Motor Production Workshop Collaborative Management System and Method

Abstract

This application relates to the field of intelligent management, and specifically relates to a collaborative management system and method for a motor production workshop. The system includes: a task assignment module, which is used to generate a production plan according to order requirements and existing resources, split the production plan into production tasks, and then assign the production tasks to the operators on each production line; a progress tracking module, which is used to upload production progress data through data collection terminals deployed at each work station and generate a real-time production progress chart based on the production progress data; a plan adjustment module, which is used to issue an alarm and generate a production plan adjustment prompt in response to detecting equipment failures; a performance evaluation module, which is used to record the working hours and task completion status of each operator and generate a performance evaluation report. In this way, it can help enterprises maintain flexibility and efficiency in a complex production environment.

Description

Technical Field

[0001] This application relates to the field of intelligent management, and specifically relates to a collaborative management system and method for a motor production workshop. Background Art

[0002] In modern manufacturing, especially in the field of motor production, with the continuous change of market demand and the explosion of order volume, production efficiency and flexibility have become key factors in enterprise competitiveness. The customized production model requires the factory to be able to quickly respond to customer needs while maintaining high-quality production standards.

[0003] Modern motor production often involves multiple processes and production lines, and each link needs to cooperate with each other to complete the production and manufacturing of motors. However, in the traditional motor production workshop management system, it usually relies on manual operation and empirical judgment, lacking intelligent data analysis and decision support. This makes it difficult to adjust and optimize the motor production plan and difficult to cope with emergencies. For example, when facing problems with production equipment, it is often difficult to detect and handle them in a timely manner, resulting in production interruptions and losses. In addition, there is insufficient information sharing between production links in the existing motor production workshop management system, resulting in poor cooperation between departments and serious information island phenomena, making the adjustment of production plans and task allocation complicated and reducing the overall production efficiency.

[0004] Therefore, a collaborative management system for a motor production workshop is desired. Summary of the Invention

[0005] This application is made in consideration of the above problems. An object of this application is to provide a collaborative management system and method for a motor production workshop.

[0006] Embodiments of this application provide a collaborative management system for a motor production workshop, which includes:

[0007] A task allocation module, configured to generate a production plan according to order requirements and existing resources, and after splitting the production plan into production tasks, allocate the production tasks to the operators of each production line;

[0008] A progress tracking module, configured to upload production progress data through data collection terminals deployed at each work station, and generate a real-time production progress chart based on the production progress data;

[0009] A plan adjustment module, configured to issue an alarm and generate a production plan adjustment prompt in response to detecting equipment failures;

[0010] A performance evaluation module, configured to record the working hours and task completion status of each operator, and generate a performance evaluation report.

[0011] For example, a collaborative management system for a motor production workshop according to an embodiment of the present application, wherein the plan adjustment module includes:

[0012] An equipment thermal infrared image acquisition unit for acquiring thermal infrared images of monitored equipment collected by the data acquisition terminal;

[0013] An equipment component image region division unit for dividing the thermal infrared image based on equipment components to obtain a set of thermal infrared component unit region images;

[0014] A thermal infrared component unit region position coding unit for performing position coding on each thermal infrared component unit region image in the set of thermal infrared component unit region images using a position coding function to obtain a set of equipment component position coding vectors;

[0015] An equipment component thermal distribution feature extraction unit for respectively extracting thermal distribution features from each thermal infrared component unit region image in the set of thermal infrared component unit region images to obtain a set of equipment component thermal distribution feature maps;

[0016] An equipment component thermal distribution feature enhancement unit for respectively passing each equipment component thermal distribution feature map in the set of equipment component thermal distribution feature maps through a feature space structure consistency self-attention cross-channel enhancement module to obtain a set of equipment component thermal distribution enhanced feature maps;

[0017] A position information equipment thermal distribution feature fusion unit for respectively fusing each corresponding equipment component thermal distribution enhanced feature map and equipment component position coding vector in the set of equipment component thermal distribution enhanced feature maps and the set of equipment component position coding vectors to obtain a set of equipment component thermal distribution feature vectors containing position information;

[0018] An equipment whole and local component object granularity temperature distribution characterization unit for inputting the set of equipment component thermal distribution feature vectors containing position information into a feature aggregation network based on feature energy local significance gating to obtain equipment whole and local component object granularity temperature distribution feature vectors;

[0019] An equipment fault detection unit for inputting the equipment whole and local component object granularity temperature distribution feature vectors into an equipment detector based on a classifier to obtain a detection result, and the detection result is used to indicate whether the monitored equipment has a fault.

[0020] For example, a collaborative management system for a motor production workshop according to an embodiment of the present application, wherein the equipment component thermal distribution feature extraction unit is used for:

[0021] Respectively pass each thermal infrared component unit area image in the set of the thermal infrared component unit area images through the device part thermal distribution feature extractor based on the dilated convolutional neural network model to obtain the set of the device part thermal distribution feature maps.

[0022] For example, in the motor production workshop collaborative management system according to the embodiment of the present application, wherein, the device part thermal distribution feature enhancement unit includes:

[0023] A layer normalization sub-unit, configured to perform layer normalization on the device part thermal distribution feature map to obtain a normalized device part thermal distribution feature map;

[0024] A point convolution processing sub-unit, configured to perform point convolution processing on the normalized device part thermal distribution feature map to obtain a device part thermal distribution state channel context correlation representation feature map;

[0025] A convolution encoding sub-unit, configured to perform convolution encoding on the device part thermal distribution state channel context correlation representation feature map to obtain a device part thermal distribution state space context correlation representation feature map;

[0026] A replication sub-unit, configured to replicate the device part thermal distribution state space context correlation representation feature map to obtain a device part thermal distribution state backup space context correlation representation feature map;

[0027] A first feature shape reshaping sub-unit, configured to perform feature shape reshaping on the device part thermal distribution state channel context correlation representation feature map, the device part thermal distribution state space context correlation representation feature map, and the device part thermal distribution state backup space context correlation representation feature map to obtain a device part thermal distribution state channel context correlation representation feature matrix, a device part thermal distribution state space context correlation representation feature matrix, and a device part thermal distribution state backup space context correlation representation feature matrix;

[0028] A covariance matrix calculation sub-unit, configured to calculate the device part thermal distribution state cross-channel cross-covariance matrix between the device part thermal distribution state channel context correlation representation feature matrix and the device part thermal distribution state space context correlation representation feature matrix;

[0029] An activation sub-unit, configured to activate the device part thermal distribution state cross-channel cross-covariance matrix using the Softmax function to obtain a device part thermal distribution state feature global interaction attention matrix;

[0030] A product calculation subunit for calculating the product between the backup space context correlation representation feature matrix of the thermal distribution state of the device component and the global interaction attention matrix of the thermal distribution state of the device component to obtain an attention-enhanced feature representation matrix of the thermal distribution state of the device component;

[0031] A second feature shape reshaping subunit for reshaping the feature shape of the attention-enhanced feature representation matrix of the thermal distribution state of the device component to obtain a thermal distribution enhanced feature map of the device component.

