CN-121997267-A - Collaborative decision determining method, device, equipment and medium for equipment

CN121997267ACN 121997267 ACN121997267 ACN 121997267ACN-121997267-A

Abstract

The application provides a collaborative decision-making determining method, device, equipment and medium of equipment, wherein the method comprises the steps of carrying out feature enhancement and weighted fusion on a plurality of initial feature vectors of the equipment to obtain target fusion feature data, inputting the target fusion feature data into a disturbance recognition model which is built in advance, determining a target disturbance type corresponding to the equipment and a disturbance intensity value corresponding to the target disturbance type from a plurality of preset disturbance types, determining an objective function to be optimized and a target constraint condition based on the disturbance type and the disturbance intensity value, and determining a collaborative decision-making scheme aiming at the equipment by utilizing the objective function to be optimized and the target constraint condition. By the method and the device, the accuracy and timeliness of equipment decision in a real dynamic environment are improved.

Inventors

LIU XUEFEI
YIN ZHIFENG
ZHANG JINKAI
WANG HAILIN
SU ZHIHAO
WANG XIJUN

Assignees

中国电子信息产业集团有限公司第六研究所

Dates

Publication Date: 20260508
Application Date: 20260129

Claims (10)

1. A collaborative decision-making determination method of equipment, the collaborative decision-making determination method comprising: Acquiring initial data of equipment under various modal types, and extracting features of the initial data to obtain a plurality of initial feature vectors; Performing feature enhancement and weighted fusion on the plurality of initial feature vectors to obtain target fusion feature data; Inputting the target fusion characteristic data into a disturbance recognition model which is built in advance, and determining a target disturbance type corresponding to the equipment from a plurality of preset disturbance types and a disturbance intensity value corresponding to the target disturbance type; And determining an objective function to be optimized and a constraint condition based on the disturbance type and the disturbance intensity value, and determining a collaborative decision scheme aiming at the equipment by utilizing the objective function to be optimized and the constraint condition.
2. The collaborative decision-making method according to claim 1, wherein the feature enhancement and weighted fusion of the plurality of initial feature vectors to obtain target fused feature data comprises: calculating a modal feature quality assessment factor corresponding to each initial feature vector based on the signal-to-noise ratio corresponding to each initial feature vector; Determining a feature enhancement vector corresponding to each initial feature vector based on a modal feature quality evaluation factor corresponding to each initial feature vector and a feature enhancement formula; And carrying out weighted fusion on a plurality of feature enhancement vectors by utilizing feature weights corresponding to each mode type to obtain the target fusion feature data, wherein the feature weights are dynamically updated based on feature correlation and decision contribution.
3. The collaborative decision-making method of claim 2, wherein the feature weights for each modality type are calculated by: For each mode type, calculating the mode characteristic correlation between the mode type and other mode types, and determining a characteristic correlation gradient based on the mode characteristic correlation; Calculating a feature importance score corresponding to the mode type, and determining a decision sharing degree gradient based on the feature importance score; And determining the feature weight corresponding to the modality type based on the feature correlation gradient and the decision sharing gradient.
4. The collaborative decision-making determination method of claim 2, wherein after determining the collaborative decision-making scheme, the collaborative decision-making determination method further comprises: and acquiring decision execution data of the collaborative decision scheme in an actual execution process, calculating a decision error based on the decision execution data, and adjusting the characteristic weight of each mode type based on the decision error.
5. The collaborative decision-making method of claim 1, wherein the disturbance recognition model is trained by: Acquiring a training data set, wherein the training data comprises fusion characteristic sample data corresponding to sample data under a plurality of modal types and disturbance sample labels corresponding to the sample data, and the disturbance sample labels comprise disturbance type labels and disturbance intensity sample values; inputting the fusion characteristic sample data into a disturbance identification original model, and determining a disturbance prediction label corresponding to the sample data; and determining a loss function according to the disturbance sample label and the disturbance prediction label, and performing iterative training on the disturbance recognition original model based on the loss function until a preset training completion condition is met, so as to obtain the trained disturbance recognition model.
6. A collaborative decision-making apparatus of equipment, characterized in that the collaborative decision-making apparatus comprises: The feature vector generation module is used for acquiring initial data of the equipment under various modal types, and extracting features of the initial data to obtain a plurality of initial feature vectors; The fusion characteristic data determining module is used for carrying out characteristic enhancement and weighted fusion on the plurality of initial characteristic vectors to obtain target fusion characteristic data; The disturbance type identification module is used for inputting the target fusion characteristic data into a disturbance identification model which is built in advance, and determining a target disturbance type corresponding to the equipment from a plurality of preset disturbance types and a disturbance intensity value corresponding to the target disturbance type; And the decision scheme generation module is used for determining an objective function to be optimized and a constraint condition based on the disturbance type and the disturbance intensity value, and determining a collaborative decision scheme aiming at the equipment by utilizing the objective function to be optimized and the constraint condition.
7. The collaborative decision-making device of claim 6, wherein the fused feature data determination module, when configured to perform feature enhancement and weighted fusion on a plurality of the initial feature vectors to obtain target fused feature data, is further configured to: calculating a modal feature quality assessment factor corresponding to each initial feature vector based on the signal-to-noise ratio corresponding to each initial feature vector; Determining a feature enhancement vector corresponding to each initial feature vector based on a modal feature quality evaluation factor corresponding to each initial feature vector and a feature enhancement formula; And carrying out weighted fusion on a plurality of feature enhancement vectors by utilizing feature weights corresponding to each mode type to obtain the target fusion feature data, wherein the feature weights are dynamically updated based on feature correlation and decision contribution.
8. The collaborative decision-making apparatus of claim 7, wherein the fused feature data determination module is further configured to calculate feature weights for each modality type by: For each mode type, calculating the mode characteristic correlation between the mode type and other mode types, and determining a characteristic correlation gradient based on the mode characteristic correlation; Calculating a feature importance score corresponding to the mode type, and determining a decision sharing degree gradient based on the feature importance score; And determining the feature weight corresponding to the modality type based on the feature correlation gradient and the decision sharing gradient.
9. An electronic device comprising a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication via the bus when the electronic device is in operation, the machine-readable instructions being executable by the processor to perform the steps of the collaborative decision-making method of an equipment device of any one of claims 1 to 5.
10. A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, performs the steps of the collaborative decision-making method of an equipment device according to any of claims 1 to 5.

