Electric power material distribution management system and method based on big data

Abstract

The invention relates to the technical field of electric power material distribution management, particularly discloses an electric power material distribution management system and method based on big data, and aims to solve the problems of dispersed material data, inaccurate demand prediction, low distribution efficiency, improper inventory management and the like in an existing system. According to the specific technical scheme, the method comprises the steps of integrating material names, material description information and supplier information, and establishing a unified electric power material database; carrying out preliminary classification and refined classification on the materials by using a clustering algorithm, and adding a label for each type of materials; historical demand data are analyzed, a demand prediction model is constructed, and the prediction precision is improved; dynamic optimal distribution path planning is realized by utilizing a geographic information system technology and a linear programming algorithm, and the distribution efficiency is improved; the stock is dynamically adjusted by calculating the safe stock and the reorder cycle, and stock overstock or shortage is avoided; the distribution plan is dynamically adjusted according to the inventory level and the demand frequency, the distribution process is monitored in real time, and on-time distribution is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of power material distribution management, and in particular to a power material distribution management system and method based on big data. Background Art

[0002] With the rapid development of power grid construction, the distribution management of power materials has become increasingly important. Traditional power material distribution management methods have many shortcomings, such as scattered material information, inaccurate demand forecasts, and unreasonable distribution routes. With the development of big data technology and geographic information systems, it has become possible to use big data to manage power material distribution, which can effectively improve distribution efficiency and management level.

[0003] At present, the power material distribution management system has deficiencies in many aspects: material data management can only collect basic material information, and lacks integrated management of material description information and supplier information; demand forecasting mainly relies on empirical estimates and lacks scientific data support, resulting in low forecast accuracy; distribution route planning is based on static data and cannot be adjusted in real time, resulting in low distribution efficiency; traditional inventory management methods make it difficult to accurately grasp the actual demand for materials, which easily leads to inventory backlogs or shortages.

[0004] The present invention proposes an electric power material distribution management system and method based on big data, which has the following advantages: by integrating material names, material description information and supplier information, a unified electric power material database is established to provide reliable data support for subsequent processing; by conducting in-depth analysis of historical demand data, a scientific demand forecasting model is constructed to improve forecasting accuracy; by using geographic information system technology and linear programming algorithms, dynamic optimal distribution path planning is achieved to improve distribution efficiency; by calculating safety stock and reorder cycle, dynamic adjustment of inventory is achieved to avoid inventory backlog or shortage problems. Summary of the invention

[0005] The purpose of the present invention is to provide a power material distribution management system and method based on big data. The specific technical scheme includes: first, obtaining power material data in power grid engineering projects, including material names, material description information and supplier information, and establishing a unified power material database through integration; using a clustering algorithm to preliminarily classify the materials, and further refine the classification according to the keywords in the material description information, adding labels to each type of material to identify its main characteristics; extracting historical demand data of various types of materials in a certain period of time in the past from the power material database, extracting feature data related to the demand, and performing correlation analysis on the feature data, identifying the main factors affecting the demand, integrating these features to obtain time series data of material demand, and using the ARIMA model based on the time series data to build a material demand forecasting model to predict the demand for various types of materials in a future period of time, and using the seasonal decomposition method to adjust the forecasting model; according to the results of the forecasting model, analyzing the demand frequency of various types of materials in the future time period, statistically analyzing the quantity demand distribution of high-frequency demand materials and low-frequency demand materials, and setting the quantity according to the demand frequency and quantity demand. Thresholds are used to distinguish between large quantities of materials and small quantities of materials; materials with high frequency and high demand are combined into one package, or materials that are related or complementary to each other are combined into one package, and the real-time demand data of small quantities of materials are combined into one package; the status information of power grid materials is obtained, a network diagram is constructed using geographic information system technology to generate the best distribution path, the minimum transportation energy consumption is solved through a linear programming algorithm, the optimal distribution plan is found, and the distribution strategy of large quantities of materials is optimized based on this; the demand data of small quantities of materials in the past period of time are extracted, the demand mean and standard deviation are calculated, the reorder cycle is determined, and the safety stock is calculated to ensure that the inventory is neither too much nor too little. When the actual inventory is lower than the safety stock, an early warning message is issued to remind the replenishment of inventory, and the distribution plan is dynamically adjusted according to the inventory level and demand frequency, and the demand changes are responded to in a timely manner. Priorities are set according to the urgency of the demand, emergency demands are given priority, and the delivery time window is set to ensure that the materials arrive on time. The optimal distribution path is planned using the path optimization algorithm, and real-time adjustments are made to cope with sudden demands. The time progress of the distribution process is monitored in real time to ensure that the delivery time matches the demand time.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for managing power material distribution based on big data comprises the following steps:

[0008] S1: Obtain the power material data of the power grid project and establish a unified power material database through integration;

[0009] Classify the materials in the power material database;

[0010] S2: Obtain historical demand data of various materials in power grid projects within a certain period of time in the past, and build a material demand prediction model;

[0011] S3: According to the prediction results of the prediction model, the quantity demand distribution of high-frequency demand materials and low-frequency demand materials is obtained;

[0012] Set quantity thresholds based on quantity demand distribution to distinguish between large and small quantities of materials;

[0013] Combined with the demand frequency, large-volume materials and small-volume materials are divided into standard packages to obtain large-volume material standard packages and small-volume material standard packages;

[0014] S4: Optimize the distribution strategy of bulk materials based on bulk material standard packages;

[0015] Based on small batch material standard packages, the small batch material distribution strategy is optimized.

[0016] As a further solution of the present invention: the power material data includes material name, material description information and supplier information;

[0017] Material description information includes material model and parameters;

[0018] Supplier information includes remaining quantity and geographic location of supplies.

[0019] As a further solution of the present invention: based on the electric power material database, the materials are preliminarily classified according to the names of the electric power materials, and the materials are finely classified according to the material description information.

[0020] As a further solution of the present invention: the specific process of constructing the material demand forecasting model is:

[0021] Obtain historical demand record data for various materials within a certain period of time in the past;

[0022] The past certain period of time includes, but is not limited to, the past year and three years;

[0023] Use time series analysis methods to identify demand trends in historical demand record data for various materials, and extract demand-related feature data, including date, time, material category, and demand quantity;

[0024] All characteristic data extracted from the power material database are integrated and processed in chronological order to obtain the time series data of material demand;

[0025] Build a material demand forecasting model based on the time series data of material demand;

[0026] Based on the material demand forecasting model, the seasonal decomposition method is used to decompose the time series into three parts: trend, seasonality and residual;

[0027] Adjust material demand forecast models based on seasonal components.

[0028] As a further solution of the present invention: based on the prediction results of the material demand prediction model, the demand frequency and quantity demand of various materials are obtained;

[0029] Based on the demand frequency of materials, materials are divided into high-frequency demand materials and low-frequency demand materials;

[0030] Conduct statistical analysis on the demand for high-frequency and low-frequency demand materials to obtain the demand distribution of different materials.

[0031] As a further solution of the present invention: the process of dividing large batches of materials and small batches of materials into standard packages is:

[0032] Based on large quantities of materials, materials of the same or similar types are combined into one package, which is recorded as a large quantity material package;

[0033] Combine related or complementary materials to form a package, which is recorded as a large-volume material package;

[0034] Based on small batch materials, collect real-time demand data of small batch materials, extract the current actual demand situation, and obtain the actual demand quantity;

[0035] According to the actual demand, small batches of materials are combined into a standard package, which is recorded as a small batch material standard package.

[0036] As a further solution of the present invention: based on large quantities of material packages, the status information of power grid materials and the historical distribution path and time information are obtained;

[0037] Among them, the status information of power grid materials includes warehouse location information, distribution point location information and remaining quantity information;

[0038] Using GIS technology, a network map of all distribution points and warehouses was constructed;

[0039] Generate paths from each warehouse to each distribution point based on the network graph;

[0040] The path from each warehouse to each distribution point is calculated to obtain the total transportation energy consumption from each warehouse to each distribution point;

[0041] The total transport energy consumption is analyzed and calculated to minimize the energy consumption, and the optimal route with the minimum total transport energy consumption is generated.

