A thermal control system evaluation method and device

Abstract

The present application relates to a method and device for evaluating a thermal control system, and to the field of spacecraft technology. The method comprises: determining key indicator data corresponding to a target component, obtaining a normal sample data set and a fault sample data set of the thermal control system for the key indicator data, performing feature extraction and analysis processing on the key indicator data based on the normal sample data set and the fault sample data set to obtain an indicator evaluation result of the target component, and determining a state evaluation result of the thermal control system based on the indicator evaluation result, thereby being able to effectively evaluate the health status of the thermal control system and its key components, and solving the problem that the prior art lacks means for effectively evaluating the health status of the entire spacecraft thermal control system and its components.

Description

Technical Field

[0001] The present application relates to the field of spacecraft technology, and in particular to a method and device for evaluating a thermal control system. Background Art

[0002] At present, the space station, as a manned spacecraft, is one of the important directions for the innovation and development of aerospace technology. As a key component of the space station, the thermal control system of the spacecraft thermal control subsystem must not only ensure that the spacecraft thermal control subsystem can achieve the overall required performance and functions, but also ensure that the sealed cabin where the astronauts are located meets certain temperature and humidity conditions, as well as the gas flow rate in the cabin. That is, to ensure that its structural components, instruments and equipment are in a suitable temperature range in the space environment and can work normally. If the thermal control system fails in orbit, it will seriously affect the safety of the space station.

[0003] The existing health management methods for spacecraft thermal control systems mainly use automatic interpretation technology to achieve abnormal interpretation of parameters and events. However, the existing health management methods for spacecraft thermal control systems cannot fully utilize test data, resulting in the inability to timely grasp the health status of the thermal control system and its key components, and lack of effective means to evaluate the health status of the entire thermal control system and its components. Summary of the invention

[0004] The present application provides a method and device for evaluating a thermal control system, which can effectively evaluate the health status of the thermal control system and its key components, and solve the problem that the prior art lacks effective and comprehensive evaluation means.

[0005] In a first aspect, the present application provides a method for evaluating a thermal control system, comprising:

[0006] Determine the key indicator data corresponding to the target component;

[0007] According to the key indicator data, a normal sample data set and a fault sample data set of the thermal control system are obtained;

[0008] Based on the normal sample data set and the fault sample data set, feature extraction and analysis processing are performed on the key indicator data to obtain an indicator evaluation result of the target component;

[0009] Based on the indicator evaluation result, a state evaluation result of the thermal control system is determined.

[0010] Optionally, the key indicator data corresponding to the target component is determined, including:

[0011] Obtaining preset index parameters corresponding to the target component;

[0012] Based on the indicator parameters, data is collected from the data interface corresponding to the target component in a preset data collection method to obtain the key indicator data.

[0013] Optionally, before acquiring the normal sample data set and the fault sample data set of the thermal control system, the method further includes:

[0014] Based on the index parameters, a simulation test is performed on the thermal control system to obtain the normal sample data set;

[0015] Based on the indicator parameters and in combination with preset operating condition information, a fault simulation process is performed on the thermal control system to obtain the fault sample data set, wherein the fault sample data set includes fault sample data corresponding to the operating condition information.

[0016] Optionally, the performing feature extraction and analysis processing on the key indicator data based on the normal sample data set and the fault sample data set to obtain the indicator evaluation result of the target component includes:

[0017] Constructing a curve based on the normal sample data set and the fault sample data set to obtain slip curve information;

[0018] Extracting state features according to the key indicator data and the slip curve information to obtain state information corresponding to the key indicator data;

[0019] Based on the status information, an indicator evaluation result of the target component is determined.

[0020] Optionally, the constructing a curve based on the normal sample data set and the fault sample data set to obtain the slip curve information includes: constructing a curve based on the normal sample data set to obtain first slip curve information; constructing a curve based on the fault sample data set to obtain second slip curve information;

[0021] The extracting state features based on the key indicator data and the slip curve information to obtain the state information corresponding to the key indicator data includes: extracting state features based on the first slip curve information, the second slip curve information and the key indicator data according to a preset feature analysis method to obtain the state information corresponding to the key indicator data.

[0022] Optionally, the state feature extraction is performed using the first slip curve information, the second slip curve information and the key indicator data according to a preset feature analysis method to obtain the state information corresponding to the key indicator data, including:

[0023] If the feature analysis method is a vector machine analysis method, a slip mean parameter and a slip variance parameter are determined based on the first slip curve information and the second slip curve information;

[0024] Based on the slip mean parameter and the slip variance parameter, feature extraction is performed through a preset analysis model to obtain spatial distribution information;

[0025] Extract key parameter information from the key indicator data;

[0026] The state information is obtained by performing statistical analysis based on the key parameter information and the spatial distribution information.

[0027] Optionally, the state feature extraction is performed using the first slip curve information, the second slip curve information and the key indicator data according to a preset feature analysis method to obtain the state information corresponding to the key indicator data, including:

[0028] If the characteristic analysis method is a dynamic threshold analysis method, determining a sliding mean parameter and a sliding standard deviation parameter based on the first sliding curve information and the second sliding curve information;

[0029] Determining a target interval threshold based on the sliding mean parameter and the sliding standard deviation parameter;

[0030] The status information is obtained by performing statistical analysis based on the key indicator data and the target interval threshold.

[0031] Optionally, determining the index evaluation result of the target component based on the state information includes:

[0032] Performing affiliation mapping on the status information according to a preset mapping method to determine indicator evaluation level information;

[0033] Fuzzy mapping is performed based on the indicator evaluation level information to obtain the indicator evaluation result.

[0034] Optionally, determining the state evaluation result of the thermal control system based on the indicator evaluation result includes:

[0035] Obtaining a target state corresponding to the target component;

[0036] For the target state, using a preset adaptive weight method in combination with the indicator evaluation result, determine the weight information of the target component;

[0037] Performing weighted processing based on the weight information and the indicator evaluation result to obtain a subsystem evaluation result corresponding to the target component;

[0038] Based on the subsystem evaluation results, an evaluation is performed in accordance with a preset evaluation method to obtain a state evaluation result of the thermal control system.

[0039] In a second aspect, the present application provides an evaluation device for a thermal control system, comprising:

[0040] A key indicator data determination module is used to determine the key indicator data corresponding to the target component;

[0041] A sample data set determination module, used for acquiring a normal sample data set and a fault sample data set of the thermal control system according to the key indicator data;

[0042] An indicator evaluation result determination module is used to perform feature extraction and analysis processing on the key indicator data based on the normal sample data set and the fault sample data set to obtain an indicator evaluation result of the target component;

[0043] A status evaluation result determination module is used to determine the status evaluation result of the thermal control system based on the indicator evaluation result.

[0044] In summary, the embodiments of the present application determine the key indicator data corresponding to the target component, obtain the normal sample data set and the fault sample data set of the thermal control system for the key indicator data, perform feature extraction and analysis on the key indicator data based on the normal sample data set and the fault sample data set, and obtain the indicator evaluation result of the target component; based on the indicator evaluation result, determine the state evaluation result of the thermal control system, so as to effectively evaluate the health status of the thermal control system and its key components, thereby solving the problem that the prior art lacks the means to effectively evaluate the health status of the entire spacecraft thermal control system and its components. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0046] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0047] Figure 1 A schematic diagram of a flow chart of a method for evaluating a thermal control system provided in an embodiment of the present application;

[0048] Figure 2 It is a schematic flow chart of the steps of a method for evaluating a thermal control system provided by an optional embodiment of the present application;

[0049] Figure 3 is a flow chart of a method for evaluating a thermal control system provided by an optional embodiment of the present application;

[0050] Figure 4 is a flow chart of a method for evaluating a thermal control system provided by an optional embodiment of the present application;

[0051] Figure 5 A structural block diagram of an evaluation device for a thermal control system provided in an embodiment of the present application;

[0052] Figure 6 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0054] In related technologies, high reliability is a basic requirement for manned spacecraft. For the thermal control system, its reliability directly affects the life of the entire space station. The space station has a long time in orbit and high reliability requirements. If the thermal control system fails in orbit, it will seriously affect the safety of the space station. Fault monitoring and health status assessment of the thermal control system are key technologies for active reliability design and on-orbit maintenance.