Description

Collaborative decision determining method, device, equipment and medium for equipment Technical Field The application relates to the technical field of intelligent decision making of equipment, in particular to a collaborative decision making determining method, device, equipment and medium of equipment. Background With the increasing demand for intelligent management of equipment and equipment, existing decision systems are facing dual challenges of multi-source data fusion and complex disturbance handling. In application scenes such as military logistics, emergency guarantee and the like, equipment decision-making process relates to multi-source heterogeneous information such as material state monitoring data, environment sensing data, business system data and the like, and the multi-mode characteristic is obvious. The current mainstream technical scheme mainly adopts a characteristic splicing or linear weighting method based on fixed weight to perform data fusion processing, however, the method has obvious limitations that on one hand, the inherent characteristic difference between different mode data cannot be fully considered, and on the other hand, an effective evaluation mechanism for the dynamic importance of the data is lacking, so that the problems of information loss or characteristic conflict and the like of the fused characteristic space are caused. Meanwhile, the decision system also needs to deal with challenges of various dynamic disturbance factors. The traditional robust optimization method is generally used for constructing a model based on a preset single disturbance hypothesis, and the processing mode is difficult to accurately identify the type characteristics and the intensity distribution of disturbance events in an actual application scene, so that the adaptability of a generated decision scheme in a real complex environment is obviously insufficient. Disclosure of Invention In view of the above, the present application aims to provide a collaborative decision determining method, a device, equipment and a medium for equipment, which realize accurate characterization of multi-source heterogeneous data and robust decision under multi-type disturbance. The method effectively overcomes the defects of large multi-mode fusion information loss and model construction based on only a single disturbance hypothesis in the prior art, and improves the accuracy and timeliness of equipment decision in a real dynamic environment. In a first aspect, an embodiment of the present application provides a collaborative decision determining method of equipment, where the collaborative decision determining method includes: Acquiring initial data of equipment under various modal types, and extracting features of the initial data to obtain a plurality of initial feature vectors; Performing feature enhancement and weighted fusion on the plurality of initial feature vectors to obtain target fusion feature data; Inputting the target fusion characteristic data into a disturbance recognition model which is built in advance, and determining a target disturbance type corresponding to the equipment from a plurality of preset disturbance types and a disturbance intensity value corresponding to the target disturbance type; and determining an objective function to be optimized and a target constraint condition based on the disturbance type and the disturbance intensity value, and determining a collaborative decision scheme aiming at the equipment by utilizing the objective function to be optimized and the target constraint condition. Further, the performing feature enhancement and weighted fusion on the plurality of initial feature vectors to obtain target fusion feature data includes: calculating a modal feature quality assessment factor corresponding to each initial feature vector based on the signal-to-noise ratio corresponding to each initial feature vector; Determining a feature enhancement vector corresponding to each initial feature vector based on a modal feature quality evaluation factor corresponding to each initial feature vector and a feature enhancement formula; And carrying out weighted fusion on a plurality of feature enhancement vectors by utilizing feature weights corresponding to each mode type to obtain the target fusion feature data, wherein the feature weights are dynamically updated based on feature correlation and decision contribution. Further, the feature weight corresponding to each modality type is calculated by the following steps: For each mode type, calculating the mode characteristic correlation between the mode type and other mode types, and determining a characteristic correlation gradient based on the mode characteristic correlation; Calculating a feature importance score corresponding to the mode type, and determining a decision sharing degree gradient based on the feature importance score; And determining the feature weight corresponding to the modality type based on the feature corr