[0032] For example, in the motor production workshop collaborative management system according to an embodiment of the present application, wherein the covariance matrix calculation subunit is used for:

[0033] Calculating the matrix multiplication between the channel context correlation representation feature matrix of the thermal distribution state of the device component and the space context correlation representation feature matrix of the thermal distribution state of the device component to obtain a channel-space context correlation feature matrix of the thermal distribution state of the device component;

[0034] Calculating the element-wise division between the channel-space context correlation feature matrix of the thermal distribution state of the device component and a hyperparameter to obtain the cross-channel cross-covariance matrix of the thermal distribution state of the device component.

[0035] For example, in the motor production workshop collaborative management system according to an embodiment of the present application, wherein the location information device thermal distribution feature fusion unit is used for:

[0036] Performing global average pooling on each thermal distribution enhanced feature map in the set of thermal distribution enhanced feature maps of the device component to obtain a set of thermal distribution enhanced feature vectors of the device component;

[0037] Fusing each corresponding device component position encoding vector and thermal distribution enhanced feature vector in the set of device component position encoding vectors and the set of thermal distribution enhanced feature vectors of the device component respectively to obtain a set of device component thermal distribution feature vectors containing location information.

[0038] For example, in the motor production workshop collaborative management system according to an embodiment of the present application, wherein the device global-local component object granularity temperature distribution characterization unit includes:

[0039] A feature energy level coefficient calculation subunit for determining the feature energy level coefficient of each device component thermal distribution feature vector containing location information based on the maximum value, minimum value, mean value, and variance of each device component thermal distribution feature vector containing location information in the set of device component thermal distribution feature vectors containing location information to obtain a sequence of device component thermal distribution feature energy level coefficients containing location information;

[0040] A local neighborhood mean calculation subunit, configured to determine the scale of the local neighborhood of the thermal distribution of the device component including position information, use the scale of the local neighborhood of the thermal distribution of the device component including position information as the mean calculation range, and calculate the local neighborhood mean of the thermal distribution characteristic energy level coefficients of each device component including position information to obtain a sequence of thermal distribution characteristic energy level significance descriptors of the device component including position information;

[0041] A mask subunit, configured to input the sequence of thermal distribution characteristic energy level significance descriptors of the device component including position information into a mask module based on a gating function to obtain a sequence of masked gating probabilistic thermal distribution characteristic energy level significance descriptors of the device component including position information;

[0042] A weighted sum calculation subunit, configured to use the sequence of masked gating probabilistic thermal distribution characteristic energy level significance descriptors of the device component including position information as a sequence of weights, and calculate the position-wise weighted sum of the set of thermal distribution feature vectors of the device component including position information to obtain the granular temperature distribution feature vector of the entire local component object of the device.

[0043] For example, in the motor production workshop collaborative management system according to an embodiment of the present application, wherein the characteristic energy level coefficient calculation subunit is configured to:

[0044] Calculate the maximum value, minimum value, average value, and variance of the thermal distribution feature vector of the device component including position information to obtain the maximum value of the thermal distribution of the device component including position information, the minimum value of the thermal distribution of the device component including position information, the average value of the thermal distribution of the device component including position information, and the variance of the thermal distribution of the device component including position information;

[0045] Calculate the sum of the variance of the thermal distribution of the device component including position information and a hyperparameter to obtain a first local energy statistical factor of the thermal distribution of the device component including position information;

[0046] Calculate the difference between the maximum value of the thermal distribution of the device component including position information and the average value of the thermal distribution of the device component including position information to obtain the maximum deviation value of the thermal distribution of the device component including position information;

[0047] Calculate the difference between the minimum value of the thermal distribution of the device component including position information and the average value of the thermal distribution of the device component including position information to obtain the minimum deviation value of the thermal distribution of the device component including position information;

[0048] Calculate the sum of the squares of the maximum deviation value and the minimum deviation value of the thermal distribution of the device component including position information and add it to the hyperparameter to obtain a second local energy statistical factor of the thermal distribution of the device component including position information;

[0049] Calculate the division between the local energy statistical factor of the thermal distribution of the first device component containing location information and the local energy statistical factor of the thermal distribution of the second device component containing location information to obtain the characteristic energy level coefficient of the thermal distribution of the device component containing location information.

[0050] For example, in the motor production workshop collaborative management system according to an embodiment of the present application, wherein the masking subunit is configured to:

[0051] Using the negative number of each device component thermal distribution characteristic energy level significance descriptor in the sequence of device component thermal distribution characteristic energy level significance descriptors containing location information as the exponent, calculate the exponential function value with the natural constant e as the base to obtain the sequence of device component thermal distribution characteristic energy level class support significance descriptors;

[0052] Calculate the position-wise summation between the sequence of device component thermal distribution characteristic energy level class support significance descriptors containing location information and the constant 1 to obtain the linearly modulated sequence of device component thermal distribution characteristic energy level class support significance descriptors;

[0053] Calculate the reciprocal of each linearly modulated device component thermal distribution characteristic energy level class support significance descriptor in the linearly modulated sequence of device component thermal distribution characteristic energy level class support significance descriptors to obtain the sequence of device component thermal distribution gated probabilistic characteristic energy level significance descriptors;

[0054] Perform masking processing on each device component thermal distribution gated probabilistic characteristic energy level significance descriptor in the sequence of device component thermal distribution gated probabilistic characteristic energy level significance descriptors containing location information to obtain the sequence of masked gated probabilistic device component thermal distribution characteristic energy level significance descriptors;

[0055] Wherein, performing masking processing on each device component thermal distribution gated probabilistic characteristic energy level significance descriptor in the sequence of device component thermal distribution gated probabilistic characteristic energy level significance descriptors containing location information to obtain the sequence of masked gated probabilistic device component thermal distribution characteristic energy level significance descriptors includes:

[0056] In response to the device component thermal distribution gated probabilistic characteristic energy level significance descriptor being greater than a predetermined threshold, take the original value of the device component thermal distribution gated probabilistic characteristic energy level significance descriptor, otherwise set it to 0.

[0057] Embodiments of the present application also provide a collaborative management method for a motor production workshop, which includes:

[0058] Generate a production plan according to order requirements and existing resources, and after splitting the production plan into production tasks, assign the production tasks to the operators on each production line;

[0059] Upload production progress data through data acquisition terminals deployed at each work station, and generate a real-time production progress chart based on the production progress data;

[0060] In response to detecting a device failure, issue an alarm and generate a production plan adjustment prompt;

[0061] Record the working hours and task completion status of each operator, and generate a performance evaluation report.

[0062] According to the collaborative management system and method for a motor production workshop in the embodiments of the present application, through the information interaction of the task assignment module, progress tracking module, plan adjustment module and performance evaluation module, more intelligent and collaborative motor production management can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application and do not limit the present application.

[0064] Figure 1 Shows an application architecture diagram of the collaborative management system for a motor production workshop in an embodiment of the present application;

[0065] Figure 2 Shows a structural diagram of the collaborative management system for a motor production workshop in an embodiment of the present application;

[0066] Figure 3 Shows a structural diagram of the plan adjustment module of the collaborative management system for a motor production workshop in an embodiment of the present application;

[0067] Figure 4 Shows a flowchart of the collaborative management method for a motor production workshop in an embodiment of the present application; and

[0068] Figure 5 Shows an application scenario diagram of the collaborative management system for a motor production workshop in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0069] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all embodiments. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall also fall within the scope of protection of the present application.

