Discrete device data acquisition and analysis system and method based on the Internet of Things

Abstract

This application relates to the field of intelligent data collection and analysis, and specifically discloses a discrete device data collection and analysis system and method based on the Internet of Things, which uses artificial intelligence technology based on the field of deep learning to extract and encode the operating conditions of logistics equipment and material flow information to generate an optimized material distribution plan. In this way, by intelligently collecting and analyzing the data of the logistics distribution process, the data can be fully utilized to improve the efficiency of material distribution and production scheduling.

Description

Technical Field

[0001] The present application relates to the field of intelligent data collection and analysis, and in particular to a discrete device data collection and analysis system and method based on the Internet of Things. Background Art

[0002] Discrete manufacturing is a production method. Corresponding to "continuous production", it is called "discrete production". Discrete manufacturing is the mainstream production method in today's society and the theme of manufacturing management. From the perspective of the processing process, the production process of discrete manufacturing enterprises is a complex process composed of different parts processing sub-processes in parallel or in series, and the process contains more changes and uncertainties. In this sense, the process control of discrete manufacturing enterprises is more complex and changeable. The product production of enterprises is divided into two production modes: large-scale production oriented to demand and small-scale production oriented to inventory, which are characterized by complex product structure and high output.

[0003] Take automobile manufacturing as an example. During the automobile production process, automobile manufacturers break down the production process into multiple independent processes or stages. Each process or stage can be completed independently, and then these independently completed parts are assembled together to finally form a complete automobile product.

[0004] Each workstation on the automobile production line is responsible for completing a specific task or process, such as body welding, paint spraying, assembly, etc. This production method is more flexible than a continuous production line, and can be adjusted and changed according to demand. It is also easier to adapt to the production of different car models and configurations. In the automobile production process, there are many discrete devices that need to collect and analyze data, such as painting equipment, assembly line equipment, quality inspection equipment, logistics equipment, etc. For logistics equipment, it includes automatic guided vehicles, conveyor lines and other equipment. Traditional monitoring of the operation of logistics equipment is often done manually, and often a simple analysis of the data is performed, and the efficiency of material distribution and production scheduling is low.

[0005] Therefore, an optimized discrete device data acquisition and analysis solution is expected. Summary of the invention

[0006] In order to solve the above technical problems, this application is proposed. The embodiment of this application provides a discrete device data collection and analysis system and method based on the Internet of Things, which uses artificial intelligence technology based on the field of deep learning to extract and encode the operating conditions of logistics equipment and material flow information to generate an optimized material distribution plan. In this way, by intelligently collecting and analyzing the data of the logistics distribution process, the data can be fully utilized to improve the efficiency of material distribution and production scheduling.

[0007] According to one aspect of the present application, a discrete device data acquisition and analysis system based on the Internet of Things is provided, which includes:

[0008] Material distribution data acquisition module, used to obtain the operation status of logistics equipment and material flow information;

[0009] A logistics equipment operation feature extraction module is used to analyze the operation status of the logistics equipment and obtain a graph structure logistics equipment operation feature matrix;

[0010] A material flow feature extraction module, used for processing the material flow information to obtain a graph structure material flow feature matrix;

[0011] The material distribution optimization plan generation module is used to comprehensively analyze the graph structure logistics equipment operation characteristic matrix and the graph structure material flow characteristic matrix to generate an optimized material distribution plan.

[0012] According to another aspect of the present application, a discrete device data collection and analysis method based on the Internet of Things is also provided, which includes:

[0013] Obtain the operation status of logistics equipment and material flow information;

[0014] Analyze the operation status of the logistics equipment to obtain a graph structure logistics equipment operation characteristic matrix;

[0015] Processing the material flow information to obtain a graph structure material flow feature matrix;

[0016] The graph structure logistics equipment operation characteristic matrix and the graph structure material flow characteristic matrix are comprehensively analyzed to generate an optimized material distribution plan.

[0017] In summary, the discrete device data collection and analysis system and method based on the Internet of Things provided by this application uses artificial intelligence technology based on the field of deep learning to extract and encode the operating conditions of logistics equipment and material flow information to generate an optimized material distribution plan. In this way, by intelligently collecting and analyzing the data of the logistics distribution process, the data can be fully utilized to improve the efficiency of material distribution and production scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work. In the drawings, the same reference numerals generally represent the same components or steps.

[0019] Figure 1 It is a block diagram of a discrete device data acquisition and analysis system based on the Internet of Things according to an embodiment of the present application.

[0020] Figure 2 It is a block diagram of a logistics equipment operation feature extraction module in a discrete equipment data acquisition and analysis system based on the Internet of Things according to an embodiment of the present application.

[0021] Figure 3 It is a block diagram of a logistics equipment operation characteristic measurement unit in a discrete equipment data collection and analysis system based on the Internet of Things according to an embodiment of the present application.

