A comprehensive capability evaluation method for space target monitoring system

Abstract

The present invention relates to the field of aerospace technology, and in particular to a method for evaluating the comprehensive capability of a space target monitoring system, comprising step S1, collecting heterogeneous data of the space target monitoring system; step S2, obtaining an error and omission set of each abnormal situation; step S3, generating a heterogeneous data matrix; step S4, outputting an evaluation result; step S5, outputting an intelligent optimization decision; step S6, outputting a Pareto optimal solution; step S7, performing a decision execution simulation, and performing a decision execution judgment; step S8, optimizing the space target monitoring system; step S9, providing feedback on the comprehensive capability evaluation effect of the space target monitoring system; step S10, performing optimization; and step S11, increasing the dimension of the heterogeneous data of the space target monitoring system. The present invention evaluates and optimizes the space target monitoring system, and improves the accuracy of the comprehensive capability evaluation of the space target monitoring system and the monitoring efficiency of the space target monitoring system.

Description

Technical Field

[0001] The present invention relates to the field of aerospace technology, and in particular to a comprehensive capability evaluation method for a space target monitoring system. Background Art

[0002] The space target monitoring system still has significant deficiencies at the current technical level. First, the accuracy of its comprehensive capability assessment is low, and it is unable to fully and accurately capture and analyze the dynamic information of space targets. This defect directly affects the effectiveness and reliability of the monitoring system, and may lead to false alarms or missed alarms. In addition, the existing space target monitoring system has not been fully optimized during the design and operation process, resulting in its overall low monitoring efficiency. Insufficient optimization may be reflected in poor algorithm performance, slow data processing speed, and low sensor accuracy. These technical shortcomings not only limit the real-time monitoring capabilities of the system, but may also affect the accuracy and timeliness of decision-making at critical moments. Therefore, it is urgent to make comprehensive technical upgrades and optimizations, including improving sensor accuracy, enhancing data processing capabilities, optimizing algorithm design, etc., to improve the overall performance and efficiency of the space target monitoring system and ensure that it can provide reliable and efficient monitoring services in complex space environments.

[0003] Chinese Patent Publication No.: CN116089249B discloses a method and device for evaluating the comprehensive capability of a monitoring system for space target monitoring. First, multiple monitoring systems including at least one monitoring equipment system are preset; then the system parameters of each monitoring equipment system in the monitoring system are obtained; then, the time resource scheduling data and monitoring resource scheduling data of each monitoring equipment system are obtained based on the system parameters of each monitoring equipment system, which are used to indicate the continuity in time and space when monitoring space targets; finally, the comprehensive capability evaluation value is obtained based on the time resource scheduling data and monitoring resource scheduling data of each monitoring equipment system, so as to select the best monitoring system from multiple preset monitoring systems to monitor space targets. However, this scheme fails to evaluate the space target monitoring system in real time, and fails to optimize the space target monitoring system. The comprehensive capability evaluation accuracy of the space target monitoring system is low, and the monitoring efficiency of the space target monitoring system is low. Summary of the invention

[0004] To this end, the present invention proposes a comprehensive capability evaluation method for a space target monitoring system, so as to overcome the problems in the prior art of low accuracy in comprehensive capability evaluation of the space target monitoring system and low monitoring efficiency of the space target monitoring system due to failure to optimize the space target monitoring system.

[0005] To achieve the above object, the present invention provides a comprehensive capability evaluation method for a space target monitoring system, comprising:

[0006] Step S1, collecting heterogeneous data of the space target monitoring system;

[0007] Step S2, obtaining an error and omission set of each abnormal situation according to the heterogeneous data of the space target monitoring system;

[0008] Step S3, generating a heterogeneous data matrix according to the error and omission sets of each abnormal situation;

[0009] Step S4, inputting the heterogeneous data matrix into the deep learning model to evaluate the comprehensive capability of the space target monitoring system, and outputting the evaluation result;

[0010] Step S5, inputting the heterogeneous data matrix into the optimization decision model according to the evaluation result, and outputting the intelligent optimization decision through the optimization decision model;

[0011] Step S6, running a multi-objective optimization algorithm according to the output of the intelligent optimization decision, and outputting a Pareto optimal solution;

[0012] Step S7, performing decision execution simulation according to the intelligent optimization decision and the Pareto optimal solution, and making a decision execution judgment according to the simulation results;

[0013] Step S8, optimizing the space target monitoring system according to the decision execution judgment result;

[0014] Step S9, providing feedback on the comprehensive capability evaluation effect of the space target monitoring system according to the simulation results and the optimization results;

[0015] Step S10, optimizing the multi-objective optimization algorithm, the optimization decision model and the deep learning model according to the feedback results;

[0016] Step S11, increasing the dimension of the heterogeneous data of the space target monitoring system according to the optimization effect.

[0017] Furthermore, in step S2, when making an error or omission judgment, the orbit target tracking comparison value is compared with the orbit target preset value to identify an abnormal situation of the orbit target, the space debris monitoring comparison value is compared with the space debris impact preset value to identify an abnormal situation of space debris impact, and the space environment monitoring comparison value is compared with the preset electromagnetic interference value to identify an abnormal situation of electromagnetic interference;

[0018] In step S2, when making an error or omission judgment, the minimum orbit target abnormal neighborhood time difference t1min of each orbit target abnormality is compared with the preset neighborhood time difference t0, and the error or omission of the orbit target abnormality is judged according to the comparison result, wherein:

[0019] When t1min≤t0, it is determined that there is no error or omission in the abnormal situation of the orbit target;

[0020] When t1min>t0, it is determined that there is an error or omission in the orbit target abnormality, and the orbit target abnormality is added to the orbit target abnormality error or omission set R1;

[0021] Obtain the difference y1 of the orbital target tracking comparison value without any error or omission, and compare it with the difference y10 of the preset orbital target tracking comparison value. According to the comparison result, the matching of the orbital target abnormality without any error or omission and the monitoring orbital target abnormality output by the space target monitoring system corresponding to the minimum orbital target abnormality neighborhood time difference is judged, where:

[0022] When y1≤y10, it is determined that the orbital target anomaly situation without any omissions and the monitored orbital target anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference match each other, and the monitored orbital target anomaly situation is used as the matched orbital target anomaly situation;

[0023] When y1>y10, it is determined that the orbital target anomaly situation without any omission does not match the monitored orbital target anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference, and it is compensated to the orbital target anomaly situation omission set R1;

[0024] The minimum space debris impact anomaly neighborhood time difference t2min of each space debris impact anomaly is compared with the preset neighborhood time difference t0, and the error and omission of the space debris impact anomaly is judged according to the comparison result, where:

[0025] When t2min≤t0, it is determined that there is no error or omission in the abnormal situation of space debris impact;

[0026] When t2min>t0, it is determined that there is an error in the space debris impact anomaly, and the space debris impact anomaly is added to the space debris impact anomaly error set R2;

[0027] The difference y2 of the space debris monitoring comparison value without any omission is obtained, and it is compared with the difference y20 of the preset space debris monitoring comparison value. According to the comparison result, the matching of the space debris impact anomaly situation without any omission and the monitoring space debris impact anomaly situation output by the space target monitoring system corresponding to the time difference of the minimum space debris impact anomaly neighborhood is judged, wherein:

[0028] When y2≤y20, it is determined that the space debris impact anomaly situation without any omissions and the monitored space debris impact anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference match each other, and it is used as the matching space debris impact anomaly situation;

[0029] When y2>y20, it is determined that the space debris impact anomaly situation without any omission does not match the monitored space debris impact anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference, and it is compensated to the space debris impact anomaly situation omission set R2;

[0030] The minimum electromagnetic interference abnormal neighborhood time difference t3min of each electromagnetic interference abnormal situation is compared with the preset neighborhood time difference t0, and the error and omission of the electromagnetic interference abnormal situation is judged according to the comparison result, where:

[0031] When t3min≤t0, it is determined that there is no error or omission in the abnormal electromagnetic interference situation;

[0032] When t3min>t0, it is determined that the electromagnetic interference abnormality has an error or omission, and the electromagnetic interference abnormality is added to the electromagnetic interference abnormality error or omission set R3;

[0033] The difference y3 of the electromagnetic interference monitoring comparison value without any error or omission is obtained, and it is compared with the difference y30 of the preset electromagnetic interference monitoring comparison value. According to the comparison result, the matching of the electromagnetic interference abnormality output by the space target monitoring system corresponding to the electromagnetic interference abnormality without any error or omission and its minimum electromagnetic interference abnormality neighborhood time difference is judged, where:

[0034] When y3≤y30, it is determined that the electromagnetic interference anomaly without any error or omission and the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the minimum electromagnetic interference anomaly neighborhood time difference match each other, and it is used as the matched electromagnetic interference anomaly;

[0035] When y3>y30, it is determined that the electromagnetic interference anomaly without any omission does not match the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the minimum electromagnetic interference anomaly neighborhood time difference, and it is compensated to the electromagnetic interference anomaly error set R3;

[0036] The error omission sets of each abnormal situation include an orbit target abnormal situation error omission set R1, a space debris impact abnormal situation error omission set R2 and an electromagnetic interference abnormal situation error omission set R3.

[0037] Furthermore, in the step S3, when generating the heterogeneous data matrix, the number n1 of the orbital target abnormality error set R1, the number n2 of the space debris impact abnormality error set R2, the number n3 of the electromagnetic interference abnormality error set R3, the orbital target tracking comparison value u1 in the error set R1, the space debris monitoring comparison value u2 in the error set R2, and the space environment monitoring comparison value u3 in the error set R3 are obtained, and the minimum orbital target abnormality neighborhood time difference t1min of each orbital target abnormality without error or omission, the orbit without error or omission, and the orbit target tracking comparison value u1 in the error set R1, the space debris monitoring comparison value u2 in the error set R2, and the space environment monitoring comparison value u3 in the error set R3 are obtained. The difference y1 of target tracking contrast value, the minimum space debris impact anomaly neighborhood time difference t2min of each space debris impact anomaly without any error or omission, the difference y2 of space debris monitoring contrast value without any error or omission, the minimum electromagnetic interference anomaly neighborhood time difference t3min of each electromagnetic interference anomaly without any error or omission and the difference y3 of electromagnetic interference monitoring contrast value without any error or omission generate a heterogeneous data matrix [n1, n2, n3, u1, u2, u3, t1min, y1, t2min, y2, t3min, y3].

[0038] Furthermore, in step S4, when evaluating the comprehensive capability of the space target monitoring system, a neural network model is constructed, and the neural network model is trained according to a historical heterogeneous evaluation data training set, and the trained neural network model is verified by a historical heterogeneous evaluation data verification set. When the verification accuracy of the neural network model does not reach 98%, a new historical heterogeneous evaluation data training set is used to train the neural network model until the verification accuracy of the neural network model reaches 98%. The neural network model is then output as a deep learning model, the heterogeneous data matrix is ​​input into the deep learning model, and the deep learning model outputs the evaluation result.

[0039] Furthermore, in step S5, when outputting the intelligent optimization decision, an intelligent optimization judgment is performed according to the evaluation result, wherein:

[0040] When the evaluation result shows that the comprehensive capability of the space target monitoring system is excellent, it is determined that intelligent optimization will not be performed;

[0041] When the evaluation result shows that the comprehensive capability of the space target monitoring system is good, it is determined to perform intelligent optimization;

[0042] When the evaluation result shows that the comprehensive capability of the space target monitoring system is poor, it is determined to perform intelligent optimization;

[0043] When deciding to perform intelligent optimization, the heterogeneous data matrix is ​​input into the optimization decision model, and the optimization decision model outputs the intelligent optimization decision.

[0044] Furthermore, in step S6, when the Pareto optimal solution is output, the operation of the multi-objective optimization algorithm is judged according to the intelligent optimization decision output g, wherein:

[0045] When g = 1, it is determined that the multi-objective optimization algorithm is not run;

[0046] When g>1, it is determined to run the multi-objective optimization algorithm, and output the Pareto optimal solution through the multi-objective optimization algorithm.

[0047] Furthermore, in step S7, when making a decision execution judgment, a decision execution simulation is performed according to the intelligent optimization decision and the Pareto optimal solution, wherein:

[0048] When the intelligent optimization decision output g=1, the intelligent optimization decision output by the optimization decision model is input into the decision execution simulation model, and the simulation data is output;

[0049] When the output of the intelligent optimization decision g>1, the Pareto optimal solution is input into the decision execution simulation model, and the simulation data is output;

[0050] The simulation data is input into the deep learning model and the evaluation result is output. The evaluation result is used as the simulation result, and the decision execution judgment is made according to the simulation result, where:

[0051] If the simulation result shows that the comprehensive capability of the space target monitoring system is excellent, the decision is made;

[0052] If the simulation result shows that the comprehensive capability of the space target monitoring system is good, it is determined that no decision execution will be made;

[0053] If the simulation result shows that the comprehensive capability of the space target monitoring system is poor, it is determined that no decision will be executed.

[0054] Furthermore, in step S8, when the decision execution judgment result is to perform decision execution, the controller executes the intelligent optimization decision and Pareto optimal solution corresponding to the decision execution judgment result to optimize the space target monitoring system.

[0055] Furthermore, in step S9, when performing multi-layer feedback, the evaluation result of the comprehensive capability of the space target monitoring system within the execution cycle is obtained, which is used as the optimization result, and the evaluation effect of the comprehensive capability of the space target monitoring system is fed back according to the simulation result and the optimization result, wherein:

[0056] When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is excellent, the feedback is that the comprehensive capability evaluation effect of the space target monitoring system is excellent;

[0057] When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is good, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor;

[0058] When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is poor, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor;

[0059] When the simulation result shows that the comprehensive capability of the space target monitoring system is good, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor;

[0060] When the simulation result shows that the comprehensive capability of the space target monitoring system is poor, the feedback indicates that the comprehensive capability evaluation effect of the space target monitoring system is poor.

