Spacecraft control system fault identification method and device based on multi-source information fusion

Abstract

The application discloses a spacecraft control system fault identification method and device based on multi-source information fusion. The method comprises the following steps: multi-source fusion of data in the development and testing process of a spacecraft and historical data in orbit to obtain a plurality of fault samples; each fault sample is provided with a fault label, and each fault label is used for representing a fault mode corresponding to the fault sample; each fault sample is normalized to obtain a normalized fault sample; the normalized fault sample is subjected to clustering processing based on a preset clustering algorithm to obtain a plurality of clusters and a fault mode of each cluster; each cluster comprises at least one core object; for in-orbit data at any moment, the distance between the in-orbit data and the core object in each cluster is calculated to determine the cluster to which the in-orbit data belongs based on the calculated distance; and the fault mode corresponding to the cluster to which the in-orbit data belongs is determined as the identification result of the in-orbit data. The application can improve the accuracy of spacecraft fault identification.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fault diagnosis, in particular to a spacecraft control system fault identification method and device based on multi-source information fusion. BACKGROUND

[0002] At present, when identifying faults of a spacecraft control system, whether autonomous identification on the satellite or auxiliary identification on the ground, a model-based method or a signal processing method is mainly used. The two methods generally identify faults according to the appearance of on-orbit faults, without considering the fault knowledge contained in the data of product development, testing process and on-orbit historical data, resulting in waste of useful information.

[0003] Therefore, there is an urgent need for a spacecraft control system fault identification method and device based on multi-source information fusion to solve the above problems. SUMMARY

[0004] The present application provides a spacecraft control system fault identification method and device based on multi-source information fusion, which can improve the accuracy of spacecraft fault identification. The technical solution is as follows:

[0005] On the one hand, a spacecraft control system fault identification method based on multi-source information fusion is provided, which comprises:

[0006] Data of a spacecraft in the development and testing process and on-orbit historical data are multi-source fused to obtain a plurality of fault samples; each fault sample is provided with a fault label, and each fault label is used to represent the fault mode corresponding to the fault sample;

[0007] Each fault sample is normalized to obtain a normalized fault sample;

[0008] The normalized fault sample is clustered based on a preset clustering algorithm to obtain a plurality of clusters and the fault mode of each cluster; wherein each cluster includes at least one core object;

[0009] For the on-orbit data of a spacecraft at any time, the distance between the on-orbit data and the core object in each cluster is calculated to determine the cluster to which the on-orbit data belongs based on the calculated distance;

[0010] The fault mode corresponding to the cluster to which the on-orbit data belongs is determined as the identification result of the on-orbit data.

[0011] On the other hand, a spacecraft control system fault identification device based on multi-source information fusion is provided, which comprises:

[0012] A fusion unit is configured to fuse data of the spacecraft in a development and test process and in-orbit historical data to obtain a plurality of fault samples; each of the fault samples is provided with a fault label, and each of the fault labels is used to represent a fault mode corresponding to the fault sample;

[0013] A processing unit is configured to perform normalization processing on each of the fault samples to obtain normalized fault samples;

[0014] A clustering unit is configured to perform clustering processing on the normalized fault samples based on a preset clustering algorithm to obtain a plurality of clusters and a fault mode of each cluster; each cluster includes at least one core object;

[0015] A calculation unit is configured to calculate a distance between in-orbit data of the spacecraft at any moment and the core object in each cluster based on the in-orbit data, and determine a cluster to which the in-orbit data belongs based on the calculated distance;

[0016] A determination unit is configured to determine a fault mode corresponding to the cluster to which the in-orbit data belongs as an identification result of the in-orbit data.

[0017] In another aspect, a computer readable storage medium is provided, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the spacecraft control system fault identification method based on multi-source information fusion.

[0018] In another aspect, a computer program product is provided, and the computer program product includes a computer program, and the computer program is executed by a processor to implement the steps of the spacecraft control system fault identification method based on multi-source information fusion.