[0070] The terms used in this specification are those general terms that are currently widely used in the art in consideration of the functions of the present application. However, these terms may change according to the intentions of those of ordinary skill in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in such cases, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be construed as simple names, but rather based on the meanings of the terms and the overall description of the present application.

[0071] Although the present application makes various references to certain modules in the systems according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or the server. The modules are merely illustrative, and different aspects of the system and method may use different modules.

[0072] Flowcharts are used in the present application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. On the contrary, various steps may be processed in reverse order or simultaneously as needed. At the same time, other operations may also be added to these processes, or one or several operations may be removed from these processes.

[0073] Figure 1 The application architecture diagram of the collaborative management system in the motor production workshop according to the embodiments of the present application is shown, including a server 100 and a terminal device 200.

[0074] The terminal device 200 and the server 100 can be connected via the Internet to achieve communication with each other. Optionally, the above Internet uses standard communication technologies and / or protocols. The Internet is usually the Internet, but can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network. In some embodiments, technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. are used to represent the data exchanged through the network. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can be used to encrypt all or some of the links. In other embodiments, custom and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0075] The server 100 can provide various network services for the terminal device 200. Among them, the server 100 can be a single server, a server cluster composed of several servers, or a cloud computing center. Specifically, the server 100 can include a processor 110 (Central Processing Unit, CPU), a memory 120, an input device 130, an output device 140, etc. The input device 130 can include a keyboard, a mouse, a touch screen, etc. The output device 140 can include a display device, such as a liquid crystal display (LCD), a cathode ray tube (CRT), etc.

[0076] The memory 120 can include a read-only memory (ROM) and a random access memory (RAM), and provide the program instructions and data stored in the memory 120 to the processor 110. In the embodiments of the present application, the memory 120 can be used to store the corresponding methods of the motor production workshop collaborative management system in the embodiments of the present application.

[0077] The processor 110 is configured to execute the steps of any one of the motor production workshop collaborative management methods in the embodiments of the present application according to the program instructions obtained by calling the program instructions stored in the memory 120.

[0078] In addition, the application architecture diagram in the embodiments of the present application is for more clearly illustrating the technical solutions in the embodiments of the present application, and does not constitute a limitation on the technical solutions provided in the embodiments of the present application. Of course, for other application architectures and business applications, the technical solutions provided in the embodiments of the present application are equally applicable to similar problems.

[0079] Next, a non-limiting description of the motor production workshop collaborative management system provided according to at least one embodiment of the present application will be given through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflict, so as to obtain new examples or embodiments, and these new examples or embodiments also fall within the protection scope of the present application.

[0080] It should be understood that the production plan is the basis for ensuring the smooth production process. In the motor production workshop, a reasonable production plan can effectively balance resource allocation, equipment utilization rate, and production efficiency. By monitoring and adjusting the production plan in real time, an enterprise can minimize production delays, ensure on-time delivery, and the quality of motor products.

[0081] Based on this, in the technical solution of the present application, a motor production workshop collaborative management system is proposed, as Figure 2 shown, which includes: a task assignment module 810, configured to generate a production plan according to order requirements and existing resources, and after splitting the production plan into production tasks, assign the production tasks to the operators on each production line; a progress tracking module 820, configured to upload production progress data through data collection terminals deployed at each work station, and generate a real-time production progress chart based on the production progress data; a plan adjustment module 830, configured to issue an alarm and generate a production plan adjustment prompt in response to detecting a device failure; a performance evaluation module 840, configured to record the working hours and task completion status of each operator, and generate a performance evaluation report. In this motor production workshop collaborative management system, more intelligent and collaborative motor production management can be achieved through the information interaction of the task assignment module 810, the progress tracking module 820, the plan adjustment module 830, and the performance evaluation module 840.

[0082] Specifically, in the above-mentioned collaborative management system for the motor production workshop, considering that equipment failures are one of the main factors affecting production efficiency, it is necessary to conduct real-time and effective detection of equipment failures in the motor production workshop, so as to promptly discover potential failures and issue alarms, help operators take prompt measures, and adjust the production plan to ensure the quality of motor products while enabling the motor production workshop to complete production on time. In this way, more intelligent collaborative management of the motor production workshop can be achieved, helping the enterprise maintain flexibility and efficiency in a complex production environment.

[0083] Among them, as Figure 3 shown, the plan adjustment module 830 includes: an equipment thermal infrared image acquisition unit 831 for acquiring the thermal infrared image of the monitored equipment collected by the data acquisition terminal; an equipment component image region division unit 832 for dividing the thermal infrared image based on equipment components to obtain a set of thermal infrared component unit region images; a thermal infrared component unit region position encoding unit 833 for using a position encoding function to perform position encoding on each thermal infrared component unit region image in the set of thermal infrared component unit region images to obtain a set of equipment component position encoding vectors; an equipment component thermal distribution feature extraction unit 834 for respectively extracting thermal distribution features from each thermal infrared component unit region image in the set of thermal infrared component unit region images to obtain a set of equipment component thermal distribution feature maps; an equipment component thermal distribution feature enhancement unit 835 for respectively passing each equipment component thermal distribution feature map in the set of equipment component thermal distribution feature maps through a feature space structure consistency self-attention cross-channel enhancement module to obtain a set of equipment component thermal distribution enhanced feature maps; a position information equipment thermal distribution feature fusion unit 836 for respectively fusing each corresponding equipment component thermal distribution enhanced feature map and equipment component position encoding vector in the set of equipment component thermal distribution enhanced feature maps and the set of equipment component position encoding vectors to obtain a set of equipment component thermal distribution feature vectors containing position information; an equipment global and local component object granularity temperature distribution characterization unit 837 for inputting the set of equipment component thermal distribution feature vectors containing position information into a feature aggregation network based on feature energy local saliency gating to obtain an equipment global and local component object granularity temperature distribution feature vector; an equipment failure detection unit 838 for inputting the equipment global and local component object granularity temperature distribution feature vector into an equipment detector based on a classifier to obtain a detection result, and the detection result is used to indicate whether the monitored equipment has a failure.

[0084] Specifically, in the plan adjustment module 830, first, obtain the thermal infrared image of the monitored device collected by the data collection terminal. Then, considering that in the thermal infrared image, the temperature distribution characteristic information of each component of the monitored device is included, and the temperature distributions of these different device components can better reflect which components have abnormal conditions. Based on this, in the technical solution of this application, perform image region division based on device components on the thermal infrared image to obtain a set of thermal infrared component unit region images. By performing region division based on device components on the thermal infrared image, the thermal distribution information of the entire device can be refined to each specific component, and this refinement enables subsequent analysis to be carried out for each component, facilitating the identification of which components have abnormal thermal distributions.

