Intelligent management method and system for mechanical and electrical engineering equipment

Abstract

This application relates to the field of intelligent management of production equipment, and provides an intelligent management method and system for electromechanical engineering equipment. It first acquires production task information for multiple production tasks and equipment information for the electromechanical engineering equipment to be assigned. Next, it initializes the allocation scheme for the multiple production tasks on the equipment to be assigned. Finally, it manages the processing priority of the multiple production tasks based on the production task information and the equipment information. This allows for more flexible task allocation and effectively improves the adaptability and flexibility of equipment management.

Description

Technical Field

[0001] This application relates to the field of intelligent management of production equipment, and more specifically, to an intelligent management method and system for electromechanical engineering equipment. Background Technology

[0002] With the development of information technology, technologies such as sensors, the Internet of Things, and big data analytics have made information-based management and control of electromechanical equipment possible, solving problems such as cumbersome manual inspections, slow fault response, and high maintenance costs in traditional management. However, existing management and control systems, due to their use of predefined scheduling methods, struggle to adapt to changes in production in real time when faced with multivariable and complex systems, impacting efficiency and management effectiveness.

[0003] To address the aforementioned technical issues, patent CN118095580B proposes an information-based management and control system for electromechanical engineering equipment. This system collects production tasks and equipment information, generates and optimizes task allocation schemes, iterates continuously until the globally optimal solution is found, and finally outputs the best task allocation scheme. This solves the problems of cumbersome manual inspections and untimely fault handling in traditional equipment management.

[0004] The aforementioned patent updates the allocation speed of each task by combining the global optimal solution, the individual optimal solution, the allocated content, and the allocation speed. While this method can quickly adjust strategies to cope with changing environments and conditions, its lack of in-depth analysis of contextual semantic relationships makes it relatively less adaptable and flexible when dealing with complex production and equipment management scenarios. Furthermore, this method primarily focuses on the current optimized value and historical best solutions, failing to fully consider multi-dimensional factors such as task dependencies and equipment operating status. This makes the optimization results potentially incomplete and inaccurate, making it difficult to meet the demands of highly dynamic and interconnected production environments. Such single-dimensional optimization may lead to unreasonable resource allocation, affecting overall production efficiency and equipment utilization.

[0005] Therefore, there is a need for an optimized intelligent management system for electromechanical engineering equipment. Summary of the Invention

[0006] This application addresses the shortcomings of existing technologies by providing an intelligent management method and system for electromechanical engineering equipment.

[0007] According to one aspect of this application, an intelligent management system for electromechanical engineering equipment is provided, comprising:

[0008] The production task information acquisition module is used to acquire production task information from multiple production tasks to obtain a collection of production task information.

[0009] The equipment information acquisition module is used to acquire equipment information of the electromechanical engineering equipment to be allocated;

[0010] The allocation speed initialization module is used to initialize the allocation scheme of the multiple production tasks on the electromechanical engineering equipment to be allocated;

[0011] The production task allocation speed adjustment module is used to manage the processing priority of the multiple production tasks based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated.

[0012] The production task allocation speed adjustment module includes: a production task information encoding unit, used to semantically encode and associate the set of production task information to obtain a set of semantic association features of production task information; a device information encoding unit, used to semantically encode the device information of the electromechanical engineering device to be allocated to obtain device information semantic encoding features; a production task device semantic interaction response unit, used to perform principal component feature semantic interaction response on the set of semantic association features of production task information and the semantic encoding features of device information to obtain a set of production task information-device information semantic interaction response features; and a processing priority management unit, used to generate a task processing priority allocation result based on the set of production task information-device information semantic interaction response features, and to sort the processing priorities of the multiple production tasks.

[0013] In the aforementioned intelligent management system for electromechanical engineering equipment, the production task information encoding unit includes: a production task information semantic encoding subunit, used to perform semantic encoding on the set of production task information to obtain a set of production task information semantic encoding feature vectors; and a production task information semantic association feature extraction subunit, used to extract a set of production task information semantic association feature vectors by passing the set of production task information semantic encoding feature vectors through a converter-based context encoder, as the set of production task information semantic association features.

[0014] In the above-mentioned intelligent management system for electromechanical engineering equipment, the equipment information encoding unit to be allocated is used to: perform semantic encoding on the equipment information of the electromechanical engineering equipment to be allocated to obtain the equipment information semantic encoding feature vector as the equipment information semantic encoding feature.

[0015] In the aforementioned intelligent management system for electromechanical engineering equipment, the production task equipment semantic interaction response unit includes: a production task information equipment information principal component analysis subunit, used to perform principal component analysis on the semantic association feature vector of the production task information and the semantic encoding feature vector of the equipment information to obtain a set of semantic principal component feature components of the production task information and a set of semantic principal component feature components of the equipment information; and a production task information-equipment information optimal matching subunit, used to perform optimal matching interaction between the set of semantic principal component feature components of the production task information and the set of semantic principal component feature components of the equipment information to obtain a production task information-equipment information semantic interaction response feature vector as the set of semantic interaction response features of the production task information-equipment information.

[0016] In the aforementioned intelligent management system for electromechanical engineering equipment, the optimal matching subunit for production task information and equipment information includes: a secondary subunit for optimal matching of production task information and equipment information, used as query vectors each of the semantic principal component features of the production task information in the set of semantic principal component features of the production task information, and using the set of semantic principal component features of the equipment information as a query library, to match the equipment information semantic principal component features that best match each of the query vectors from the query library to obtain a set of optimal matching pairs of {production task information semantic principal component features; equipment information semantic principal component features}; and multi-scale interaction of production task information and equipment information. The second-level subunit is used to input each of the best-matching pairs of {production task information semantic principal component feature components; equipment information semantic principal component feature components} from the set of best-matching pairs of {production task information semantic principal component feature components; equipment information semantic principal component feature components} into the multi-scale interactive response coupling module to obtain a set of multi-scale interactive coupling representation vectors of the best-matching pairs of production task information and equipment information semantics; the second-level subunit for generating semantic interactive response features of production task information and equipment information is used to cascade the set of multi-scale interactive coupling representation vectors of the best-matching pairs of production task information and equipment information semantics to obtain the semantic interactive response feature vector of production task information and equipment information.

