Raw material supply and demand flow monitoring system and method based on multi-source data fusion

Abstract

The present application relates to the field of data processing, disclose a raw material supply and demand flow direction monitoring system and method based on multi-source data fusion, the system includes information collection module, flow direction prediction module, monitoring module, flow direction analysis module and supply and demand flow direction generation module, obtain the traffic logistics data of raw materials and carry out noise point cleaning, obtain standardized data, predict the supply and demand flow direction of raw materials according to standardized data, obtain supply and demand flow direction prediction graph; the transport flow direction of raw materials is monitored, and flow direction data is obtained; according to the flow direction data, the supply and demand flow direction of raw materials is analyzed, and the supply and demand flow direction actual graph is obtained; the supply and demand flow direction prediction graph is corrected, and the supply and demand flow direction graph is obtained, the supply and demand flow direction real-time monitoring model of raw materials is constructed, the supply and demand flow direction of multi-source data fusion raw materials is monitored in real time according to the supply and demand flow direction real-time monitoring model, and the supply and demand flow direction is obtained.The present application improves the monitoring ability of raw materials and the ability to process fragmented data in monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of data processing, in particular to a raw material supply and demand flow direction monitoring system and method based on multi-source data fusion. BACKGROUND

[0002] In the field of raw material supply and demand flow direction monitoring today, the existing monitoring system exposes many problems, which seriously restricts the monitoring effect and the efficient development of related business. The data sources of the existing monitoring system are extremely extensive, but the data formats of different sources are greatly different. Production enterprises usually record the yield and delivery information of raw materials in the form of Excel tables, and transportation companies store transportation tracks and cargo loading and unloading data in self-defined database formats. This incompatibility in format makes data encounter many difficulties in the integration process, greatly aggravates the degree of data fragmentation, and makes it difficult to build a unified and coherent data set for subsequent analysis work.

[0003] The existing monitoring system has data update delay at many supply and demand flow direction nodes, and cannot timely reflect the dynamic changes of the raw material supply and demand flow direction. This leads to difficulties in responding to traffic congestion, vehicle breakdown and other sudden conditions during transportation, and in the aspect of inventory monitoring, the inventory data of some warehouses is updated only once or twice a day, which cannot reflect the changes of inventory in and out of the warehouse in real time. When the inventory fluctuates abnormally, the enterprise cannot timely detect and take corresponding measures, which seriously affects the supply guarantee capability of raw materials. These problems make the information collected by the raw material supply and demand flow direction monitoring system present a fragmented state, and further lead to weak monitoring capability of raw materials. SUMMARY

[0004] The present application provides a raw material supply and demand flow direction monitoring system and method based on multi-source data fusion, which mainly aims to solve the problems of fragmented information collection and weak raw material monitoring capability.

[0005] To achieve the above purpose, the raw material supply and demand flow direction monitoring system based on multi-source data fusion provided by the present application is characterized in that the system comprises an information collection module, a flow direction prediction module, a monitoring module, a flow direction analysis module and a supply and demand flow direction generation module, wherein:

[0006] The information collection module is used to obtain traffic logistics data of raw materials;

[0007] The flow direction prediction module is used to clean up noise points of the traffic logistics data to obtain standardized data of the traffic logistics data, and predict the supply and demand flow direction of the raw materials according to the standardized data to obtain a supply and demand flow direction prediction graph of the raw materials;

[0008] The monitoring module is configured to monitor the transportation flow direction of the raw materials to obtain flow direction data of the raw materials.

[0009] The flow direction analysis module is configured to analyze the supply-demand flow direction of the raw materials according to the flow direction data to obtain a supply-demand flow direction actual graph of the raw materials.

[0010] The supply-demand flow direction generation module is configured to correct the supply-demand flow direction prediction graph based on the supply-demand flow direction actual graph to obtain a supply-demand flow direction graph of the raw materials, construct a supply-demand flow direction real-time monitoring model of the raw materials according to the supply-demand flow direction graph and the flow direction data, and monitor the multi-source data fused supply-demand flow direction of the raw materials in real time according to the supply-demand flow direction real-time monitoring model to obtain the supply-demand flow direction of the raw materials.

[0011] In a preferred embodiment, the information collection module, when performing the acquisition of the traffic logistics data of the raw materials, is specifically configured to:

[0012] collect the traffic logistics data of the raw materials through an industry data platform, wherein the traffic logistics data includes manufacturing industry chain flow direction data, supply chain traffic logistics data, flow direction data of the raw materials, and manufacturing product traffic logistics data.

[0013] In a preferred embodiment, the flow direction prediction module, when performing the prediction of the supply-demand flow direction of the raw materials according to the standardized data to obtain a supply-demand flow direction prediction graph of the raw materials, is specifically configured to:

[0014] input the standardized data and an input weight matrix into a sigmoid activation function perform calculation to obtain current time new information of the standardized data;

[0015] input the standardized data and a forgetting weight matrix into a sigmoid activation function perform calculation to obtain last time retained information of the standardized data;

[0016] input the standardized data and an output weight matrix into a sigmoid activation function perform calculation to obtain an information output degree of the standardized data;

[0017] perform prediction based on the current time new information, the last time retained information, the information output degree, and a hidden state calculation formula to obtain the supply-demand flow direction prediction graph of the raw materials, wherein the hidden state calculation formula is:

[0018]

[0019] In the formula, a prediction graph of the raw material, an information output degree, retained information of the last time, newly added information of the current time, a time factor, a calculation symbol of element-by-element multiplication.