[0019] The spacecraft control system fault identification method based on multi-source information fusion provided by the embodiments of the present application fully considers data of the spacecraft in a development and test process and in-orbit historical data, and the fault samples determined through multi-source information fusion contain more useful information. Therefore, through clustering analysis on the fault samples, the cluster and the fault mode of each cluster obtained are more accurate. Finally, based on the obtained cluster, in-orbit data is identified to obtain more accurate results. Therefore, the present application can improve the accuracy of spacecraft fault identification. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0021] Figure 1 is a flow chart of a spacecraft control system fault identification method based on multi-source information fusion provided by an embodiment of the present application;

[0022] Figure 2 is a structural diagram of a spacecraft control system fault identification device based on multi-source information fusion provided by an embodiment of the present application;

[0023] Figure 3 is a hardware architecture diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in connection with the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the protection scope of the present application.

[0025] The specific implementation of the above concept will be described below.

[0026] Referring to Figure 1 The embodiments of the present application provide a spacecraft control system fault identification method based on multi-source information fusion, which comprises the following steps:

[0027] In step 100, data in the development and testing process of a spacecraft and historical data in orbit are fused to obtain a plurality of fault samples. Each fault sample is provided with a fault label, and each fault label is used to represent a fault mode corresponding to the fault sample.

[0028] In step 102, each fault sample is normalized to obtain a normalized fault sample.

[0029] In step 104, the normalized fault sample is clustered based on a preset clustering algorithm to obtain a plurality of clusters and a fault mode of each cluster. Each cluster includes at least one core object.

[0030] In step 106, for spacecraft in-orbit data at any time, the distance between the in-orbit data and the core object in each cluster is calculated to determine the cluster to which the in-orbit data belongs based on the calculated distance.

[0031] In step 108, the fault mode corresponding to the cluster to which the in-orbit data belongs is determined as the identification result of the in-orbit data.

[0032] In this embodiment, the data of the spacecraft in the development and test process and the in-orbit historical data are fully considered, the fault samples determined through multi-source information fusion contain more useful information. Therefore, through clustering analysis on the fault samples, the obtained clusters and the fault modes of each cluster are more accurate. Finally, based on the obtained clusters, the in-orbit data are subjected to fault identification, and the obtained result is more accurate. It can be seen that the application can improve the accuracy of spacecraft fault identification.

[0033] The execution manner of each step shown in the above is described below. Figure 1 The execution manner of each step shown in the above is described below.

[0034] Firstly, for step 100 and step 102:

[0035] In the above steps, the fault label is used to represent the fault mode corresponding to the fault sample. For example, the fault label of a certain fault sample is 1, which represents that the fault mode is fault mode 1; the fault label 2 represents fault mode 2, etc. Each fault mode represents a different hazard level, for example, some faults are single-point faults, which will have a serious impact on the spacecraft once they occur, while the occurrence of some faults has little impact on the spacecraft. Therefore, when performing fault identification, the hazard level of the fault needs to be fully considered, and the fault of the key component needs to be focused on. Generally, the hazard level is closely related to the fault severity, whether it is a single-point fault, the fault occurrence probability, the fault component, etc., and is generally determined by the spacecraft FMEA.

[0036] In addition, since the formats of the data of the spacecraft in the development and test process and the in-orbit historical data are different, the data need to be standardized for subsequent data analysis.

[0037] For step 104, the preset clustering algorithm is DBSCAN algorithm, the normalized fault samples are subjected to clustering processing based on the preset clustering algorithm, and a plurality of clusters and the fault mode of each cluster are obtained, including:

[0038] The DBSCAN algorithm is optimized based on the particle swarm algorithm, and the optimal field radius and the optimal minimum number of neighbors of the DBSCAN algorithm are obtained;