[0017] In the aforementioned intelligent management system for electromechanical engineering equipment, the second-level sub-unit for optimal matching of production task information and equipment information is used to: extract predetermined semantic principal component features of production task information from the set of semantic principal component features of production task information; calculate the Mahalanobis distance between the predetermined semantic principal component features of production task information and each semantic principal component feature of equipment information in the set of semantic principal component features of equipment information to obtain a set of matching distances, wherein the pair {semantic principal component features of production task information; semantic principal component features of equipment information} with the smallest matching distance in the set of matching distances is taken as the optimal matching pair.

[0018] In the aforementioned intelligent management system for electromechanical engineering equipment, the second-level subunit for multi-scale interaction between production task information and equipment information is used to: calculate the L2 norms of the semantic principal component features of the production task information and the semantic principal component features of the equipment information in the best matching pair of {semantic principal component features of production task information; semantic principal component features of equipment information} to obtain the norm values ​​of production task information and equipment information; extract the maximum value from the norm values ​​of production task information and equipment information, as well as the modulation parameters, to obtain the interaction maximum value; and calculate the positional subtraction, positional multiplication, and positional addition between the semantic principal component features of the production task information and the semantic principal component features of the equipment information to obtain the semantic principal component difference vector and the semantic principal component multiplication vector of production task information and equipment information. The semantic principal components of production task information and equipment information are summed into a vector; the difference vector of the semantic principal components of production task information and equipment information, the dot product vector of the semantic principal components of production task information and equipment information, and the summed vector of the semantic principal components of production task information and equipment information are concatenated to obtain a concatenated vector of semantic principal components of production task information and equipment information; the concatenated vector of semantic principal components of production task information and equipment information is then subjected to one-dimensional convolutional encoding to obtain a convolutional encoded vector of semantic principal components of production task information and equipment information; the convolutional encoded vector of semantic principal components of production task information and equipment information is subjected to maximum pooling based on a local window, and the resulting semantic interaction representation vector of production task information and equipment information is divided positionally by the maximum interaction value to obtain the multi-scale interaction coupling representation vector of the semantic best match of production task information and equipment information.

[0019] In the aforementioned intelligent management system for electromechanical engineering equipment, the processing priority management unit includes: a production task allocation speed decoding subunit, used to obtain a set of production task allocation speed decoding values ​​based on the set of semantic interaction response features between production task information and equipment information; a task processing priority allocation result generation subunit, used to arrange each production task allocation speed decoding value in the set of production task allocation speed decoding values ​​in descending order to obtain a task processing priority allocation result; and a production task processing sorting subunit, used to sort the processing priorities of the multiple production tasks based on the task processing priority allocation result.

[0020] In the aforementioned intelligent management system for electromechanical engineering equipment, the production task allocation speed decoding subunit is used to: pass each production task information-equipment information semantic interaction response feature vector in the set of production task information-equipment information semantic interaction response feature vectors through a decoder-based task priority allocator to obtain a set of production task allocation speed decoding values.

[0021] According to another aspect of this application, an intelligent management method for electromechanical engineering equipment is provided, comprising:

[0022] Obtain production task information from multiple production tasks to obtain a set of production task information;

[0023] Obtain equipment information for electromechanical engineering equipment to be allocated;

[0024] The allocation scheme for the multiple production tasks on the electromechanical engineering equipment to be allocated is initialized;

[0025] The processing priority of the multiple production tasks is managed based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated.

[0026] The process of managing the processing priority of the multiple production tasks based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated includes: semantically encoding and associating the set of production task information to obtain a set of semantic association features of production task information; semantically encoding the equipment information of the electromechanical engineering equipment to be allocated to obtain equipment information semantic encoding features; performing principal component feature semantic interaction response on the set of semantic association features of production task information and the semantic encoding features of equipment information to obtain a set of semantic interaction response features of production task information-equipment information; generating a task processing priority allocation result based on the set of semantic interaction response features of production task information-equipment information, and sorting the processing priority of the multiple production tasks.

[0027] This application, by adopting the above technical solution, has significant technical effects:

[0028] The intelligent management method and system for electromechanical engineering equipment provided in this application employs artificial intelligence-based information analysis and encoding to semantically encode and correlate the information of each production task with context. It also semantically encodes the information of the equipment to be allocated. Based on the principal component semantic interaction response features between the semantic correlation features of each production task and the semantic encoding features of the equipment information, it intelligently obtains the decoding value of the production task allocation speed and adjusts the processing priority of each production task. This approach better adapts to complex and ever-changing production and equipment management scenarios, improves adaptability and flexibility in dynamic environments, and more comprehensively considers multi-dimensional factors such as the dependencies between production tasks and the working status of equipment, thus providing a more reasonable and flexible solution for task allocation. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0030] Figure 1 This is a system block diagram of an intelligent management system for electromechanical engineering equipment according to an embodiment of this application.

[0031] Figure 2 This is a block diagram of the production task allocation speed adjustment module in the intelligent management system for electromechanical engineering equipment according to an embodiment of this application.

[0032] Figure 3 This is a schematic diagram of the data flow in the production task allocation speed adjustment module of the intelligent management system for electromechanical engineering equipment according to an embodiment of this application.

[0033] Figure 4 This is a block diagram of the production task information encoding unit in the intelligent management system for electromechanical engineering equipment according to an embodiment of this application.

[0034] Figure 5 This is a block diagram of the semantic interaction response unit for production task equipment in the intelligent management system for electromechanical engineering equipment according to an embodiment of this application.

[0035] Figure 6 This is a block diagram of the processing priority management unit in the intelligent management system for electromechanical engineering equipment according to an embodiment of this application.

[0036] Figure 7This is a flowchart of an intelligent management method for electromechanical engineering equipment according to an embodiment of this application. Detailed Implementation

[0037] Hereinafter, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments of this application. It should be understood that this application is not limited to the exemplary embodiments described herein.