[0085] Furthermore, in order to be able to associate the thermal distribution characteristics of different device components with specific component positions, thereby improving the accuracy of device fault detection and facilitating more precise device fault maintenance subsequently, in the technical solution of this application, further use a position encoding function to perform position encoding on each thermal infrared component unit region image in the set of thermal infrared component unit region images to obtain a set of device component position encoding vectors. Specifically, by using the position encoding function to encode each thermal infrared component unit region image, it is beneficial to combine the position information of each component with its thermal characteristics, which helps to quickly locate the problem component during analysis and fault detection and improve the efficiency of fault troubleshooting.

[0086] Then, respectively pass each thermal infrared component unit region image in the set of thermal infrared component unit region images through a device part thermal distribution feature extractor based on an atrous convolutional neural network model for feature mining to respectively extract the thermal distribution characteristic information of the device components in each thermal infrared part unit region image, thereby obtaining a set of device component thermal distribution feature maps.

[0087] Correspondingly, the device component thermal distribution feature extraction unit 834 is used to: respectively pass each thermal infrared component unit region image in the set of thermal infrared component unit region images through a device part thermal distribution feature extractor based on an atrous convolutional neural network model to obtain a set of device component thermal distribution feature maps.

[0088] It should be understood that each device component thermal distribution feature map in the set of device component thermal distribution feature maps contains thermal distribution semantics and feature information about each device component. However, the device component thermal distribution feature map not only contains the key thermal distribution features of each device component, but also contains some irrelevant noise interference and redundant information, which contribute less to the subsequent device fault detection task. Based on this, in the technical solution of the present application, each device component thermal distribution feature map in the set of device component thermal distribution feature maps is further passed through a feature space structure consistency self-attention cross-channel enhancement module to obtain a set of device component thermal distribution enhanced feature maps. The design of the feature space structure consistency self-attention cross-channel enhancement module aims to preserve the structure of the feature space, which means that during the processing, the overall layout and relationship of the device component thermal distribution feature map will not be destroyed. This is particularly important for understanding the thermal distribution feature map because these feature maps reflect the thermal states and mutual relationships of the various components of the device. Therefore, through the processing of the feature space structure consistency self-attention cross-channel enhancement module, it is possible to capture the channel context correlation information by utilizing the interaction of the feature channels while preserving the spatial structure of the device component thermal distribution features, so that the device component thermal distribution enhanced feature map after feature enhancement can, while preserving the feature space structure, fuse the channel correlation information between feature structures in units of the feature space structure to balance the robustness and richness of feature expression. It is worth mentioning that through cross-channel interaction, the module can capture the context correlation information between the device component thermal distribution feature channels. This information is crucial for understanding the mutual influence between device components, especially during fault detection.

[0089] Accordingly, the device component thermal distribution feature enhancement unit 835 includes: a layer normalization subunit for performing layer normalization on the device component thermal distribution feature map to obtain a normalized device component thermal distribution feature map; a point convolution processing subunit for performing point convolution processing on the normalized device component thermal distribution feature map to obtain a device component thermal distribution state channel context correlation representation feature map; a convolutional encoding subunit for performing convolutional encoding on the device component thermal distribution state channel context correlation representation feature map to obtain a device component thermal distribution state space context correlation representation feature map; a replication subunit for replicating the device component thermal distribution state space context correlation representation feature map to obtain a device component thermal distribution state backup space context correlation representation feature map; a first feature shape reshaping subunit for performing feature shape reshaping on the device component thermal distribution state channel context correlation representation feature map, the device component thermal distribution state space context correlation representation feature map, and the device component thermal distribution state backup space context correlation representation feature map to obtain a device component thermal distribution state channel context correlation representation feature matrix, a device component thermal distribution state space context correlation representation feature matrix, and a device component thermal distribution state backup space context correlation representation feature matrix; a covariance matrix calculation subunit for calculating a device component thermal distribution state cross-channel cross-covariance matrix between the device component thermal distribution state channel context correlation representation feature matrix and the device component thermal distribution state space context correlation representation feature matrix; an activation subunit for activating the device component thermal distribution state cross-channel cross-covariance matrix using the Softmax function to obtain a device component thermal distribution state feature global interaction attention matrix; a product calculation subunit for calculating the product between the device component thermal distribution state backup space context correlation representation feature matrix and the device component thermal distribution state feature global interaction attention matrix to obtain a device component thermal distribution state attention enhancement feature representation matrix; a second feature shape reshaping subunit for performing feature shape reshaping on the device component thermal distribution state attention enhancement feature representation matrix to obtain a device component thermal distribution enhancement feature map.

[0090] Among them, the covariance matrix calculation subunit is used to: calculate the matrix multiplication between the device component thermal distribution state channel context correlation representation feature matrix and the device component thermal distribution state space context correlation representation feature matrix to obtain a device component thermal distribution state channel-space context correlation feature matrix; calculate the element-wise division between the device component thermal distribution state channel-space context correlation feature matrix and a hyperparameter to obtain the device component thermal distribution state cross-channel cross-covariance matrix.

[0091] In a specific example, the device component thermal distribution feature strengthening unit 835 is configured to: process each device component thermal distribution feature map in the set of device component thermal distribution feature maps through the feature space structure consistency self-attention cross-channel strengthening module respectively according to the following self-attention feature strengthening formula to obtain the set of device component thermal distribution strengthened feature maps; wherein, the self-attention feature strengthening formula is:

[0092] F ln = Layer Normalization(F i )

[0093] F Q = Conv 1×1 (F ln )

[0094] F K = Conv 3×3 (Conv 1×1 (F ln ))

[0095] F V = Copy(F K )

[0096] M Q = Reshape(F Q )

[0097] M K = Reshape(F K )

[0098] M V = Reshape(F V )

[0099]

[0100]

[0101] Wherein, F i is the device component thermal distribution feature map, Layer Normalization(·) represents performing layer normalization operation on the feature map, F ln is the normalized device component thermal distribution feature map, Conv 1×1 (·) is the point convolution operation, F Q is the device component thermal distribution state channel context association representation feature map, Conv 3×3 (·) is the dilated convolution operation with a convolution kernel of 3×3, F K is the device component thermal distribution state space context association representation feature map, Copy(·) is the copy operation, F VFor the feature map of the backup space context association representation of the thermal distribution state of the device component, Reshape(·) represents the feature shape reshaping process, M Q For the feature matrix of the channel context association representation of the thermal distribution state of the device component, M K For the feature matrix of the space context association representation of the thermal distribution state of the device component, M V For the feature matrix of the backup space context association representation of the thermal distribution state of the device component, For matrix multiplication, θ is a hyperparameter, softmac(·) is the softmax activation function, M a For the global interaction attention matrix of the thermal distribution state features of the device component, F a For each thermal distribution enhancement feature map in the set of thermal distribution enhancement feature maps of the device component.

[0102] Furthermore, after performing global average pooling on each thermal distribution enhancement feature map in the set of thermal distribution enhancement feature maps of the device component to obtain a set of thermal distribution enhancement feature vectors, the corresponding device component position encoding vectors and thermal distribution enhancement feature vectors in each group of the set of device component position encoding vectors and the set of thermal distribution enhancement feature vectors are respectively fused to obtain a set of thermal distribution feature vectors of the device component containing position information. It should be understood that global average pooling converts the high-dimensional feature map into a low-dimensional feature vector by calculating the global average of each feature matrix along the channel dimension of each thermal distribution enhancement feature map. This dimensionality reduction process not only reduces the computational complexity but also retains the important thermal distribution feature information of the device component. In addition, since the device component position encoding vector contains the position information of each component in the overall device. Combining this information with the thermal distribution enhancement feature vector can provide context for the feature vector, enabling subsequent processing to better understand the position relationship and thermal distribution transfer situation between components, thereby better performing the thermal distribution situation recognition and fault detection of the device component.