[0055] One of the concepts of the embodiments of the present application is to propose a method for evaluating a thermal control system, by determining key indicator data corresponding to a target component, and obtaining a normal sample data set and a fault sample data set of the thermal control system for the key indicator data, and performing feature extraction and analysis on the key indicator data based on the normal sample data set and the fault sample data set to obtain an indicator evaluation result of the target component, and then determining a state evaluation result of the thermal control system based on the indicator evaluation result, thereby achieving an effective evaluation of the health status of the entire thermal control system and its components, making full use of test data such as sample data sets, so as to achieve the purpose of timely grasping the health status of the thermal control system and its key components, and solving the problem that the prior art lacks means for effectively evaluating the health status of the entire spacecraft thermal control system and its components.

[0056] To facilitate the understanding of the embodiments of the present application, further explanation will be given below in conjunction with the drawings and specific embodiments. The embodiments do not constitute a limitation on the embodiments of the present application.

[0057] Figure 1 A schematic diagram of a flow chart of a method for evaluating a thermal control system provided in an embodiment of the present application. Figure 1 As shown, the evaluation method of the thermal control system provided in the embodiment of the present application may specifically include the following steps:

[0058] Step 110, determining key indicator data corresponding to the target component.

[0059] Specifically, the target component may include a component that needs to be evaluated, and the embodiment of the present application does not limit this; the key indicator data may include indicator data to be evaluated, and the indicator data to be evaluated may be indicator data collected from the target component and need to be evaluated. The target component may correspond to one or more key indicator data, and the embodiment of the present application does not limit this. Specifically, the embodiment of the present application may determine the component that needs to be evaluated as the target component, and the target component may correspond to one or more indicators, and data collection may be performed on the target component according to the indicator to obtain the key indicator data of the indicator.

[0060] In a specific implementation, the thermal control system in the embodiment of the present application may include a spacecraft thermal control system, that is, the evaluation method of the thermal control system provided in the embodiment of the present application may be applied to the spacecraft thermal control system, so as to evaluate the spacecraft thermal control system through the evaluation method of the thermal control system, and realize the health status evaluation of the spacecraft thermal control system. Specifically, the spacecraft thermal control system may include at least one subsystem, such as a system controller subsystem, a ventilation subsystem, a fluid circuit subsystem, and an electric heating temperature control subsystem, each of which may include at least one component, such as a fluid circuit subsystem may include components such as an external circuit pump and a controller, and the embodiment of the present application does not limit this. When evaluating the thermal control system, the components included in the subsystem may be used as target components, and then the corresponding indicators of the target components may be determined. For example, with the external circuit pump and controller as the target components, its indicators may include the external circuit temperature control value and the external circuit pump group speed, and this example does not limit this. Data collection can be performed based on the indicator through the data interface corresponding to the external circuit pump and the controller, and the indicator data of the external circuit temperature control value and the indicator data of the external circuit pump group speed are collected as key indicator data.

[0061] In actual processing, the embodiment of the present application can preset a collection time, collect data on the target component within the collection time, and obtain key indicator data for a certain period of time, so that feature extraction and analysis can be performed on the key indicator data later.

[0062] Step 120: acquiring a normal sample data set and a fault sample data set of the thermal control system with respect to the key indicator data.

[0063] Specifically, the fault sample data set may include fault sample data under different working conditions, wherein the working conditions may include high temperature, low temperature, manned and unmanned working conditions, etc., and the embodiments of the present application do not limit this. Specifically, after the embodiments of the present application determine the key indicator data, the normal sample data set and the fault sample data set of the thermal control system may be obtained for the key indicator data. For example, in the case where the target component is an external circuit pump and a controller, the normal sample data set and the fault sample data set corresponding to the external circuit pump and the controller may be obtained for the key indicator data corresponding to the external circuit pump and the controller, wherein the normal sample data set may include the normal sample data corresponding to each indicator, such as the normal sample data corresponding to the external circuit temperature control value and the normal sample data corresponding to the speed of the external circuit pump group, and the fault sample data set may include the fault sample data corresponding to each indicator under different working conditions, such as the fault sample data corresponding to the external circuit temperature control value and the fault sample data corresponding to the speed of the external circuit pump group under different working conditions.

[0064] Step 130: Based on the normal sample data set and the fault sample data set, feature extraction and analysis are performed on the key indicator data to obtain an indicator evaluation result of the target component.

[0065] Specifically, the index evaluation result of the target component may include the component evaluation level and the component health, such as the component evaluation level can be divided into five levels such as severe danger (YV), danger (VX), severe degradation (YT), degradation (TH) and health (JK), which is not limited in the embodiment of the present application. Specifically, the embodiment of the present application can generate a normal parameter curve corresponding to the normal sample data set, a fault parameter curve corresponding to the fault sample data set and a parameter curve to be evaluated corresponding to each indicator in the key indicator data based on the normal sample data set, the fault sample data set and the key indicator data, respectively, and then combine the various parameter curves to perform feature extraction analysis, analyze the data distribution of the key indicator data under normal conditions and fault conditions, and then determine the index evaluation level corresponding to each key indicator data corresponding to the target component, and determine the component evaluation level of the target component according to the index evaluation level as the index evaluation result. Among them, the index evaluation level can include danger (VX), degradation (TH) and health (JK). It can effectively utilize test data such as normal sample data sets and fault sample data sets to improve the accuracy of the evaluation.

[0066] It should be noted that, in the embodiment of the present application, when the component evaluation level of the target component is JK, it can be determined that the target component is in a normal state (i.e., a healthy state), and the target component is a normal component at this time; when the component evaluation level of the target component is non-JK (including YV, VX, YT and TH), it can be determined that the target component is in an abnormal state (i.e., a faulty state), and the target component is an abnormal component (i.e., a faulty component), and the embodiment of the present application does not limit this.

[0067] For example, for the external loop pump and controller, taking the external loop temperature control value as an example, a curve can be constructed based on the normal sample data, fault sample data and key indicator data corresponding to the external loop temperature control value to obtain the normal sample curve of the external loop temperature control value in the normal state (constructed based on the normal sample data), the fault sample curve in the abnormal state (constructed based on the fault sample data) and the indicator curve corresponding to the key indicator data. Subsequently, the normal sample curve and the fault sample curve corresponding to the external loop temperature control value can be analyzed to determine that the temperature assignment of the external loop temperature control value in the normal state is a smooth curve, and the temperature assignment decreases in the fault state, indicating that the temperature control ability of the external loop pump and controller in the fault state decreases, and the external loop temperature control value fluctuates significantly. Therefore, the normal sample curve, fault sample curve and indicator curve corresponding to the external loop temperature control value can be analyzed to determine the indicator evaluation level corresponding to the external loop temperature control value. Specifically, feature extraction and analysis can be performed based on the normal sample curve and the fault sample curve in combination with the index curve to determine whether the key indicator data corresponding to the external loop temperature control value is normal state data or abnormal state data, and then the distribution of the key indicator data corresponding to the external loop temperature control value in normal state and abnormal state can be determined, and the health degree corresponding to the external loop temperature control value can be determined based on the distribution, and the index evaluation level corresponding to the external loop temperature control value can be determined based on the health degree corresponding to the external loop temperature control value. For the speed of the external loop pump group, the curve construction and analysis can be performed in accordance with the analysis method of the external loop temperature control value to determine whether the key indicator data corresponding to the speed of the external loop pump group is normal state data or abnormal state data, and then the distribution of the key indicator data corresponding to the speed of the external loop pump group in normal state and abnormal state can be determined, and the health degree corresponding to the speed of the external loop pump group can be determined based on the distribution, and then the index evaluation level corresponding to the speed of the external loop pump group can be determined based on the health degree corresponding to the speed of the external loop pump group. Subsequently, the component health and component evaluation level corresponding to the target component can be determined based on the index evaluation level corresponding to the external loop temperature control value and the index evaluation level corresponding to the external loop pump group speed, and the component health and component evaluation level can be used as the index evaluation results corresponding to the external loop pump and controller.