[0038] With the continuous advancement of information technology, the application of technologies such as sensors, the Internet of Things, and big data analytics has made the information management of electromechanical equipment a reality. This effectively solves problems inherent in traditional management methods, such as high labor intensity from manual inspections, slow fault response times, and high maintenance costs. However, existing management systems often employ preset scheduling strategies, making it difficult to adapt to the multivariate and complex changes in the production process, thus affecting management efficiency and effectiveness.

[0039] To address the aforementioned technical issues, patent CN118095580B proposes an information-based management and control system for electromechanical engineering equipment. This system can collect production task and equipment status information, and generate and optimize task allocation schemes. Through continuous iteration, the system can find the globally optimal task allocation scheme, effectively solving problems such as heavy manual inspection burden and untimely fault response in traditional equipment management.

[0040] While the method described in this patent can quickly adjust strategies to adapt to environmental changes, its adaptability and flexibility are limited when dealing with complex production and equipment management scenarios due to a lack of in-depth analysis of contextual semantics. Furthermore, the method primarily focuses on current optimization results and historical best solutions, failing to adequately consider multi-dimensional factors such as task dependencies and the actual operating status of equipment. This may result in incomplete and inaccurate optimization results, making it difficult to meet the demands of highly dynamic and interconnected production environments. Single-dimensional optimization can lead to unreasonable resource allocation, thereby affecting overall production efficiency and equipment utilization.

[0041] Based on this, this application proposes an intelligent management system for electromechanical engineering equipment. Figure 1 This is a system block diagram of an intelligent management system for electromechanical engineering equipment according to an embodiment of this application. Figure 1As shown, the intelligent management system 100 for electromechanical engineering equipment includes: a production task information acquisition module 110, used to acquire production task information of multiple production tasks to obtain a set of production task information; an equipment information acquisition module 120, used to acquire equipment information of electromechanical engineering equipment to be allocated; an allocation speed initialization module 130, used to initialize the allocation scheme of the multiple production tasks on the electromechanical engineering equipment to be allocated; and a production task allocation speed adjustment module 140, used to manage the processing priority of the multiple production tasks based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated.

[0042] In this embodiment, the production task information acquisition module 110 and the equipment information acquisition module 120 are respectively used to acquire production task information of multiple production tasks to obtain a set of production task information and to acquire equipment information of electromechanical engineering equipment to be allocated. It should be understood that the production task information of each production task includes task number, task type, processing priority, etc., and a set of production task information can be obtained by organizing the production task information of multiple production tasks. The equipment information of the electromechanical engineering equipment to be allocated specifically includes equipment number, equipment type, processing capacity, current status, etc. Here, by acquiring the production task information of multiple production tasks, it can be analyzed in subsequent steps, helping the system understand the urgency of different production tasks and identify the dependencies between different production tasks, thereby helping to optimize task scheduling. By acquiring and analyzing the equipment information of the electromechanical engineering equipment to be allocated, it can help the system better understand the equipment's processing capacity and status, thereby optimizing task allocation and avoiding resource waste and equipment overload. In summary, by combining real-time production task and equipment information, the system can flexibly respond to changes in the production environment, adjust task priorities and equipment allocation schemes in a timely manner, thereby improving overall production efficiency.

[0043] In this embodiment, the allocation speed initialization module 130 is used to initialize the allocation scheme of the multiple production tasks on the electromechanical engineering equipment to be allocated. It should be noted that this application proposes the concept of allocation speed, which refers to the degree of adjustment of the processing priority of production tasks on the electromechanical engineering equipment. It should be understood that initializing the allocation scheme of the multiple production tasks in this application specifically means initially setting the allocation speed of the task allocation scheme for each of the multiple production tasks to 0. By resetting the allocation speed of different production tasks to zero, a unified baseline state can be provided for subsequent task scheduling and priority adjustment, thereby helping to avoid scheduling chaos caused by inconsistent initial states of different tasks.

[0044] In this embodiment of the application, the production task allocation speed adjustment module 140 is used to manage the processing priority of the multiple production tasks based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated.

[0045] Accordingly, in the production task allocation speed adjustment module, the technical concept of this application is to use an artificial intelligence-based information analysis and encoding method to semantically encode the information of each production task and establish semantic associations between contexts, and to semantically encode the information of the equipment to be allocated. Based on the principal component semantic interaction response features between the semantic association features of each production task information and the semantic encoding features of the equipment information, the decoding value of the production task allocation speed is intelligently obtained, and the processing priority of each production task is adjusted. This better adapts to complex and ever-changing production and equipment management scenarios, improves adaptability and flexibility in dynamic environments, and can more comprehensively consider multi-dimensional factors such as the dependencies between production tasks and the working status of equipment, thereby providing a more reasonable and flexible solution for task allocation.

[0046] Specifically, Figure 2 This is a block diagram of the production task allocation speed adjustment module in the intelligent management system for electromechanical engineering equipment according to an embodiment of this application. Figure 3 This is a schematic diagram of the data flow in the production task allocation speed adjustment module of the intelligent management system for electromechanical engineering equipment according to an embodiment of this application. Figure 2 and Figure 3 As shown, the production task allocation speed adjustment module 140 includes: a production task information encoding unit 141, used to perform semantic encoding and association on the set of production task information to obtain a set of semantic association features of production task information; a device information encoding unit 142, used to perform semantic encoding on the device information of the electromechanical engineering equipment to be allocated to obtain device information semantic encoding features; a production task device semantic interaction response unit 143, used to perform principal component feature semantic interaction response on the set of semantic association features of production task information and the semantic encoding features of device information to obtain a set of production task information-device information semantic interaction response features; and a processing priority management unit 144, used to generate a task processing priority allocation result based on the set of production task information-device information semantic interaction response features, and to sort the processing priorities of the multiple production tasks.

[0047] In this embodiment of the application, the production task information encoding unit 141 is used to perform semantic encoding and association on the set of production task information to obtain a set of semantic association features of production task information. Specifically, Figure 4 This is a block diagram of the production task information encoding unit in the intelligent management system for electromechanical engineering equipment according to an embodiment of this application. Figure 4 As shown, the production task information encoding unit 141 includes: a production task information semantic encoding subunit 1411, used to perform semantic encoding on the set of production task information to obtain a set of production task information semantic encoding feature vectors; and a production task information semantic association feature extraction subunit 1412, used to pass the set of production task information semantic encoding feature vectors through a converter-based context encoder to obtain a set of production task information semantic association feature vectors as the set of production task information semantic association features.