[0020] In a preferred embodiment, the monitoring module, when performing monitoring on the transportation flow direction of the raw material, obtains flow direction data of the raw material, and is specifically configured to:

[0021] filtering according to importance of the traffic logistics data, to obtain important nodes of the traffic logistics data, and taking the important nodes as supply-demand flow direction nodes of the raw material;

[0022] monitoring the transportation flow direction of the raw material at the supply-demand flow direction nodes, to obtain multi-source data of the supply-demand flow direction nodes;

[0023] standardizing different formats of the multi-source data, to obtain standardized data of the multi-source data;

[0024] classifying according to data importance of the standardized data, to obtain classified data of the standardized data, wherein the classified data includes ordinary data and core data;

[0025] performing consensus benefit calculation based on the classified data and a warrant dynamic consensus algorithm, to obtain flow direction data of the raw material.

[0026] In a preferred embodiment, the monitoring module, when performing monitoring on the transportation flow direction of the raw material, obtains flow direction data of the raw material, and is specifically configured to:

[0027] performing consensus benefit calculation based on the ordinary data and the warrant dynamic consensus algorithm, layer calculation, to obtain transportation information of the raw material, wherein, the layer calculation formula is:

[0028]

[0029] in the formula, the transportation information, the ordinary data, time information of supply-demand flow direction node data, reliability of supply-demand flow direction node data, the supply-demand flow direction node.

[0030] performing consensus benefit calculation based on the core data and the warrant dynamic consensus algorithm, layer calculation, to obtain cooperation information of the raw materials, wherein, The layer calculation formula is:

[0031]

[0032] In the formula, The cooperation information, The preparation stage, The supply-demand flow direction node, The sequence number of the monitoring data, The core data, The ordered tuple is a symbolic expression;

[0033] The transport information and the cooperation information are integrated to obtain flow direction data of the raw materials.

[0034] In a preferred embodiment, when the flow direction analysis module performs supply-demand flow direction analysis of the raw materials according to the flow direction data to obtain a supply-demand flow direction actual graph of the raw materials, it is specifically used for:

[0035] The flow direction data is subjected to noise cleaning to obtain standard data of the flow direction data;

[0036] The flow direction supply-demand information of the standard data is extracted;

[0037] The flow direction supply-demand information is subjected to standardized coding to obtain supply-demand flow direction data of the raw materials;

[0038] The supply-demand flow direction actual graph of the raw materials is constructed based on the supply-demand flow direction data.

[0039] In a preferred embodiment, when the supply-demand flow direction generation module performs feasibility correction of the supply-demand flow direction prediction graph based on the supply-demand flow direction actual graph to obtain a supply-demand flow direction graph of the raw materials, it is specifically used for:

[0040] The flow direction information of the supply-demand flow direction prediction graph is compared and identified based on the supply-demand flow direction actual graph to obtain a difference identification point of the supply-demand flow direction actual graph to the supply-demand flow direction prediction graph;

[0041] The supply-demand flow direction actual graph is subjected to flow direction path correction based on the difference identification point to obtain a supply-demand flow direction graph of the raw materials.

[0042] In a preferred embodiment, when the supply-demand flow direction generation module performs construction of a supply-demand flow direction real-time monitoring model of the raw materials according to the supply-demand flow direction graph and the flow direction data, and performs real-time monitoring of the supply-demand flow direction of the raw materials based on multi-source data fusion according to the supply-demand flow direction real-time monitoring model to obtain a supply-demand flow direction of the raw materials, it is specifically used for:

[0043] identify the flow direction features of the raw material flow direction data to obtain key features of the raw material flow direction data;

[0044] construct a supply-demand flow direction real-time monitoring model of the raw material based on the key features and the supply-demand flow direction graph.

[0045] In a preferred embodiment, the supply-demand flow direction generation module, when performing real-time monitoring on the multi-source data fused raw material supply-demand flow direction according to the supply-demand flow direction real-time monitoring model to obtain the supply-demand flow direction of the raw material, is specifically used for:

[0046] construct a flow direction network of the raw material based on the flow direction data;

[0047] analyze the node properties and flow direction information in the flow direction network to obtain the supply-demand relationship of the raw material;

[0048] perform flow direction analysis based on the supply-demand relationship and the supply-demand flow direction real-time monitoring model to obtain the supply-demand flow direction of the raw material.

[0049] In order to solve the above problems, the application also provides a raw material supply-demand flow direction monitoring method based on multi-source data fusion, characterized in that the method comprises:

[0050] S1, obtaining traffic logistics data of raw materials;

[0051] S2, performing noise point cleaning on the traffic logistics data to obtain standardized data of the traffic logistics data, and predicting the supply-demand flow direction of the raw materials according to the standardized data to obtain a supply-demand flow direction prediction graph of the raw materials;

[0052] S3, monitoring the transportation flow direction of the raw materials to obtain flow direction data of the raw materials;

[0053] S4, analyzing the supply-demand flow direction of the raw materials according to the flow direction data to obtain a supply-demand flow direction actual graph of the raw materials;

[0054] S5, performing feasibility correction on the supply-demand flow direction prediction graph based on the supply-demand flow direction actual graph to obtain a supply-demand flow direction graph of the raw materials, constructing a supply-demand flow direction real-time monitoring model of the raw materials according to the supply-demand flow direction graph and the flow direction data, and performing real-time monitoring on the multi-source data fused raw material supply-demand flow direction according to the supply-demand flow direction real-time monitoring model to obtain the supply-demand flow direction of the raw materials.

[0055] Compared with the prior art, the application has the following beneficial effects:

[0056] 1. The information collection module uses advanced collection technology to extensively and deeply collect data of each link in the manufacturing industry chain, including the transportation path of raw materials from the production area to the processing workshop, and the detailed traffic flow of manufactured products from the factory to various markets through various transportation modes, etc. In the subsequent processing stage, the flow prediction module first performs detailed cleaning work on the collected massive data, then performs standardized processing on the data according to strict industry standards and specifications, the flow analysis module accurately extracts key information, and at the same time, through unified data standards, whether the data from the sensor or the manually entered data can follow the same format specification, thereby successfully overcoming the data format difference problem between different data sources, and effectively reducing the data fragmentation phenomenon.