[0068] Step 140: Determine a state evaluation result of the thermal control system based on the indicator evaluation result.

[0069] Specifically, after determining the indicator evaluation result corresponding to the target component, the embodiment of the present application can determine the status evaluation result of the thermal control system based on the indicator evaluation result, and structurally decompose the health evaluation problem of the thermal control system performance, select key components and important indicators to form a health evaluation indicator system, and realize the health status evaluation of the thermal control system and its components from the indicator level, component level, and system level step by step. It can accurately evaluate the health of each component and the health of the system, and assist in locating the root component of the abnormal problem, effectively solving the problem of lack of means for evaluating the health status of the thermal control system and its components, and can timely and effectively obtain the health status of the thermal control system and its components, thereby improving the accuracy and effectiveness of the evaluation.

[0070] For example, a spacecraft thermal control system may include multiple subsystems, each subsystem may include multiple components, each component may be evaluated as a target component, and the indicator evaluation result corresponding to each target component may be determined. Subsequently, the subsystem evaluation result may be determined based on the indicator evaluation result corresponding to the target component, and then the health status of the entire spacecraft thermal control system may be determined based on the subsystem evaluation result to obtain a status evaluation result.

[0071] It can be seen that the embodiment of the present application determines the key indicator data corresponding to the target component, obtains the normal sample data set and the fault sample data set of the thermal control system for the key indicator data, and then performs feature extraction and analysis on the key indicator data based on the normal sample data set and the fault sample data set to obtain the indicator evaluation result of the target component, and then determines the state evaluation result of the thermal control system based on the indicator evaluation result, so as to effectively evaluate the health status of the thermal control system and its key components, thereby solving the problem that the prior art lacks the means to effectively evaluate the health status of the entire spacecraft thermal control system and its components.

[0072] Reference Figure 2 , shows a schematic flow chart of the steps of a thermal control system evaluation method provided by an optional embodiment of the present application. The thermal control system evaluation method may specifically include the following steps:

[0073] Step 210, determining key indicator data corresponding to the target component.

[0074] In a specific implementation, the embodiment of the present application can construct an index system for the evaluation of the thermal control system in advance. Figure 3, an indicator system can be constructed for the health status assessment of the spacecraft thermal control system, such as selecting the evaluable subsystems and key components that affect the health of the thermal control system, as well as key indicator parameters that can reflect the health of the key components. In this way, the health assessment problem structure of the spacecraft thermal control system performance can be decomposed through the spacecraft thermal control system health status assessment algorithm, and the key components and important indicators can be selected to establish a health assessment indicator system. The indicator system can be used for subsequent assessment to obtain the assessment results.

[0075] As an example, refer to Table 1, which shows the thermal control system evaluation index system provided in the embodiment of the present application. Specifically, the subsystems mainly involved in the evaluation of the thermal control system may include four subsystems, namely, the system controller, the ventilation system, the fluid circuit, and the electric heating temperature control system, which are not limited in this example. Among them, each subsystem can have corresponding components, and the components can correspond to one or more indicators, which are not limited in this example. When evaluating the thermal control system, the components in the subsystem can be used as target components, and the indicator data of each indicator in the target component can be used as key indicator data.

[0076]

[0077] Table 1

[0078] In an optional embodiment, the embodiment of the present application determines the key indicator data corresponding to the target component, which may specifically include: obtaining the preset indicator parameters corresponding to the target component; based on the indicator parameters, collecting data from the data interface corresponding to the target component according to the preset data collection method to obtain the key indicator data. Among them, the indicator parameters may include one or more indicators, which is not limited by the embodiment of the present application. The preset data collection method may include the data sampling frequency corresponding to the indicator parameter, which is not limited by the embodiment of the present application.

[0079] Specifically, the embodiment of the present application can collect data from the data interface corresponding to the target component to obtain raw data, and use the raw data as key indicator data. Since the sampling frequency of raw data is high and the sampling frequencies of different indicator parameters and different indicators are different, the corresponding data sampling frequencies can be preset for different indicator parameters or indicators in advance as a preset data collection method. Among them, the data sampling frequencies corresponding to different indicator parameters can be different or the same. For example, the sampling frequency can be reduced to minutes, that is, data is collected once every minute, and the collected data is unified into minute data.

[0080] For example, indicators can be preset for different components as preset indicator parameters. When determining the key indicator data corresponding to the target component, the preset indicator parameters corresponding to the target component can be obtained. Based on the indicator parameters, data acquisition is performed on the data interface of the target component at the data sampling frequency corresponding to the indicator parameters to obtain the original data corresponding to the indicator as the key indicator data. For example, two indicators can be corresponding to the external circuit pump and controller, namely, the external circuit temperature control value (as indicator 1) and the external circuit pump group speed (as indicator 2). Then, data acquisition can be performed through the data interface of the external circuit pump and controller according to the preset sampling frequency to obtain the indicator data corresponding to indicator 1 and the indicator data corresponding to indicator 2 as the key indicator data corresponding to the external circuit pump and controller; for another example, for the ventilation system, the components included are ventilation ducts, and the ventilation ducts can correspond to two indicators, namely, the speed of the circulation fan in the cabin (indicator 3) and the speed of the equipment heat dissipation fan (indicator 4). The indicator data corresponding to indicator 3 and the indicator data corresponding to indicator 4 can be used as the key indicator data corresponding to the ventilation duct.

[0081] Step 220: Based on the indicator parameters, a simulation test is performed on the thermal control system to obtain the normal sample data set.

[0082] Specifically, the embodiment of the present application can perform a simulation test on the thermal control system based on the indicator parameters to obtain a normal sample data set.

[0083] For example, refer to Figure 3 , the aerospace thermal control system can be simulated and tested to obtain a normal sample data set.

[0084] Step 230 , based on the indicator parameters and in combination with preset operating condition information, a fault simulation process is performed on the thermal control system to obtain the fault sample data set.

[0085] Wherein, the fault sample data machine includes fault sample data corresponding to the operating condition information.

[0086] In a specific implementation, the embodiment of the present application can perform fault simulation processing on the thermal control system according to the index parameters under different working conditions to obtain a fault sample data set, and then make full use of test data such as the normal sample data set and the fault sample data set.

[0087] As an example, see Figure 3 , the fault simulation test of the entire spacecraft thermal control system can be carried out under high temperature, low temperature, manned and unmanned working conditions, and the fault sample data set under different working conditions can be obtained.

[0088] Step 240 : constructing a curve based on the normal sample data set and the fault sample data set to obtain slip curve information.

[0089] Specifically, the embodiment of the present application can construct a curve based on a normal sample data set and a fault sample data set. For example, the normal sample data corresponding to each indicator can be extracted from the normal sample data set based on the indicator, and the curve is constructed according to the normal sample data to obtain a normal slip curve; the curve is constructed according to the fault sample data to obtain a fault slip curve. The normal slip curve and the fault slip curve can then be used as slip curve information.

[0090] In an optional embodiment, the embodiment of the present application constructs a curve based on the normal sample data set and the fault sample data set to obtain slip curve information, which may specifically include: constructing a curve based on the normal sample data set to obtain first slip curve information; constructing a curve based on the fault sample data set to obtain second slip curve information. The first slip curve information may include a normal slip curve, and the second slip curve information may include a fault slip curve, which is not limited in the embodiment of the present application.

[0091] Step 250: extract state features based on the key indicator data and the slip curve information to obtain state information corresponding to the key indicator data.