[0022] Figure 4 The present invention is a flowchart of a discrete device data collection and analysis method based on the Internet of Things according to an embodiment of the present application. DETAILED DESCRIPTION

[0023] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings, and the technical solutions in the embodiments of the present application will be described clearly and completely. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Figure 1 FIG. 1 is a block diagram of a discrete device data acquisition and analysis system based on the Internet of Things according to an embodiment of the present application. Figure 1 As shown, according to the embodiment of the present application, the discrete device data acquisition and analysis system 100 based on the Internet of Things includes: a material distribution data acquisition module 110, which is used to obtain the operation status and material flow information of the logistics equipment; a logistics equipment operation feature extraction module 120, which is used to analyze the operation status of the logistics equipment to obtain a graph structure logistics equipment operation feature matrix; a material flow feature extraction module 130, which is used to process the material flow information to obtain a graph structure material flow feature matrix; and a material distribution optimization plan generation module 140, which is used to comprehensively analyze the graph structure logistics equipment operation feature matrix and the graph structure material flow feature matrix to generate an optimized material distribution plan.

[0025] In the above-mentioned discrete device data acquisition and analysis system 100 based on the Internet of Things, the material distribution data acquisition module 110 is used to obtain the operation status of logistics equipment and material flow information. As mentioned in the above-mentioned background technology, logistics equipment includes automatic guided vehicles, conveyor lines and other equipment. Traditional monitoring of the operation status of logistics equipment is often done manually, and often a simple analysis of the data is performed, and the efficiency of material distribution and production scheduling is low. Therefore, an optimized discrete device data acquisition and analysis solution is expected.

[0026] In response to the above technical problems, a discrete equipment data collection and analysis system based on the Internet of Things is proposed, which uses artificial intelligence technology based on the field of deep learning to extract and encode the operating conditions of logistics equipment and material flow information to generate an optimized material distribution plan. In this way, by intelligently collecting and analyzing the data of the logistics distribution process, the data can be fully utilized to improve the efficiency of material distribution and production scheduling.

[0027] Specifically, in the technical solution of the present application, by combining the material network and deep learning technology, the material distribution plan is optimized to improve the efficiency of logistics distribution and production scheduling. The operation data and material flow information of logistics equipment (such as AGV (automatic guided vehicle), conveyor line, etc.) are collected in real time through the Internet of Things technology to construct a data set. The operation status of the logistics equipment includes: the operation time, downtime, failure time, operation speed, load condition, and energy consumption data of the equipment. The load condition refers to the weight or quantity of materials currently carried by the logistics equipment. The material flow information includes: the position of the material on the production line, the number of materials at the material position, the material flow direction, and the arrival time of the material. The position of the material on the production line includes the starting point, destination, and current position of the material. Then, based on the deep learning technology, the collected data is feature extracted and encoded, and the operation status of the logistics equipment and the material flow information are converted into feature vectors. Using these feature vectors, the system can apply the optimization algorithm to generate an optimized material distribution plan, taking into account the equipment operation status and material flow information to improve the distribution efficiency. At the same time, the system can monitor the situation in the logistics distribution process in real time, and adjust and optimize according to the real-time data to ensure the smooth progress of the distribution process.

[0028] Specifically, first, obtain the operation status of logistics equipment and material flow information. By connecting logistics equipment to the Internet of Things platform, data sharing and remote monitoring between devices can be realized, and the operation status of equipment can be obtained in real time. At the same time, using radio frequency identification technology (RFID) to mark materials can achieve real-time tracking and management of material flow information. In this way, enterprises can achieve comprehensive monitoring and management of the logistics process, thereby improving production efficiency, reducing costs and improving production processes.

[0029] In the above-mentioned discrete device data acquisition and analysis system 100 based on the Internet of Things, the logistics equipment operation feature extraction module 120 is used to analyze the operation status of the logistics equipment and obtain a graph-structured logistics equipment operation feature matrix.

[0030] Figure 2 FIG. 1 is a block diagram of a logistics equipment operation feature extraction module in a discrete equipment data collection and analysis system based on the Internet of Things according to an embodiment of the present application. Figure 2 As shown, the logistics equipment operation feature extraction module 120 includes: a logistics equipment operation data encoding unit 121, which is used to obtain multiple logistics equipment operation feature vectors by passing the operation status of the logistics equipment through a converter-based logistics equipment semantic feature encoder; a logistics equipment operation feature measurement unit 122, which is used to calculate the distance between every two logistics equipment operation feature vectors in the multiple logistics equipment operation feature vectors to obtain a logistics equipment similarity measurement matrix; a multi-scale logistics equipment operation measurement feature extraction unit 123, which is used to obtain a logistics equipment operation similarity measurement feature matrix by using a logistics equipment operation feature extractor with convolution kernels of different sizes; a matrixing unit 124, which is used to arrange the multiple logistics equipment operation feature vectors in two dimensions into a logistics equipment operation feature matrix; and a logistics equipment operation graph structure feature encoding unit 125, which is used to pass the logistics equipment operation similarity measurement feature matrix and the logistics equipment operation feature matrix through a graph neural network to obtain a graph structure logistics equipment operation feature matrix.