[0057] 2. The monitoring module comprehensively and deeply analyzes the importance of traffic logistics data, uses scientific evaluation methods to accurately determine key nodes, and sets up supply and demand flow nodes according to these nodes. These nodes are like information hubs that can comprehensively collect flow data from different transportation modes and different time periods, greatly expanding the monitoring range. At the same time, the supply and demand flow generation module will closely combine the actual supply and demand flow map, carefully compare and correct the previously generated prediction map, and through continuous optimization and adjustment, build a highly accurate real-time monitoring model, thereby enhancing the response ability and real-time monitoring ability to sudden conditions. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 The system architecture diagram of the raw material supply and demand flow monitoring system based on multi-source data fusion provided by an embodiment of the present application is provided.

[0059] Figure 2 The flowchart of the raw material supply and demand flow monitoring method based on multi-source data fusion provided by an embodiment of the present application is provided.

[0060] The implementation, functional characteristics and advantages of the present application will be further described with reference to the accompanying drawings. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme of the embodiments of the present application will be described clearly and completely below with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of the present application.

[0062] The terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the description of the application and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be understood that the terms "comprises" and / or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0063] The word "if" can be interpreted to mean "upon" or "when," also to mean "in response to the determination," or "in response to the occurrence" of stated condition or event. Similarly, the phrases "if it is determined" or "if a determination is made," can be interpreted to mean "upon making a determination," or "upon the determination," or "when making a determination," or "when the determination," is made.

[0064] In addition, the sequence of steps in the following method embodiments is merely an example, not a strict limitation.

[0065] In fact, the server device deployed by the raw material supply and demand flow monitoring system based on multi-source data fusion can be composed of one or more devices. The raw material supply and demand flow monitoring system based on multi-source data fusion can be implemented as a business instance, a virtual machine, or a hardware device. For example, the raw material supply and demand flow monitoring system based on multi-source data fusion can be implemented as a business instance deployed on one or more devices in a cloud node. In short, the raw material supply and demand flow monitoring system based on multi-source data fusion can be understood as a software deployed on a cloud node, which is used to provide a raw material supply and demand flow monitoring system based on multi-source data fusion for each user end. Alternatively, the raw material supply and demand flow monitoring system based on multi-source data fusion can also be implemented as a virtual machine deployed on one or more devices in a cloud node. The virtual machine has application software installed for managing each user end. Alternatively, the raw material supply and demand flow monitoring system based on multi-source data fusion can also be implemented as a server composed of a plurality of same or different types of hardware devices, and one or more hardware devices are set to provide a raw material supply and demand flow monitoring system based on multi-source data fusion for each user end.

[0066] In terms of implementation, the raw material supply and demand flow monitoring system based on multi-source data fusion and the user end are mutually adaptive. That is, the raw material supply and demand flow monitoring system based on multi-source data fusion is an application installed on a cloud service platform, and the user end is a client that establishes a communication connection with the application; or the raw material supply and demand flow monitoring system based on multi-source data fusion is implemented as a website, and the user end is implemented as a web page; or the raw material supply and demand flow monitoring system based on multi-source data fusion is implemented as a cloud service platform, and the user end is implemented as an applet in an instant messaging application.

[0067] As Figure 1 Fig. 1 is a system architecture diagram of a raw material supply and demand flow monitoring system based on multi-source data fusion provided by an embodiment of the present application.

[0068] The raw material supply and demand flow monitoring system based on multi-source data fusion 100 can be set in a cloud server, and in the form of implementation, can be one or more service devices, or can be installed on a cloud (such as a server of a mobile service operator, a server cluster, etc.), or can be developed as a website. According to the function of implementation, the raw material supply and demand flow monitoring system based on multi-source data fusion 100 can include an information collection module 101, a flow prediction module 102, a monitoring module 103, a flow analysis module 104, and a supply and demand flow generation module 105. The modules of the present application can also be referred to as units, which refer to a series of computer program segments that can be executed by an electronic device processor and can complete a fixed function, which are stored in the memory of the electronic device.

[0069] In the embodiment of the present application, each of the above modules in the raw material supply and demand flow monitoring system based on multi-source data fusion can be independently implemented and called by other modules. The calling here can be understood as that a module can connect a plurality of modules of another type and provide corresponding services for the plurality of modules connected thereby. For example, the sharing evaluation module can call the same information collection module to obtain the information collected by the information collection module. Based on the above characteristics, the raw material supply and demand flow monitoring system based on multi-source data fusion provided by the embodiment of the present application can adjust the application scope of the raw material supply and demand flow monitoring system architecture based on multi-source data fusion by increasing modules and directly calling without modifying program codes, realize cluster horizontal expansion, and achieve the purpose of quickly and flexibly expanding the raw material supply and demand flow monitoring system based on multi-source data fusion. In actual application, the above modules can be set in the same device or different devices, or can be set in a virtual device, such as a service instance in a cloud server.

[0070] The following will be described in combination with specific embodiments, respectively for each component and specific work flow of the raw material supply and demand flow monitoring system based on multi-source data fusion:

[0071] The information collection module 101 is used to obtain traffic logistics data of raw materials.

[0072] In the embodiment of the present application, when the information collection module is used to obtain traffic logistics data of raw materials, it is specifically used for:

[0073] Collect traffic logistics data of raw materials through an industry data platform, wherein the traffic logistics data comprises manufacturing industry chain flow data, supply chain traffic logistics data, raw material flow data and manufacturing product traffic logistics data.

[0074] Specifically, among numerous industry data platforms, data authority, data coverage comprehensiveness and relevance to raw material traffic logistics are screened.

