquisition unit is used to acquire the state data of the aircraft's multi-subsystem, the set of excitation actions during the integrated test of the aircraft's multi-subsystem, the Pearson correlation coefficient between the excitation actions and the subsystem state, and the test duration of the excitation actions; among which, the state data of the aircraft's multi-subsystem refers to the state data of the aircraft's multi-subsystem before the integrated test. The sorting unit is used to generate a subsystem state sorting sequence based on the state data of the multi-subsystem of the aircraft. The stimulus action analysis unit is used to establish an stimulus action optimization model based on the Pearson correlation coefficient between the stimulus action and the subsystem state and the stimulus action test consumption time, and to generate the optimal stimulus action; whereby the stimulus action test consumption time refers to the average consumption time of stimulus action tests in historical data; The historical data analysis unit is used to acquire historical data on the changes in the amplitude of the stimulus action and the state of the subsystem, and to establish an impact curve on the state of the subsystem based on the historical data on the changes in the amplitude of the stimulus action and the state of the subsystem. The historical data on the changes in the amplitude of the stimulus action and the state of the subsystem refers to the changes in the state of the subsystem when the amplitude of the stimulus action changes in the historical data. The risk analysis unit is used to generate a test risk score for the aircraft subsystem based on the impact curve of the excitation action amplitude on the subsystem state and the action amplitude of the optimal excitation action. The motion amplitude analysis unit is used to establish a motion amplitude analysis model with the motion amplitude of the optimal excitation motion as the independent variable and the test risk score of the aircraft subsystem as the dependent variable, and to generate the optimal motion amplitude of the optimal excitation motion. The excitation and control unit is used to excite and control the subsystems in the subsystem state sorting sequence in sequence according to the optimal excitation action and the optimal action amplitude of the optimal excitation action.

2. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 1, characterized in that, The sorting unit specifically includes: The status assessment module is used to generate subsystem status assessment values ​​based on the status data of the aircraft's multiple subsystems. The sequence output module is used to generate a sorted sequence of subsystem states based on the subsystem state evaluation values.

3. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 2, characterized in that, The specific methods for generating the subsystem state evaluation value include: Through the formula: ; Generate subsystem state evaluation values ; In the formula, This represents the state assessment value of the i-th subsystem of the aircraft. This represents the normalized value of the j-th state data of the i-th subsystem of the aircraft. represents the weight coefficient of the j-th state data of the i-th subsystem of the aircraft, and m represents the number of state data of the i-th subsystem of the aircraft.

4. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 1, characterized in that, The incentive action analysis unit specifically includes: The correlation analysis module is used to generate correlation scores based on the Pearson correlation coefficient between the excitation action and the subsystem state. The time consumption analysis module is used to generate a time score for the incentive action based on the time consumed in the incentive action test; The optimization result output module is used to establish an optimization model for incentive actions based on the correlation score and the incentive action time score, and generate the optimal incentive action.

5. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 4, characterized in that, The specific methods for generating the relevance score include: Through the formula: ; Generate relevance scores ; In the formula, This represents the correlation score between the stimulus action k and the state of subsystem i. This represents the absolute value of the Pearson correlation coefficient between the excitation action k and the state of subsystem i. This represents the minimum absolute value of the Pearson correlation coefficient between all excitation actions and the state of subsystem i. This represents the maximum absolute value of the Pearson correlation coefficient between all excitation actions and the state of subsystem i.

6. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 4, characterized in that, The specific methods for generating the incentive action time score include: Through the formula: ; Generate motivational action time score ; In the formula, This represents the time score of the excitation action k used for subsystem i. This represents the test duration for the excitation action k used in subsystem i. This represents the threshold for test duration.

7. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 4, characterized in that, The specific expression of the incentive action optimization model is as follows: ; In the expression, This represents the optimal excitation action for subsystem i. This represents the correlation score between the stimulus action k and the state of subsystem i. G represents the time score of the excitation action k used for subsystem i, and G represents the set of excitation actions. , All are weighting coefficients, and .

8. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 3, characterized in that, The specific methods for generating the test risk score for the aircraft subsystem include: Through the formula: ; Generate test risk scores for aircraft subsystems ; In the formula, This represents the optimal incentive action. The test risk score for aircraft subsystem i when the amplitude of the action is h. This represents the state evaluation value of subsystem i. This represents the change in the subsystem state evaluation value of aircraft subsystem i when the optimal excitation amplitude is h, obtained based on the influence curve of the excitation amplitude on the subsystem state. This represents the state assessment threshold; where, when obtaining the change value of the subsystem state assessment value of aircraft subsystem i based on the influence curve of the excitation action amplitude on the subsystem state, the change in the action amplitude of the optimal excitation action is the difference between the action amplitude h and the initial action amplitude.

9. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 1, characterized in that, The specific expression for the motion amplitude analysis model is as follows: ; In the expression, This represents the optimal incentive action. The optimal range of motion, This represents the optimal incentive action. The test risk score for aircraft subsystem i when the amplitude of the action is h. This represents the optimal incentive action. The test completion score is given when the range of motion is h, and F represents the optimal motivational action. The set of range of motion, , All are weighting coefficients, and .

10. The universal excitation and control system for integrated testing of multiple subsystems of an aircraft according to claim 9, characterized in that, The specific methods for generating the test completion score include: Through the formula: ; Generate test completion score ; In the formula, This represents the optimal incentive action. The range of motion setting value, This refers to the optimal excitation action during the integrated testing of multiple subsystems of an aircraft. The minimum range of motion required.

Images