[0031] More specifically, the operating status of the logistics equipment is passed through a logistics equipment semantic feature encoder to obtain multiple logistics equipment operating feature vectors. The logistics equipment semantic feature encoder is a converter-based context encoder model that includes an embedding layer. The use of a converter-based context encoder can integrate the operating status information of multiple logistics equipment, capture the relationship and context information between the equipment, and help to more comprehensively understand the operating status of the entire logistics system. The converter-based context encoder can effectively extract and encode the operating status characteristics of the logistics equipment, and convert the original data into a more representative and interpretable feature vector, which is conducive to subsequent analysis and decision-making. By obtaining multiple logistics equipment operating feature vectors, the correlation and mutual influence between the equipment operating conditions can be better captured, which helps the system to more accurately understand the operating status of the entire logistics system.

[0032] Specifically, in an embodiment of the present application, the logistics equipment semantic feature encoder is a converter-based context encoder model including an embedding layer.

[0033] Specifically, in an embodiment of the present application, the logistics equipment operation data encoding unit 121 includes: an embedding conversion subunit, which is used to pass the operation status of the logistics equipment through an embedding layer to convert each data item in the operation status of the logistics equipment into a logistics embedding vector to obtain a sequence of logistics embedding vectors, wherein the embedding layer uses a learnable embedding matrix to embed the data items; an encoding subunit, which is used to input the sequence of logistics embedding vectors into the converter-based logistics equipment semantic feature encoder to obtain the multiple logistics equipment operation context semantic feature vectors.

[0034] Specifically, in an embodiment of the present application, the logistics equipment operation data semantic encoding subunit includes: a query vector construction secondary subunit, which is used to arrange the sequence of the logistics embedding vectors into an input vector; a vector conversion secondary subunit, which is used to convert the input vector into a query vector and a key vector respectively through a learnable embedding matrix; a self-attention secondary subunit, which is used to calculate the product between the query vector and the transposed vector of the key vector to obtain a self-attention association matrix; a normalization secondary subunit, which is used to normalize the self-attention association matrix to obtain a standardized self-attention association matrix; an attention calculation secondary subunit, which is used to input the standardized self-attention association matrix into a Softmax activation function for activation to obtain a self-attention feature matrix; and an attention application secondary subunit, which is used to multiply the self-attention feature matrix with each logistics embedding vector in the sequence of the logistics embedding vector to obtain the multiple logistics equipment operation context semantic feature vectors.

[0035] More specifically, the distance between each two logistics equipment operation feature vectors in multiple logistics equipment operation feature vectors is calculated to obtain a logistics equipment similarity measurement matrix. By calculating the distance between logistics equipment operation feature vectors, the similarity or difference between different equipment can be quantified, which helps to identify the correlation between equipment and better understand the relationship between equipment. The distance between logistics equipment operation feature vectors can reflect the similarity of equipment status. By comparing these distances, the difference in operating status between equipment can be found, which helps to identify abnormal conditions of equipment or optimize operating strategies. Based on the logistics equipment similarity measurement matrix, optimization and scheduling between equipment can be carried out, and equipment with similar operating status can be found to reasonably allocate tasks, thereby improving equipment utilization and production efficiency.

[0036] Figure 3 FIG. 1 is a block diagram of a logistics equipment operation characteristic measurement unit in a discrete equipment data collection and analysis system based on the Internet of Things according to an embodiment of the present application. Figure 3As shown, the logistics equipment operation characteristic measurement unit 122 includes: a similarity calculation subunit 11, which is used to calculate the distance between every two logistics equipment operation characteristic vectors in the multiple logistics equipment operation characteristic vectors using the following similarity calculation formula to obtain multiple logistics equipment operation similarities; wherein the similarity calculation formula is:

[0037]

[0038] Among them, V i and V j respectively represent every two logistics equipment operation feature vectors in the plurality of logistics equipment operation feature vectors, and Respectively represent the characteristic values ​​of each position of each two logistics equipment operation characteristic vectors in the multiple logistics equipment operation characteristic vectors, d(V i ,V j ) represents the logistics equipment operation similarity between every two logistics equipment operation feature vectors in the multiple logistics equipment operation feature vectors; and a two-dimensional arrangement subunit 12, which is used to two-dimensionally arrange the multiple logistics equipment operation similarities to obtain the logistics equipment similarity measurement matrix.