[0039] Each fault sample is sequentially taken as a target fault sample, a circular region is determined with the target fault sample as the center and the optimal field radius as the radius; each fault sample is traversed, for each fault sample traversed, the Euclidean distance between the fault sample and the target fault sample is calculated, and it is judged whether the calculated Euclidean distance is less than the optimal field radius; if yes, it is determined that the fault sample is in the circular region, and if no, it is determined that the fault sample is not in the circular region; until each fault sample is traversed, the number of fault samples in the circular region is obtained;

[0040] determining whether the number of fault samples in the circular region is greater than the optimal minimum number of neighbors; if yes, determining that the target fault sample is a core object, and determining all density-connected fault samples from the core object;

[0041] For each determined core object, the following is performed: traversing all core objects within the optimal domain radius of the core object, and finding all density-connected fault samples of the core objects to form a corresponding cluster, and determining a fault mode of the cluster based on a fault label of each fault sample contained in the cluster.

[0042] In this step, the DBSCAN (Density-Based Spatial Clustering of Application with Noise) algorithm can divide a region with a high enough density into a cluster according to the density of data distribution, so as to realize fault identification by judging the cluster in which the spacecraft on-orbit data is located.

[0043] In some embodiments, determining the fault mode of the cluster based on the fault label of each fault sample contained in the cluster includes:

[0044] The number of fault samples corresponding to each fault label in the cluster is counted respectively;

[0045] The fault mode represented by the fault label with the largest number of corresponding fault samples is determined as the fault mode of the cluster.

[0046] For example, a certain cluster contains 20 fault samples, of which 15 fault samples have fault label 2 and 5 fault samples have fault label 1. Since the number of fault samples corresponding to fault label 2 is the largest in the cluster, the fault mode represented by fault label 2 is determined as the fault mode of the cluster.

[0047] In some embodiments, the preset clustering algorithm is optimized based on a particle swarm algorithm to obtain an optimal domain radius and an optimal minimum number of neighbors of the clustering algorithm, including:

[0048] S1, initializing particle swarm parameters, the particle swarm parameters including a number of particles, an initial best position of the particle swarm, an initial position of each particle, an initial speed, and an initial best position of the particle;

[0049] S2, sequentially traverse each particle in the particle swarm. For each traversed particle, perform the following: sequentially treat each fault sample as a sub-target fault sample, and determine a circular region with the sub-target fault sample as the center and the current loop's domain radius as the radius; traverse other fault samples, for each traversed fault sample, calculate the Euclidean distance between the fault sample and the sub-target fault sample, and determine whether the calculated Euclidean distance is less than the current loop's domain radius; if so, determine that the fault sample is within the circular region; if not, determine that the fault sample is not within the circular region; continue until all fault samples have been traversed, obtaining the number of fault samples within the circular region; determine whether the number of fault samples within the circular region is greater than the current loop's minimum neighbor number; if so, determine that the sub-target fault sample is the core object, and identify all density-connected fault samples originating from the core object;

[0050] For each identified core object, the following steps are performed: traverse all core objects within the current loop's domain radius, find all fault samples that are density-connected to these core objects to form a corresponding cluster, and determine the fault mode of the cluster based on the fault label of each fault sample contained in the cluster.

[0051] For each identified cluster, the following steps are performed: traverse each fault sample in the cluster and determine whether the fault mode corresponding to the fault label of the traversed fault sample is consistent with the fault mode of the cluster; if yes, keep the fault sample; if no, remove the fault sample from the cluster; continue until each fault sample in the cluster has been traversed to obtain the exact number of samples in the cluster.

[0052] S3, based on the number of accurate samples in each cluster, calculate the fault identification accuracy of the current loop, and then execute S4;

[0053] S4, determine whether the fault identification accuracy of the current loop is greater than the accuracy corresponding to the current best position of the current particle. If it is greater, update the best position of the current particle. Determine whether the fault identification accuracy of the current loop is greater than the accuracy corresponding to the current best position of the current particle swarm. If it is greater, update the best position of the current particle swarm.

[0054] S5, determine whether the preset stopping condition has been met; if not, update the position and velocity of each particle, increment the iteration count by 1, and repeat S2 to S4 until the preset stopping condition is met to obtain the final best position of the particle swarm.