[0061] Furthermore, in step S10, the optimization conditions of the multi-objective optimization algorithm, the optimization decision model and the deep learning model are judged according to the feedback results, wherein:

[0062] When the intelligent optimization decision output g=1, it is determined that the optimization decision model and the deep learning model are optimized;

[0063] When the output of the intelligent optimization decision g>1, it is determined that the multi-objective optimization algorithm, the optimization decision model and the deep learning model are optimized;

[0064] In the step S11, the comprehensive capability evaluation effect of the space target monitoring system within the optimized monitoring time is obtained, and the optimization effect is judged according to the comprehensive capability evaluation effect of the space target monitoring system within the optimized monitoring time, wherein:

[0065] When the evaluation result within the optimized monitoring time is that the comprehensive capability evaluation effect of the space target monitoring system is excellent, the optimization effect is determined to be effective;

[0066] When the evaluation result within the optimized monitoring time is that the comprehensive capability evaluation effect of the space target monitoring system is poor, the optimization effect is determined to be invalid;

[0067] And according to the optimization effect, the dimension increase judgment of the heterogeneous data of the space target monitoring system is carried out, among which:

[0068] When the optimization effect is effective, it is determined that no dimension increase is performed;

[0069] When the optimization effect is invalid, it is determined to increase the dimension;

[0070] When increasing the dimension, a grid search is performed on the spare type heterogeneous data, and a cross-validation is performed. The increased dimension type heterogeneous data in the spare type heterogeneous data is screened out according to the cross-validation results, and the increased dimension type heterogeneous data is added to the heterogeneous data of the space target monitoring system for collection.

[0071] Compared with the prior art, the beneficial effects of the present invention are that the method collects heterogeneous data of the space target monitoring system through step S1 so as to evaluate the space target monitoring system, the method obtains error and omission sets of each abnormal situation according to the heterogeneous data of the space target monitoring system through step S2 so as to identify the error and omission situations of the heterogeneous data of the space target monitoring system, the method generates a heterogeneous data matrix according to the error and omission sets of each abnormal situation through step S3 so as to integrate the error and omission sets of each abnormal situation, the method inputs the heterogeneous data matrix into the deep learning model through step S4 to evaluate the comprehensive capability of the space target monitoring system, and outputs the evaluation result so as to evaluate the comprehensive capability of the space target monitoring system through the deep learning model, thereby improving the efficiency of the comprehensive capability evaluation of the space target monitoring system, the method inputs the heterogeneous data matrix into the optimization decision model according to the evaluation result through step S5, outputs the intelligent optimization decision through the optimization decision model, so as to optimize the space target monitoring system, and further improve the monitoring efficiency of the space target monitoring system, the method generates a heterogeneous data matrix according to the error and omission sets of each abnormal situation through step S6 The decision output runs a multi-objective optimization algorithm and outputs a Pareto optimal solution to facilitate the screening of the optimal intelligent optimization decision, thereby further improving the monitoring efficiency of the space target monitoring system. The method performs a decision execution simulation according to the intelligent optimization decision and the Pareto optimal solution through step S7, and makes a decision execution judgment based on the simulation results to simulate and judge the intelligent optimization decision, thereby ensuring the effectiveness of the execution of the intelligent optimization decision, preventing the optimization failure caused by the impracticality of the intelligent optimization decision, and improving the monitoring efficiency of the space target monitoring system. The method optimizes the space target monitoring system according to the decision execution judgment result through step S8, and provides feedback on the comprehensive capability evaluation effect of the space target monitoring system according to the simulation results and the optimization results through step S9 to ensure the optimization effect. The method optimizes the multi-objective optimization algorithm, the optimization decision model and the deep learning model according to the feedback results through step S10, and increases the dimension of the heterogeneous data of the space target monitoring system according to the optimization effect through step S11, thereby further optimizing and adjusting the evaluation process of the space target monitoring system, thereby improving the evaluation efficiency of the space target monitoring system. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 Schematic diagram of the flow of the comprehensive capability evaluation method of the space target monitoring system of this embodiment. DETAILED DESCRIPTION

[0073] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0074] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0075] It should be noted that, in the description of the present invention, terms such as "up", "down", "left", "right", "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is merely for the convenience of description and does not indicate or imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present invention.

[0076] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0077] See also Figure 1 As shown, it is a flow chart of the comprehensive capability evaluation method of the space target monitoring system of this embodiment, and the method includes:

[0078] Step S1, collecting heterogeneous data of the space target monitoring system;

[0079] Step S2, obtaining an error and omission set of each abnormal situation according to the heterogeneous data of the space target monitoring system;

[0080] Step S3, generating a heterogeneous data matrix according to the error and omission sets of each abnormal situation;

[0081] Step S4, inputting the heterogeneous data matrix into the deep learning model to evaluate the comprehensive capability of the space target monitoring system, and outputting the evaluation result;

[0082] Step S5, inputting the heterogeneous data matrix into the optimization decision model according to the evaluation result, and outputting the intelligent optimization decision through the optimization decision model;

[0083] Step S6, running a multi-objective optimization algorithm according to the output of the intelligent optimization decision, and outputting a Pareto optimal solution;

[0084] Step S7, performing decision execution simulation according to the intelligent optimization decision and the Pareto optimal solution, and making a decision execution judgment according to the simulation results;

[0085] Step S8, optimizing the space target monitoring system according to the decision execution judgment result;

[0086] Step S9, providing feedback on the comprehensive capability evaluation effect of the space target monitoring system according to the simulation results and the optimization results;

[0087] Step S10, optimizing the multi-objective optimization algorithm, the optimization decision model and the deep learning model according to the feedback results;

[0088] Step S11, increasing the dimension of the heterogeneous data of the space target monitoring system according to the optimization effect.

[0089] Specifically, the method is set in an evaluation terminal of a space target monitoring system to evaluate the space target monitoring system so as to control the monitoring effect of the space target monitoring system in real time, optimize the space target monitoring system, and improve the monitoring efficiency of the space target monitoring system. The method collects heterogeneous data of the space target monitoring system through step S1 to facilitate the evaluation of the space target monitoring system. The method obtains an error and omission set of each abnormal situation according to the heterogeneous data of the space target monitoring system through step S2 to facilitate the identification of the error and omission of the heterogeneous data of the space target monitoring system. The method is Step S3 generates a heterogeneous data matrix according to the error and omission sets of each abnormal situation, so as to integrate the error and omission sets of each abnormal situation. The method inputs the heterogeneous data matrix into the deep learning model through step S4 to evaluate the comprehensive capability of the space target monitoring system, and outputs the evaluation result, so as to evaluate the comprehensive capability of the space target monitoring system through the deep learning model, thereby improving the efficiency of the comprehensive capability evaluation of the space target monitoring system. The method inputs the heterogeneous data matrix into the optimization decision model according to the evaluation result through step S5, and outputs the intelligent optimization decision through the optimization decision model to evaluate the space target monitoring system. The method further improves the monitoring efficiency of the space target monitoring system by optimizing the space target monitoring system. The method runs the multi-objective optimization algorithm according to the output of the intelligent optimization decision through step S6, and outputs the Pareto optimal solution, so as to screen out the optimal intelligent optimization decision, thereby further improving the monitoring efficiency of the space target monitoring system. The method simulates the execution of the decision according to the intelligent optimization decision and the Pareto optimal solution through step S7, and makes a decision execution judgment according to the simulation result, so as to simulate and judge the intelligent optimization decision, ensure the effectiveness of the execution of the intelligent optimization decision, prevent the optimization failure caused by the impracticality of the intelligent optimization decision, and improve the monitoring efficiency of the space target monitoring system. The method optimizes the space target monitoring system according to the decision execution judgment result through step S8, and feeds back the comprehensive capability evaluation effect of the space target monitoring system according to the simulation result and the optimization result through step S9 to ensure the optimization effect. The method optimizes the multi-objective optimization algorithm, the optimization decision model and the deep learning model according to the feedback result through step S10, and increases the dimension of the heterogeneous data of the space target monitoring system according to the optimization effect through step S11, so as to further optimize and adjust the evaluation process of the space target monitoring system, thereby improving the evaluation efficiency of the space target monitoring system.