[0103] Correspondingly, the position information device thermal distribution feature fusion unit 836 is used for: performing global average pooling on each thermal distribution enhancement feature map in the set of thermal distribution enhancement feature maps of the device component to obtain a set of thermal distribution enhancement feature vectors; respectively fusing the corresponding device component position encoding vectors and thermal distribution enhancement feature vectors in each group of the set of device component position encoding vectors and the set of thermal distribution enhancement feature vectors to obtain the set of thermal distribution feature vectors of the device component containing position information.

[0104] Furthermore, since each of the device component thermal distribution feature vectors containing location information in the set of device component thermal distribution feature vectors containing location information respectively includes the thermal distribution semantic features of each device component with location information, and the overall state and fault conditions of the device are usually jointly determined by the local states of multiple components, this is crucial for subsequent device fault detection tasks. Based on this, in the technical solution of this application, the set of device component thermal distribution feature vectors containing location information is further input into a feature aggregation network based on local saliency gating of feature energy to obtain a device full-local component object granularity temperature distribution feature vector. The feature aggregation network based on local saliency gating of feature energy can quantify the energy distribution of each device component thermal distribution feature vector containing location information, so as to identify the most important part for subsequent fault detection tasks in a specific scenario. In this way, during the aggregation process of the thermal distribution features of each device component containing location information, it is possible to automatically focus on those with saliency (i.e., features that have a greater impact on the overall device state), and by dynamically adjusting the weights of different features, use the gating mechanism to highlight important features and suppress redundant information during the feature aggregation process, thereby improving the effectiveness of feature aggregation. In this way, it is possible to improve the model's ability to understand and process data, as well as the generalization ability of the model, thereby enhancing its stability and reliability in device fault detection in complex environments.

[0105] Accordingly, the device full-part object granularity temperature distribution characterization unit 837 includes: a characteristic energy level coefficient calculation sub-unit, configured to determine, based on the maximum value, minimum value, mean value, and variance of each device component thermal distribution feature vector including position information in the set of device component thermal distribution feature vectors including position information, the characteristic energy level coefficient of each device component thermal distribution feature vector including position information to obtain a sequence of device component thermal distribution characteristic energy level coefficients including position information; a local neighborhood mean calculation sub-unit, configured to determine the scale of the local neighborhood of the device component thermal distribution including position information, and use the scale of the local neighborhood of the device component thermal distribution including position information as the mean calculation range, and calculate the local neighborhood mean of each device component thermal distribution characteristic energy level coefficient including position information respectively to obtain a sequence of device component thermal distribution characteristic energy level significance descriptors including position information; a masking sub-unit, configured to input the sequence of device component thermal distribution characteristic energy level significance descriptors including position information into a masking module based on a gating function to obtain a sequence of masked gating probabilistic device component thermal distribution characteristic energy level significance descriptors including position information; a weighted sum calculation sub-unit, configured to use the sequence of masked gating probabilistic device component thermal distribution characteristic energy level significance descriptors including position information as a sequence of weights, and calculate the position-wise weighted sum of the set of device component thermal distribution feature vectors including position information to obtain the device full-part object granularity temperature distribution feature vector.

[0106] Among them, the characteristic energy level coefficient calculation subunit is configured to: calculate the maximum value, minimum value, average value, and variance of the device component thermal distribution feature vector including position information to obtain the device component thermal distribution maximum value including position information, the device component thermal distribution minimum value including position information, the device component thermal distribution average value including position information, and the device component thermal distribution variance including position information; calculate the sum of the device component thermal distribution variance including position information and a hyperparameter to obtain a first device component thermal distribution local energy statistical factor including position information; calculate the difference between the device component thermal distribution maximum value including position information and the device component thermal distribution average value including position information to obtain the device component thermal distribution maximum deviation value including position information; calculate the difference between the device component thermal distribution minimum value including position information and the device component thermal distribution average value including position information to obtain the device component thermal distribution minimum deviation value including position information; calculate the sum of the squares of the device component thermal distribution maximum deviation value including position information and the device component thermal distribution minimum deviation value including position information and then add the hyperparameter to obtain a second device component thermal distribution local energy statistical factor including position information; calculate the division between the first device component thermal distribution local energy statistical factor including position information and the second device component thermal distribution local energy statistical factor including position information to obtain the device component thermal distribution characteristic energy level coefficient.

[0107] Among them, the mask sub-unit is used to: take the negative number of each device component thermal distribution characteristic energy level significance descriptor containing position information in the sequence of device component thermal distribution characteristic energy level significance descriptors containing position information as the exponential power, calculate the exponential function value with the natural constant e as the base to obtain a sequence of device component thermal distribution characteristic energy level class support significance descriptors containing position information; calculate the position-wise sum of the sequence of device component thermal distribution characteristic energy level class support significance descriptors containing position information and the constant 1 to obtain a linearly modulated sequence of device component thermal distribution characteristic energy level class support significance descriptors containing position information; calculate the reciprocal of each linearly modulated device component thermal distribution characteristic energy level class support significance descriptor in the linearly modulated sequence of device component thermal distribution characteristic energy level class support significance descriptors containing position information to obtain a sequence of device component thermal distribution gated probability characteristic energy level significance descriptors containing position information; perform masking processing on each device component thermal distribution gated probability characteristic energy level significance descriptor in the sequence of device component thermal distribution gated probability characteristic energy level significance descriptors containing position information to obtain the sequence of masked gated probability device component thermal distribution characteristic energy level significance descriptors containing position information. Among them, performing masking processing on each device component thermal distribution gated probability characteristic energy level significance descriptor in the sequence of device component thermal distribution gated probability characteristic energy level significance descriptors containing position information to obtain the sequence of masked gated probability device component thermal distribution characteristic energy level significance descriptors containing position information includes: in response to the device component thermal distribution gated probability characteristic energy level significance descriptor containing position information being greater than a predetermined threshold, taking the original value of the device component thermal distribution gated probability characteristic energy level significance descriptor containing position information, otherwise setting it to 0.