[0075] Specifically, the background information of the platform is viewed, the reputation in the industry is understood, and whether the data source channel is reliable is evaluated to determine whether the platform has a special board or service for data of manufacturing industry chain, supply chain, raw material flow and manufacturing product traffic logistics.

[0076] For example, some data platforms focusing on the logistics industry may have more detailed data on the transportation link, but lack data on the production and supply link of raw materials in the manufacturing industry chain; and comprehensive industrial data platforms may involve all aspects, but the depth may differ. Through comparison and analysis of multiple platforms, 2-3 industry data platforms most suitable for the demand are determined.

[0077] Specifically, after performing a search operation on the industry data platform, the platform presents relevant data results according to the set search conditions. The data results are preliminarily reviewed to confirm the completeness and accuracy of the data, and finally the traffic logistics data of raw materials are obtained.

[0078] For example, check whether there are missing values and abnormal values in the data, and whether the format of the data meets the requirements of subsequent processing. For data meeting the requirements, download the data in a suitable format (such as CSV, Excel, etc.) according to the download function provided by the platform.

[0079] The flow direction prediction module 102 is configured to clean noise points of the traffic logistics data to obtain standardized data of the traffic logistics data, predict the supply and demand flow direction of the raw materials according to the standardized data, and obtain a supply and demand flow direction prediction graph of the raw materials.

[0080] In the embodiment of the present application, the standardized data and the input weight matrix are input into a sigmoid activation function to obtain the current time new information of the standardized data;

[0081] The standardized data and the forgetting weight matrix are input into a sigmoid activation function to obtain the last time retained information of the standardized data;

[0082] input the standardized data and an output weight matrix into a sigmoid activation function perform calculation to obtain an information output degree of the standardized data

[0083] perform prediction based on the newly added information at the current moment, the retained information at the last moment, the information output degree and a hidden state calculation formula to obtain a supply-demand flow direction prediction graph of the raw material, wherein the hidden state calculation formula is:

[0084]

[0085] In the formula, is the prediction graph of the raw material, is the information output degree, is the retained information at the last moment, is the newly added information at the current moment, is a time factor, is a calculation symbol of element-by-element multiplication.

[0086] Specifically, the newly added information at the current moment is obtained by inputting the standardized data and an input weight matrix in the warrant dynamic consensus platform into a sigmoid activation function , and it reflects a part of information newly entering the system at the current moment and possibly affecting the supply-demand flow direction of the raw material.

[0087] Specifically, the retained information at the last moment is also obtained by inputting the standardized data and a forgetting weight matrix in the warrant dynamic consensus platform into a sigmoid activation function , and it represents a part of information selected to be retained from the data at the last moment and used to participate in the prediction of the supply-demand flow direction of the raw material at the current moment.

[0088] Specifically, the information output degree is also obtained by inputting the standardized data and an input weight matrix in the warrant dynamic consensus platform into a sigmoid activation function , and it determines the degree of information output from the memory unit to the hidden state.

[0089] The monitoring module 103 is configured to monitor the transportation flow direction of the raw material to obtain flow direction data of the raw material.

[0090] In the embodiment of the present application, when the monitoring module monitors the transportation flow direction of the raw material to obtain the flow direction data of the raw material, it is specifically configured to:

[0091] screen the traffic logistics data according to the importance of the traffic logistics data to obtain important nodes of the traffic logistics data, and take the important nodes as the supply-demand flow direction nodes of the raw material.

[0092] The supply-demand flow direction node monitors the transportation flow direction of the raw materials, and obtains multi-source data of the supply-demand flow direction node;

[0093] The multi-source data is standardized in different formats to obtain standardized data of the multi-source data;

[0094] The standardized data is classified according to the data importance, and classified data of the standardized data is obtained, wherein the classified data includes ordinary data and core data;

[0095] The classified data and the warrant dynamic consensus algorithm are used for consensus benefit calculation to obtain the flow direction data of the raw materials.

[0096] The classified data and the warrant dynamic consensus algorithm are used for consensus benefit calculation to obtain the flow direction data of the raw materials.

[0097] The ordinary data and the warrant dynamic consensus algorithm are used for layer calculation to obtain the transportation information of the raw materials, wherein, The layer calculation formula is:

[0098]

[0099] In the formula, is the transportation information, is the ordinary data, is time information of supply-demand flow direction node data, is reliability of supply-demand flow direction node data, is the supply-demand flow direction node.

[0100] The ordinary data and the warrant dynamic consensus algorithm are used for layer calculation to obtain the cooperation information of the raw materials, wherein, The layer calculation formula is:

[0101]

[0102] In the formula, is the cooperation information, is the preparation stage, is the supply-demand flow direction node, is a serial number of monitoring data, is the core data, is a ordered tuple expression symbol;

[0103] The transportation information and the cooperation information are integrated to obtain the flow direction data of the raw materials.

[0104] Specifically, a comprehensive and in-depth importance assessment is carried out on the collected traffic logistics data, and a scientific importance assessment system is constructed, which comprehensively considers the busy degree of the transportation route involved in the data, the value of the transported goods, the influence degree on the overall supply chain stability and other factors.

[0105] For example, the transportation route data connecting large raw material production bases and core processing enterprises is given a higher importance score because it carries a large amount of transportation of key raw materials and is crucial to the operation of the industrial chain; while some occasional branch transportation route data has a relatively low importance score.

[0106] Further, by scoring and sorting each data, the nodes corresponding to the data with higher scores are selected, and these nodes are determined as the supply and demand flow nodes of raw materials, which play a key role in the supply and demand flow process of raw materials.

[0107] Specifically, at the supply and demand flow nodes, various advanced monitoring means and equipment are deployed, and high-precision GPS positioning devices are used to track the position information of transportation vehicles at the nodes in real time, ensuring that the real-time dynamics of raw material transportation can be accurately mastered; various sensors are installed in the warehouses and loading and unloading areas of the nodes to collect data such as loading and unloading time, quantity, and state of raw materials.