[0090] Specifically, in step S1, when the heterogeneous data of the space target monitoring system are collected, the heterogeneous data of the space target monitoring system include orbital target tracking comparison values, space debris monitoring comparison values ​​and space environment monitoring comparison values. The orbital target tracking comparison values ​​refer to the satellite position coordinates and the real-time satellite movement speed obtained through external monitoring of the space target monitoring system. In this embodiment, the satellite position coordinates and the real-time satellite movement speed are collected in real time through positioning sensors and speed sensors. The space debris monitoring comparison values ​​refer to the force values ​​of the satellite being hit by space debris obtained through external monitoring of the space target monitoring system. In this embodiment, the space debris monitoring comparison values ​​are collected through pressure sensors. The space environment monitoring comparison values ​​refer to the electromagnetic interference values ​​obtained through external monitoring of the space target monitoring system. In this embodiment, the space environment monitoring comparison values ​​are collected through electromagnetic sensors.

[0091] Specifically, step S1 obtains heterogeneous data of the space target monitoring system through external monitoring of the space target monitoring system, so as to evaluate and judge the monitoring capability of the space target monitoring system independently of the space target monitoring system, and evaluates the comprehensive capability of the space target monitoring system through subsequent analysis of the heterogeneous data of the space target monitoring system.

[0092] Specifically, in step S2, when making error and omission judgments, the orbital target tracking comparison value is compared with the orbital target preset value to identify orbital target abnormalities, the space debris monitoring comparison value is compared with the space debris impact preset value to identify space debris impact abnormalities, and the space environment monitoring comparison value is compared with the preset electromagnetic interference value to identify electromagnetic interference abnormalities.

[0093] Specifically, the orbital target preset value refers to a preset value that can identify the orbital target abnormality. This embodiment does not limit the identification process of the orbital target abnormality. Those skilled in the art can freely set it according to actual conditions, and only need to meet the identification requirements of the orbital target abnormality. For example, the orbital target preset value can be set to a preset threshold value. When the orbital target tracking comparison value does not fall within the orbital target preset value, the orbital target abnormality is determined to be an orbital target abnormality. The space debris impact preset value refers to a preset value that can identify the space debris impact abnormality. This embodiment does not limit the identification process of the space debris impact abnormality. Those skilled in the art can freely set it according to actual conditions, and only need to meet The need for identifying abnormal space debris impact situations can be met. For example, the preset space debris impact value can be set as a preset threshold value. When the space debris monitoring comparison value does not fall within the preset space debris impact value, the space debris impact abnormality is determined to be a space debris impact abnormality. The preset electromagnetic interference value refers to a preset value that can identify the electromagnetic interference abnormality. This embodiment does not limit the identification process of the electromagnetic interference abnormality. Technical personnel in this field can freely set it according to actual conditions. It only needs to meet the need for identifying the electromagnetic interference abnormality. For example, the preset electromagnetic interference value can be set as a preset threshold value. When the space environment monitoring comparison value does not fall within the preset electromagnetic interference value, the electromagnetic interference abnormality is determined to be an electromagnetic interference abnormality.

[0094] Specifically, in step S2, when making an error or omission judgment, the minimum orbit target abnormal neighborhood time difference t1min of each orbit target abnormality is compared with the preset neighborhood time difference t0, and the error or omission of the orbit target abnormality is judged according to the comparison result, wherein:

[0095] When t1min≤t0, it is determined that there is no error or omission in the abnormal situation of the orbit target;

[0096] When t1min>t0, it is determined that there is an error or omission in the orbit target abnormality, and the orbit target abnormality is added to the orbit target abnormality error or omission set R1;

[0097] Obtain the difference y1 of the orbital target tracking comparison value without any error or omission, and compare it with the difference y10 of the preset orbital target tracking comparison value. According to the comparison result, the matching of the orbital target abnormality without any error or omission and the monitoring orbital target abnormality output by the space target monitoring system corresponding to the minimum orbital target abnormality neighborhood time difference is judged, where:

[0098] When y1≤y10, it is determined that the orbital target anomaly situation without any omissions and the monitored orbital target anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference match each other, and the monitored orbital target anomaly situation is used as the matched orbital target anomaly situation;

[0099] When y1>y10, it is determined that the orbital target anomaly situation without any omission does not match the monitored orbital target anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference, and it is compensated to the orbital target anomaly situation omission set R1;

[0100] The minimum space debris impact anomaly neighborhood time difference t2min of each space debris impact anomaly is compared with the preset neighborhood time difference t0, and the error and omission of the space debris impact anomaly is judged according to the comparison result, where:

[0101] When t2min≤t0, it is determined that there is no error or omission in the abnormal situation of space debris impact;

[0102] When t2min>t0, it is determined that there is an error in the space debris impact anomaly, and the space debris impact anomaly is added to the space debris impact anomaly error set R2;

[0103] The difference y2 of the space debris monitoring comparison value without any omission is obtained, and it is compared with the difference y20 of the preset space debris monitoring comparison value. According to the comparison result, the matching of the space debris impact anomaly situation without any omission and the monitoring space debris impact anomaly situation output by the space target monitoring system corresponding to the time difference of the minimum space debris impact anomaly neighborhood is judged, wherein:

[0104] When y2≤y20, it is determined that the space debris impact anomaly situation without any omissions and the monitored space debris impact anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference match each other, and it is used as the matching space debris impact anomaly situation;

[0105] When y2>y20, it is determined that the space debris impact anomaly situation without any omission does not match the monitored space debris impact anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference, and it is compensated to the space debris impact anomaly situation omission set R2;

[0106] The minimum electromagnetic interference abnormal neighborhood time difference t3min of each electromagnetic interference abnormal situation is compared with the preset neighborhood time difference t0, and the error and omission of the electromagnetic interference abnormal situation is judged according to the comparison result, where:

[0107] When t3min≤t0, it is determined that there is no error or omission in the abnormal electromagnetic interference situation;

[0108] When t3min>t0, it is determined that the electromagnetic interference abnormality has an error or omission, and the electromagnetic interference abnormality is added to the electromagnetic interference abnormality error or omission set R3;

[0109] The difference y3 of the electromagnetic interference monitoring comparison value without any error or omission is obtained, and it is compared with the difference y30 of the preset electromagnetic interference monitoring comparison value. According to the comparison result, the matching of the electromagnetic interference abnormality output by the space target monitoring system corresponding to the electromagnetic interference abnormality without any error or omission and its minimum electromagnetic interference abnormality neighborhood time difference is judged, where:

[0110] When y3≤y30, it is determined that the electromagnetic interference anomaly without any error or omission and the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the minimum electromagnetic interference anomaly neighborhood time difference match each other, and it is used as the matched electromagnetic interference anomaly;

[0111] When y3>y30, it is determined that the electromagnetic interference anomaly without any omission does not match the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the minimum electromagnetic interference anomaly neighborhood time difference, and it is compensated to the electromagnetic interference anomaly error set R3;

[0112] The error omission sets of each abnormal situation include an orbit target abnormal situation error omission set R1, a space debris impact abnormal situation error omission set R2 and an electromagnetic interference abnormal situation error omission set R3.