[0108] In a specific example, the device global and local component object granularity temperature distribution characterization unit 837 is used to: input the set of device component thermal distribution characteristic vectors containing position information into the feature aggregation network based on feature energy local significance gating and process it according to the following feature aggregation formula to obtain the device global and local component object granularity temperature distribution characteristic vector; where the feature aggregation formula is:

[0109] X = {x1, x2,..., x k ,..., x n}

[0110]

[0111] w si = mask(w i )

[0112]

[0113] Y = {y1, y2,..., y k ,..., y n}, y i = w si ·x i

[0114]

[0115] where X is the set of thermal distribution feature vectors of the device components containing location information, x i is the i-th thermal distribution feature vector of the device components containing location information in the set of thermal distribution feature vectors of the device components containing location information, x k and x n are the k-th and n-th thermal distribution feature vectors of the device components containing location information in the set of thermal distribution feature vectors of the device components containing location information respectively, μ i is the mean of the i-th thermal distribution feature vector of the device components containing location information, σ i 2 is the variance of the i-th thermal distribution feature vector of the device components containing location information, ε is a hyperparameter, max(x i ) represents the maximum value in the i-th thermal distribution feature vector of the device components containing location information, min(x i ) represents the minimum value in the i-th thermal distribution feature vector of the device components containing location information, e i is the characteristic energy level coefficient of the i-th thermal distribution feature vector of the device components containing location information, Num r represents the number of characteristic energy coefficients in the local neighborhood centered on the characteristic energy level coefficient e i , e q is the characteristic energy level coefficient of the q-th thermal distribution feature vector of the device components containing location information, e ci is the characteristic energy level significance descriptor of the i-th thermal distribution feature vector of the device components containing location information, w i is the gated probabilistic characteristic energy level significance descriptor of the i-th thermal distribution feature vector of the device components containing location information, mask(·) is a masking operation, θ is a predetermined threshold, w si is the masked gated probabilistic characteristic energy level significance descriptor of the i-th thermal distribution feature vector of the device components containing location information, y iFor the saliency-enhanced device component thermal distribution feature vector corresponding to the thermal distribution feature vector of the i-th device component containing position information, n is the number of feature vectors in the set of saliency-enhanced device component thermal distribution feature vectors containing position information, and f is the device global and local component object granularity temperature distribution feature vector.

[0116] Furthermore, input the device global and local component object granularity temperature distribution feature vector into a device detector based on a classifier to obtain a detection result, and the detection result is used to indicate whether there is a fault in the monitored device. That is to say, use the aggregated feature of the temperature distribution of all components of the device for classification processing, so as to detect the device faults in the motor production workshop, so as to timely discover potential faults and issue alarms, help the operator quickly take measures, and adjust the production plan to ensure the quality of the motor products while enabling the motor production workshop to complete production on time. In this way, more intelligent collaborative management of the motor production workshop can be realized, helping the enterprise to maintain flexibility and efficiency in a complex production environment.

[0117] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistics regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistics regression or SVM can also be used, but it requires multiple binary classifications to form a multi-class classification, which is prone to errors and low efficiency. The commonly used multi-class classification method is the Softmax classification function.

[0118] In a preferred example, when the set of device component position encoding vectors and the set of device component thermal distribution enhanced feature vectors respectively represent the image semantic position encoding feature and the image semantic space self-attention channel distribution feature of the local image semantic space domain based on the thermal infrared image, the feature aggregation network based on feature energy local saliency gating will perform distribution aggregation based on the feature energy saliency of the local image semantic space domain under the global image semantic space domain, so that the obtained device global and local component object granularity temperature distribution feature vector has insufficient fine-grained semantic aggregation expression caused by the difference in energy saliency aggregation. Therefore, it is expected to improve the detailed semantic aggregation expression effect of the device global and local component object granularity temperature distribution feature vector based on the difference in image semantic representation.

[0119] In a preferred example, inputting the device global and local component object granularity temperature distribution feature vector into a device detector based on a classifier to obtain a detection result includes:

[0120] Calculate the sum of the absolute values of the eigenvalues of the granular temperature distribution eigenvector of all components of the device to obtain the first granular temperature distribution and modulation value of all components of the device, and calculate the square root of the sum of the squares of the eigenvalues of the granular temperature distribution eigenvector of all components of the device to obtain the second granular temperature distribution and modulation value of all components of the device;

[0121] After subtracting the granular temperature distribution eigenvector of all components of the device from the second granular temperature distribution and modulation value of all components of the device, perform dot multiplication with the number of eigenvalues of the granular temperature distribution eigenvector of all components of the device and the reciprocal of the first granular temperature distribution and modulation value of all components of the device respectively, and take the reciprocal of each eigenvalue to obtain the first granular temperature distribution phase conversion vector of all components of the device;

[0122] After subtracting the granular temperature distribution eigenvector of all components of the device from the first granular temperature distribution and modulation value of all components of the device, perform dot multiplication with the square root of the number of eigenvalues of the granular temperature distribution eigenvector of all components of the device and the reciprocal of the second granular temperature distribution and modulation value of all components of the device respectively, and take the reciprocal of each eigenvalue to obtain the second granular temperature distribution phase conversion vector of all components of the device;

[0123] Subtract the dot multiplication vector of the second granular temperature distribution phase conversion vector of all components of the device and the weighted hyperparameter from the first granular temperature distribution phase conversion vector of all components of the device to obtain the optimized granular temperature distribution eigenvector of all components of the device;

[0124] Input the optimized granular temperature distribution eigenvector of all components of the device into the device detector based on the classifier to obtain the detection result.

[0125] Here, the optimization of the granular temperature distribution eigenvector V of all components of the device is expressed as:

[0126]

[0127] v i ∈V∈R n

[0128] where V is the granular temperature distribution eigenvector of all components of the device, R represents the set of real numbers, and v i represents the eigenvalue at the i-th position of the granular temperature distribution eigenvector of all components of the device, n represents the number of eigenvalues of the granular temperature distribution eigenvector of all components of the device, α represents the first granular temperature distribution and modulation value of all components of the device, β represents the second granular temperature distribution and modulation value of all components of the device, and ⊙ represents dot multiplication by position. Indicates subtraction by position, (·) ⊙-1 Indicates calculating the reciprocal of each eigenvalue of the eigenvector. V1 represents the phase conversion vector of the temperature distribution of the whole and local component object granularity of the first device, V2 represents the phase conversion vector of the temperature distribution of the whole and local component object granularity of the second device, ω represents the weighted hyperparameter, and V' represents the optimized eigenvector of the temperature distribution of the whole and local component object granularity of the device.

[0129] Correspondingly, in the preferred example, the difference between the eigenvalues of the eigenvector of the temperature distribution of the whole and local component object granularity of the device with respect to the different sum and modulation representation of the eigenvector set of the whole eigenvector of the temperature distribution of the whole and local component object granularity of the device is used as the semantic change intensity information, and a class phase conversion corresponding to the position-based intensity modulation is performed through different sum and modulation representation forms, so that the aggregation enhancement of semantic change phase perception can enhance the axial aggregation receptive field along the feature aggregation direction by performing a space translation operation based on alternating stacking under the scale balance of the eigenvector set of the temperature distribution of the whole and local component object granularity of the device, thereby enhancing the perception effect of the aggregated semantics of the eigenvector of the temperature distribution of the whole and local component object granularity of the device on the detailed semantic change, enhancing the expression effect of the eigenvector of the temperature distribution of the whole and local component object granularity of the device, and enhancing the accuracy of the detection result obtained by passing it through the device detector based on the classifier. In this way, the equipment failures in the motor production workshop can be detected in real time and effectively, so as to timely discover potential failures and issue alarms, help the operators take measures quickly, and adjust the production plan, so as to ensure the quality of the motor products while enabling the motor production workshop to complete production on time.