[0055] S6 uses the neighborhood radius and minimum neighbor number corresponding to the final best position of the particle swarm as the optimal neighborhood radius and optimal minimum neighbor number of the clustering algorithm.

[0056] In this embodiment, it is assumed that the size of the particle swarm is n, the position of the i-th particle is represented as x id = [x i1 i2 ], where x i1 =Eps, x i2 =MinPts, i=1, 2, … n, d=1, 2. The velocity of the i-th particle is represented as v id , the best position experienced by the i-th particle is represented as pid, and the best position experienced by the particle swarm is represented as p gd ; Eps is the field radius of the DBSCAN algorithm; and MinPts is the minimum number of neighbors of the DBSCAN algorithm.

[0057] The i-th particle then updates its own velocity and position according to the following formula:

[0058] v id = ωv id + c1r1(p id -x id ) + c2r2(p gd -x id )

[0059] x id =x id +v id

[0060] where i=1, 2, … n, d=1, 2; ω, c1, and c2 are constants, usually ω=1, c1=c2=2, γ1 and γ2 are random numbers between 0 and 1, and the initial values of x id and v id are generally randomly given.

[0061] It should be noted that, in the parameter optimization of this embodiment, the fault recognition accuracy of the fault hazard level is fully considered, and the accuracy of the optimization result can be improved.

[0062] It should also be noted that the following method is used to determine whether two parameters are density-connected:

[0063] Suppose p and q are any two points in the data set, and density accessibility mainly means that there is at least one data point k satisfying: p∈N eps (k), q∈N eps (k), where N eps (k) represents the set of neighbor points with k as the center and Eps as the neighborhood radius. If p and q are density accessible to the same point, then p and q are density connected.

[0064] In some embodiments, the fault recognition rate is calculated using the following formula:

[0065] ​

[0066] In the formula, c is a fault recognition rate; j = 1, 2,..., N, N is the number of fault modes; w j is a hazard level of the jth fault mode; N j is the total number of fault samples contained in the cluster corresponding to the jth fault mode; is the number of correct fault samples contained in the cluster corresponding to the jth fault mode.

[0067] In this step, the hazard level is closely related to the fault severity, whether it is a single-point fault, the fault occurrence probability, the fault component to which the fault belongs, and the like, and is generally determined by the spacecraft FMEA.

[0068] In some embodiments, the preset stop condition is that:

[0069] the accuracy corresponding to the best position of the current cycle of the particle swarm is greater than an accuracy threshold value;

[0070] and / or the number of iterations is greater than a number threshold value.

[0071] Finally, for steps 106 and 108:

[0072] Step 106 includes: traversing each cluster, and for each cluster traversed, performing:

[0073] calculating the distance between the on-orbit data and each core object in the cluster; and determining whether there is a core object that satisfies the distance from the on-orbit data to the core object being less than the optimal field radius; if so, determining that the on-orbit data belongs to the cluster; and if not, traversing the next cluster until the cluster to which the on-orbit data belongs is determined.

[0074] After the cluster to which the on-orbit data belongs is determined, the fault mode corresponding to the cluster is the fault mode of the on-orbit data, that is, the recognition result. For example, if the fault mode of the cluster is a single-point fault, then the fault mode of the on-orbit data is a single-point fault.

[0075] As shown in Figure 2 , Figure 3 An embodiment of the present application provides a spacecraft control system fault recognition device based on multi-source information fusion. The device embodiment can be realized by software, or realized by hardware or a combination of software and hardware. From the hardware layer, as shown in Figure 2 is a hardware architecture diagram of a computing device on which the spacecraft control system fault recognition device based on multi-source information fusion is provided, in addition to Figure 2In addition to the processor, the memory, the network interface, and the non-volatile memory shown, the computing device in which the apparatus in the embodiments is generally also comprised of other hardware, such as a forwarding chip responsible for processing packets, and the like. For example, in the case of software implementation, as shown in Figure 3 As a logical sense of the apparatus, it is formed by the CPU of the computing device in which it is located reading the corresponding computer program in the non-volatile memory into the memory and running.