[0113] Specifically, the orbital target abnormality neighborhood time difference of each orbital target abnormality refers to the distance difference between the time point of each orbital target abnormality and the time point of the monitored orbital target abnormality output by the space target monitoring system when the time point of each orbital target abnormality and the time point of the monitored orbital target abnormality output by the adjacent space target monitoring system are mapped on the time axis. The minimum orbital target abnormality neighborhood time difference of each orbital target abnormality refers to the distance difference closer to the time point of the monitored orbital target abnormality output by the space target monitoring system adjacent to the time point of each orbital target abnormality. The preset neighborhood time difference refers to a preset value of the time difference reflecting the existence of errors and omissions in orbital target abnormalities, errors and omissions in space debris impact abnormalities, and errors and omissions in electromagnetic interference abnormalities. For example, the preset neighborhood time difference t0=30s is set. The errors and omissions of the orbital target abnormality refer to the errors and omissions of the orbital target abnormality that are not recognized by the space target monitoring system. The orbital target abnormality error and omission set refers to the orbital target abnormality error and omission set storing the orbital target abnormality errors and omissions of the space target monitoring system. The data collection of the orbital target abnormal conditions, the difference of the orbital target tracking comparison value refers to the difference between the orbital target tracking comparison value corresponding to the time point of each orbital target abnormal condition and the orbital target tracking comparison value of the monitoring orbital target abnormal condition output by the space target monitoring system adjacent to the time point of each orbital target abnormal condition, the difference of the preset orbital target tracking comparison value refers to the preset value of the orbital target tracking comparison value reflecting the mismatch between the monitoring orbital target abnormal condition output by the space target monitoring system corresponding to the orbital target abnormal condition without any errors or omissions and the minimum orbital target abnormal neighborhood time difference, for example, the difference of the preset orbital target tracking comparison value can be set to 5, the matching condition of the monitoring orbital target abnormal condition output by the space target monitoring system corresponding to the orbital target abnormal condition without any errors or omissions and the minimum orbital target abnormal neighborhood time difference refers to whether the monitoring orbital target abnormal condition output by the space target monitoring system corresponding to the orbital target abnormal condition without any errors or omissions and the minimum orbital target abnormal neighborhood time difference is the same abnormal cause and whether they can match each other;

[0114] The minimum space debris impact anomaly neighborhood time difference refers to the minimum value of the distance difference between the time point of each orbital target anomaly and the time point of the monitoring orbital target anomaly output by the adjacent space target monitoring system when the time point of each space debris impact anomaly and the time point of the space debris impact anomaly output by the space target monitoring system are mapped on the time axis. The space debris impact anomaly error and omission situation refers to the error and omission situation of the space target monitoring system where there is an unrecognized space debris impact anomaly. The space debris impact anomaly error and omission set refers to the data collection storing the space debris impact anomaly that is missed by the space target monitoring system. The difference of the space debris monitoring comparison value without error and omission refers to the space debris monitoring comparison value corresponding to the time point of each space debris impact anomaly and the time of each space debris impact anomaly. The difference in space debris monitoring comparison values ​​of the space debris impact anomaly output by the space target monitoring system with adjacent points is closer in time point, the difference in the preset space debris monitoring comparison values ​​refers to the preset value of space debris impact reflecting that the space debris impact anomaly without any errors and the space debris impact anomaly output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference do not match, the matching situation of the space debris impact anomaly monitored by the space target monitoring system corresponding to the space debris impact anomaly without any errors and the minimum space debris impact anomaly neighborhood time difference refers to whether the space debris impact anomaly output by the space target monitoring system corresponding to the space debris impact anomaly without any errors and the minimum space debris impact anomaly neighborhood time difference is the same anomaly cause and whether they can match each other;

[0115] The minimum electromagnetic interference anomaly neighborhood time difference of each electromagnetic interference anomaly refers to the minimum value of the distance difference between the time point of each electromagnetic interference anomaly and the time point of the monitoring electromagnetic interference anomaly output by the space target monitoring system when both are mapped on the time axis. The error or omission of the electromagnetic interference anomaly refers to the error or omission of the space target monitoring system where there is an electromagnetic interference anomaly that is not identified. The difference in the electromagnetic interference monitoring comparison value without error or omission refers to the space environment monitoring comparison value corresponding to the time point of each electromagnetic interference anomaly and the electromagnetic environment monitoring comparison value output by the space target monitoring system adjacent to the time point of each orbital target anomaly. The difference in space environment monitoring comparison values ​​that are closer to the time point of the interference anomaly, the difference in the preset electromagnetic interference monitoring comparison values ​​refers to the preset value of the space environment monitoring comparison value that reflects the mismatch between the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the electromagnetic interference anomaly without any errors or omissions and its minimum electromagnetic interference anomaly neighborhood time difference, the matching situation of the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the electromagnetic interference anomaly without any errors or omissions and its minimum electromagnetic interference anomaly neighborhood time difference refers to whether the electromagnetic interference anomaly output by the space target monitoring system corresponding to the electromagnetic interference anomaly without any errors or omissions and its minimum electromagnetic interference anomaly neighborhood time difference are the same abnormal cause and whether they can match each other.

[0116] Specifically, in step S3, when generating the heterogeneous data matrix, the number n1 of the orbital target abnormality error set R1, the number n2 of the space debris impact abnormality error set R2, the number n3 of the electromagnetic interference abnormality error set R3, the orbital target tracking comparison value u1 in the error set R1, the space debris monitoring comparison value u2 in the error set R2, and the space environment monitoring comparison value u3 in the error set R3 are obtained, and the minimum orbital target abnormality neighborhood time difference t1min of each orbital target abnormality without error or omission, the orbit without error or omission, and the orbit target tracking comparison value u1 in the error set R1, the space debris monitoring comparison value u2 in the error set R2, and the space environment monitoring comparison value u3 in the error set R3 are obtained. The difference y1 of target tracking contrast value, the minimum space debris impact anomaly neighborhood time difference t2min of each space debris impact anomaly without any error or omission, the difference y2 of space debris monitoring contrast value without any error or omission, the minimum electromagnetic interference anomaly neighborhood time difference t3min of each electromagnetic interference anomaly without any error or omission and the difference y3 of electromagnetic interference monitoring contrast value without any error or omission generate a heterogeneous data matrix [n1, n2, n3, u1, u2, u3, t1min, y1, t2min, y2, t3min, y3].

[0117] Specifically, the number n1 of the orbital target abnormal situation error omission set R1 refers to the number of data in the orbital target abnormal situation error omission set, the number n2 of the space debris impact abnormal situation error omission set R2 refers to the number of data in the space debris impact abnormal situation error omission set, and the number n3 of the electromagnetic interference abnormal situation error omission set R3 refers to the number of data in the electromagnetic interference abnormal situation error omission set.