[0039] More specifically, the logistics equipment similarity measurement matrix is ​​extracted by a logistics equipment operation feature extractor using convolution kernels of different sizes to obtain a logistics equipment operation similarity measurement feature matrix. The logistics equipment operation feature extractor is a convolutional neural network model using convolution kernels of different sizes. Convolutional neural network (CNN) can effectively extract features from the logistics equipment similarity measurement matrix, and realize multi-level feature abstraction through convolution kernels of different sizes, which helps to capture the complex relationships and patterns in the logistics equipment similarity measurement matrix. Convolution kernels of different sizes can provide receptive fields of different sizes, so as to better capture the characteristics of data in local areas, which helps to identify local patterns and structures in the logistics equipment similarity measurement matrix. The CNN model reduces the number of model parameters by sharing parameters, improves the generalization ability of the model, and can effectively process high-dimensional data, which is suitable for processing complex data such as the logistics equipment similarity measurement matrix. Convolution kernels of different sizes can capture the operation feature information of logistics equipment of different scales. By fusing these logistics equipment operation features, the similarity relationship between logistics equipment can be more comprehensively described, and the representation ability of the similarity measurement feature matrix can be improved.

[0040] Specifically, in an embodiment of the present application, the logistics equipment operation feature extractor is a convolutional neural network model using convolution kernels of different sizes.

[0041] Specifically, in an embodiment of the present application, the multi-scale logistics equipment operation metric feature extraction unit 123 is used to: use each layer of the logistics equipment operation feature extractor to respectively perform the following on the input data in the forward pass of the layer: use a first convolution kernel to convolve the input data to obtain a first convolution feature map; use a second convolution kernel to convolve the first convolution feature map to obtain a second convolution feature map, wherein the size of the first convolution kernel is larger than the size of the second convolution kernel; perform pooling on the second convolution feature map to obtain a pooled feature map; and perform activation on the pooled feature map to obtain an activation feature map; wherein the output of the last layer of the logistics equipment operation feature extractor is the logistics equipment operation similarity metric feature matrix, and the input of the first layer of the logistics equipment operation feature extractor is the logistics equipment similarity metric matrix.

[0042] Specifically, in an embodiment of the present application, the size of the first convolution kernel is 5×5, and the size of the second convolution kernel is 3×3.

[0043] More specifically, multiple logistics equipment operation feature vectors are arranged in two dimensions into a logistics equipment operation feature matrix. Arranging the logistics equipment operation feature vectors into a matrix form can better structure the data, so that the data presents a clear row-column relationship, which is convenient for subsequent data processing and analysis. The logistics equipment operation feature matrix can better show the relationship between different features, help extract the patterns and laws between features, and provide a basis for subsequent feature analysis and modeling.

[0044] Furthermore, the logistics equipment operation similarity measurement feature matrix and the logistics equipment operation feature matrix are combined through the graph neural network (GNN) to obtain the graph structure logistics equipment operation feature matrix. The similarity relationship between logistics equipment can be expressed as a graph structure, in which the logistics equipment is the node and the similarity measurement feature is the edge between the nodes. By constructing a graph structure, the complex similarity relationship between logistics equipment can be better captured. At the same time, the graph neural network can consider the association between the node and its neighboring nodes, which helps to share information and features in the entire logistics equipment network, so as to better capture the complex relationship and pattern between logistics equipment. In this way, the similarity measurement features and logistics equipment features are combined together, and the fusion and integration of features can be achieved through the processing of the graph neural network, so as to obtain a more representative and rich graph structure logistics equipment operation feature matrix. The graph neural network can dynamically learn the relationship and feature representation between nodes, adapt to the changes and evolution of the similarity relationship between different logistics equipment, and improve the generalization ability and adaptability of the model.

[0045] Specifically, in an embodiment of the present application, the graph neural network performs graph structure data encoding on the logistics equipment operation similarity measurement feature matrix and the logistics equipment operation feature matrix through learnable weight parameters to obtain the graph structure logistics equipment operation feature matrix containing global topological space similarity features and logistics equipment operation association features.

[0046] In the above-mentioned discrete device data acquisition and analysis system 100 based on the Internet of Things, the material flow feature extraction module 130 is used to process the material flow information and obtain a graph structure material flow feature matrix. In the technical solution of the present application, the process of processing the material flow information and obtaining the graph structure material flow feature matrix is ​​similar to processing the operation status of logistics equipment, and the method can refer to the process of processing the operation status of logistics equipment. Both the material flow information and the operation status of logistics equipment involve the association and interaction between data, so similar methods can be used to process these data, such as constructing a graph structure, extracting features, etc. Similar to the relationship between logistics equipment, material flow can also be regarded as a graph structure, in which the material flow path can be represented as an edge in the graph, and the relationship between material flow nodes can be represented as nodes in the graph. In this way, the graph neural network can effectively transfer and aggregate information on the graph structure, and by considering the relationship and interaction between material flow nodes, the pattern and characteristics of material flow can be better captured.