[0042] As a further solution of the present invention: based on the small batch material package, the historical demand record of the small batch material is obtained to obtain the demand data in the past period of time, wherein the past period of time includes but is not limited to one year and two years;

[0043] Identify and extract the demand data of small batch materials in the past period of time to obtain the daily demand and total number of days in the past period of time;

[0044] Based on the daily demand and total number of days for small batches of materials in the past period, the average demand for small batches of materials in the past period is calculated;

[0045] The standard deviation of demand for small batches of materials over a period of time in the past is calculated through the demand mean;

[0046] Based on the historical demand records of small batch materials, the demand fluctuations of small batch materials are extracted to obtain the reorder cycle of small batch materials;

[0047] Based on the demand standard deviation and reorder cycle of small batch materials, the safety stock of small batch materials is calculated.

[0048] As a further solution of the present invention: based on the safety inventory of small batch materials, the actual inventory of small batch materials is obtained;

[0049] Compare actual inventory of small-volume supplies with safety stock;

[0050] When the actual inventory of small-batch materials is lower than the safety inventory, an early warning message is issued to remind you to replenish the inventory;

[0051] When the actual inventory of small-batch materials is higher than or equal to the safety stock, the distribution plan is dynamically adjusted according to the actual inventory level and demand frequency, and the priority is set according to the urgency of demand.

[0052] As a further solution of the present invention: a power material distribution management system based on big data, characterized by comprising a data collection module, a data processing module, a model building module and a material package division and strategy formulation module;

[0053] The data collection module is used to obtain the power material data of the power grid project and establish a unified power material database through integration;

[0054] The data processing module is used to classify the materials in the power material database;

[0055] The model building module is used to obtain the historical demand data of various materials in the power grid project within a certain period of time in the past, and build a material demand prediction model;

[0056] According to the prediction results of the prediction model, the quantity demand distribution of high-frequency demand materials and low-frequency demand materials is obtained;

[0057] The material package division and strategy formulation module is used to set quantity thresholds according to quantity demand distribution and distinguish between large and small quantities of materials;

[0058] Combined with the demand frequency, large-volume materials and small-volume materials are divided into standard packages to obtain large-volume material standard packages and small-volume material standard packages;

[0059] Optimize the distribution strategy of bulk materials based on bulk material standard packages;

[0060] Based on small batch material standard packages, the small batch material distribution strategy is optimized.

[0061] Beneficial effects of the present invention:

[0062] (1) First, by integrating material names, material description information and supplier information, a unified power material database is established. This not only provides comprehensive and structured data support, but also solves the problem of scattered material data and lack of integrated management in the existing system, improving data availability and management efficiency. Secondly, by conducting in-depth analysis of historical demand data and building a demand forecasting model, the accuracy of demand forecasting is improved, solving the problem that traditional forecasting methods mainly rely on empirical estimates and have low forecasting accuracy, making demand forecasting more scientific and reliable. In addition, by using geographic information system technology and linear programming algorithms, dynamic optimal distribution path planning is realized, which improves distribution efficiency and solves the problem that existing distribution path planning is based on static data and cannot be adjusted in real time, thereby improving distribution flexibility and efficiency.

[0063] (2) By calculating the safety stock and reorder cycle, dynamic adjustment of inventory is achieved, avoiding the problem of inventory backlog or shortage, solving the problem that traditional inventory management methods are difficult to accurately grasp the actual demand for materials and easily cause inventory backlog or shortage, and improving the accuracy of inventory management. At the same time, a clustering algorithm is used to preliminarily classify materials, and further refine the classification based on the keywords in the material description information, adding labels to each type of material, improving the accuracy of material classification and the convenience of management, solving the problem of unclear material classification and inconvenient management, and making material management more systematic and standardized. In addition, by processing demand data through time series processing methods (such as moving average and smoothing methods) and seasonal decomposition methods, the accuracy of demand forecasting is improved, solving the problem that seasonal changes and special circumstances are not fully considered in demand forecasting, and making the forecast results closer to actual demand;

[0064] (3) By identifying special events in historical records and their impact on demand and adding event variables to the prediction model, the robustness of demand forecasting is improved, the problem of not considering the impact of emergencies in demand forecasting is solved, and the prediction results are more reliable. At the same time, the distribution plan is dynamically adjusted according to the inventory level and demand frequency, and the priority is set according to the urgency of the demand, which improves the response speed and service level of distribution, solves the problem of fixed distribution plans and inability to respond to changes in demand in a timely manner, and enhances the flexibility and responsiveness of the distribution system. In addition, by real-time monitoring of the time progress of the distribution process and using the path optimization algorithm to plan the optimal distribution path, it is ensured that the distribution time matches the demand time, solves the problem of lack of real-time monitoring and path optimization in the distribution process, and improves the punctuality rate and service quality of distribution. In summary, the present invention significantly improves the overall performance and service level of the power material distribution management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The present invention will be further described below in conjunction with the accompanying drawings.