[0108] Further, by data interfacing with the logistics management system of transportation enterprises, transportation plans, vehicle scheduling and other related information are obtained, and the data collected by these different monitoring methods are complementary to each other.

[0109] In general, the multi-source data of the supply and demand flow nodes obtained ultimately reflects the transportation flow situation of raw materials at the supply and demand flow nodes from all aspects.

[0110] Specifically, in the process of obtaining traffic logistics data from multiple ways, different source data may have deviations, and the warrant dynamic consensus algorithm allows each participating node to agree on the data.

[0111] For example, when determining the raw material flow data, the data of multiple nodes such as transportation enterprises, warehousing enterprises and production enterprises needs to reach a consensus, and the algorithm ensures that only data verified consistent by multiple nodes is recognized as valid, avoiding the mixing of false or false data, improving data reliability, and providing a solid foundation for subsequent analysis and decision-making based on these data.

[0112] Further, the warrant dynamic consensus algorithm uses different calculation levels (such as POS layer and PBFT layer) according to the importance of the data (such as ordinary data and core data), and for ordinary data, the POS layer is used for rapid processing, and the proof-of-stake mechanism is used to efficiently calculate the transportation information according to factors such as time information and reliability of the data.

[0113] Further, for core data, high security and accuracy are ensured by the PBFT layer, and cooperation information is obtained by complex calculation.

[0114] Overall, this layered processing improves overall data processing efficiency and meets the processing needs of different types of data.

[0115] The flow direction analysis module 104 is configured to analyze the supply and demand flow direction of the raw materials according to the flow direction data to obtain the actual supply and demand flow direction map of the raw materials.

[0116] In the embodiment of the present application, when the flow direction analysis module performs the analysis of the supply and demand flow direction of the raw materials according to the flow direction data to obtain the actual supply and demand flow direction map of the raw materials, it is specifically used for:

[0117] Noise cleaning is performed on the flow direction data to obtain standard data of the flow direction data;

[0118] The flow direction supply and demand information of the standard data is extracted;

[0119] The flow direction supply and demand information is standardized and coded to obtain the supply and demand flow direction data of the raw materials;

[0120] The actual supply and demand flow direction map of the raw materials is constructed based on the supply and demand flow direction data.

[0121] Specifically, the collected flow direction data is cleaned of noise by using professional data cleaning algorithms and tools, and by setting a reasonable data threshold range, data points deviating from the normal range, such as abnormal data of raw material transportation quantity far exceeding the reasonable maximum value of the industry or being negative, are identified and removed.

[0122] Further, the continuity of the data is checked, and missing values caused by data transmission interruption and the like are filled by interpolation method, mean filling method and the like to obtain the standard data of the flow direction information.

[0123] Overall, through a series of cleaning operations, the flow direction data that may have noise and errors is converted into standard data that meets industry standards, is accurate and coherent, providing a reliable foundation for subsequent data extraction and analysis.

[0124] Specifically, the flow direction supply and demand related information is accurately extracted from the cleaned standard data by using data mining technology and pre-set rules, and for the transportation data of the raw materials, the relationship between the transportation direction, transportation quantity and each stage supply and demand node is analyzed, and the information of which regions receive raw materials (demand side), which regions output raw materials (supply side) and the corresponding supply and demand quantity is extracted to obtain the flow direction supply and demand information.

[0125] Further, from the warehouse data, the increase and decrease changes of inventory in different periods are mined, it is judged whether the inventory accumulation is caused by oversupply or the inventory reduction is caused by strong demand, and the supply and demand dynamic information in each stage is sorted out, which is the key content to deeply understand the supply and demand flow direction of raw materials.

[0126] Specifically, the geographic position information of the supply side and the demand side is encoded according to specific geographic coding rules, such as using internationally recognized geographic coordinate coding or industry-recognized regional coding system.

[0127] Further, for numerical information such as supply and demand quantity and transportation time, the dimension is unified and converted into a specified digital coding format to obtain the supply and demand flow direction data.

[0128] In general, through such standardized coding operation, diversified supply and demand flow information is converted into a unified format that is easy for computers to recognize and process, and supply and demand flow direction data of raw materials for further analysis and visualization are obtained.

[0129] Specifically, according to the standardized coded supply and demand flow direction data, the supply and demand flow direction actual map of raw materials is visually displayed on the map, different colors, line thicknesses or icons are used to intuitively show the supply and demand places of raw materials, lines are used to connect the supply and demand sides and the line thickness is set according to the transportation volume, the transportation path of raw materials is clearly presented, and finally the supply and demand flow direction actual map of raw materials is constructed.

[0130] In general, the supply and demand flow direction actual map displays the supply and demand changes in different time periods, such as inventory level and transportation frequency changes, through a dynamic time axis.

[0131] In general, the supply and demand flow direction actual map constructed in this way can intuitively and visually present the actual flow state of raw materials in the entire supply chain, and provide intuitive data support for enterprise decision-making.

[0132] The supply and demand flow direction generation module 105 is configured to correct the supply and demand flow direction prediction map based on the supply and demand flow direction actual map, obtain the supply and demand flow direction map of the raw materials, construct the real-time monitoring model of the supply and demand flow direction of the raw materials according to the supply and demand flow direction map and the flow direction data, and monitor the raw material supply and demand flow direction based on multi-source data fusion in real time according to the real-time monitoring model of the supply and demand flow direction, to obtain the supply and demand flow direction of the raw materials.

[0133] In the embodiment of the present application, when the supply and demand flow direction generation module performs the correction of the supply and demand flow direction prediction map based on the supply and demand flow direction actual map to obtain the supply and demand flow direction map of the raw materials, it is specifically used for:

[0134] The flow direction information of the supply-demand flow direction prediction graph is compared and identified based on the supply-demand flow direction actual graph, to obtain a difference identification point of the supply-demand flow direction actual graph to the supply-demand flow direction prediction graph.