[0076] Please refer to Figure 3 The embodiment of the present application provides a spacecraft control system fault identification apparatus based on multi-source information fusion, which comprises:

[0077] The fusion unit 300 is configured to perform multi-source fusion on data in the development and test process of the spacecraft and historical data in orbit to obtain a plurality of fault samples; each fault sample is provided with a fault label, and each fault label is configured to represent a fault mode corresponding to the fault sample.

[0078] The processing unit 302 is configured to perform normalization processing on each fault sample to obtain a normalized fault sample.

[0079] The clustering unit 304 is configured to perform clustering processing on the normalized fault sample based on a preset clustering algorithm to obtain a plurality of clusters and a fault mode of each cluster; each cluster comprises at least one core object.

[0080] The computing unit 306 is configured to calculate the distance between the on-orbit data of the spacecraft at any moment and the core object in each cluster based on the on-orbit data, and determine the cluster to which the on-orbit data belongs based on the calculated distance.

[0081] The determining unit 308 is configured to determine the fault mode corresponding to the cluster to which the on-orbit data belongs as the identification result of the on-orbit data.

[0082] In some embodiments, the preset clustering algorithm is a DBSCAN algorithm.

[0083] The clustering unit 304 is configured to perform the following operations:

[0084] The DBSCAN algorithm is optimized based on a particle swarm algorithm to obtain an optimal field radius and an optimal minimum number of neighbors of the DBSCAN algorithm.

[0085] each fault sample as a target fault sample, and determining a circular region with the target fault sample as a center and the optimal field radius as a radius; traversing other fault samples, for each fault sample traversed, calculating a Euclidean distance between the fault sample and the target fault sample, and judging whether the calculated Euclidean distance is less than the optimal field radius; if yes, determining that the fault sample is in the circular region, and if no, determining that the fault sample is not in the circular region; until each fault sample is traversed, obtaining a number of fault samples in the circular region;

[0086] judging whether the number of fault samples in the circular region is greater than the optimal minimum neighborhood number; if yes, determining that the target fault sample is a core object, and determining all density-connected fault samples from the core object;

[0087] for each core object determined, performing: traversing all core objects within the optimal field radius of the core object, and finding all density-connected fault samples of the core objects to form a corresponding cluster, and determining a fault mode of the cluster based on a fault label of each fault sample included in the cluster.

[0088] In some embodiments, determining the fault mode of the cluster based on the fault label of each fault sample included in the cluster includes:

[0089] respectively counting a number of fault samples corresponding to each fault label in the cluster;

[0090] determining a fault mode represented by a fault label corresponding to the largest number of fault samples as the fault mode of the cluster.

[0091] In some embodiments, the preset clustering algorithm is optimized based on a particle swarm algorithm to obtain the optimal field radius and the optimal minimum neighborhood number of the clustering algorithm, including:

[0092] S1, initializing particle swarm parameters, the particle swarm parameters including a number of particles, an initial best position of the particle swarm, an initial position of each particle, an initial speed, and an initial best position of the particle;

[0093] S2, sequentially traverse each particle in the particle swarm, and for each particle traversed, sequentially take each fault sample as a sub-target fault sample, and determine a circular region with the sub-target fault sample as the center and the domain radius in the current cycle as the radius; traverse other fault samples, and for each fault sample traversed, calculate the Euclidean distance between the fault sample and the sub-target fault sample, and determine whether the calculated Euclidean distance is less than the domain radius in the current cycle; if yes, it is determined that the fault sample is in the circular region, and if no, it is determined that the fault sample is not in the circular region; until each fault sample is traversed, the number of fault samples in the circular region is obtained; determine whether the number of fault samples in the circular region is greater than the minimum neighborhood number in the current cycle; if yes, it is determined that the sub-target fault sample is a core object, and all density-connected fault samples from the core object are determined;