[0048] In this embodiment, the production task information semantic encoding subunit 1411 is used to semantically encode the set of production task information to obtain a set of production task information semantic encoding feature vectors. It should be understood that the set of production task information contains a large amount of semantic information, such as task descriptions, processing priorities, and resource requirements. To better capture and extract the deep semantic meanings and contextual relationships in each piece of production task information, the technical solution of this application performs semantic encoding on the set of production task information to obtain a set of production task information semantic encoding feature vectors. Specifically, in a specific implementation of this embodiment, the BERT model can be used to semantically encode the set of production task information to obtain the set of production task information semantic encoding feature vectors.

[0049] In this embodiment, the semantic association feature extraction subunit 1412 for production task information is used to obtain a set of semantic association feature vectors for production task information by passing the set of semantically encoded feature vectors for production task information through a converter-based context encoder. Accordingly, considering that different features in the set of semantically encoded feature vectors for production task information have different semantic associations, and that the converter model can effectively capture long-distance dependencies between features at different positions in the input feature sequence through a self-attention mechanism, this means that the model can capture the complex relationships and contextual information between semantically encoded features of production task information. Based on this, in the technical solution of this application, the set of semantically encoded feature vectors for production task information is passed through a converter-based context encoder to capture and mine the semantic association relationships of production task information across different semantic spans, thereby obtaining a set of semantically associated feature vectors for production task information.

[0050] In this embodiment, the equipment information encoding unit 142 is used to semantically encode the equipment information of the electromechanical engineering equipment to be allocated to obtain equipment information semantic encoding features. Specifically, in this embodiment, the equipment information encoding unit is used to: semantically encode the equipment information of the electromechanical engineering equipment to be allocated to obtain an equipment information semantic encoding feature vector as the equipment information semantic encoding features. It should be understood that the equipment information to be allocated expresses semantic information about the equipment, such as the equipment's capability range, equipment number, and processing status. Therefore, in the technical solution of this application, the equipment information of the electromechanical engineering equipment to be allocated is semantically encoded to capture the deep semantics in the equipment information and obtain the equipment information semantic encoding feature vector. In particular, in a specific implementation of this embodiment, the BERT model is used to semantically encode the equipment information of the electromechanical engineering equipment to be allocated, thereby obtaining the equipment information semantic encoding feature vector.

[0051] In this embodiment, the production task equipment semantic interaction response unit 143 is used to perform principal component feature semantic interaction response on the set of semantic association features of the production task information and the semantic encoding features of the equipment information to obtain a set of semantic interaction response features between production task information and equipment information. Specifically, Figure 5 This is a block diagram of the semantic interaction response unit for production task equipment in an intelligent management system for electromechanical engineering equipment according to an embodiment of this application. Figure 5 As shown, the production task equipment semantic interaction response unit 143 includes: a production task information equipment information principal component analysis subunit 1431, used to perform principal component analysis on the production task information semantic association feature vector and the equipment information semantic encoding feature vector to obtain a set of production task information semantic principal component feature components and a set of equipment information semantic principal component feature components; and a production task information-equipment information optimal matching subunit 1432, used to perform production task information-equipment information optimal matching interaction on the set of production task information semantic principal component feature components and the set of equipment information semantic principal component feature components to obtain a production task information-equipment information semantic interaction response feature vector as the set of production task information-equipment information semantic interaction response features.

[0052] It should be understood that each semantic association feature vector of production task information and each semantic encoding feature vector of equipment information in the set of semantic association feature vectors of production task information contain key semantic information, and there is a semantic correlation between production task information and equipment information. Therefore, in order to achieve more refined and efficient matching between production tasks and equipment, to identify which task features are most relevant to the equipment features, and to optimize the matching accordingly, thereby improving the allocation of processing priorities, in the technical solution of this application, principal component feature semantic interaction response is performed on each semantic association feature vector of production task information and each semantic encoding feature vector of equipment information in the set of semantic association feature vectors of production task information to obtain a set of semantic interaction response feature vectors of production task information-equipment information as the set of semantic interaction response features of production task information-equipment information.

[0053] In detail, firstly, principal component analysis (PCA) is performed on the semantic association feature vector of the production task information and the semantic encoding feature vector of the equipment information to obtain the sets of semantic principal component feature components of the production task information and the equipment information. Specifically, PCA is a commonly used dimensionality reduction technique that projects the original feature vectors into a new coordinate system to reveal the inherent structure hidden in high-dimensional data, preserving most of the variance information of the original data while reducing the dimensionality of the features. In other words, the resulting sets of principal components represent the main parts of the original feature space that best reflect the changes in the data.

[0054] Then, the set of semantic principal component features of the production task information and the set of semantic principal component features of the equipment information are subjected to optimal matching interaction to obtain the semantic interaction response feature vector of production task information-equipment information. Specifically, using each semantic principal component feature of the production task information as a query vector and the set of semantic principal component features of the equipment information as a query library, optimal matching is performed on each semantic principal component feature of the production task information to obtain a set of best matching pairs of {semantic principal component features of production task information; semantic principal component features of equipment information}. Such pairing is not merely a simple element association, but an attempt to establish a one-to-one mapping relationship between the two sets of features, thereby clarifying the key semantic relationship between production task information and equipment information and selecting the most representative feature combination. In particular, in this technical solution, the Mahalanobis distance between two principal component features is calculated as the matching value between them, and the pair with the smallest distance is taken as the best matching pair. Subsequently, to more deeply identify and explore the profound semantic connections and multi-dimensional, multi-scale interactions between each best-matching pair, a multi-scale interactive response coupling module was used to perform multi-level interactions and capture on each best-matching pair, resulting in a set of multi-scale interactive coupling representation vectors for the semantic best-matching pairs of production task information and equipment information. Finally, the set of vectors representing the best-matching pairs was cascaded to comprehensively and fully characterize the semantic matching between production tasks and equipment, forming a compact and information-rich semantic interactive response feature vector for production task information and equipment information.