[0135] The flow direction path of the supply-demand flow direction actual graph is corrected based on the difference identification point, to obtain the supply-demand flow direction graph of the raw material.

[0136] The supply-demand flow direction generation module, in the execution of constructing the real-time monitoring model of the supply-demand flow direction of the raw material according to the supply-demand flow direction graph and the flow direction data, and in the real-time monitoring of the supply-demand flow direction of the raw material based on the multi-source data fusion according to the real-time monitoring model of the supply-demand flow direction of the raw material, specifically comprises:

[0137] The flow direction characteristics of the raw material flow direction data are identified to obtain the key characteristics of the raw material flow direction data;

[0138] The real-time monitoring model of the supply-demand flow direction of the raw material is constructed based on the key characteristics and the supply-demand flow direction graph.

[0139] The supply-demand flow direction generation module, in the execution of constructing the real-time monitoring model of the supply-demand flow direction of the raw material according to the supply-demand flow direction real-time monitoring model, and in the real-time monitoring of the supply-demand flow direction of the raw material based on the multi-source data fusion, specifically comprises:

[0140] The flow direction network of the raw material is constructed based on the flow direction data;

[0141] The supply-demand relationship of the raw material is obtained by analyzing the node properties and flow direction information in the flow direction network;

[0142] The supply-demand flow direction of the raw material is obtained by flow direction analysis based on the supply-demand relationship and the real-time monitoring model of the supply-demand flow direction.

[0143] Specifically, the flow direction information of the supply-demand flow direction actual graph and the supply-demand flow direction prediction graph is compared and identified point by point and path by path.

[0144] Further, for the transportation route of the raw material in the actual graph and the prediction graph, whether the starting point, the ending point and the intermediate key node are consistent is checked, and if there is a position deviation or inconsistent route, the difference position is recorded.

[0145] Further, in terms of flow data, the actual transportation volume is compared with the predicted transportation volume, and when the number difference between the two exceeds a certain threshold value, the flow data point is marked as a difference point, and at the same time, the transportation time node is focused on, such as the difference between the actual transportation arrival time and the predicted time, and the points with inconsistent time are also included in the difference identification range.

[0146] Overall, by comprehensive comparative analysis, accurately find the differences between the actual and predicted flow direction of supply and demand, form a detailed list of difference identification points, provide the basis for subsequent flow path correction.

[0147] Specifically, according to the difference identification points, the flow path of the actual supply and demand flow direction is corrected, for the difference point of the transportation route, if the actual transportation route shows that a raw material from A to C, and the predicted graph is from A to B, and the actual route is more accurate, then adjust the transportation route in the actual graph to A to C. For the flow difference point, if the actual transportation volume is larger than the predicted volume, increase the flow display on the transportation path in the actual graph accordingly.

[0148] Further, for the transportation time difference point, if the actual arrival time is later than the predicted time, update the time information in the actual graph, and analyze the reasons that may cause the time delay, such as traffic congestion, vehicle failure, etc., record these reasons as note information in the actual graph, and obtain the supply and demand flow direction graph of the raw material.

[0149] Overall, by processing each difference point one by one, the flow path of the actual supply and demand flow direction is more accurately reflected the real supply and demand flow of raw materials, and finally the corrected raw material supply and demand flow direction is obtained, providing more reliable decision support for enterprises.

[0150] Specifically, the obtained raw material flow data is analyzed in detail, first, pay attention to the time series characteristics in the data.

[0151] For example, the time interval of raw materials starting from the production area, arriving at each processing area and consumption area at different time points, through time series analysis to identify the rules of raw material transportation cycle, determine the peak period and the trough period.

[0152] Further, analyze the spatial characteristics, study the geographical route of raw materials from the production area to the consumption area, identify the key transportation nodes and transportation routes, and judge which routes are the main transportation channels and which nodes play a pivotal role in the transportation process.

[0153] Further, the flow characteristics are mined, the transportation volume of raw materials at different stages is counted, and it is determined that there is a large amount of raw material flow in which link, and the correlation between these flow changes and market demand, seasonal factors, etc.

[0154] Overall, through comprehensive analysis of time, space, flow and other characteristics, the key features that can accurately reflect the essence and rules of raw material flow are extracted, and a key feature set is formed, which provides core data support for subsequent monitoring model construction.

[0155] Specifically, based on the acquired flow data, each transportation starting point, transfer point and terminal point in the data is set as a network node.

[0156] For example, the origin of raw materials, logistics hubs in transit, and final processing plants or sales terminals all become nodes.

[0157] Further, by analyzing the transportation path and frequency of raw materials between different nodes in the flow data, a flow network is obtained by connecting each node with a directed edge, where the direction of the directed edge represents the flow direction of the raw materials, and the weight of the edge is determined according to the transportation volume or transportation frequency and other factors.

[0158] For example, if the transportation volume of raw materials on a certain route is large, the weight of the corresponding directed edge is set to be high.

[0159] In general, the flow network converts discrete flow data into an intuitive and structured structure, fully displaying the flow path and relationship of raw materials in the entire transportation process, and providing a clear framework for subsequent analysis.

[0160] Specifically, the properties and flow information of the nodes in the constructed flow network are analyzed in depth. From the node properties, those nodes with a large inflow but a small outflow of raw materials are likely to be demand ends, such as large manufacturing enterprises, where a large amount of raw materials flow in for production and processing.

[0161] Further, nodes with a large outflow but a small inflow of raw materials are usually supply ends, such as raw material mining sites or large production bases. By counting the inflow and outflow of each node, the supply and demand relationship is further quantified.