[0118] Specifically, in step S4, when evaluating the comprehensive capability of the space target monitoring system, a neural network model is constructed, and the neural network model is trained according to a historical heterogeneous evaluation data training set, and the trained neural network model is verified by a historical heterogeneous evaluation data verification set. When the verification accuracy of the neural network model does not reach 98%, a new historical heterogeneous evaluation data training set is used to train the neural network model until the verification accuracy of the neural network model reaches 98%. The neural network model is then output as a deep learning model, the heterogeneous data matrix is ​​input into the deep learning model, and the deep learning model outputs the evaluation result.

[0119] Specifically, the historical heterogeneous evaluation data training set refers to a data set used to train a neural network model, such as historical data in the form of heterogeneous data matrices and evaluation results corresponding to the heterogeneous data matrices; the historical heterogeneous evaluation data verification set refers to a data set for verifying the neural network model; the newly added historical heterogeneous evaluation data training set refers to a newly added historical heterogeneous evaluation data training set that is different from the previous historical heterogeneous evaluation data training set; the deep learning model refers to an algorithm model that can output evaluation results after inputting a heterogeneous data matrix after training; the evaluation result refers to the result output by the deep learning model, with the comprehensive capability of the space target monitoring system as the content, including the comprehensive capability of the space target monitoring system is excellent, the comprehensive capability of the space target monitoring system is good, and the comprehensive capability of the space target monitoring system is poor.

[0120] Specifically, in step S5, when outputting the intelligent optimization decision, an intelligent optimization judgment is performed according to the evaluation result, wherein:

[0121] When the evaluation result shows that the comprehensive capability of the space target monitoring system is excellent, it is determined that intelligent optimization will not be performed;

[0122] When the evaluation result shows that the comprehensive capability of the space target monitoring system is good, it is determined to perform intelligent optimization;

[0123] When the evaluation result shows that the comprehensive capability of the space target monitoring system is poor, it is determined to perform intelligent optimization;

[0124] When deciding to perform intelligent optimization, the heterogeneous data matrix is ​​input into the optimization decision model, and the optimization decision model outputs the intelligent optimization decision.

[0125] Specifically, the optimization decision model refers to an algorithm model whose content is to optimize the comprehensive capabilities of the space target monitoring system, and the intelligent optimization decision refers to an intelligent optimization method whose content is to optimize the comprehensive capabilities of the space target monitoring system and is output by the optimization decision model.

[0126] Specifically, in step S6, when the Pareto optimal solution is output, the operation of the multi-objective optimization algorithm is judged according to the intelligent optimization decision output g, where:

[0127] When g = 1, it is determined that the multi-objective optimization algorithm is not run;

[0128] When g>1, it is determined to run the multi-objective optimization algorithm, and output the Pareto optimal solution through the multi-objective optimization algorithm.

[0129] Specifically, the intelligent optimization decision output refers to the number of intelligent optimization decisions output by the optimization decision model, and the multi-objective optimization algorithm refers to an algorithm for optimizing and screening multiple intelligent optimization decisions. This embodiment does not limit the specific type of the multi-objective optimization algorithm. Technical personnel in this field can freely set it according to actual conditions, and only need to meet the needs of optimizing and screening multiple intelligent optimization decisions. For example, the multi-objective optimization algorithm can be set to the NSGA-II multi-objective evolutionary algorithm, and the Pareto optimal solution refers to the intelligent optimization decision with the best effect among multiple intelligent optimization decisions.

[0130] Specifically, in step S7, when making a decision execution judgment, a decision execution simulation is performed according to the intelligent optimization decision and the Pareto optimal solution, wherein:

[0131] When the intelligent optimization decision output g=1, the intelligent optimization decision output by the optimization decision model is input into the decision execution simulation model, and the simulation data is output;

[0132] When the output of the intelligent optimization decision g>1, the Pareto optimal solution is input into the decision execution simulation model, and the simulation data is output;

[0133] The simulation data is input into the deep learning model and the evaluation result is output. The evaluation result is used as the simulation result, and the decision execution judgment is made according to the simulation result, where:

[0134] If the simulation result shows that the comprehensive capability of the space target monitoring system is excellent, the decision is made;

[0135] If the simulation result shows that the comprehensive capability of the space target monitoring system is good, it is determined that no decision execution will be made;

[0136] If the simulation result shows that the comprehensive capability of the space target monitoring system is poor, it is determined that no decision will be executed.

[0137] Specifically, the decision execution simulation model refers to a simulation model that simulates the intelligent optimization decision output by the optimization decision model, and the simulation data refers to the data obtained after simulating the intelligent optimization decision output by the decision execution simulation model.

[0138] Specifically, in step S8, when the decision execution judgment result is to perform decision execution, the controller executes the intelligent optimization decision and Pareto optimal solution corresponding to the decision execution judgment result to optimize the space target monitoring system.

[0139] Specifically, the controller refers to a device that controls the space target monitoring system according to the intelligent optimization decision and Pareto optimal solution corresponding to the decision execution judgment result. This embodiment does not limit the setting position of the controller. Technical personnel in this field can freely set it according to actual conditions. It only needs to meet the control requirements of the space target monitoring system. For example, the controller can be installed in the processing terminal of the space target monitoring system.

[0140] Specifically, in step S9, when performing multi-layer feedback, the evaluation result of the comprehensive capability of the space target monitoring system within the execution cycle is obtained, which is used as the optimization result, and the evaluation effect of the comprehensive capability of the space target monitoring system is fed back according to the simulation result and the optimization result, wherein:

[0141] When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is excellent, the feedback is that the comprehensive capability evaluation effect of the space target monitoring system is excellent;

[0142] When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is good, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor;

[0143] When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is poor, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor;

[0144] When the simulation result shows that the comprehensive capability of the space target monitoring system is good, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor;

[0145] When the simulation result shows that the comprehensive capability of the space target monitoring system is poor, the feedback indicates that the comprehensive capability evaluation effect of the space target monitoring system is poor.

[0146] Specifically, the execution cycle refers to the preset time interval after the controller executes the intelligent optimization decision and Pareto optimal solution corresponding to the decision execution judgment result. The evaluation result of the comprehensive capability of the space target monitoring system within the execution cycle refers to the result obtained by evaluating the comprehensive capability of the optimized space target monitoring system through a neural network model based on simulation results and optimization results.

[0147] Specifically, in step S10, the optimization conditions of the multi-objective optimization algorithm, the optimization decision model and the deep learning model are judged according to the feedback results, wherein:

[0148] When the intelligent optimization decision output g=1, it is determined that the optimization decision model and the deep learning model are optimized;

[0149] When the intelligent optimization decision output g>1, it is determined to optimize the multi-objective optimization algorithm, the optimization decision model and the deep learning model.

[0150] It can be understood that the present embodiment does not limit the specific method of optimizing the multi-objective optimization algorithm, the optimization decision model and the deep learning model. Technical personnel in this field can freely set it according to the actual situation, and it is only necessary to meet the optimization requirements of the multi-objective optimization algorithm, the optimization decision model and the deep learning model. For example, the optimization decision model and the deep learning model can be set to be optimized and trained by adding expert data.