[0130] Based on the above embodiments, refer to Figure 4 As shown, it is a flowchart of a collaborative management method for a motor production workshop in an embodiment of the present application. For example, this collaborative management method for a motor production workshop can be executed by a server, and the server can be Figure 1 the server 100 shown in Figure 4 As shown, according to the collaborative management method for a motor production workshop in an embodiment of the present application, it includes the steps: S510, generating a production plan according to the order requirements and existing resources, splitting the production plan into production tasks, and then allocating the production tasks to the operators on each production line; S520, uploading production progress data through the data acquisition terminals deployed at each work station, and generating a real-time production progress chart based on the production progress data; S530, in response to detecting an equipment failure, issuing an alarm and generating a production plan adjustment prompt; S540, recording the working hours and task completion status of each operator, and generating a performance evaluation report.

[0131] Here, those skilled in the art can understand that the specific operations of each step in the above-mentioned collaborative management method for the motor production workshop have been described in detail in the description of the motor production workshop collaborative management system 800 with reference to Figures 2 to 3 above, and thus, the repeated description thereof will be omitted.

[0132] Figure 5 FIG. is an application scenario diagram of the motor production workshop collaborative management system according to an embodiment of the present application. As Figure 5 shown, in this application scenario, first, a thermal infrared image of the monitored device collected by the data acquisition terminal (for example, Figure 5 D shown in ) is input into a server (for example, Figure 5 S shown in ) deployed with the motor production workshop collaborative management algorithm, where the server can use the motor production workshop collaborative management algorithm to process the thermal infrared image to obtain a detection result indicating whether there is a fault in the monitored device.

[0133] Based on the above embodiments, an electronic device according to another exemplary embodiment is further provided in the embodiments of the present application. In some possible implementation manners, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the motor production workshop collaborative management method in the above embodiments can be implemented.

[0134] For example, taking the server 100 in the present application Figure 1 as an example of the electronic device for illustration, the processor in this electronic device is the processor 110 in the server 100, and the memory in this electronic device is the memory 120 in the server 100.

[0135] The embodiments of the present application further provide a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are run by a processor, the motor production workshop collaborative management method according to the embodiments of the present application described with reference to the above drawings can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory, etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc.

[0136] Embodiments of the present application also provide a computer program product or a computer program, which includes computer-executable instructions stored in a computer-readable storage medium. The processor of the computer device reads the computer-executable instructions from the computer-readable storage medium, and the processor executes the computer-executable instructions, so that the computer device executes the collaborative management method for the motor production workshop according to the embodiments of the present application.

[0137] Those skilled in the art can understand that the content disclosed in the present application can have various variations and improvements. For example, the various devices or components described above can be implemented by hardware, or by software, firmware, or some or all of the combinations of the three.

[0138] In addition, although the present application makes various references to certain units in the system according to the embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0139] Those of ordinary skill in the art can understand that all or part of the steps in the above method can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a magnetic disk, or an optical disc, etc. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiments can be implemented in the form of hardware or in the form of a software functional module. The present application is not limited to any specific combination of hardware and software.

[0140] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in a general dictionary should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and should not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0141] The above is an explanation of the present application and should not be regarded as a limitation thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will easily understand that many modifications can be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A motor production workshop collaborative management system, characterized in that: include: A task allocation module is used to generate a production plan based on order requirements and existing resources, split the production plan into production tasks, and then allocate the production tasks to operators of each production line; A progress tracking module, used to upload production progress data through a data acquisition terminal deployed at each workstation, and generate a real-time production progress chart based on the production progress data; A plan adjustment module, for responding to detection of equipment failure, issuing an alarm and generating a production plan adjustment prompt; A performance evaluation module, used to record the working hours and task completion of each operator and generate a performance evaluation report; Wherein, the plan adjustment module includes: An equipment thermal infrared image acquisition unit, used to acquire a thermal infrared image of the monitored equipment acquired by the data acquisition terminal; An equipment component image region division unit, used for performing image region division on the thermal infrared image based on equipment components to obtain a set of unit region images of thermal infrared components; A thermal infrared component unit area position coding unit, used for using a position coding function to perform position coding on each thermal infrared component unit area image in the set of thermal infrared component unit area images to obtain a set of device component position coding vectors; The device component thermal distribution feature extraction unit is used to extract thermal distribution features of each thermal infrared component unit area image in the set of thermal infrared component unit area images to obtain a set of device component thermal distribution feature maps; A device component thermal distribution feature enhancement unit, used for respectively performing a feature space structure consistency self-attention cross-channel enhancement module on each device component thermal distribution feature map in the set of device component thermal distribution feature maps to obtain a set of device component thermal distribution enhancement feature maps; A location information equipment thermal distribution feature fusion unit, used to fuse each corresponding set of equipment component thermal distribution enhancement feature maps and equipment component position encoding vectors in the set of equipment component thermal distribution enhancement feature maps and the set of equipment component position encoding vectors to obtain a set of equipment component thermal distribution feature vectors containing location information; A device global component object granular temperature distribution characterization unit, used for inputting the set of device component thermal distribution feature vectors containing position information into a feature aggregation network based on feature energy local significance gating to obtain a device global component object granular temperature distribution feature vector; An equipment fault detection unit, used for inputting the equipment global component object granular temperature distribution feature vector into a classifier-based equipment detector to obtain a detection result, wherein the detection result is used for indicating whether the monitored equipment has a fault; Wherein, the equipment component thermal distribution characteristic strengthening unit comprises: A layer normalization subunit, used for performing layer normalization on the device component thermal distribution characteristic map to obtain a normalized device component thermal distribution characteristic map; A point convolution processing subunit, used for performing point convolution processing on the normalized device component thermal distribution feature map to obtain a device component thermal distribution state channel context association representation feature map; A convolutional encoding subunit, configured to perform convolutional encoding on the device component thermal distribution state channel context association representation feature map to obtain a device component thermal distribution state space context association representation feature map; A copying subunit, used for copying the equipment component thermal distribution state space context association representation feature map to obtain the equipment component thermal distribution state backup space context association representation feature map; The first characteristic shape reshaping subunit is used to perform characteristic shape reshaping on the equipment component thermal distribution state channel context association representation feature map, the equipment component thermal distribution state space context association representation feature map and the equipment component thermal distribution state backup space context association representation feature map to obtain an equipment component thermal distribution state channel context association representation feature matrix, an equipment component thermal distribution state space context association representation feature matrix and an equipment component thermal distribution state backup space context association representation feature matrix; A covariance matrix calculation subunit, used to calculate a cross-channel cross-covariance matrix of thermal distribution states of equipment components between a channel context association representation feature matrix of thermal distribution states of equipment components and a spatial context association representation feature matrix of thermal distribution states of equipment components; An activation subunit, used for activating the cross-channel cross-covariance matrix of the thermal distribution state of the equipment components using a Softmax function to obtain a global interactive attention matrix of the thermal distribution state features of the equipment components; A product calculation subunit, used for calculating the product between the equipment component thermal distribution state backup space context association representation feature matrix and the equipment component thermal distribution state feature global interactive attention matrix to obtain the equipment component thermal distribution state attention enhancement feature representation matrix; The second characteristic shape reshaping subunit is used to perform characteristic shape reshaping on the attention enhancement feature representation matrix of the thermal distribution state of the equipment component to obtain the equipment component thermal distribution enhancement feature map.