[0162] Further, for the flow information, the flow path and flow change of raw materials from the supply nodes to the demand nodes are observed. If a demand node has multiple supply nodes delivering raw materials to it, the supply proportion of each supply node can be analyzed to determine the importance of different supply nodes in meeting the demand node. By comprehensively analyzing these results, the supply and demand relationship of raw materials is accurately obtained.

[0163] Specifically, the supply and demand relationship of raw materials obtained from the previous analysis is combined with the constructed supply and demand flow real-time monitoring model to perform in-depth flow analysis. The supply nodes, demand nodes and supply and demand quantities between them in the supply and demand relationship are input into the real-time monitoring model. The existing time, space, flow and other feature analysis modules in the model are used to analyze the real-time monitored data, such as the current inventory level of each node, the real-time position and transportation status of the transportation vehicle, and other information. The flow of raw materials at different time periods and different spatial positions is dynamically analyzed to obtain the supply and demand flow of the raw materials.

[0164] For example, according to the model, the changes of the supply path and the supply amount of the downstream demand node caused by the increase of the production of a certain supply node in a certain period of time are predicted, so that the supply and demand flow direction of the raw material in the entire supply chain is accurately obtained, and accurate decision basis for the enterprise to reasonably arrange production, transportation and storage and other links is provided.

[0165] Referring to Figure 2 Fig. 1 is a flowchart of a raw material supply and demand flow direction monitoring method based on multi-source data fusion provided by an embodiment of the present application. In this embodiment, the raw material supply and demand flow direction monitoring method based on multi-source data fusion comprises the following steps:

[0166] Obtaining traffic logistics data of a raw material;

[0167] Cleaning noise points of the traffic logistics data to obtain standardized data of the traffic logistics data, and predicting a supply and demand flow direction of the raw material according to the standardized data to obtain a supply and demand flow direction prediction graph of the raw material;

[0168] Monitoring a transportation flow direction of the raw material to obtain flow direction data of the raw material;

[0169] Analyzing the supply and demand flow direction of the raw material according to the flow direction data to obtain a supply and demand flow direction actual graph of the raw material;

[0170] Correcting the supply and demand flow direction prediction graph based on the supply and demand flow direction actual graph to obtain a supply and demand flow direction graph of the raw material, constructing a supply and demand flow direction real-time monitoring model of the raw material according to the supply and demand flow direction graph and the flow direction data, and monitoring the raw material supply and demand flow direction based on multi-source data fusion in real time according to the supply and demand flow direction real-time monitoring model to obtain the supply and demand flow direction of the raw material.

[0171] The present application solves the problem of information fragmentation by virtue of the advantages of multi-source data fusion. The information collection module widely collects data such as manufacturing industry chain, raw material and manufacturing product traffic flow direction, etc. The subsequent flow direction prediction module cleans and standardizes the data, the flow direction analysis module extracts key information and integrates, while unifying the data standard, overcoming the problem of data format difference, effectively reducing data fragmentation, providing coherent data set for analysis, the monitoring module determines the node through the importance analysis of the traffic logistics data and sets up the supply and demand flow direction node, collects the flow direction data comprehensively, widens the monitoring range, the supply and demand flow direction generation module corrects the prediction graph according to the supply and demand flow direction actual graph, and constructs the real-time monitoring model, timely reflects the dynamic change, and enhances the response and real-time monitoring ability to the sudden situation.

[0172] It is apparent for those skilled in the art that the present application is not limited to the details of the above-described exemplary embodiments, and the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application.

[0173] Thus, the embodiments should be considered in all respects as illustrative and not restrictive, the scope of the application being indicated by the appended claims rather than by the description given above, therefore all changes which come within the meaning and range of equivalency of the claims are intended to be embraced therein. Any reference signs in the claims should not be construed as limiting the scope of the claims.

[0174] Embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Among them, artificial intelligence (Artificial Intelligence, AI) is to use digital computers or digital computer controlled machines to simulate, extend and expand human intelligence, perceive needs, acquire knowledge and use knowledge to obtain the best results. Theory, method, technology and application system.

[0175] In addition, it is clear that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claims can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, not any particular order.

[0176] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A raw material supply and demand flow monitoring system based on multi-source data fusion, characterized in that, The system comprises an information collection module, a flow direction prediction module, a monitoring module, a flow direction analysis module, and a supply-demand flow direction generation module, wherein: The information collection module is configured to obtain traffic logistics data of raw materials. The flow direction prediction module is configured to clean noise points from the traffic logistics data to obtain standardized data of the traffic logistics data, and predict the supply-demand flow direction of the raw materials based on the standardized data to obtain a supply-demand flow direction prediction graph of the raw materials. inputting the standardized data and the input weight matrix into a sigmoid activation function carrying out calculation to obtain the current time new information of the standardized data; inputting the standardized data and the forgetting weight matrix into a sigmoid activation function performing calculation to obtain the last time reserved information of the standardized data; inputting the standardized data and the output weight matrix into a sigmoid activation function performing a calculation to obtain an information output degree of the standardized data; The monitoring module is configured to monitor the transportation flow direction of the raw materials to obtain flow direction data of the raw materials. ； In the formula, is a prediction map of the raw material, is the information output degree, is the last time reserved information, is the current time new information, is a time factor, is a calculation symbol of element-by-element multiplication; The important nodes of the traffic logistics data are obtained by screening based on the importance of the traffic logistics data, and the important nodes are used as the supply-demand flow direction nodes of the raw materials. The transportation flow direction of the raw materials is monitored at the supply-demand flow direction nodes to obtain multi-source data of the supply-demand flow direction nodes. The multi-source data is standardized in different formats to obtain standardized data of the multi-source data. The classification data of the standardized data is obtained by classifying based on the data importance of the standardized data, wherein the classification data comprises ordinary data and core data. The flow direction data of the raw materials is obtained by consensus equity calculation based on the classification data and a dynamic consensus algorithm of a warrant. The transportation information and the cooperation information are integrated to obtain the flow direction data of the raw materials. Based on the common data and the warrant dynamic consensus algorithm Layer calculation, get the raw material transportation information, wherein, The formula of the layer calculation is: ； In the formula, is the transport information, is the common data, is time information of the supply-demand flow direction node data, is reliability of the supply-demand flow direction node data, is the supply-demand flow direction node; Based on the core data and the warrant dynamic consensus algorithm Layer calculation, get the cooperation information of the raw materials, wherein, The formula for layer calculation is: ； In the formula, is the cooperation information, is the preparation stage, is the supply-demand flow direction node, is the serial number of the monitoring data, is the core data, is a ordered tuple expression symbol; The flow direction analysis module is configured to analyze the supply-demand flow direction of the raw materials based on the flow direction data to obtain a supply-demand flow direction actual graph of the raw materials. The supply-demand flow direction generation module is configured to correct the supply-demand flow direction prediction graph based on the supply-demand flow direction actual graph to obtain a supply-demand flow direction graph of the raw materials, construct a supply-demand flow direction real-time monitoring model of the raw materials based on the supply-demand flow direction graph and the flow direction data, and monitor the multi-source data fusion supply-demand flow direction of the raw materials in real time based on the supply-demand flow direction real-time monitoring model to obtain the supply-demand flow direction of the raw materials. When the information collection module is used to obtain traffic logistics data of raw materials, it is specifically configured to:

2. The raw material supply and demand flow monitoring system based on multi-source data fusion according to claim 1, characterized in that, Collect traffic logistics data of raw materials through an industry data platform, wherein the traffic logistics data comprises manufacturing industry chain flow direction data, supply chain traffic logistics data, raw material flow direction data, and manufacturing product traffic logistics data. When the flow direction analysis module is used to analyze the supply-demand flow direction of the raw materials based on the flow direction data to obtain a supply-demand flow direction actual graph of the raw materials, it is specifically configured to:

3. The raw material supply and demand flow monitoring system based on multi-source data fusion according to claim 1, characterized in that, Clean noise points from the flow direction data to obtain standard data of the flow direction data; Extract flow direction supply-demand information of the standard data; Standardize the flow direction supply-demand information to obtain supply-demand flow direction data of the raw materials; Construct the supply-demand flow direction actual graph of the raw materials based on the supply-demand flow direction data. ​ 4. The raw material supply and demand flow monitoring system based on multi-source data fusion according to claim 1, characterized in that, The supply and demand flow direction generation module, when performing feasibility correction on the supply and demand flow direction prediction graph based on the supply and demand flow direction actual graph to obtain the supply and demand flow direction graph of the raw materials, is specifically used for: comparing and identifying the flow direction information of the supply and demand flow direction prediction graph based on the supply and demand flow direction actual graph to obtain difference identification points of the supply and demand flow direction actual graph on the supply and demand flow direction prediction graph; correcting the flow direction path of the supply and demand flow direction actual graph based on the difference identification points to obtain the supply and demand flow direction graph of the raw materials.

5. The raw material supply and demand flow monitoring system based on multi-source data fusion according to claim 1, characterized in that, The supply and demand flow direction generation module, when performing the supply and demand flow direction real-time monitoring model of the raw materials based on the supply and demand flow direction graph and the flow direction data, and performing real-time monitoring on the multi-source data fused supply and demand flow direction of the raw materials based on the supply and demand flow direction real-time monitoring model to obtain the supply and demand flow direction of the raw materials, is specifically used for: identifying the flow direction features of the raw material flow direction data to obtain key features of the raw material flow direction data; constructing the supply and demand flow direction real-time monitoring model of the raw materials based on the key features and the supply and demand flow direction graph.

6. The raw material supply and demand flow monitoring system based on multi-source data fusion according to claim 1, characterized in that, The supply and demand flow direction generation module, when performing the supply and demand flow direction real-time monitoring model of the raw materials based on the supply and demand flow direction graph and the flow direction data, and performing real-time monitoring on the multi-source data fused supply and demand flow direction of the raw materials based on the supply and demand flow direction real-time monitoring model to obtain the supply and demand flow direction of the raw materials, is specifically used for: constructing the flow direction network of the raw materials based on the flow direction data; analyzing the node properties and flow direction information in the flow direction network to obtain the supply and demand relationship of the raw materials; performing flow direction analysis based on the supply and demand relationship and the supply and demand flow direction real-time monitoring model to obtain the supply and demand flow direction of the raw materials.

7. A raw material supply and demand flow direction monitoring method based on multi-source data fusion, used for realizing the raw material supply and demand flow direction monitoring system based on multi-source data fusion of claim 1, the method comprising: S1, obtaining traffic logistics data of raw materials; S2, performing noise point cleaning on the traffic logistics data to obtain standardized data of the traffic logistics data, and predicting the supply and demand flow direction of the raw materials based on the standardized data to obtain a supply and demand flow direction prediction graph of the raw materials; S3, monitoring the transportation flow direction of the raw materials to obtain flow direction data of the raw materials; S4, analyzing the supply and demand flow direction of the raw materials based on the flow direction data to obtain a supply and demand flow direction actual graph of the raw materials; S5, performing feasibility correction on the supply and demand flow direction prediction graph based on the supply and demand flow direction actual graph to obtain a supply and demand flow direction graph of the raw materials, constructing a supply and demand flow direction real-time monitoring model of the raw materials based on the supply and demand flow direction graph and the flow direction data, and performing real-time monitoring on the multi-source data fused supply and demand flow direction of the raw materials based on the supply and demand flow direction real-time monitoring model to obtain the supply and demand flow direction of the raw materials.

Images