[0094] For each core object determined, traverse all core objects within the domain radius in the current cycle of the core object, and find all fault samples connected in density with these core objects to form a corresponding cluster, and determine the fault mode of the cluster based on the fault label of each fault sample contained in the cluster;

[0095] For each cluster determined, traverse each fault sample in the cluster, and determine whether the fault mode corresponding to the fault label of the traversed fault sample is consistent with the fault mode of the cluster; if yes, the fault sample is retained; if no, the fault sample is removed from the cluster; until each fault sample in the cluster is traversed, the number of accurate samples in the cluster is obtained;

[0096] S3, based on the number of accurate samples in each cluster, calculate the fault recognition accuracy rate in the current cycle, and perform S4;

[0097] S4, determine whether the fault recognition accuracy rate in the current cycle is greater than the accuracy rate corresponding to the current best position of the current particle, if greater, update the best position of the current particle; determine whether the fault recognition accuracy rate in the current cycle is greater than the accuracy rate corresponding to the current cycle best position of the current particle swarm, if greater, update the best position of the current particle swarm;

[0098] S5, determine whether a predetermined stop condition is reached; if no, update the position and speed of each particle, increment the iteration number by 1, and loop execute S2-S4 until the predetermined stop condition is reached, and the final best position of the particle swarm is obtained;

[0099] S6, take the domain radius and the minimum neighborhood number corresponding to the final best position of the particle swarm as the optimal domain radius and the optimal minimum neighborhood number of the clustering algorithm.

[0100] In some embodiments, the fault identification rate is calculated using the following formula:

[0101]

[0102] wherein c is the fault identification rate; j = 1, 2, …, N, N is the number of fault modes; w j is the hazard level of the jth fault mode; N j is the total number of fault samples contained in the cluster corresponding to the jth fault mode; is the number of correct fault samples contained in the cluster corresponding to the jth fault mode.

[0103] In some embodiments, the preset stopping condition is:

[0104] the accuracy corresponding to the best position of the current cycle of the particle swarm is greater than the accuracy threshold;

[0105] and / or the number of iterations is greater than the number threshold.

[0106] In some embodiments, the computing unit 306 is configured to perform the following operations:

[0107] traverse each cluster, and for each cluster traversed, perform the following operations:

[0108] calculate the distance between the on-orbit data and each core object in the cluster; and determine whether there is a core object that satisfies the condition that the distance between the on-orbit data and the core object is less than the optimal domain radius; if there is, determine that the on-orbit data belongs to the cluster; if not, traverse the next cluster until the cluster to which the on-orbit data belongs is determined.

[0109] It should be noted that the device for identifying faults of a spacecraft control system based on multi-source information fusion provided in the above embodiments is only used as an example for the division of the above functional modules, and in actual applications, the above functions can be completed by different functional modules according to needs, i.e., the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the device for identifying faults of a spacecraft control system based on multi-source information fusion provided in the above embodiments and the method for identifying faults of a spacecraft control system based on multi-source information fusion belong to the same concept, and the specific implementation process is described in detail in the method embodiments, which will not be repeated here.

[0110] Embodiments of the present application also provide a computer device, which refers to Figure 3 the computer device includes a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, which is loaded and executed by the processor to implement the method for identifying faults of a spacecraft control system based on multi-source information fusion provided in the above method embodiments.

[0111] The embodiment of the present application further provides a computer readable storage medium, wherein at least one instruction, at least one program, a code set or an instruction set are stored on the computer readable storage medium, and the at least one instruction, the at least one program, the code set or the instruction set are loaded and executed by a processor to implement the spacecraft control system fault identification method based on multi-source information fusion provided by any of the foregoing method embodiments.

[0112] The embodiment of the present application further provides a computer program product, wherein the computer program product comprises a computer program, and a processor of a computer device reads the computer program from a computer readable storage medium, and the processor executes the computer program to enable the computer device to execute the spacecraft control system fault identification method based on multi-source information fusion according to any of the foregoing embodiments.