[0092] Specifically, the state information may include the health level corresponding to the key indicator data, which is not limited in the present embodiment. Specifically, the present embodiment may construct a curve based on the key indicator data to obtain the indicator curve information corresponding to the key indicator data, and then may combine the indicator curve information and the slip curve information to extract the state feature, such as referring to Figure 3 , characteristic intervals can be extracted based on multivariate statistical analysis for the index curve information and the slip curve information, and the corresponding health of the key index data can be obtained as the status information. The evaluation of the health status at the index level can make full use of the test data and improve the evaluation accuracy of the thermal control system.

[0093] In a specific implementation, the embodiment of the present application may preset a plurality of feature analysis methods, such as the feature analysis method may include a vector machine analysis method and a dynamic threshold analysis method, wherein the vector machine analysis method may correspond to the vector machine evaluation method, and the dynamic threshold analysis method may correspond to the dynamic threshold evaluation method, and the embodiment of the present application does not limit this. Specifically, referring to Figure 3After determining the key indicator data and the slip curve information, the key indicator data and the slip curve information can be indexed and evaluated by a vector machine analysis method or a dynamic threshold analysis method. For example, the key indicator data and the slip curve information can be extracted from the characteristic interval, and the indicator curve information can be obtained by curve construction of the key indicator data. The sliding window can be set for the indicator curve information and the slip curve information. The sliding window can be a dynamic interval. The curve data corresponding to the dynamic interval can be extracted from the indicator curve information and the slip curve information in a dynamic interval manner to achieve characteristic interval extraction. Subsequently, multivariate statistical analysis can be performed on the curve data to obtain the state information corresponding to the key indicator data.

[0094] In an optional embodiment, the embodiment of the present application performs state feature extraction based on the key indicator data and the slip curve information to obtain state information corresponding to the key indicator data, which may specifically include: according to a preset feature analysis method, using the first slip curve information, the second slip curve information and the key indicator data to perform state feature extraction to obtain state information corresponding to the key indicator data.

[0095] Optionally, the above-mentioned state feature extraction is performed using the first slip curve information, the second slip curve information and the key indicator data according to a preset feature analysis method to obtain the state information corresponding to the key indicator data, which may specifically include the following sub-steps:

[0096] Sub-step 2501: If the feature analysis method is a vector machine analysis method, a slip mean parameter and a slip variance parameter are determined based on the first slip curve information and the second slip curve information.

[0097] Specifically, the slip mean parameter may include a normal slip mean and a fault slip mean, and the slip variance parameter may include a normal slip variance and a fault slip variance, which is not limited in the embodiment of the present application. Wherein, both the normal slip mean and the normal slip variance can be determined based on the first slip curve information, and both the fault slip mean and the fault slip variance can be determined based on the second slip curve information, which is not limited in the embodiment of the present application. Specifically, in an embodiment of the present application, when the feature analysis method is a vector machine analysis method, data statistics can be performed based on the first slip curve information and the second slip curve information using a sliding window method. For example, the first slip curve information and the second slip curve can be valued in a dynamic interval manner, and mean calculation and variance calculation can be performed based on the extracted data to obtain a normal slip mean and a normal slip variance corresponding to the first slip curve information and a fault slip mean and a fault slip variance corresponding to the second slip curve information, and the normal slip mean and the fault slip mean are used as slip mean parameters, and the normal slip variance and the fault slip variance are used as slip variance parameters.

[0098] Sub-step 2502: Based on the sliding mean parameter and the sliding variance parameter, feature extraction is performed through a preset analysis model to obtain spatial distribution information.

[0099] Specifically, the preset analysis model may include an outlier detection (One-Class SVM) model, which is not limited in the embodiment of the present application; the spatial distribution information may include a normal spatial distribution function and an abnormal spatial distribution function, which is not limited in the embodiment of the present application. Specifically, after determining the slip mean parameter and the slip variance parameter, the embodiment of the present application may input the normal slip mean in the slip mean parameter and the normal slip variance in the slip variance parameter into a pre-trained One-Class SVM model, and extract the normal spatial distribution function of the normal slip mean and the normal slip variance through the One-Class SVM model; input the fault slip mean in the slip mean parameter and the fault slip variance in the slip variance parameter into a pre-trained One-Class SVM model, and extract the abnormal spatial distribution function of the fault slip mean and the fault slip variance through the One-Class SVM model. Subsequently, the normal spatial distribution function and the abnormal spatial distribution function may be used as spatial distribution information.

[0100] Sub-step 2503, extracting key parameter information from the key indicator data.

[0101] Specifically, the key parameter information may include the indicator sliding variance and the indicator sliding mean, which is not limited in the embodiments of the present application. Specifically, the embodiments of the present application may be based on the indicator curve corresponding to the key indicator data, perform dynamic interval value taking on the indicator curve in a sliding window manner, perform mean and variance calculation on the data in the dynamic interval, obtain the indicator sliding variance and the indicator sliding mean, and use the indicator sliding variance and the indicator sliding mean as the key parameter information.

[0102] Sub-step 2504: Perform statistical analysis based on the key parameter information and the spatial distribution information to obtain the status information.

[0103] Specifically, the embodiment of the present application can input key parameter information into a pre-trained One-Class SVM model, such as the indicator slip variance and the indicator slip mean can be input into the One-Class SVM model, the indicator slip variance and the indicator slip mean are calculated by the One-Class SVM model and input into the normal space distribution function, the distance from the indicator slip variance and the indicator slip mean to the normal distribution space is calculated, and the data points within the normal distribution space are marked as normal; the indicator slip variance and the indicator slip mean are calculated by the One-Class SVM model and input into the abnormal space distribution function, the distance from the indicator slip variance and the indicator slip mean to the abnormal distribution space is calculated, and the data points within the abnormal distribution space (that is, the various data of the indicator) are marked as abnormal. Subsequently, the number of data points divided into normal and the number of data points divided into abnormal can be counted, and the proportion of abnormal data points in the overall data (that is, the entire indicator data) is counted according to the number of normal data points and the number of abnormal data points, and the health of the key indicator data is determined based on the proportion, that is, the indicator health is determined as state information. It can make full use of test data and combine it with vector machines to quickly obtain the health of various indicators. While improving the accuracy of spacecraft thermal control system assessment, it can also achieve rapid assessment of the health status of the spacecraft thermal control system.

[0104] In a specific implementation, the embodiment of the present application can pre-construct a One-Class SVM classification model, and can perform model training on the One-Class SVM classification model. Specifically, a normal sample data set and a fault sample data set can be used as training sets to perform model training on the One-Class SVM classification model, and a trained One-Class SVM classification model can be obtained as a preset classification model.

[0105] As an example, see Figure 4 Taking index 1 and index 2 as examples, we can perform statistical calculations on the normal sample data set corresponding to index 1 to obtain the normal sliding mean corresponding to index 1 and the normal slip variance Perform statistical calculation on the fault sample data corresponding to indicator 1 to obtain the fault slip mean corresponding to indicator 1 and fault slip variance Then as well as Input into the trained One-Class SVM classification model and extract The corresponding spatial distribution function and The corresponding spatial distribution function F std1 ,extract The corresponding spatial distribution function Fmean2 ,and The corresponding spatial distribution function F std2 Then, the indicator sliding mean 1 and indicator sliding variance 1 corresponding to indicator 1 can be determined by using the short-time series method, and the indicator sliding mean 1 is input into the spatial distribution function F mean1 Calculate its distance to the healthy distribution space to obtain the evaluation result. Based on the evaluation result, determine the points in the healthy distribution space, mark the points in the healthy distribution space as normal points, and input the index sliding mean 1 into the spatial distribution function F mean2 Calculate its distance to the abnormal distribution space, mark the points within the abnormal distribution space as abnormal points, and calculate the percentage P of normal points based on the number of abnormal points and the number of normal points. mean ; Input the index slip variance 1 into the spatial distribution function F std1 Calculate its distance to the healthy distribution space, mark the points within the healthy distribution space as normal points, and input the index sliding variance 1 into the spatial distribution function F std2 Calculate its distance to the abnormal distribution space, mark the points within the abnormal distribution space as abnormal points, and calculate the percentage P of normal points based on the number of abnormal points and the number of normal points. std , take P mean and P std The minimum value in is taken as the health of indicator 1, and the health of indicator 1 is taken as the status information of indicator 1. Similarly, for indicator 2, the statistical analysis method of indicator 1 can be referred to, and statistical analysis can be performed on indicator 2, the normal sample data set corresponding to indicator 2, and the fault sample data set to obtain the health of indicator 2, and the health of indicator 2 is taken as the status information of indicator 2. Subsequently, the indicator evaluation result of the target component can be determined based on the status information of indicator 1 and the status information of indicator 2.