[0151] Specifically, in step S11, the comprehensive capability evaluation effect of the space target monitoring system within the optimized monitoring time is obtained, and the optimization effect is judged according to the comprehensive capability evaluation effect of the space target monitoring system within the optimized monitoring time, wherein:

[0152] When the evaluation result within the optimized monitoring time is that the comprehensive capability evaluation effect of the space target monitoring system is excellent, the optimization effect is determined to be effective;

[0153] When the evaluation result within the optimized monitoring time is that the comprehensive capability evaluation effect of the space target monitoring system is poor, the optimization effect is determined to be invalid;

[0154] And according to the optimization effect, the dimension increase judgment of the heterogeneous data of the space target monitoring system is carried out, among which:

[0155] When the optimization effect is effective, it is determined that no dimension increase is performed;

[0156] When the optimization effect is invalid, it is determined to increase the dimension;

[0157] When increasing the dimension, a grid search is performed on the spare type heterogeneous data, and a cross-validation is performed. The increased dimension type heterogeneous data in the spare type heterogeneous data is screened out according to the cross-validation results, and the increased dimension type heterogeneous data is added to the heterogeneous data of the space target monitoring system for collection.

[0158] Specifically, the optimization monitoring time refers to the preset time interval for monitoring the optimization effect, such as the optimization monitoring time can be set to 24 hours, the dimension increase refers to increasing the dimension of the heterogeneous data of the space target monitoring system, the spare type heterogeneous data refers to the preset dimension of the heterogeneous data of the space target monitoring system used for preparing the dimension increase, the grid search refers to, the cross-validation refers to the technology used to evaluate the performance of the statistical model and better estimate the performance of the model on unseen data in order to prevent overfitting and underfitting of the model. This embodiment does not limit the screening method of the dimension increase type heterogeneous data in the spare type heterogeneous data. Technical personnel in this field can freely set it according to actual conditions, and only need to meet the screening requirements of the spare type heterogeneous data, such as setting the spare type heterogeneous data to be screened by leave-one-out cross-validation.

[0159] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A comprehensive capability evaluation method for a space target monitoring system, characterized in that: include: Step S1, collecting heterogeneous data of the space target monitoring system; Step S2, obtaining an error and omission set of each abnormal situation according to the heterogeneous data of the space target monitoring system; Step S3, generating a heterogeneous data matrix according to the error and omission sets of each abnormal situation; Step S4, inputting the heterogeneous data matrix into the deep learning model to evaluate the comprehensive capability of the space target monitoring system, and outputting the evaluation result; Step S5, inputting the heterogeneous data matrix into the optimization decision model according to the evaluation result, and outputting the intelligent optimization decision through the optimization decision model; Step S6, running a multi-objective optimization algorithm according to the output of the intelligent optimization decision, and outputting a Pareto optimal solution; Step S7, performing decision execution simulation according to the intelligent optimization decision and the Pareto optimal solution, and making a decision execution judgment according to the simulation results; Step S8, optimizing the space target monitoring system according to the decision execution judgment result; Step S9, providing feedback on the comprehensive capability evaluation effect of the space target monitoring system according to the simulation results and the optimization results; Step S10, optimizing the multi-objective optimization algorithm, the optimization decision model and the deep learning model according to the feedback results; Step S11, increasing the dimension of the heterogeneous data of the space target monitoring system according to the optimization effect; In step S2, when making an error or omission judgment, the orbit target tracking comparison value is compared with the orbit target preset value to identify an abnormal situation of the orbit target, the space debris monitoring comparison value is compared with the space debris impact preset value to identify an abnormal situation of space debris impact, and the space environment monitoring comparison value is compared with the preset electromagnetic interference value to identify an abnormal situation of electromagnetic interference; In step S2, when making an error or omission judgment, the minimum orbit target abnormal neighborhood time difference t1min of each orbit target abnormality is compared with the preset neighborhood time difference t0, and the error or omission of the orbit target abnormality is judged according to the comparison result, wherein: When t1min≤t0, it is determined that there is no error or omission in the abnormal situation of the orbit target; When t1min>t0, it is determined that there is an error or omission in the orbit target abnormality, and the orbit target abnormality is added to the orbit target abnormality error or omission set R1; Obtain the difference y1 of the orbital target tracking comparison value without any error or omission, and compare it with the difference y10 of the preset orbital target tracking comparison value. According to the comparison result, the matching of the orbital target abnormality without any error or omission and the monitoring orbital target abnormality output by the space target monitoring system corresponding to the minimum orbital target abnormality neighborhood time difference is judged, where: When y1≤y10, it is determined that the orbital target anomaly situation without any omissions and the monitored orbital target anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference match each other, and the monitored orbital target anomaly situation is used as the matched orbital target anomaly situation; When y1>y10, it is determined that the orbital target anomaly situation without any omission does not match the monitored orbital target anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference, and it is compensated to the orbital target anomaly situation omission set R1; The minimum space debris impact anomaly neighborhood time difference t2min of each space debris impact anomaly is compared with the preset neighborhood time difference t0, and the error and omission of the space debris impact anomaly is judged according to the comparison result, where: When t2min≤t0, it is determined that there is no error or omission in the abnormal situation of space debris impact; When t2min>t0, it is determined that there is an error in the space debris impact anomaly, and the space debris impact anomaly is added to the space debris impact anomaly error set R2; The difference y2 of the space debris monitoring comparison value without any omission is obtained, and it is compared with the difference y20 of the preset space debris monitoring comparison value. According to the comparison result, the matching of the space debris impact anomaly situation without any omission and the monitoring space debris impact anomaly situation output by the space target monitoring system corresponding to the time difference of the minimum space debris impact anomaly neighborhood is judged, wherein: When y2≤y20, it is determined that the space debris impact anomaly situation without any omissions and the monitored space debris impact anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference match each other, and it is used as the matching space debris impact anomaly situation; When y2>y20, it is determined that the space debris impact anomaly situation without any omission does not match the monitored space debris impact anomaly situation output by the space target monitoring system corresponding to the minimum orbital target anomaly neighborhood time difference, and it is compensated to the space debris impact anomaly situation omission set R2; The minimum electromagnetic interference abnormal neighborhood time difference t3min of each electromagnetic interference abnormal situation is compared with the preset neighborhood time difference t0, and the error and omission of the electromagnetic interference abnormal situation is judged according to the comparison result, where: When t3min≤t0, it is determined that there is no error or omission in the abnormal electromagnetic interference situation; When t3min>t0, it is determined that the electromagnetic interference abnormality has an error or omission, and the electromagnetic interference abnormality is added to the electromagnetic interference abnormality error or omission set R3; The difference y3 of the electromagnetic interference monitoring comparison value without any error or omission is obtained, and it is compared with the difference y30 of the preset electromagnetic interference monitoring comparison value. According to the comparison result, the matching of the electromagnetic interference abnormality output by the space target monitoring system corresponding to the electromagnetic interference abnormality without any error or omission and its minimum electromagnetic interference abnormality neighborhood time difference is judged, where: When y3≤y30, it is determined that the electromagnetic interference anomaly without any error or omission and the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the minimum electromagnetic interference anomaly neighborhood time difference match each other, and it is used as the matched electromagnetic interference anomaly; When y3>y30, it is determined that the electromagnetic interference anomaly without any omission does not match the monitored electromagnetic interference anomaly output by the space target monitoring system corresponding to the minimum electromagnetic interference anomaly neighborhood time difference, and it is compensated to the electromagnetic interference anomaly error set R3; The error omission sets of each abnormal situation include an orbit target abnormal situation error omission set R1, a space debris impact abnormal situation error omission set R2 and an electromagnetic interference abnormal situation error omission set R3.