[0113] For the convenience of description, the foregoing system or device is described in various modules or units in terms of functions. Of course, the functions of the units can be implemented in one or more software and / or hardware in the implementation of the present application.

[0114] From the description of the foregoing embodiments, those skilled in the art can clearly understand that the present application can be implemented by means of software and a necessary general hardware platform. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, and the computer software product can be stored in a storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments of the present application.

[0115] Finally, it should be noted that, in this document, the relationship terms such as first, second, third and fourth are merely used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or device including the element.

[0116] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, some improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as the protection scope of the present application.

Claims

1. A method for identifying faults of a spacecraft control system based on multi-source information fusion, characterized in that, The method comprises: Multi-source fusion is performed on data of a spacecraft in a development and test process and historical data in orbit to obtain a plurality of fault samples; each of the fault samples is provided with a fault label, and each of the fault labels is used to represent a fault mode corresponding to the fault sample; Each of the fault samples is subjected to normalization processing to obtain normalized fault samples; Based on a preset clustering algorithm, the normalized fault samples are subjected to clustering processing to obtain a plurality of clusters and a fault mode of each cluster; each cluster includes at least one core object; For spacecraft in-orbit data at any moment, the distance between the in-orbit data and the core object in each cluster is calculated to determine the cluster to which the in-orbit data belongs based on the calculated distance; The fault mode corresponding to the cluster to which the in-orbit data belongs is determined as the identification result of the in-orbit data; The preset clustering algorithm is a DBSCAN algorithm; The clustering processing of the normalized fault samples based on the preset clustering algorithm to obtain a plurality of clusters and a fault mode of each cluster comprises: The DBSCAN algorithm is optimized based on a particle swarm algorithm to obtain an optimal field radius and an optimal minimum number of neighbors of the DBSCAN algorithm; Each fault sample is sequentially taken as a target fault sample, and a circular region is determined with the target fault sample as the center and the optimal field radius as the radius; other fault samples are traversed, and for each fault sample traversed, the Euclidean distance between the fault sample and the target fault sample is calculated, and it is determined whether the calculated Euclidean distance is less than the optimal field radius; if yes, it is determined that the fault sample is in the circular region, and if no, it is determined that the fault sample is not in the circular region; until each fault sample is traversed, the number of fault samples in the circular region is obtained; It is determined whether the number of fault samples in the circular region is greater than the optimal minimum number of neighbors; if yes, it is determined that the target fault sample is a core object, and all density-connected fault samples starting from the core object are determined; For each determined core object, all core objects within the optimal field radius of the core object are traversed, and all fault samples density-connected with the core objects are found to form a corresponding cluster, and a fault mode of the cluster is determined based on the fault label of each fault sample contained in the cluster; The optimization of the preset clustering algorithm based on the particle swarm algorithm to obtain an optimal field radius and an optimal minimum number of neighbors of the clustering algorithm comprises: S1, initialize particle swarm parameters, including the number of particles, the initial best position of the particle swarm, the initial position, the initial speed and the initial best position of each particle; S2, sequentially traversing each particle in the particle group, for each particle traversed, performing: sequentially taking each fault sample as a sub-target fault sample, and determining a circular region with the sub-target fault sample as the center and the domain radius of the current cycle as the radius; traversing other fault samples, for each fault sample traversed, calculating the Euclidean distance between the fault sample and the sub-target fault sample, and judging whether the calculated Euclidean distance is less than the domain radius of the current cycle; if yes, it is judged that the fault sample is in the circular region, and if no, it is judged that the fault sample is not in the circular region; until each fault sample is traversed, the number of fault samples in the circular region is obtained; judging whether the number of fault samples in the circular region is greater than the minimum neighborhood number of the current cycle; if yes, it is judged that the sub-target fault sample is the core object, and all density-connected fault samples from the core object are determined; for each core object determined, performing: traversing all core objects within the domain radius of the current cycle of the core object, and finding all fault samples connected in density with these core objects to form a corresponding cluster, and determining the fault mode of the cluster based on the fault label of each fault sample contained in the cluster; for each cluster determined, performing: traversing each fault sample in the cluster, and judging whether the fault mode corresponding to the fault label of the traversed fault sample is consistent with the fault mode of the cluster; if yes, the fault sample is retained; if no, the fault sample is removed from the cluster; until each fault sample in the cluster is traversed, the number of accurate samples in the cluster is obtained; S3, based on the number of accurate samples in each cluster, calculating the fault recognition accuracy of the current cycle, and performing S4; S4, judging whether the fault recognition accuracy of the current cycle is greater than the accuracy corresponding to the current best position of the current particle, if greater, updating the best position of the current particle; judging whether the fault recognition accuracy of the current cycle is greater than the accuracy corresponding to the current cycle best position of the current particle group, if greater, updating the best position of the current particle group; S5, judging whether the preset stopping condition is reached; if not, updating the position and speed of each particle, incrementing the iteration number by 1, and cyclically performing S2-S4 until the preset stopping condition is reached, obtaining the final best position of the particle group; S6, taking the domain radius and the minimum neighborhood number corresponding to the final best position of the particle group as the optimal domain radius and the optimal minimum neighborhood number of the clustering algorithm; The distance between the on-orbit data of the spacecraft at any moment and each core object in the cluster is calculated to determine the cluster to which the on-orbit data belongs based on the calculated distance, comprising: traversing each cluster, for each cluster traversed, performing: calculating the distance between the on-orbit data and each core object in the cluster; and judging whether there is a core object that satisfies the distance from the on-orbit data to the core object is less than the optimal domain radius; if yes, it is judged that the on-orbit data belongs to the cluster; if no, the next cluster is traversed until the cluster to which the on-orbit data belongs is determined.