[0106] Sub-step 2505: If the feature analysis method is a dynamic threshold analysis method, a sliding mean parameter and a sliding standard deviation parameter are determined based on the first sliding curve information and the second sliding curve information.

[0107] Specifically, the slip standard deviation parameter may include a normal slip standard deviation and a fault slip standard deviation, which is not limited in the embodiment of the present application. In the embodiment of the present application, when the feature analysis mode is a dynamic threshold analysis mode, a data statistical analysis is performed based on the first slip curve information to obtain a normal slip mean and a normal slip standard deviation, and a data statistical analysis is performed based on the second slip curve information to obtain a fault slip mean and a fault slip standard deviation, and the normal slip mean and the fault slip mean are used as slip mean parameters, and the normal slip standard deviation and the fault slip standard deviation are used as slip standard deviation parameters.

[0108] Sub-step 2506, determining a target interval threshold based on the sliding mean parameter and the sliding standard deviation parameter.

[0109] Specifically, the target interval threshold can be used to determine the upper and lower limits of data fluctuations, and the target interval threshold can be used to determine whether the key indicator data is normal data or abnormal data, and the embodiments of the present application do not limit this. Specifically, the embodiments of the present application can determine the normal threshold value based on the normal slip mean and the normal slip standard deviation, determine the abnormal threshold value based on the fault slip mean and the fault slip standard deviation, and then determine the target interval threshold value based on the normal threshold threshold value and the abnormal threshold threshold value.

[0110] In a specific implementation, the embodiment of the present application can determine the target interval threshold according to the ±3σ principle after determining the slip mean and the slip standard deviation, where σ can be the slip standard deviation, and the embodiment of the present application is not limited to this. Specifically, taking the normal threshold as an example, the normal slip mean can be subtracted by 3 times the normal slip standard deviation to obtain the lower threshold corresponding to the normal data, and the normal slip mean can be added to 3 times the normal slip standard deviation to obtain the upper threshold corresponding to the normal data. The normal threshold is determined based on the upper and lower thresholds corresponding to the normal data, such as the normal threshold can be [μ1-3σ1, μ1+3σ1], where μ1 can be the normal slip mean, and σ1 can be the normal slip standard deviation. Similarly, taking the abnormal threshold as an example, the abnormal threshold can be [μ2-3σ2, μ2+3σ2], where μ2 can be the fault slip mean, and σ2 can be the fault slip standard deviation. The normal threshold and the abnormal threshold are used as the target interval threshold. Subsequently, it can be determined whether each data in the key indicator data is normal data or abnormal data based on the target interval threshold.

[0111] Sub-step 2507, performing statistical analysis based on the key indicator data and the target interval threshold to obtain the status information.

[0112] Specifically, after determining the target interval threshold, the embodiment of the present application can analyze each data in the key indicator data with the target interval threshold. For example, the key indicator data can be statistically analyzed according to the normal threshold and the abnormal threshold. The key indicator data that falls within the normal threshold is normal data, and the number of normal data points in the key indicator data that are normal data is determined; the key indicator data that falls within the abnormal threshold is abnormal data, and the number of abnormal data points in the key indicator data that are abnormal data is determined. Subsequently, the proportion of the number of normal data points can be determined based on the number of normal data points and the number of abnormal data points, and then the status information corresponding to the key indicator data can be determined. It can make full use of the test data, and combine with the dynamic threshold method to quickly calculate the health of each indicator. While improving the accuracy of the spacecraft thermal control system evaluation, it can also achieve rapid health status evaluation of the spacecraft thermal control system.

[0113] As an example, taking indicators 3 and 4 as examples, the upper and lower limits of normal data fluctuations, i.e., the normal threshold value, are determined according to the operating state of indicator 3 under normal conditions, and the upper and lower limits of fault data fluctuations, i.e., the abnormal threshold value, are determined according to the operating state of indicator 3 under fault conditions. Then, the number of normal data points belonging to the normal threshold range and the number of abnormal data points belonging to the abnormal threshold range in the key indicator data corresponding to indicator 3 are determined. According to the number of normal data points and the number of abnormal data points, the total data volume is obtained, and the proportion of the number of normal data points in the total data volume is counted to determine the health and health level corresponding to indicator 3, and the state information corresponding to indicator 3 is determined based on the health and health level. Similarly, for indicator 4, the method of indicator 3 can also be referred to to determine the number of normal data points belonging to the normal threshold range and the number of abnormal data points belonging to the abnormal threshold range in the key indicator data corresponding to indicator 4, and the total data volume is obtained according to the number of normal data points and the number of abnormal data points, and the proportion of the number of normal data points in the total data volume is counted to determine the health and health level corresponding to indicator 4, and the state information corresponding to indicator 4 is determined based on the health and health level.

[0114] Step 260: Determine an indicator evaluation result of the target component based on the status information.

[0115] Specifically, after determining the status information, the embodiment of the present application can determine the indicator evaluation result of the target component based on the status information to achieve component-level health status evaluation.

[0116] In a specific implementation, the indicator evaluation results may include component evaluation levels and component health. A preset correspondence between component evaluation levels and component health may be provided in advance. For example, the correspondence may be: YV=10, VX=30, VT=50, TH=70, JK=90. Of course, different correspondences between component evaluation levels and component health may be provided according to actual needs. That is, the correspondence between component evaluation levels and component health in the embodiment of the present application is not limited.

[0117] In a specific implementation, the embodiment of the present application can perform fuzzy mapping of component evaluation level and indicator evaluation level for the target component and the indicator corresponding to the target component, obtain fuzzy mapping rules, and generate a fuzzy rule table based on the fuzzy mapping rules. Specifically, after determining the indicator health corresponding to the key indicator data, the indicator evaluation level can be determined according to the indicator health, and then the indicator evaluation level corresponding to each key indicator data of the target component can be combined to determine the component evaluation level corresponding to the target component according to the fuzzy rule table, and then the component evaluation level can be determined for the component evaluation level by using the preset corresponding relationship between the component evaluation level and the component health, and the component evaluation level and the component health are used as the indicator evaluation result of the target component.

[0118] As an example, fuzzy rules can be preset for the external circuit pump and controller and the indicators (including indicator 1 and indicator 2) corresponding to the external circuit pump and controller, and the fuzzy rules may include: ① The higher the indicator health of the external circuit pump group speed, the higher the indicator health of the external circuit temperature control value, the higher the component health of the external circuit pump group and controller; ② The indicator health of the external circuit pump group speed is moderate, the indicator health of the external circuit temperature control value is moderate, and the component health of the external circuit pump and controller is moderate; ③ The lower the indicator health of the external circuit pump group speed, the lower the indicator health of the external circuit temperature control value, the lower the component health of the external circuit pump and controller. A fuzzy rule table can then be established based on the fuzzy rules.