[0047] In the above-mentioned discrete device data acquisition and analysis system 100 based on the Internet of Things, the material distribution optimization plan generation module 140 is used to comprehensively analyze the graph structure logistics equipment operation characteristic matrix and the graph structure material flow characteristic matrix to generate an optimized material distribution plan.

[0048] Specifically, in the embodiment of the present application, the material distribution optimization plan generation module 140 includes: a material distribution scheduling feature fusion unit, which is used to fuse the graph structure logistics equipment operation feature matrix and the graph structure material flow feature matrix to obtain a material distribution scheduling feature matrix; a compensation optimization unit, which is used to perform quasi-coherent compensation on the material distribution scheduling feature matrix based on feature implicit expression to obtain a compensated material distribution scheduling feature matrix; and a material distribution optimization plan generation unit, which is used to pass the compensated material distribution scheduling feature matrix through a generator to generate an optimized material distribution plan.

[0049] More specifically, the graph structure logistics equipment operation feature matrix and the graph structure material flow feature matrix are integrated to obtain the material distribution scheduling feature matrix. The integration of logistics equipment operation characteristics and material flow characteristics can comprehensively consider the operation of equipment and material flow in the logistics system, so as to better understand the operation status of the entire logistics system. The integration of different types of feature matrices can make the obtained material distribution scheduling feature matrix richer and more comprehensive, covering different aspects of information in the logistics system, which helps to improve the accuracy of scheduling decisions. At the same time, by integrating logistics equipment operation characteristics and material flow characteristics, various constraints in the logistics system, such as equipment operation status, material flow path, etc., can be comprehensively considered to provide a more reasonable feature representation for material distribution scheduling.

[0050] Specifically, in the embodiment of the present application, the material distribution scheduling feature fusion unit is used to: fuse the graph structure logistics equipment operation feature matrix and the graph structure material flow feature matrix with the following cascade formula to obtain the material distribution scheduling feature matrix; wherein the cascade formula is:

[0051] M=Concat[M1,M2]

[0052] Among them, M1 represents the graph structure logistics equipment operation characteristic matrix, M2 represents the graph structure material flow characteristic matrix, M represents the material distribution scheduling characteristic matrix, and Concat[·,·] represents the cascade function.

[0053] In particular, in the technical solution of the present application, considering the use of convolution kernels of different sizes for feature extraction, multiple levels of features may be extracted from the original data, and these features may be redundant, that is, the same information is repeatedly represented at different levels. When processing graph structure data, graph neural networks consider the connection relationship between nodes, which may introduce additional features, which may increase the dimension of the material distribution scheduling feature matrix finally obtained. The acquisition and processing of material flow information may introduce features that are related to but not completely the same as the operation of logistics equipment, and these features may cause redundancy when fused. When fusing the logistics equipment operation feature matrix and the material flow feature matrix, since the two may contain similar or repeated information, this may lead to redundancy and overlap in the material distribution scheduling feature matrix. Different operating states of logistics equipment and different modes of material flow may be similar in nature, such as similar failure modes or similar material flow paths, and these similarities may lead to overlaps between categories when representing features. The complexity of the logistics system itself may also lead to high dimensionality and redundancy of the feature matrix. For example, logistics equipment may operate under different environmental conditions, or material flow may be affected by multiple factors, which may be considered repeatedly when representing features. Based on this, in the technical solution of the present application, the material distribution scheduling feature matrix is ​​subjected to quasi-coherent compensation based on implicit feature expression, so as to eliminate high-dimensional and redundant features in the material distribution scheduling feature matrix.

[0054] Specifically, in an embodiment of the present application, the compensation optimization unit includes: a feature implicit expression generation subunit, which is used to use the principal component analysis method to process the material distribution scheduling feature matrix to map the material distribution scheduling feature matrix to a low-dimensional feature space to obtain a generated feature implicit expression; a class coherence calculation subunit, which is used to calculate the class coherence matrix of the generated feature implicit expression; and a class correlation compensation subunit, which is used to perform class correlation compensation on the generated feature implicit expression based on the class coherence matrix to obtain a compensated material distribution scheduling feature matrix.

[0055] Specifically, in the embodiment of the present application, the class coherence calculation subunit is used to calculate the class coherence matrix of the implicit expression of the generated feature using the following class coherence calculation formula; wherein the class coherence calculation formula is:

[0056]

[0057] Wherein, V represents the implicit expression of the generated feature, u represents the mean of the implicit expression of the generated feature, T represents the transpose of the vector, L represents the scale of the implicit expression of the generated feature, and C represents the class coherence matrix

[0058] Here, for the material distribution scheduling feature matrix, firstly, by projecting the data into a low-dimensional feature space, the computational efficiency of the model is improved, the algorithm complexity is reduced, and the calculation speed is accelerated. Then, with the goal of enhancing robustness, the implicit expression of the features is used to capture the intrinsic structure of the data and enhance the anti-interference ability of the model. Next, the overlap between samples of different categories is reduced through class coherence compensation, thereby improving the accuracy of the generation algorithm. Furthermore, in the feature selection stage, class coherence compensation helps identify the most relevant features, thereby optimizing the generation effect. At the same time, the implicit expression of features also provides insights for enhancing the interpretability of the generation model, making it easier for users to understand the data structure.