[0066] Figure 1 It is a flowchart of specific steps in a power material distribution management system and method based on big data of the present invention;

[0067] Figure 2 is a flow chart for distinguishing large batches of materials from small batches of materials in the present invention;

[0068] Figure 3 It is a flow chart for optimizing the bulk material distribution strategy in the present invention. DETAILED DESCRIPTION

[0069] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] Example 1

[0071] See also Figure 1 As shown, the present invention is a power material distribution management method based on big data, comprising the following steps:

[0072] S1: Obtain electronic material related data from power grid engineering projects and integrate them to obtain a power material database;

[0073] Classify the materials in the power material database;

[0074] S2: Obtain historical demand data of various materials in power grid projects within a certain period of time in the past, and build a material demand prediction model;

[0075] S3: Based on the material demand forecasting model, the demand for various materials in a certain period of time in the future is analyzed to obtain the demand frequency and quantity demand for various materials in a certain period of time in the future;

[0076] Count the quantity demand distribution of high-frequency demand materials and low-frequency demand materials;

[0077] Set quantity thresholds based on demand frequency to differentiate between large and small quantities of materials;

[0078] Combined with the quantity requirements, large-volume materials and small-volume materials are divided into standard packages to obtain large-volume material standard packages and small-volume material standard packages;

[0079] S4: Based on the bulk material packages, the status information of the power grid materials is obtained to build a network diagram;

[0080] Based on the network graph, the paths from each warehouse to each distribution point are generated, and the total transportation energy consumption from each warehouse to each distribution point is obtained;

[0081] Solve the minimum energy consumption for the total transportation energy consumption and generate the best path with the minimum total transportation energy consumption;

[0082] Optimize the delivery routes of large quantities of materials based on the best path;

[0083] Based on the standard packages of small batch materials, the demand data of small batch materials in the past period is extracted, and the standard deviation of demand and reorder cycle of small batch materials are calculated;

[0084] The safety stock of small batch materials is calculated through the standard deviation of demand and reorder cycle of small batch materials;

[0085] The safety stock level of small batch materials is processed with the actual stock of small batch materials to obtain the stock level of small batch materials;

[0086] Dynamically adjust distribution plans based on inventory levels and demand frequency for small-batch materials.

[0087] Example 2

[0088] This embodiment is a detailed description of steps S1-S2 in embodiment 1;

[0089] Obtain power material data for power grid projects;

[0090] The power material data includes: the name of the power material, the material description information corresponding to the material, and the supplier information corresponding to the material;

[0091] Material description information includes: material model and material parameters;

[0092] Supplier information includes: remaining quantity of power grid materials and supplier’s geographical location;

[0093] Integrate and process the power material data of the power grid project to obtain a power material database;

[0094] Based on the power material database, a clustering algorithm is used to preliminarily classify the materials according to their names, such as "cable", "transformer", "circuit breaker", etc.;

[0095] Based on the material description information, the keywords in the material description information, including material models and material parameters, are extracted by using NLP technology;

[0096] Clustering is performed based on the extracted keywords, that is, materials are classified according to different models and parameters to form material categories, such as dividing cables of different specifications into different subcategories;

[0097] Obtain historical demand record data for various materials within a certain period of time in the past;

[0098] The past certain period of time includes, but is not limited to, the past year and three years;

[0099] Use time series analysis methods to identify demand trends in historical demand record data for various materials, and extract demand-related feature data, including date, time, material category, and demand quantity;

[0100] It should be noted that the purpose of mining historical record data is to find out the law of material demand, so as to provide basic data support for prediction;

[0101] All characteristic data extracted from the power material database are integrated and processed in chronological order to obtain the time series data of material demand;

[0102] Based on the time series data of material demand, the ARIMA model is used to build a material demand forecasting model;