[0119] For example, referring to Table 2, for the external circuit pump and controller, a fuzzy rule table (i.e., Table 2) can be established based on the health level corresponding to indicator 3 and the health level corresponding to indicator 4, combined with fuzzy rules. After determining the indicator evaluation level 1 corresponding to indicator 1 and the indicator evaluation level 2 corresponding to indicator 2, the component evaluation level of the external circuit pump and controller can be determined based on the indicator evaluation level 1 and the indicator evaluation level 2, combined with Table 2. Exemplarily, taking the indicator evaluation level corresponding to the speed of the external circuit pump group as VX, and the indicator evaluation level corresponding to the external circuit temperature control value as VX, the component evaluation level corresponding to the external circuit pump and controller can be determined as YV in combination with Table 2.

[0120]

[0121] Table 2

[0122] It should be noted that in actual processing, the fuzzy rule table can be set according to actual evaluation requirements, and this example does not limit the fuzzy rule table.

[0123] Optionally, the step of determining the index evaluation result of the target component based on the state information may specifically include the following sub-steps:

[0124] Sub-step 2601, performing affiliation mapping on the status information according to a preset mapping method to determine indicator evaluation level information.

[0125] Specifically, the indicator evaluation level information may include the probability of key indicator data at different indicator evaluation levels, which is not limited in the embodiments of the present application; the preset mapping method may include a membership function mapping method, wherein the membership function mapping method may be to use a preset membership function formula to perform membership mapping on the state information to determine the indicator evaluation level, which is not limited in the embodiments of the present application. Specifically, the embodiments of the present application may perform membership mapping on the state information using a preset membership function formula according to the preset membership function mapping method, that is, use a preset membership function to perform membership calculation on the indicator health degree, obtain the membership degree corresponding to the key indicator data, and determine the probability of the key indicator data at different indicator evaluation levels based on the membership degree as the indicator level information.

[0126] In a specific implementation, the embodiment of the present application can preset a corresponding membership function for the indicator. After determining the status information corresponding to the key indicator data, the preset membership function and mapping formula can be used to perform membership mapping on the indicator health to obtain the membership corresponding to the key indicator data.

[0127] As an example, the membership function formula corresponding to the external loop pump and controller can be formula (1), which is not limited in this example. Assuming that the index health X1=60 corresponding to index 1 and the index health X2=70 corresponding to index 2, X1=60 and X2=70 can be substituted into formula (1) for membership calculation, and the membership of index 1 corresponding to 0.8 and 0.2 is obtained, that is, the probability that index 1 is in a degraded state (ie TH) is 0.8, and the probability that index 1 is in a healthy state (ie JK) is 0.2, thereby determining the index evaluation level information corresponding to index 1. The membership of index 2 is 0.6 and 0.4, that is, the probability that index 2 is in a healthy state (ie JK) is 0.6, and the probability that it is in a degraded state (ie TH) is 0.4, thereby determining the index evaluation level information corresponding to index 2.

[0128] Wherein, formula (1) is as follows:

[0129]

[0130] Sub-step 2602, performing fuzzy mapping based on the indicator evaluation level information to obtain the indicator evaluation result.

[0131] Specifically, the embodiments of the present application can perform mapping calculations based on the indicator evaluation level information to determine the component evaluation level corresponding to the target component and the probability that the target component is of the component evaluation level. Subsequently, the component evaluation level with the largest probability value can be taken as the component evaluation level corresponding to the target component according to the size of the probability. Then, based on the component evaluation level combined with the preset fuzzy rules, the component health corresponding to the target component can be obtained, and the component evaluation level and component health corresponding to the target component can be used as the indicator evaluation result.

[0132] As an example, the probability of indicator 1 being in TH is 0.8, the probability of being in JK is 0.2, the probability of indicator 2 being in JK is 0.6, and the probability of being in TH is 0.4. The indicator evaluation level and probability corresponding to indicator 1 can be combined with the indicator evaluation level and probability corresponding to indicator 2, and combined with fuzzy rule table 2 to obtain the result formula (2). According to the result formula (2), it can be determined that there may be three situations for the external loop pump and controller: the probabilities of being in a degraded state (i.e., TH) are 0.6, 0.4, and 0.2, respectively, and the probability of being in a healthy state (i.e., JK) is 0.2. The state corresponding to the maximum probability is a degraded state. Therefore, the component evaluation level of the external loop pump and controller at this time is determined to be TH. Based on the component evaluation level combined with the corresponding relationship, the component health of the external loop pump and controller at this time is determined to be 70, that is, when the health of indicator 1 is 60 and the health of indicator 2 is 70, the component health of the external loop pump and controller is 70.

[0133] Wherein, formula (2) is as follows:

[0134]

[0135] In actual processing, in the embodiment of the present application, before performing fuzzy mapping, fuzzy mapping rules can be pre-constructed. The fuzzy mapping (i.e., fuzzy discrimination) can be performed based on fuzzy rules under fuzzy sets. Fuzzy discrimination includes domain division, determination of fuzzy sets, solving membership functions, fuzzy rule reasoning, and calculation of support, etc. The embodiment of the present application does not limit this. Specifically, the characteristic indicators for fuzzy discrimination can be determined first, and the domain of each indicator change can be determined. For example, given a simple sample set with two inputs and one output: Among them, x1 and x2 are input attribute features, and y is the output attribute. Then the membership function and fuzzy region division can be determined. Assume that the domains of input attribute features x1, x2 and output attribute y are and [y - ,y +], each attribute domain can be divided into (2N+1) attribute fuzzy sets (also called fuzzy regions, where N is a positive integer, and different attribute domain divisions can have different N values). A function (i.e., membership function, whose value range is [0,1]) is defined on each attribute fuzzy set to indicate the degree to which the sample data belongs to the attribute fuzzy set (called membership). Common membership functions include triangular functions, trapezoidal functions, etc. If N=2, the domain of the input attribute or output attribute is divided into 5 fuzzy regions; if N=3, the domain of the input attribute or output attribute is divided into 7 fuzzy regions. Then, generate fuzzy rules on the given training data set: Calculate and i (where i = 1, 2...n, n is a positive integer) the membership degree of the corresponding attribute fuzzy set, respectively find and i The attribute fuzzy set with the maximum membership degree is used as and i The attribute fuzzy set to which it belongs. For example, The fuzzy set with the maximum membership is B1, then Belongs to the B1 fuzzy set. Finally, fuzzy rules are extracted for the given training sample set. For example: Rule1:IFx1is in B1 and x2is in S1,Then y is in B1;→Rule(B1,S1;B1) and Rule2:IFx1is in B1 and x2is in CE,Then y is in B1;→Rule(B1,CE;B1) and so on; in this way, n rules are extracted from n samples, each rule consists of an IF part and a THEN part, and then after all the fuzzy rules are obtained, these rules can be placed in the fuzzy rule base. When in use, fuzzy mapping calculations are performed on the actual input features (such as the health corresponding to indicator 1 and indicator 2) to obtain the indicator evaluation level of the target component.

[0136] Step 270: Obtain the target state corresponding to the target component.

[0137] Specifically, the target state corresponding to the target state can be used to determine the fault severity corresponding to the target component, which is not limited in the present embodiment. Specifically, the present embodiment determines the target state corresponding to the target component, and then determines the fault severity corresponding to the target component based on the target state.

[0138] In specific implementation, the embodiment of the present application can preset fault severity for different target components according to actual evaluation requirements. Taking the fluid circuit subsystem as an example, the fluid circuit subsystem includes eight components: external circuit pump and controller, temperature control valve, pressure sensor, external circuit compensator, etc. Among them, the fault severity corresponding to the external circuit pump and controller and the medium temperature pump group and controller is relatively high, and other components can be preset with lower fault severity. This example does not limit the setting of fault severity.