[0055] Specifically, in this embodiment, the optimal matching subunit for production task information and equipment information includes: a secondary subunit for optimal matching of production task information and equipment information, used as query vectors each of the semantic principal component features of production task information in the set of semantic principal component features of production task information and as a query library the set of semantic principal component features of equipment information, to match the equipment information semantic principal component features that best match each of the query vectors from the query library to obtain a set of optimal matching pairs of {semantic principal component features of production task information; semantic principal component features of equipment information}; and a secondary subunit for multi-scale interaction of production task information and equipment information. The unit is used to input each of the best matching pairs of the {production task information semantic principal component feature components; equipment information semantic principal component feature components} from the set of best matching pairs of {production task information semantic principal component feature components; equipment information semantic principal component feature components} into the multi-scale interactive response coupling module to obtain a set of multi-scale interactive coupling representation vectors of the best matching pairs of production task information and equipment information semantics; the secondary sub-unit for generating semantic interactive response features of production task information and equipment information is used to cascade the set of multi-scale interactive coupling representation vectors of the best matching pairs of production task information and equipment information semantics to obtain the semantic interactive response feature vector of production task information and equipment information.

[0056] More specifically, in this embodiment of the application, the secondary subunit for optimal matching of production task information and equipment information is used to: extract predetermined semantic principal component features of production task information from the set of semantic principal component features of production task information; calculate the Mahalanobis distance between the predetermined semantic principal component features of production task information and each semantic principal component feature of equipment information in the set of semantic principal component features of equipment information to obtain a set of matching distances, wherein the pair {semantic principal component features of production task information; semantic principal component features of equipment information} with the smallest matching distance in the set of matching distances is taken as the optimal matching pair.

[0057] More specifically, in this embodiment, the production task information-equipment information multi-scale interaction secondary subunit is used to: calculate the L2 norm of the production task information semantic principal component feature component and the equipment information semantic principal component feature component in the best matching pair of {production task information semantic principal component feature component; equipment information semantic principal component feature component} to obtain the production task information norm value and the equipment information norm value; extract the maximum value from the production task information norm value, the equipment information norm value and the modulation parameter to obtain the interaction maximum value; calculate the positional subtraction, positional multiplication and positional addition between the production task information semantic principal component feature component and the equipment information semantic principal component feature component to obtain the production task information-equipment information semantic principal component difference vector, the production task information-equipment information semantic principal component dot product vector and the production task information-equipment information semantic principal component difference vector. The semantic principal components of production task information and equipment information are summed into a vector; the difference vector of the semantic principal components of production task information and equipment information, the dot product vector of the semantic principal components of production task information and equipment information, and the summed vector of the semantic principal components of production task information and equipment information are concatenated to obtain a concatenated vector of semantic principal components of production task information and equipment information; the concatenated vector of semantic principal components of production task information and equipment information is then subjected to one-dimensional convolutional encoding to obtain a convolutional encoded vector of semantic principal components of production task information and equipment information; the convolutional encoded vector of semantic principal components of production task information and equipment information is then subjected to maximum pooling based on a local window, and the resulting semantic interaction representation vector of production task information and equipment information is divided positionally by the maximum interaction value to obtain the multi-scale interaction coupling representation vector of the best matching pair of semantic principal components of production task information and equipment information.

[0058] In this embodiment of the application, specifically, the production task equipment semantic interaction response unit is used to: perform principal component feature semantic interaction response processing on the production task information semantic association feature vector and the equipment information semantic encoding feature vector using the following formula, thereby obtaining a production task information-equipment information semantic interaction response feature vector; wherein, the formula is:

[0059] C1=U1Λ1U1 T

[0060] C2=U2Λ2U2 T

[0061] U1 = [v 11 ,v 12 ,…,v 1m ]

[0062]

[0063] U2 = [v 21 ,v 22 ,…,v2m ]

[0064]

[0065] v f =[v p1 ;v p2 ;...;v pi ...;v pm ]

[0066] Where C1 and C2 are the sample covariance matrices of production task information and equipment information, respectively, and U1 and U2 are the principal component orthogonal matrices of production task information and equipment information, respectively. 11 ,v 12 ,…,v 1m For each semantic principal component feature component of the production task information in the set of semantic principal component feature components, v 21 ,v 22 ,…,v 2m Let λ be the semantic principal component feature component of each device information in the set of semantic principal component feature components of the device information, Λ1 and Λ2 be the diagonal matrix of production task information and device information respectively, and λ be the semantic principal component feature component of each device information. 11 , λ 1m Let λ be the weight values ​​of the 1st and mth principal component features of the device information semantics in the set of device information semantics principal component features, respectively. 21 , λ 2m U1 represents the weight values ​​of the 1st and mth principal component features of the device information semantics in the set of device information semantics principal component features, respectively. T and U2 T The transposes of U1 and U2 are respectively, v 1i S is the i-th semantic principal component feature of production task information in the set of semantic principal component feature components of the production task information. ij For v 1i and v 2j The covariance matrix between them, arg j min(·) returns the j value corresponding to the minimum value, where k is the minimum approximate matching value. ⊙ and These represent positional differences, positional dot products, and positional additions, respectively. [·;·;·] represents concatenation. conv1D(·) is a one-dimensional convolutional coding operation. MaxPool(·) is a max pooling operation. max(·,·,·) extracts the maximum value. ‖·‖2 is the L2 norm of the vector, ε is the modulation parameter, and v p1 v p2 v pi and v pmThese are the 1st, 2nd, 1st, and 1mth multi-scale interactive coupling representation vectors of the best semantic matching pairs of production task information and equipment information, respectively, in the set of such vectors. f It is the semantic interaction response feature vector of the production task information-equipment information.