[0103] Among them, the ARIMA model, a time series analysis model, can be used to build a forecasting model for power materials, which is used to predict the demand and frequency of various materials in the future;

[0104] Based on the material demand forecasting model, the seasonal decomposition method is used to decompose the time series into three parts: trend, seasonality and residual;

[0105] Adjust the material demand forecast model based on seasonal components to reflect the impact of seasonal changes;

[0106] It should be noted that seasonal adjustment is to take into account the impact of seasonal factors on demand and ensure more accurate forecasts. By seasonally decomposing time series data, the seasonal components in the demand for power materials can be identified. Adjusting the forecast model based on seasonal components can reflect the impact of seasonal changes on material demand.

[0107] Use the adjusted material demand forecasting model to predict the demand for various materials within a certain period of time in the future, and obtain the demand frequency and quantity requirements for various materials.

[0108] Example 3

[0109] This embodiment is a detailed description of step S3 in embodiment 1;

[0110] See also Figure 2 As shown in the figure, based on the demand frequency and quantity requirements of the above-mentioned types of materials, the materials are divided into large-volume materials and small-volume materials. Specifically:

[0111] Based on the demand frequency of the material, the demand frequency of the material is compared with a demand frequency threshold;

[0112] If the demand frequency of a material is greater than or equal to the demand frequency threshold, it means that the material is a high-frequency demand material, and the material is marked as a high-frequency demand material;

[0113] If the demand frequency of a material is less than the demand frequency threshold, it means that the material is a low-frequency demand material, and the material is marked as a low-frequency demand material;

[0114] Based on the quantity demand of various materials, statistical analysis is performed on the demand for high-frequency and low-frequency materials to obtain the demand distribution of different materials;

[0115] Based on the demand distribution of different materials, set quantity thresholds according to demand frequency and quantity requirements;

[0116] Among them, a higher threshold is set for materials with high frequency and high demand, and a lower threshold is set for materials with low frequency or low demand. For example, the quantity threshold of materials with low frequency or low demand is set to 80% of the quantity threshold of materials with high frequency and high demand;

[0117] It should be noted that the quantity threshold is regularly adjusted based on actual demand and market changes. For example, seasonal changes, grid expansion and other factors may affect the setting of the threshold;

[0118] Based on the set quantity threshold, various types of materials are divided into large-volume materials and small-volume materials;

[0119] Based on large quantities of materials, materials of the same or similar types are combined into one package, which is recorded as a large-volume material package. For example, all types of cables are classified as one type of material in a package.

[0120] Combine related or complementary materials into a package, which is recorded as a bulk material package. For example, distribute cables together with matching connectors, terminals and other materials;

[0121] It should be noted that materials of the same type usually have similar transportation requirements. After merging, the transportation plan can be optimized, the number of transportations and costs can be reduced. Combining related materials can reduce the cost of multiple distributions, while ensuring that related materials can arrive at the same time, thereby improving construction efficiency.

[0122] Based on small batch materials, collect real-time demand data of small batch materials, understand the current actual demand situation, and obtain the actual demand;

[0123] According to the actual demand, small batches of materials are combined into a standard package, which is recorded as a small batch material standard package.

[0124] Example 4

[0125] This embodiment is a detailed description of step S4 in embodiment 1;

[0126] See also Figure 3 As shown, based on the above classification of large-volume materials and small-volume materials, further analysis is conducted on large-volume materials and small-volume materials to optimize the distribution strategy. Specifically:

[0127] Based on the bulk material packages, obtain the status information of power grid materials and the historical distribution path and time information;

[0128] Among them, the status information of power grid materials includes warehouse location information, distribution point location information and remaining quantity information;

[0129] Using GIS technology, a network map of all distribution points and warehouses was constructed;

[0130] Generate paths from each warehouse to each distribution point based on the network graph;

[0131] By formula Calculate the total transportation energy consumption Z from each warehouse to each distribution point;

[0132] Among them, c ij represents the unit energy consumption from warehouse i to distribution point j, x ij represents the delivery volume from warehouse i to delivery point j, i is the number of warehouses, i = 1, 2, ..., m, j is the number of delivery points, j = 1, 2, ..., n, adding constraints to warehouse capacity Add constraints to delivery point requirements Add constraints for non-negative states x ij ≥0, si is the inventory of warehouse i, dj is the demand of distribution point j, W is the number of warehouses, and D is the number of distribution points;