[0139] In actual processing, the worst health principle can be used to determine the state of the subsystem corresponding to the target component. Specifically, for a target component with a higher fault severity, when the target component fails, the component with the lowest health among the target components in the subsystem is determined, and the health corresponding to the component with the lowest health is assigned to the subsystem using the worst health principle. For example, when the external circuit pump and controller and / or the medium temperature pump group and controller fail, the health corresponding to the component with the lowest health among the eight components included in the fluid circuit subsystem can be assigned to the fluid circuit subsystem; when the target component with a higher fault severity does not fail, etc., an adaptive weight method can be used to amplify the weight of the faulty component and reduce the weight of the healthy component to obtain the weight information of the target component, that is, execute step 280.

[0140] Step 280, for the target state, using a preset adaptive weight method in combination with the indicator evaluation result to determine the weight information of the target component.

[0141] Specifically, the adaptive weight method can be used to determine the weight corresponding to each target component, and the embodiments of the present application are not limited to this. Specifically, the embodiments of the present application can use the adaptive weight method combined with the index evaluation results of the target component when the target component with higher fault severity has not failed, set a larger weight for the failed target component, set a smaller weight for the healthy component, and obtain the weight of the target component as the weight information of the target component.

[0142] As an example, assuming that components with higher fault severity in the fluid circuit subsystem do not fail, the number n of faulty components (n is a positive integer) can be determined, and then the weight k of each component in the fluid circuit subsystem can be calculated by the formula: k=1 / (1+n) as the weight information of the target component.

[0143] Step 290: Perform weighted processing based on the weight information and the indicator evaluation result to obtain a subsystem evaluation result corresponding to the target component.

[0144] Specifically, the subsystem evaluation result may include the subsystem health, which is not limited in the embodiments of the present application. Specifically, after determining the weight corresponding to the target component, the embodiments of the present application may perform a weighted calculation based on the weight of the target component combined with the component health corresponding to the target component to obtain the subsystem health corresponding to the target component as the subsystem evaluation result.

[0145] For example, the component health m of each faulty component in the fluid circuit subsystem can be determined g , combine the component health of all faulty components to generate the faulty component health set M g =[m g1 , m g2 , ..., m gn ], and the number of normal components and the component health corresponding to the normal components can be determined, and the average health of the normal components can be calculated based on the number of normal components and the component health corresponding to the normal components. After determining the weight k, the subsystem health calculation formula can be used: The health M corresponding to the fluid circuit subsystem is obtained as the fluid circuit subsystem evaluation result.

[0146] Step 300: Based on the subsystem evaluation result, an evaluation is performed in accordance with a preset evaluation method to obtain a state evaluation result of the thermal control system.

[0147] Specifically, the preset evaluation method may include the worst health principle, and the embodiments of the present application do not limit this. Specifically, the embodiments of the present application may use the worst health principle to evaluate each subsystem evaluation result, select the subsystem with the lowest health, and use the health of the subsystem as the health of the thermal control system. The health level of the thermal control system is determined based on the health as the status evaluation result of the thermal control system. The health status evaluation of the thermal control system and its components is implemented step by step from the indicator level, component level, and system level. The health status of each component and the health of the system can be accurately evaluated, and the root component of the abnormal problem can be assisted in locating. It effectively solves the problem of the lack of means for evaluating the health status of spacecraft thermal control systems and components in the prior art, and can timely and effectively obtain the health status of spacecraft thermal control systems and components, thereby improving the accuracy and effectiveness of the evaluation.

[0148] For example, referring to Table 1, the thermal control system may include four subsystems, namely, a fluid circuit, a system controller, a ventilation system, and an electric heating temperature control system. Since the health status of each subsystem has a particularly important impact on the overall system, the worst health principle can be adopted to select the lowest health status of the four subsystems as the health status of the thermal control system to achieve an evaluation of the health status of the thermal control system.

[0149] In summary, the embodiment of the present application determines the key indicator data corresponding to the target component, performs a simulation test on the thermal control system based on the indicator parameters, obtains a normal sample data set, performs fault simulation processing on the thermal control system based on the indicator parameters and in combination with preset working condition information, obtains a fault sample data set, constructs a curve based on the normal sample data set and the fault sample data set to obtain slip curve information, extracts state features based on the key indicator data and the slip curve information to obtain state information corresponding to the key indicator data, determines the indicator evaluation result of the target component based on the state information, and then obtains the target state corresponding to the target component, determines the weight information of the target component based on the preset adaptive weight method in combination with the indicator evaluation result for the target state, performs weighted processing based on the weight information in combination with the indicator evaluation result to obtain the subsystem evaluation result corresponding to the target component, and can make full use of test data such as normal sample data and fault sample data, and effectively evaluates the health status of the thermal control system and its components by combining the test data with the indicator data, thereby solving the problem that the prior art lacks means for effectively evaluating the health status of the entire spacecraft thermal control system and its components.

[0150] It should be noted that, for the purpose of simple description, the method embodiments are expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited to the described order of actions, because according to the embodiments of the present application, certain steps may be performed in other orders or simultaneously.

[0151] like Figure 5 As shown, the embodiment of the present application further provides an evaluation device 500 for a thermal control system, comprising:

[0152] A key indicator data determination module 510 is used to determine the key indicator data corresponding to the target component;

[0153] A sample data set determination module 520 is used to obtain a normal sample data set and a fault sample data set of the thermal control system according to the key indicator data;

[0154] An index evaluation result determination module 530 is used to perform feature extraction and analysis processing on the key index data based on the normal sample data set and the fault sample data set to obtain an index evaluation result of the target component;

[0155] The state evaluation result determination module 540 is used to determine the state evaluation result of the thermal control system based on the indicator evaluation result.

[0156] Optionally, the key indicator data determination module includes:

[0157] An indicator parameter acquisition submodule is used to obtain preset indicator parameters corresponding to the target component;

[0158] The data acquisition submodule is used to acquire data from the data interface corresponding to the target component based on the indicator parameters in accordance with a preset data acquisition method to obtain the key indicator data.

[0159] Optionally, the evaluation device of the thermal control system further includes:

[0160] A simulation test module, used for performing a simulation test on the thermal control system based on the index parameters to obtain the normal sample data set;

[0161] A simulation processing module is used to perform fault simulation processing on the thermal control system based on the indicator parameters and in combination with preset operating condition information to obtain the fault sample data set, wherein the fault sample data set includes fault sample data corresponding to the operating condition information.

[0162] Optionally, the indicator evaluation result determination module includes:

[0163] A curve construction submodule, used to construct a curve based on the normal sample data set and the fault sample data set to obtain slip curve information;

[0164] A state feature extraction submodule, used to extract state features according to the key indicator data and the slip curve information, and obtain state information corresponding to the key indicator data;

[0165] The indicator evaluation result determination submodule is used to determine the indicator evaluation result of the target component based on the state information.

[0166] Optionally, the curve construction submodule includes:

[0167] A first curve construction unit, configured to construct a curve based on the normal sample data set to obtain first slip curve information;

[0168] The second curve construction unit is used to construct a curve based on the fault sample data set to obtain a second slip curve information.

[0169] Optionally, the state feature extraction submodule includes:

[0170] The state information determination unit is used to extract state features by using the first slip curve information, the second slip curve information and the key indicator data in accordance with a preset feature analysis method, so as to obtain state information corresponding to the key indicator data.

[0171] Optionally, the state information determining unit includes:

[0172] a first parameter determination subunit, configured to determine a slip mean parameter and a slip variance parameter based on the first slip curve information and the second slip curve information when the feature analysis method is a vector machine analysis method;

[0173] A spatial distribution information determination subunit, configured to extract features through a preset analysis model based on the slip mean parameter and the slip variance parameter to obtain spatial distribution information;

[0174] A key indicator parameter information extraction subunit, used to extract key parameter information from the key indicator data;

[0175] The first statistical analysis subunit is used to perform statistical analysis based on the key parameter information and the spatial distribution information to obtain the state information.