[0067] In this embodiment, the processing priority management unit 144 is used to generate a task processing priority allocation result based on the set of semantic interaction response features between the production task information and equipment information, and to sort the processing priorities of the multiple production tasks. Specifically, Figure 6 This is a block diagram of a processing priority management unit in an intelligent management system for electromechanical engineering equipment according to an embodiment of this application. Figure 6 As shown, the processing priority management unit 144 includes: a production task allocation speed decoding subunit 1441, used to obtain a set of production task allocation speed decoding values ​​based on the set of semantic interaction response features of the production task information-equipment information; a task processing priority allocation result generation subunit 1442, used to arrange each production task allocation speed decoding value in the set of production task allocation speed decoding values ​​in descending order to obtain a task processing priority allocation result; and a production task processing sorting subunit 1443, used to sort the processing priorities of the multiple production tasks based on the task processing priority allocation result.

[0068] In this embodiment, the production task allocation speed decoding subunit 1441 is used to obtain a set of production task allocation speed decoding values ​​based on the set of semantic interaction response features of the production task information-equipment information. Specifically, in this embodiment, the production task allocation speed decoding subunit is used to: pass each production task information-equipment information semantic interaction response feature vector in the set of production task information-equipment information semantic interaction response feature vectors through a decoder-based task priority allocator to obtain the set of production task allocation speed decoding values. That is, the set of production task information-equipment information semantic interaction response features obtained by semantic interaction response using each production task information semantic association feature vector in the set of production task information semantic association feature vectors and the device information semantic encoding feature vector is decoded to obtain the set of production task allocation speed decoding values. Specifically, tasks with higher speed decoding values ​​may have higher priority, indicating that they are more likely to be adapted to the currently allocated equipment.

[0069] In this embodiment, the task processing priority allocation result generation subunit 1442 and the production task processing sorting subunit 1443 are respectively used to arrange the production task allocation speed decoding values ​​in the set of production task allocation speed decoding values ​​in descending order to obtain the task processing priority allocation result, and to sort the processing priority of the multiple production tasks based on the task processing priority allocation result. Larger decoding values ​​generally indicate higher priority for the production task, meaning it needs to be processed faster. By arranging the production task allocation speed decoding values ​​in descending order, it is clear which production tasks are more urgent in the current production environment. Based on the task processing priority allocation result obtained from the arrangement, a clear execution order can be provided for the production equipment, and by prioritizing high-priority tasks, it can ensure that limited resources are used most effectively, reducing production losses caused by task delays. This better adapts to complex and ever-changing production and equipment management scenarios, improves adaptability and flexibility in dynamic environments, and can more comprehensively consider multi-dimensional factors such as the dependencies between production tasks and equipment workload prediction, thereby providing a more reasonable and flexible solution for task allocation.

[0070] Specifically, when the semantic association feature vector of the production task information and the semantic encoding feature vector of the equipment information represent the semantic encoding features of the production task information and the equipment information to be allocated, respectively, the sparsity of the interaction matching due to the uneven distribution of the principal component granularity of the semantic encoding features under different task source semantics will lead to the loss of local response coupling. That is, it will lead to the simple repetition of set coupling semantics in the local semantic domain, affecting the set semantic logical dependency and reducing the distinguishability accuracy of each production task allocation speed decoding value in the set of production task allocation speed decoding values.

[0071] Preferably, obtaining a set of production task allocation speed decoding values ​​by passing each production task information-equipment information semantic interaction response feature vector in the set of production task information-equipment information semantic interaction response feature vectors through a decoder-based task priority allocator includes: concatenating the various production task information-equipment information semantic interaction response feature vectors into a concatenated production task information-equipment information semantic interaction response feature vector; determining the concatenated correlation response matrix and the concatenated distance response matrix of the production task information-equipment information semantic interaction response based on the square root of the correlation value of the feature values ​​and the L2 distance value of the concatenated production task information-equipment information semantic interaction response feature vector: wherein, this process can be expressed by the formula:

[0072]

[0073] vi ,v j ∈V∈R 1×L

[0074] Where V is the cascaded feature vector of semantic interaction response between production task information and equipment information, and L represents the number of feature values ​​in V. i v is the feature value at the i-th position in the cascaded feature vector of semantic interaction response between production task information and equipment information. j M1(i,j) is the feature value at the j-th position in the cascaded feature vector of semantic interaction response between production task information and equipment information, M2(i,j) is the feature value at the (i,j)-th position in the cascaded correlation response matrix of semantic interaction response between production task information and equipment information, and M2(i,j) is the feature value at the (i,j)-th position in the cascaded distance response matrix of semantic interaction response between production task information and equipment information.

[0075] The production task information-equipment information semantic interaction response cascaded feature vector is obtained by matrix multiplication with the production task information-equipment information semantic interaction response cascaded association response matrix; the production task information-equipment information semantic interaction response cascaded distance response matrix is ​​obtained by matrix multiplication with the transpose of the production task information-equipment information semantic interaction response cascaded feature vector; the production task information-equipment information semantic interaction response cascaded distance response vector is obtained by dot product of the production task information-equipment information semantic interaction response cascaded association response matrix and the production task information-equipment information semantic interaction response cascaded distance response matrix, and then multiplied with the transpose of the production task information-equipment information semantic interaction response cascaded feature vector. Matrix multiplication is performed to obtain the concatenated bias vector of semantic interaction response between production task information and equipment information; the concatenated associated response vector, the concatenated distance response vector, and the concatenated bias vector are then summed to obtain an optimized concatenated feature vector of semantic interaction response between production task information and equipment information; this optimized feature vector is then split into individual optimized feature vectors of semantic interaction response between production task information and equipment information; finally, each optimized feature vector is passed through the decoder-based task priority allocator to obtain a set of decoded values ​​for the production task allocation speed.

[0076] Here, the optimized production task information-equipment information semantic interaction response cascade feature vector is represented as:

[0077]

[0078] V1∈R 1×L

[0079] V2∈R L×1

[0080]

[0081] Wherein, V is the concatenated feature vector of semantic interaction response between production task information and equipment information, and L represents the number of feature values ​​in V. This represents matrix multiplication, where M1 is the semantic interaction response cascade association matrix of the production task information and equipment information. This indicates addition based on location points, where M2 is the cascaded distance response matrix of the semantic interaction response between the production task information and equipment information, and V T V is the transpose of V, V1 is the dot-matrix vector between the cascaded associated response vector of the semantic interaction response between production task information and equipment information and the cascaded distance response vector of the semantic interaction response between production task information and equipment information, ⊙ represents dot-matrix by position, V2 is the cascaded bias vector of the semantic interaction response between production task information and equipment information, and V' is the optimized cascaded feature vector of the semantic interaction response between production task information and equipment information.