[0059] Specifically, in an embodiment of the present application, the class-related compensation subunit includes: a project creation secondary subunit, which is used to create a Spring Web project; a class-related compensation algorithm writing secondary subunit, which is used to write a class-related compensation algorithm in the Spring Web project, wherein the class-related compensation algorithm is to multiply the implicit expression of the generated feature by the class coherence matrix; a Spring Controller class creation secondary subunit, which is used to create a SpringController class to process HTTP requests and responses; an input encoding secondary subunit, which is used to write a method in the SpringController class to accept the implicit expression of the generated feature and the class coherence matrix as input; and a routing configuration secondary subunit, which is used to configure routing to map HTTP requests to the Spring Controller class.

[0060] More specifically, using the Spring Web mechanism to deploy and implement the class-related compensation subunits can achieve: 1. Easy development and maintenance: The Spring framework provides a large number of functions and components, making the development of Web applications easier and more efficient. Spring Boot is particularly suitable for quickly building Web applications, reducing the workload of developers. 2. RESTful API support: The Spring framework provides good support for RESTful APIs, and it is easy to create RESTful-style APIs to process data requests and responses. This is very useful for processing feature data and returning processing results. 3. Dependency injection: The dependency injection function of the Spring framework can help manage component dependencies in the application, making the code more modular and easy to maintain. 4. Integration testing: The Spring framework provides good support for integration testing, and you can easily write and run test cases to verify the functionality of the application. 5. Security: Spring Security is an important module of the Spring framework, which can help you implement authentication, authorization and other security functions to protect your Web applications from malicious attacks. 6. Extensibility: The Spring framework is a very flexible and extensible framework that can easily integrate other libraries and tools to meet different needs. In general, using the Spring Web mechanism makes it easier for you to develop, deploy, and maintain Web applications, while providing rich features and tools to support you in implementing feature compensation functions and providing a robust infrastructure to build your applications.

[0061] More specifically, the code portion of the class-dependent compensation subunit is as follows:

[0062]

[0063]

[0064] Then, the compensated material distribution scheduling feature matrix is ​​used by the generator to generate an optimized material distribution plan. The optimized material distribution plan refers to the best material distribution path, time arrangement and resource utilization plan generated by an intelligent algorithm based on multiple factors (such as material demand, equipment status, transportation cost, time limit, etc.) to achieve the optimization of the logistics distribution process. The material distribution plan can be automatically generated by the generator without human intervention, which improves efficiency and reduces the possibility of human error. At the same time, the generator can generate a variety of material distribution plans, taking into account different constraints and optimization goals, thereby providing decision makers with more choices and helping to find better solutions. The material distribution plan generated by the generator can help make more optimized decisions based on the information of the material distribution scheduling feature matrix and the optimization algorithm, thereby improving the efficiency and performance of the logistics system.

[0065] Specifically, in one embodiment of the present application, the model of the generator can be a generative adversarial network model or a variational autoencoder model, which trains the compensated material distribution scheduling feature matrix to map the compensated material distribution scheduling feature matrix to a representation of the optimized material distribution plan.

[0066] In summary, the discrete device data collection and analysis system based on the Internet of Things according to the embodiment of the present application has been explained, which uses artificial intelligence technology based on the field of deep learning to extract and encode the operating conditions of logistics equipment and material flow information to generate an optimized material distribution plan. In this way, by intelligently collecting and analyzing the data of the logistics distribution process, the data can be fully utilized to improve the efficiency of material distribution and production scheduling.

[0067] As described above, the discrete device data acquisition and analysis system 100 based on the Internet of Things according to the embodiment of the present application can be implemented in various terminal devices, such as a discrete device data acquisition and analysis server based on the Internet of Things. In one example, the discrete device data acquisition and analysis system 100 based on the Internet of Things according to the embodiment of the present application can be integrated into the terminal device as a software module and / or a hardware module. For example, the discrete device data acquisition and analysis system 100 based on the Internet of Things can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the discrete device data acquisition and analysis system 100 based on the Internet of Things can also be one of the many hardware modules of the terminal device.

[0068] Alternatively, in another example, the discrete device data acquisition and analysis system 100 based on the Internet of Things and the terminal device may also be separate devices, and the discrete device data acquisition and analysis system 100 based on the Internet of Things may be connected to the terminal device via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0069] Based on the same inventive concept, an embodiment of the present application also provides a discrete device data collection and analysis method based on the Internet of Things, which can be used to implement the system described in the above embodiment, as described in the following embodiment.