2. The comprehensive capability evaluation method of the space target monitoring system according to claim 1, characterized in that: In the step S3, when generating the heterogeneous data matrix, the number n1 of the orbital target abnormality error set R1, the number n2 of the space debris impact abnormality error set R2, the number n3 of the electromagnetic interference abnormality error set R3, the orbital target tracking comparison value u1 in the error set R1, the space debris monitoring comparison value u2 in the error set R2, and the space environment monitoring comparison value u3 in the error set R3 are obtained, and the minimum orbital target abnormality neighborhood time difference t1min of each orbital target abnormality without error or omission, the orbital target without error or omission, and the orbital target tracking comparison value u1 in the error set R1, the space debris monitoring comparison value u2 in the error set R2, and the space environment monitoring comparison value u3 in the error set R3 are obtained. The difference y1 of the tracking comparison values, the minimum space debris impact anomaly neighborhood time difference t2min of each space debris impact anomaly without any error or omission, the difference y2 of the space debris monitoring comparison values ​​without any error or omission, the minimum electromagnetic interference anomaly neighborhood time difference t3min of each electromagnetic interference anomaly without any error or omission and the difference y3 of the electromagnetic interference monitoring comparison values ​​without any error or omission are used to generate a heterogeneous data matrix [n1, n2, n3, u1, u2, u3, t1min, y1, t2min, y2, t3min, y3].

3. The comprehensive capability evaluation method of the space target monitoring system according to claim 2 is characterized in that: In step S4, when evaluating the comprehensive capability of the space target monitoring system, a neural network model is constructed, and the neural network model is trained according to a historical heterogeneous evaluation data training set, and the trained neural network model is verified by a historical heterogeneous evaluation data verification set. When the verification accuracy of the neural network model does not reach 98%, a new historical heterogeneous evaluation data training set is added to train the neural network model until the verification accuracy of the neural network model reaches 98%. The neural network model is output as a deep learning model, the heterogeneous data matrix is ​​input into the deep learning model, and the deep learning model outputs the evaluation result.

4. The comprehensive capability evaluation method of the space target monitoring system according to claim 3 is characterized in that: In step S5, when outputting the intelligent optimization decision, an intelligent optimization judgment is performed according to the evaluation result, wherein: When the evaluation result shows that the comprehensive capability of the space target monitoring system is excellent, it is determined that intelligent optimization will not be performed; When the evaluation result shows that the comprehensive capability of the space target monitoring system is good, it is determined to perform intelligent optimization; When the evaluation result shows that the comprehensive capability of the space target monitoring system is poor, it is determined to perform intelligent optimization; When deciding to perform intelligent optimization, the heterogeneous data matrix is ​​input into the optimization decision model, and the optimization decision model outputs the intelligent optimization decision.

5. The comprehensive capability evaluation method of the space target monitoring system according to claim 4 is characterized in that: In step S6, when the Pareto optimal solution is output, the operation of the multi-objective optimization algorithm is judged according to the intelligent optimization decision output g, wherein: When g = 1, it is determined that the multi-objective optimization algorithm is not run; When g>1, it is determined to run the multi-objective optimization algorithm, and output the Pareto optimal solution through the multi-objective optimization algorithm.

6. The comprehensive capability evaluation method of the space target monitoring system according to claim 5 is characterized in that: In step S7, when making a decision execution judgment, a decision execution simulation is performed according to the intelligent optimization decision and the Pareto optimal solution, wherein: When the intelligent optimization decision output g=1, the intelligent optimization decision output by the optimization decision model is input into the decision execution simulation model, and the simulation data is output; When the output of the intelligent optimization decision g>1, the Pareto optimal solution is input into the decision execution simulation model, and the simulation data is output; The simulation data is input into the deep learning model and the evaluation result is output. The evaluation result is used as the simulation result, and the decision execution judgment is made according to the simulation result, where: If the simulation result shows that the comprehensive capability of the space target monitoring system is excellent, the decision is made; If the simulation result shows that the comprehensive capability of the space target monitoring system is good, it is determined that no decision execution will be made; If the simulation result shows that the comprehensive capability of the space target monitoring system is poor, it is determined that no decision will be executed.

7. The comprehensive capability evaluation method of the space target monitoring system according to claim 6 is characterized in that: In step S8, when the decision execution judgment result is to perform decision execution, the controller executes the intelligent optimization decision and Pareto optimal solution corresponding to the decision execution judgment result to optimize the space target monitoring system.

8. The comprehensive capability evaluation method of the space target monitoring system according to claim 7 is characterized in that: In step S9, when performing multi-layer feedback, the evaluation result of the comprehensive capability of the space target monitoring system within the execution cycle is obtained, which is used as the optimization result, and the evaluation effect of the comprehensive capability of the space target monitoring system is fed back according to the simulation result and the optimization result, wherein: When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is excellent, the feedback is that the comprehensive capability evaluation effect of the space target monitoring system is excellent; When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is good, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor; When the simulation result shows that the comprehensive capability of the space target monitoring system is excellent and the optimization result shows that the comprehensive capability of the space target monitoring system is poor, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor; When the simulation result shows that the comprehensive capability of the space target monitoring system is good, the feedback evaluation effect of the comprehensive capability of the space target monitoring system is poor; When the simulation result shows that the comprehensive capability of the space target monitoring system is poor, the feedback indicates that the comprehensive capability evaluation effect of the space target monitoring system is poor.

9. The comprehensive capability evaluation method of the space target monitoring system according to claim 8, characterized in that: In step S10, the optimization conditions of the multi-objective optimization algorithm, the optimization decision model and the deep learning model are judged according to the feedback results, wherein: When the intelligent optimization decision output g=1, it is determined that the optimization decision model and the deep learning model are optimized; When the output of the intelligent optimization decision g>1, it is determined that the multi-objective optimization algorithm, the optimization decision model and the deep learning model are optimized; In step S11, the comprehensive capability evaluation effect of the space target monitoring system within the optimized monitoring time is obtained, and the optimization effect is judged according to the comprehensive capability evaluation effect of the space target monitoring system within the optimized monitoring time, wherein: When the evaluation result within the optimized monitoring time is that the comprehensive capability evaluation effect of the space target monitoring system is excellent, the optimization effect is determined to be effective; When the evaluation result within the optimized monitoring time is that the comprehensive capability evaluation effect of the space target monitoring system is poor, the optimization effect is determined to be invalid; And according to the optimization effect, the dimension increase judgment of the heterogeneous data of the space target monitoring system is carried out, among which: When the optimization effect is effective, it is determined that no dimension increase is performed; When the optimization effect is invalid, it is determined to increase the dimension; When increasing the dimension, a grid search is performed on the spare type heterogeneous data, and a cross-validation is performed. The increased dimension type heterogeneous data in the spare type heterogeneous data is screened out according to the cross-validation results, and the increased dimension type heterogeneous data is added to the heterogeneous data of the space target monitoring system for collection.

Images