[0082] In other words, by using the cascaded feature vectors of the semantic interaction response between production task information and equipment information, based on the self-correlation matrix and self-distance matrix, as a statistically based, reference-free distribution response framework, and constructing an enhanced reverse response based on the retrieval of these cascaded feature vectors, simple repetition of feature distributions is avoided. Furthermore, by ensuring the intrinsic mapping logic of the cascaded feature vectors based on the retrieval-response context, superficial combinatorial mappings are avoided. This approach maintains intuitive response relevance while achieving a logical dependency mapping from the cascaded feature vectors to the decoding target domain, thereby improving the distinguishability and accuracy of the decoded values ​​for each production task allocation speed. This approach better adapts to complex and ever-changing production and equipment management scenarios, enhances adaptability and flexibility in dynamic environments, and more comprehensively considers multi-dimensional factors such as the dependencies between production tasks and equipment workload prediction, thus providing a more reasonable and flexible solution for task allocation.

[0083] In summary, the production task allocation speed adjustment module 140 is clearly described. It employs an AI-based information analysis and encoding method to semantically encode the information of each production task and establish semantic relationships between contexts. It also semantically encodes the information of the equipment to be allocated. Based on the principal component semantic interaction response features between the semantic relationship features of each production task and the semantic encoding features of the equipment information, it intelligently obtains the production task allocation speed decoding value and adjusts the processing priority of each production task. This better adapts to complex and ever-changing production and equipment management scenarios, improves adaptability and flexibility in dynamic environments, and more comprehensively considers multi-dimensional factors such as the dependencies between production tasks and the working status of equipment, thus providing a more reasonable and flexible solution for task allocation.

[0084] In summary, the intelligent management system 100 for electromechanical engineering equipment based on the embodiments of this application is explained. It first acquires production task information for multiple production tasks and equipment information for electromechanical engineering equipment to be assigned. Next, it initializes the allocation scheme for multiple production tasks on the equipment to be assigned. Finally, it manages the processing priority of the multiple production tasks based on the production task information and the equipment information. This allows for more flexible task allocation and effectively improves the adaptability and flexibility of equipment management.

[0085] As described above, the intelligent management system 100 for electromechanical engineering equipment according to the embodiments of this application can be implemented in various terminal devices, such as servers for intelligent management of electromechanical engineering equipment. In one example, the intelligent management system 100 for electromechanical engineering equipment according to the embodiments of this application can be integrated into a terminal device as a software module and / or a hardware module. For example, the intelligent management system 100 for electromechanical engineering equipment can be a software module in the operating system of the terminal device, or it can be an application developed for the terminal device; of course, the intelligent management system 100 for electromechanical engineering equipment can also be one of many hardware modules of the terminal device.

[0086] Alternatively, in another example, the intelligent management system 100 for electromechanical engineering equipment and the terminal device can also be separate devices, and the intelligent management system 100 for electromechanical engineering equipment can be connected to the terminal device via wired and / or wireless networks, and transmit interactive information in accordance with an agreed data format.

[0087] Figure 7 This is a flowchart of an intelligent management method for electromechanical engineering equipment according to an embodiment of this application. Figure 7As shown, the intelligent management method for electromechanical engineering equipment includes: S110, acquiring production task information of multiple production tasks to obtain a set of production task information; S120, acquiring equipment information of electromechanical engineering equipment to be allocated; S130, initializing the allocation scheme of the multiple production tasks on the electromechanical engineering equipment to be allocated; S140, managing the processing priority of the multiple production tasks based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated; wherein, step S130 includes: semantically encoding and associating the set of production task information to obtain a set of semantic association features of production task information; semantically encoding the equipment information of the electromechanical engineering equipment to be allocated to obtain equipment information semantic encoding features; performing principal component feature semantic interaction response on the set of semantic association features of production task information and the semantic encoding features of equipment information to obtain a set of semantic interaction response features of production task information-equipment information; generating a task processing priority allocation result based on the set of semantic interaction response features of production task information-equipment information, and sorting the processing priority of the multiple production tasks.

[0088] Here, those skilled in the art will understand that the specific operations of each step in the above-described intelligent management method for electromechanical engineering equipment have been referenced above. Figures 1 to 6 The description of the intelligent management system for electromechanical engineering equipment is detailed here, and therefore, its repeated description will be omitted.

[0089] In summary, the intelligent management method for electromechanical engineering equipment based on the embodiments of this application is explained. It first acquires production task information for multiple production tasks and equipment information for the electromechanical engineering equipment to be allocated. Next, it initializes the allocation scheme for the multiple production tasks on the equipment to be allocated. Finally, it manages the processing priority of the multiple production tasks based on the production task information and the equipment information. This allows for more flexible task allocation and effectively improves the adaptability and flexibility of equipment management.