[0070] Figure 4 FIG. 1 is a flow chart of a discrete device data collection and analysis method based on the Internet of Things according to an embodiment of the present application. Figure 4 As shown, according to the discrete device data collection and analysis method based on the Internet of Things embodiment of the present application, the steps include: S110, obtaining the operating status and material flow information of the logistics equipment; S120, analyzing the operating status of the logistics equipment to obtain a graph structure logistics equipment operating characteristic matrix; S130, processing the material flow information to obtain a graph structure material flow characteristic matrix; and S140, comprehensively analyzing the graph structure logistics equipment operating characteristic matrix and the graph structure material flow characteristic matrix to generate an optimized material distribution plan.

[0071] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. Here, for the discrete device data collection and analysis method based on the Internet of Things disclosed in the embodiment, those skilled in the art can understand that the specific operations of each step in the above discrete device data collection and analysis method based on the Internet of Things have been referred to above. Figures 1 to 3 It has been introduced in detail in the description of the discrete device data acquisition and analysis system based on the Internet of Things, so the description is relatively simple. For related matters, please refer to the description of the discrete device data acquisition and analysis system based on the Internet of Things, and therefore, its repeated description will be omitted.

[0072] In summary, the data collection and analysis method of discrete devices based on the Internet of Things according to the embodiment of the present application has been explained, which uses artificial intelligence technology based on the field of deep learning to extract and encode the operating conditions of logistics equipment and material flow information to generate an optimized material distribution plan. In this way, by intelligently collecting and analyzing the data of the logistics distribution process, the data can be fully utilized to improve the efficiency of material distribution and production scheduling.

[0073] In summary, after reading this detailed disclosure, it will be apparent to those skilled in the art that the aforementioned detailed disclosure may be presented only by way of example and may not be restrictive. Although not explicitly stated herein, it will be appreciated by those skilled in the art that this application is intended to encompass various reasonable changes, improvements and modifications to the embodiments. These changes, improvements and modifications are intended to be proposed by this application and are within the spirit and scope of the exemplary embodiments of this application.

[0074] In addition, certain terms in this application have been used to describe embodiments of the present application. For example, "one embodiment", "embodiment" and / or "some embodiments" mean that a particular feature, structure or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present application. Therefore, it can be emphasized and should be understood that two or more references to "embodiment" or "one embodiment" or "alternative embodiment" in various parts of this specification do not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics may be appropriately combined in one or more embodiments of the present application.

[0075] Moreover, the terms "comprises", "comprising" or any other variations thereof are intended to cover non-exclusive inclusion, so that an article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such article or device. In the absence of more restrictions, the elements defined by the phrase "comprising a ..." do not exclude the presence of additional identical elements in the article or device comprising the above elements.

[0076] It should be understood that in the foregoing description of the embodiments of the present application, in order to help understand a feature and for the purpose of simplifying the present application, the present application combines various features in a single embodiment, drawing or its description. However, this does not mean that the combination of these features is necessary. When reading the present application, it is entirely possible for a person skilled in the art to extract some of the features and understand them as separate embodiments. In other words, the embodiments in the present application can also be understood as the integration of multiple secondary embodiments. This is also true when the content of each secondary embodiment is less than all the features of a single aforementioned disclosed embodiment.

[0077] Finally, it should be understood that the embodiments of the application disclosed herein are explanations of the principles of the embodiments of the present application. Other modified embodiments are also within the scope of the present application. Therefore, the embodiments disclosed in the present application are only used as examples and are not limited. Those skilled in the art can adopt alternative configurations to realize the application in the present application according to the embodiments in the present application.

[0078] Therefore, the embodiments of the present application are not limited to the embodiments precisely described in the application.