[0176] a second parameter determination subunit, configured to determine a sliding mean parameter and a sliding standard deviation parameter based on the first sliding curve information and the second sliding curve information when the feature analysis mode is a dynamic threshold analysis mode;

[0177] A target interval threshold determination subunit, configured to determine a target interval threshold based on the slip mean parameter and the slip standard deviation parameter;

[0178] The second statistical analysis subunit is used to perform statistical analysis based on the key indicator data and the target interval threshold to obtain the status information.

[0179] Optionally, the indicator evaluation result determination submodule includes:

[0180] An indicator evaluation level information determination unit, used to perform affiliation mapping on the state information according to a preset mapping method to determine the indicator evaluation level information;

[0181] The indicator evaluation result determination unit is used to perform fuzzy mapping based on the indicator evaluation level information to obtain the indicator evaluation result.

[0182] Optionally, the status assessment result determination module includes:

[0183] A target state determination submodule, used to obtain the target state corresponding to the target component;

[0184] A weight information determination submodule, for determining the weight information of the target component according to the target state by using a preset adaptive weight method in combination with the indicator evaluation result;

[0185] A subsystem evaluation result determination submodule, used to perform weighted processing based on the weight information combined with the indicator evaluation result to obtain a subsystem evaluation result corresponding to the target component;

[0186] The state evaluation result determination submodule is used to perform evaluation according to a preset evaluation method based on the subsystem evaluation result to obtain the state evaluation result of the thermal control system.

[0187] It should be noted that the evaluation device for the thermal control system provided in the embodiment of the present application can execute the evaluation method for the thermal control system provided in any embodiment of the present application, and has the corresponding functions and beneficial effects of executing the evaluation method for the thermal control system.

[0188] In a specific implementation, the evaluation device of the thermal control system can be integrated into the device, so that the device can analyze and process the key indicator parameters, normal sample data sets and fault sample data sets corresponding to the target components to obtain the status evaluation results of the thermal control system. As an electronic device, it can effectively evaluate the health status of the spacecraft thermal control system and its components. The electronic device can be composed of two or more physical entities, or it can be composed of one physical entity. For example, the electronic device can be a personal computer (PC), a computer, a server, etc., and the embodiments of the present application do not impose specific restrictions on this.

[0189] like Figure 6 As shown, an embodiment of the present application provides an electronic device, including a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114; the memory 113 is used to store computer programs; the processor 111 is used to implement the steps of the thermal control system evaluation method provided by any of the aforementioned method embodiments when executing the program stored in the memory 113. Exemplarily, the steps of the thermal control system evaluation method may include the following steps: determining key indicator data corresponding to the target component; obtaining a normal sample data set and a fault sample data set of the thermal control system for the key indicator data; performing feature extraction and analysis processing on the key indicator data based on the normal sample data set and the fault sample data set to obtain an indicator evaluation result of the target component; and determining a state evaluation result of the thermal control system based on the indicator evaluation result.

[0190] An embodiment of the present application further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the thermal control system evaluation method provided in any one of the aforementioned method embodiments are implemented.

[0191] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0192] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features applied for herein.

Claims

1. A method for evaluating a thermal control system, characterized in that: include: Determine the key indicator data corresponding to the target component; According to the key indicator data, a normal sample data set and a fault sample data set of the thermal control system are obtained; Based on the normal sample data set and the fault sample data set, feature extraction and analysis processing are performed on the key indicator data to obtain the indicator evaluation result of the target component, including: curve construction based on the normal sample data set and the fault sample data set to obtain slip curve information; state feature extraction is performed according to the key indicator data and the slip curve information to obtain state information corresponding to the key indicator data; based on the state information, the indicator evaluation result of the target component is determined; The constructing a curve based on the normal sample data set and the fault sample data set to obtain the slip curve information includes: constructing a curve based on the normal sample data set to obtain the first slip curve information; constructing a curve based on the fault sample data set to obtain the second slip curve information; The extracting state features according to the key indicator data and the slip curve information to obtain the state information corresponding to the key indicator data includes: extracting state features according to a preset feature analysis method using the first slip curve information, the second slip curve information and the key indicator data to obtain the state information corresponding to the key indicator data; Based on the indicator evaluation result, a state evaluation result of the thermal control system is determined.

2. The method according to claim 1, characterized in that: The key indicator data corresponding to the target component is determined, including: Obtaining preset index parameters corresponding to the target component; Based on the indicator parameters, data is collected from the data interface corresponding to the target component in a preset data collection method to obtain the key indicator data.

3. The method according to claim 2, characterized in that Before obtaining the normal sample data set and the fault sample data set of the thermal control system, the method further includes: Based on the index parameters, a simulation test is performed on the thermal control system to obtain the normal sample data set; Based on the indicator parameters and in combination with preset operating condition information, a fault simulation process is performed on the thermal control system to obtain the fault sample data set, wherein the fault sample data set includes fault sample data corresponding to the operating condition information.

4. The method according to claim 1, characterized in that The method of extracting state features by using the first slip curve information, the second slip curve information and the key indicator data in accordance with a preset feature analysis method to obtain state information corresponding to the key indicator data includes: If the feature analysis method is a vector machine analysis method, a slip mean parameter and a slip variance parameter are determined based on the first slip curve information and the second slip curve information; Based on the slip mean parameter and the slip variance parameter, feature extraction is performed through a preset analysis model to obtain spatial distribution information; Extract key parameter information from the key indicator data; The state information is obtained by performing statistical analysis based on the key parameter information and the spatial distribution information.

5. The method according to claim 1, characterized in that The method of extracting state features by using the first slip curve information, the second slip curve information and the key indicator data in accordance with a preset feature analysis method to obtain state information corresponding to the key indicator data includes: If the characteristic analysis method is a dynamic threshold analysis method, determining a sliding mean parameter and a sliding standard deviation parameter based on the first sliding curve information and the second sliding curve information; Determining a target interval threshold based on the sliding mean parameter and the sliding standard deviation parameter; The status information is obtained by performing statistical analysis based on the key indicator data and the target interval threshold.

6. The method according to claim 1, characterized in that Determining the index evaluation result of the target component based on the state information includes: Performing affiliation mapping on the status information according to a preset mapping method to determine indicator evaluation level information; Fuzzy mapping is performed based on the indicator evaluation level information to obtain the indicator evaluation result.

7. The method according to claim 1, characterized in that Determining the state evaluation result of the thermal control system based on the indicator evaluation result includes: Obtaining a target state corresponding to the target component; For the target state, using a preset adaptive weight method in combination with the indicator evaluation result, determine the weight information of the target component; Performing weighted processing based on the weight information and the indicator evaluation result to obtain a subsystem evaluation result corresponding to the target component; Based on the subsystem evaluation results, an evaluation is performed in accordance with a preset evaluation method to obtain a state evaluation result of the thermal control system.

8. An evaluation device for a thermal control system, characterized in that: include: A key indicator data determination module is used to determine the key indicator data corresponding to the target component; A sample data set determination module, used for acquiring a normal sample data set and a fault sample data set of the thermal control system according to the key indicator data; The indicator evaluation result determination module is used to perform feature extraction and analysis processing on the key indicator data based on the normal sample data set and the fault sample data set to obtain the indicator evaluation result of the target component; including: constructing a curve based on the normal sample data set and the fault sample data set to obtain slip curve information; extracting state features based on the key indicator data and the slip curve information to obtain state information corresponding to the key indicator data; and determining the indicator evaluation result of the target component based on the state information; The constructing a curve based on the normal sample data set and the fault sample data set to obtain the slip curve information includes: constructing a curve based on the normal sample data set to obtain the first slip curve information; constructing a curve based on the fault sample data set to obtain the second slip curve information; The extracting state features according to the key indicator data and the slip curve information to obtain the state information corresponding to the key indicator data includes: extracting state features according to a preset feature analysis method using the first slip curve information, the second slip curve information and the key indicator data to obtain the state information corresponding to the key indicator data; A status evaluation result determination module is used to determine the status evaluation result of the thermal control system based on the indicator evaluation result.

Images