Claims

1. An intelligent management system for electromechanical engineering equipment, characterized in that, include: The production task information acquisition module is used to acquire production task information from multiple production tasks to obtain a collection of production task information. The equipment information acquisition module is used to acquire equipment information of the electromechanical engineering equipment to be allocated; The allocation speed initialization module is used to initialize the allocation scheme of the multiple production tasks on the electromechanical engineering equipment to be allocated; The production task allocation speed adjustment module is used to manage the processing priority of the multiple production tasks based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated. The production task allocation speed adjustment module includes: a production task information encoding unit, used to semantically encode and associate the set of production task information to obtain a set of semantic association features of production task information; a device information encoding unit, used to semantically encode the device information of the electromechanical engineering device to be allocated to obtain device information semantic encoding features; a production task device semantic interaction response unit, used to perform principal component feature semantic interaction response on the set of semantic association features of production task information and the semantic encoding features of device information to obtain a set of production task information-device information semantic interaction response features; and a processing priority management unit, used to generate a task processing priority allocation result based on the set of production task information-device information semantic interaction response features, and to sort the processing priorities of the multiple production tasks. The semantic interaction response unit for the production task equipment includes: The principal component analysis subunit for production task information and equipment information is used to perform principal component analysis on the semantic association feature vector of production task information and the semantic encoding feature vector of equipment information to obtain the set of semantic principal component feature components of production task information and the set of semantic principal component feature components of equipment information. The production task information-equipment information optimal matching subunit is used to perform optimal matching interaction between the set of semantic principal component feature components of the production task information and the set of semantic principal component feature components of the equipment information to obtain the semantic interaction response feature vector of the production task information-equipment information as the set of semantic interaction response features of the production task information-equipment information. The optimal matching subunit for production task information and equipment information includes: using each semantic principal component feature of production task information in the set of semantic principal component feature components of production task information as a query vector, calculating the Mahalanobis distance between it and each semantic principal component feature of equipment information in the set of semantic principal component feature components of equipment information, and taking the pair with the smallest matching distance as the best matching pair {semantic principal component feature of production task information; semantic principal component feature of equipment information}. Multiple best matching pairs are input into a multi-scale interactive response coupling module. Multi-scale interactive processing is performed on each best matching pair to obtain a multi-scale interactive coupling representation vector of the semantic best matching pair of production task information and equipment information. The multi-scale interactive processing includes: calculating the difference vector, dot product vector, and sum vector between two feature components in the best matching pair by subtracting, multiplying, and adding them by position; concatenating the difference vector, dot product vector, and sum vector; and performing one-dimensional convolutional encoding and max pooling on the concatenated vector. The set of multi-scale interactive coupling representation vectors of the semantic best matching of production task information and equipment information is concatenated to obtain the semantic interactive response feature vector of production task information and equipment information.

2. The intelligent management system for electromechanical engineering equipment according to claim 1, characterized in that, The production task information encoding unit includes: The production task information semantic encoding subunit is used to perform semantic encoding on the set of production task information to obtain a set of production task information semantic encoding feature vectors; The production task information semantic association feature extraction subunit is used to obtain a set of production task information semantic association feature vectors by passing the set of production task information semantic encoded feature vectors through a converter-based context encoder, and use this set as the set of production task information semantic association features.

3. The intelligent management system for electromechanical engineering equipment according to claim 2, characterized in that, The equipment information encoding unit to be allocated is used to: perform semantic encoding on the equipment information of the electromechanical engineering equipment to be allocated to obtain a semantic encoding feature vector of equipment information as the semantic encoding feature of equipment information.

4. The intelligent management system for electromechanical engineering equipment according to claim 3, characterized in that, The processing priority management unit includes: The production task allocation speed decoding subunit is used to obtain a set of production task allocation speed decoding values ​​based on the set of semantic interaction response features of the production task information-equipment information. The task processing priority allocation result generation subunit is used to arrange each production task allocation speed decoding value in the set of production task allocation speed decoding values ​​in descending order to obtain the task processing priority allocation result; The production task processing sorting subunit is used to sort the processing priorities of the multiple production tasks based on the task processing priority allocation results.

5. The intelligent management system for electromechanical engineering equipment according to claim 4, characterized in that, The production task allocation speed decoding subunit is used to: pass each production task information-equipment information semantic interaction response feature vector in the set of production task information-equipment information semantic interaction response feature vectors through a decoder-based task priority allocator to obtain a set of production task allocation speed decoding values.

6. A method for intelligent management of equipment used in electromechanical engineering, characterized in that, include: Obtain production task information from multiple production tasks to obtain a set of production task information; Obtain equipment information for electromechanical engineering equipment to be allocated; The allocation scheme for the multiple production tasks on the electromechanical engineering equipment to be allocated is initialized; The processing priority of the multiple production tasks is managed based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated. The process of managing the processing priority of the multiple production tasks based on the set of production task information and the equipment information of the electromechanical engineering equipment to be allocated includes: semantically encoding and associating the set of production task information to obtain a set of semantic association features of production task information; semantically encoding the equipment information of the electromechanical engineering equipment to be allocated to obtain equipment information semantic encoding features; performing principal component feature semantic interaction response on the set of semantic association features of production task information and the semantic encoding features of equipment information to obtain a set of semantic interaction response features of production task information-equipment information; and generating a task processing priority allocation result based on the set of semantic interaction response features of production task information-equipment information, and sorting the processing priority of the multiple production tasks. Specifically, the set of semantic association features of the production task information and the semantic encoding features of the equipment information are subjected to principal component feature semantic interaction response to obtain a set of semantic interaction response features of production task information-equipment information, including: Principal component analysis is performed on the semantic association feature vector of production task information and the semantic encoding feature vector of equipment information to obtain the set of semantic principal component feature components of production task information and the set of semantic principal component feature components of equipment information. The optimal matching interaction between production task information and equipment information is performed on the set of semantic principal component feature components of production task information and the set of semantic principal component feature components of equipment information to obtain a semantic interaction response feature vector of production task information and equipment information as the set of semantic interaction response features of production task information and equipment information. This includes: using each semantic principal component feature component of production task information in the set of semantic principal component feature components of production task information as a query vector, calculating the Mahalanobis distance between it and each semantic principal component feature component feature component of equipment information in the set of semantic principal component feature components of equipment information, and taking the pair with the smallest matching distance as the best match {semantic principal component feature component of production task information; semantic principal component feature component feature component of equipment information}. Pairing; inputting multiple best-matching pairs into a multi-scale interactive response coupling module, performing multi-scale interactive processing on each best-matching pair to obtain a multi-scale interactive coupling representation vector of the production task information-equipment information semantic best-matching pair; the multi-scale interactive processing includes: calculating the difference vector, dot product vector, and sum vector between two feature components in the best-matching pair by positional subtraction, positional multiplication, and positional addition; concatenating the difference vector, dot product vector, and sum vector; and performing one-dimensional convolutional encoding and max pooling on the concatenated vector; concatenating the set of multi-scale interactive coupling representation vectors of the production task information-equipment information semantic best-matching pair to obtain the production task information-equipment information semantic interactive response feature vector.

Images