Claims

1. A discrete device data acquisition and analysis system based on the Internet of Things, characterized in that: include: Material distribution data acquisition module, used to obtain the operation status of logistics equipment and material flow information; A logistics equipment operation feature extraction module is used to analyze the operation status of the logistics equipment and obtain a graph structure logistics equipment operation feature matrix; A material flow feature extraction module, used for processing the material flow information to obtain a graph structure material flow feature matrix; A material distribution optimization plan generation module is used to comprehensively analyze the graph structure logistics equipment operation characteristic matrix and the graph structure material flow characteristic matrix to generate an optimized material distribution plan; Wherein, the logistics equipment operation feature extraction module includes: A logistics equipment operation data encoding unit, used for obtaining a plurality of logistics equipment operation feature vectors by passing the operation status of the logistics equipment through a logistics equipment semantic feature encoder based on a converter; a logistics equipment operation feature measurement unit, configured to calculate the distance between every two logistics equipment operation feature vectors in the plurality of logistics equipment operation feature vectors to obtain a logistics equipment similarity measurement matrix; A multi-scale logistics equipment operation metric feature extraction unit, used for obtaining a logistics equipment operation similarity metric feature matrix by using a logistics equipment operation feature extractor with convolution kernels of different sizes; A matrixing unit, used for arranging the plurality of logistics equipment operation characteristic vectors in two dimensions into a logistics equipment operation characteristic matrix; A logistics equipment operation graph structure feature encoding unit, used for obtaining a graph structure logistics equipment operation feature matrix by passing the logistics equipment operation similarity measurement feature matrix and the logistics equipment operation feature matrix through a graph neural network; Wherein, the material distribution optimization solution generation module includes: A material distribution scheduling feature fusion unit, used to fuse the graph structure logistics equipment operation feature matrix and the graph structure material flow feature matrix to obtain a material distribution scheduling feature matrix; A compensation optimization unit, used for performing quasi-coherent compensation on the material distribution scheduling feature matrix based on feature implicit expression to obtain a compensated material distribution scheduling feature matrix; A material distribution optimization plan generating unit, used for passing the compensated material distribution scheduling characteristic matrix through a generator to generate an optimized material distribution plan; Wherein, the compensation optimization unit comprises: A feature implicit expression generating subunit, used for processing the material distribution scheduling feature matrix using a principal component analysis method to map the material distribution scheduling feature matrix into a low-dimensional feature space to obtain a generated feature implicit expression; A class coherence calculation subunit, used for calculating a class coherence matrix of the implicit expression of the generated features; A class-correlation compensation subunit, used for performing class-correlation compensation on the generated feature implicit expression based on the class-coherence matrix to obtain a compensated material distribution scheduling feature matrix; Wherein, the quasi-coherent computing subunit is used for: The class coherence matrix of the implicit expression of the generated features is calculated using the following class coherence calculation formula; Wherein, the class coherence calculation formula is: Among them, V represents the implicit expression of the generated features, u represents the mean of the implicit expression of the generated features, T represents the transpose of the vector, L represents the scale of the implicit expression of the generated features, and C represents the class coherence matrix.

2. The discrete device data acquisition and analysis system based on the Internet of Things according to claim 1 is characterized in that: The logistics equipment semantic feature encoder is a converter-based context encoder model including an embedding layer.

3. The discrete device data acquisition and analysis system based on the Internet of Things according to claim 2 is characterized in that: The logistics equipment operation characteristic measurement unit includes: A similarity calculation subunit, used to calculate the distance between every two logistics equipment operation feature vectors in the multiple logistics equipment operation feature vectors using the following similarity calculation formula to obtain multiple logistics equipment operation similarities; Wherein, the similarity calculation formula is: Among them, V i and V j respectively represent every two logistics equipment operation feature vectors in the plurality of logistics equipment operation feature vectors, and Respectively represent the characteristic values ​​of each position of each two logistics equipment operation characteristic vectors in the multiple logistics equipment operation characteristic vectors, d(V i ,V j ) represents the logistics equipment operation similarity between every two logistics equipment operation feature vectors in the multiple logistics equipment operation feature vectors; The two-dimensional arrangement subunit is used to perform two-dimensional arrangement on the operation similarities of the plurality of logistics equipment to obtain the logistics equipment similarity measurement matrix.

4. The discrete device data acquisition and analysis system based on the Internet of Things according to claim 3 is characterized in that: The logistics equipment operation feature extractor is a convolutional neural network model using convolution kernels of different sizes.

5. The discrete device data acquisition and analysis system based on the Internet of Things according to claim 4 is characterized in that: The class-related compensation subunit comprises: The project creates a secondary subunit, which is used to create a Spring Web project; A class-related compensation algorithm writing secondary subunit is used to write a class-related compensation algorithm in the Spring Web project, wherein the class-related compensation algorithm is to multiply the generated feature implicit expression by the class-related matrix; The Spring Controller class creates a secondary subunit, which is used to create a Spring Controller class to handle HTTP requests and responses; An input encoding secondary subunit is used to write a method in the Spring Controller class to accept the generated feature implicit expression and the class coherence matrix as input; The routing configuration secondary subunit is used to configure routing to map HTTP requests to the Spring Controller class.

6. A method for collecting and analyzing discrete device data based on the Internet of Things, using the discrete device data collection and analysis system based on the Internet of Things according to claim 1, characterized in that: include: Obtain the operation status of logistics equipment and material flow information; Analyze the operation status of the logistics equipment to obtain a graph structure logistics equipment operation characteristic matrix; Processing the material flow information to obtain a graph structure material flow feature matrix; The graph structure logistics equipment operation characteristic matrix and the graph structure material flow characteristic matrix are comprehensively analyzed to generate an optimized material distribution plan.

Images