2. The method of claim 1, wherein, The fault mode of the cluster is determined based on the fault label of each fault sample included in the cluster, including: The number of fault samples corresponding to each fault label in the cluster is counted respectively; The fault mode represented by the fault label with the largest number of corresponding fault samples is determined as the fault mode of the cluster.

3. The method of claim 1, wherein, The fault recognition rate is calculated using the following formula: In the formula, c For fault identification rate; j =1,2…… N , N The number of failure modes; The severity level of the j-th failure mode; For the first j The total number of fault samples contained in the cluster corresponding to each fault mode; For the first j The number of correct fault samples contained in the cluster corresponding to a fault mode.

4. The method of claim 1, wherein, The preset stopping condition is: The accuracy corresponding to the best position of the current cycle of the particle swarm is greater than the accuracy threshold; And / or, the number of iterations is greater than the number of times threshold.

5. A spacecraft control system fault identification device based on multi-source information fusion, characterized in that, The device for implementing the method of any one of claims 1-4 comprises: A fusion unit is configured to perform multi-source fusion on data of the spacecraft in the development and test process and historical data in orbit to obtain a plurality of fault samples, each of the fault samples is provided with a fault label, and each of the fault labels is configured to represent a fault mode corresponding to the fault sample; A processing unit is configured to perform normalization processing on each of the fault samples to obtain normalized fault samples; A clustering unit is configured to perform clustering processing on the normalized fault samples based on a preset clustering algorithm to obtain a plurality of clusters and a fault mode of each cluster, wherein each cluster includes at least one core object; A calculation unit is configured to calculate a distance between spacecraft in-orbit data at any moment and the core object in each cluster based on the in-orbit data, and determine a cluster to which the in-orbit data belongs based on the calculated distance; A determination unit is configured to determine a fault mode corresponding to the cluster to which the in-orbit data belongs as an identification result of the in-orbit data.

6. A computer device, comprising: The computer device comprises a memory and a processor, the memory is configured to store a computer program, and the processor is configured to execute the computer program stored in the memory to implement the steps of the method of any one of claims 1-4.

7. A computer readable storage medium characterized in that, The storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the method of any one of claims 1-4.

Images