2. The motor production workshop collaborative management system according to claim 1, characterized in that: The equipment component thermal distribution feature extraction unit is used to: Each thermal infrared component unit area image in the set of thermal infrared component unit area images is respectively passed through a device part thermal distribution feature extractor based on a hole convolutional neural network model to obtain a set of thermal distribution feature maps of the device parts.

3. The motor production workshop collaborative management system according to claim 2, characterized in that: The covariance matrix calculation subunit is used for: Calculating matrix multiplication between the channel context association representation feature matrix of the thermal distribution state of the equipment component and the spatial context association representation feature matrix of the thermal distribution state of the equipment component to obtain a channel-spatial context association feature matrix of the thermal distribution state of the equipment component; The position-wise division between the channel-space contextual association feature matrix of the thermal distribution state of the device components and the hyperparameters is calculated to obtain the cross-channel cross-covariance matrix of the thermal distribution state of the device components.

4. The motor production workshop collaborative management system according to claim 3, characterized in that: The location information device thermal distribution feature fusion unit is used to: Performing global mean pooling processing on each device component thermal distribution enhancement feature map in the set of device component thermal distribution enhancement feature maps to obtain a set of device component thermal distribution enhancement feature vectors; Each corresponding set of equipment component position encoding vectors and equipment component thermal distribution enhancement feature vectors in the set of equipment component position encoding vectors and the set of equipment component thermal distribution enhancement feature vectors are fused separately to obtain the set of equipment component thermal distribution feature vectors containing position information.

5. The motor production workshop collaborative management system according to claim 4, characterized in that: The device global component object particle size temperature distribution characterization unit includes: a characteristic energy level coefficient calculation subunit, configured to determine the characteristic energy level coefficient of each equipment component thermal distribution feature vector containing position information in the set of equipment component thermal distribution feature vectors containing position information based on the maximum value, minimum value, mean value and variance of each equipment component thermal distribution feature vector containing position information, so as to obtain a sequence of equipment component thermal distribution characteristic energy level coefficients containing position information; A local neighborhood mean calculation subunit is used to determine the scale of the local neighborhood of the thermal distribution of the equipment component containing the location information, use the scale of the local neighborhood of the thermal distribution of the equipment component containing the location information as the mean calculation range, and respectively calculate the local neighborhood mean of each thermal distribution characteristic energy level coefficient of the equipment component containing the location information to obtain a sequence of thermal distribution characteristic energy level significance descriptors of the equipment component containing the location information; A mask subunit, used for inputting the sequence of equipment component thermal distribution feature energy level significance descriptors containing position information into a mask module based on a gating function to obtain a mask-gated probabilistic sequence of equipment component thermal distribution feature energy level significance descriptors containing position information; The weighted sum calculation subunit is used to calculate the position-weighted sum of the set of thermal distribution feature vectors of the equipment components containing the position information by using the sequence of the mask-gated probabilistic energy level significance descriptors of the equipment components containing the position information as the sequence of weights to obtain the global component object granularity temperature distribution feature vector of the equipment.

6. The motor production workshop collaborative management system according to claim 5, characterized in that: The characteristic energy level coefficient calculation subunit is used for: Calculating the maximum value, minimum value, average value and variance of the thermal distribution feature vector of the equipment component containing the position information to obtain the maximum value of the thermal distribution of the equipment component containing the position information, the minimum value of the thermal distribution of the equipment component containing the position information, the average value of the thermal distribution of the equipment component containing the position information and the variance of the thermal distribution of the equipment component containing the position information; Calculate the sum of the thermal distribution variance of the equipment component containing the position information and the hyperparameter to obtain a first local energy statistical factor of thermal distribution of the equipment component containing the position information; Calculating the difference between the maximum value of the thermal distribution of the equipment component containing the position information and the average value of the thermal distribution of the equipment component containing the position information to obtain the maximum deviation value of the thermal distribution of the equipment component containing the position information; Calculating the difference between the minimum value of the thermal distribution of the equipment component containing the position information and the average value of the thermal distribution of the equipment component containing the position information to obtain a minimum deviation value of the thermal distribution of the equipment component containing the position information; Calculating the square sum of the maximum deviation value of the thermal distribution of the equipment component containing the position information and the minimum deviation value of the thermal distribution of the equipment component containing the position information and adding the sum to the hyperparameter to obtain a second local energy statistical factor of the thermal distribution of the equipment component containing the position information; The division between the first local energy statistical factor of thermal distribution of equipment components containing position information and the second local energy statistical factor of thermal distribution of equipment components containing position information is calculated to obtain a characteristic energy level coefficient of thermal distribution of equipment components containing position information.

7. The motor production workshop collaborative management system according to claim 6, characterized in that: The mask subunit is used for: Taking the negative number of each device component thermal distribution feature energy level significance descriptor containing location information in the sequence of device component thermal distribution feature energy level significance descriptors containing location information as an exponential power, calculating the exponential function value with the natural constant e as the base to obtain a sequence of device component thermal distribution feature energy level class support significance descriptors containing location information; Calculating the positional sum of the sequence of the equipment component thermal distribution feature energy level class support significant descriptors containing the position information and a constant 1 to obtain a linearly modulated sequence of the equipment component thermal distribution feature energy level class support significant descriptors containing the position information; Calculating the inverse of each linear modulation device component thermal distribution feature energy level class support significance descriptor in the sequence of device component thermal distribution feature energy level class support significance descriptors containing linear modulation position information to obtain a sequence of device component thermal distribution gated probabilistic feature energy level significance descriptors containing position information; Performing masking processing on each device component thermal distribution gated probabilistic feature energy level significance descriptor containing position information in the sequence of device component thermal distribution gated probabilistic feature energy level significance descriptors containing position information to obtain the sequence of masked gated probabilistic device component thermal distribution feature energy level significance descriptors containing position information; Among them, performing masking processing on each device component thermal distribution gated probabilistic feature energy level significance descriptor containing location information in the sequence of device component thermal distribution gated probabilistic feature energy level significance descriptors containing location information to obtain the sequence of masked gated probabilistic device component thermal distribution feature energy level significance descriptors containing location information, including: In response to the device component thermal distribution gated probabilistic feature energy level significance descriptor containing the location information being greater than a predetermined threshold, the original value of the device component thermal distribution gated probabilistic feature energy level significance descriptor containing the location information is taken, otherwise it is set to 0.

8. A motor production workshop collaborative management method, using the motor production workshop collaborative management system according to claim 1, characterized in that: include: Generate a production plan based on order requirements and existing resources, split the production plan into production tasks, and assign the production tasks to operators of each production line; Uploading production progress data through a data acquisition terminal deployed at each workstation, and generating a real-time production progress chart based on the production progress data; In response to detecting equipment failure, an alarm is issued and a production plan adjustment prompt is generated; Record the working hours and task completion of each operator and generate a performance evaluation report.

Images