[0133] Based on the total transportation energy consumption from each warehouse to each distribution point, a linear programming algorithm is used to solve the minimum transportation energy consumption from each warehouse to each distribution point, find the optimal distribution plan, and generate the best path with the minimum total transportation energy consumption;

[0134] It should be noted that, according to the type and quantity of materials, large quantities of materials are aggregated into one or more standard packages, and the best route for centralized distribution is planned to reduce the total number of transportation times. Centralized distribution of large quantities of materials can ensure the safety and integrity of the materials during the distribution process;

[0135] Based on the bid packages of small batch materials, obtain the historical demand records of small batch materials and obtain the demand data in the past period of time, where the past period of time includes but is not limited to one year and two years;

[0136] Based on the demand data of small batch materials in the past period, the daily demand and total number of days in the past period are extracted;

[0137] By formula Calculate the average demand for small batches of materials over the past period of time μD, where D k is the demand on the kth day, N is the total number of days, k is the sequence number of days, k = 1, 2, ..., N;

[0138] Based on the mean demand μD of small batch materials in the past period of time, the standard deviation σD of the demand of small batch materials in the past period of time is calculated by the standard deviation formula;

[0139] Based on the historical demand records of small batch materials, the demand fluctuations of small batch materials are extracted to obtain the reorder cycle G of small batch materials;

[0140] By formula Calculate the safety stock L of small batch materials;

[0141] It should be noted that by calculating the safety stock of small batch materials, it is possible to ensure that the inventory is neither too much nor too little, thereby improving the efficiency and economic benefits of inventory management;

[0142] Based on the safety stock L, the actual inventory of small batch materials is compared with the safety stock;

[0143] If the actual inventory of small-batch materials is less than the actual inventory, an early warning message will be issued to remind relevant personnel to replenish the inventory;

[0144] Based on the inventory level and demand frequency of small batch materials, the distribution plan is dynamically adjusted according to the demand for small batch materials, so that the distribution plan can respond to changes in demand in a timely manner;

[0145] Set priorities according to the urgency of the needs and give priority to urgent needs;

[0146] Obtain customer needs and set delivery time windows based on the characteristics of small batch materials so that materials can arrive on time. At the same time, flexibly adjust the time windows according to actual conditions to cope with unexpected demands;

[0147] Obtain the warehouse location and distribution point location of small batch materials, and use the path optimization algorithm to plan the optimal distribution path. In this process, the real-time traffic and environmental conditions are input into the path planning, and the distribution path is adjusted in real time.

[0148] Monitor the time progress of the delivery process in real time to ensure that the delivery time matches the demand time.

[0149] Example 5

[0150] The present invention is a power material distribution management system based on big data, including a data collection module, a data processing module, a model building module and a material package division and strategy formulation module;

[0151] The data collection module is used to obtain the power material data of the power grid project and establish a unified power material database through integration;

[0152] The data processing module is used to perform preliminary and detailed classification of materials in the power material database and add labels to each type of material;

[0153] The model building module is used to obtain the historical demand data of various materials in the power grid project within a certain period of time in the past, and build a material demand prediction model;

[0154] According to the prediction results of the prediction model, the quantity demand distribution of high-frequency demand materials and low-frequency demand materials is obtained;

[0155] The material package division and strategy formulation module is used to set quantity thresholds according to quantity demand distribution and distinguish between large and small quantities of materials;

[0156] Combined with the demand frequency, large-volume materials and small-volume materials are divided into standard packages to obtain large-volume material standard packages and small-volume material standard packages;

[0157] Optimize the distribution strategy of bulk materials based on bulk material standard packages;

[0158] Based on small batch material standard packages, the small batch material distribution strategy is optimized.

[0159] The working principle of the present invention is as follows: first, the power material data in the power grid engineering project is obtained, including the material name, material description information and supplier information, and a unified power material database is established through integration; the materials are preliminarily classified using a clustering algorithm, and further refined according to the keywords in the material description information, and labels are added to each type of material to identify its main characteristics; the historical demand data of various types of materials in a certain period in the past are extracted from the power material database, the characteristic data related to the demand are extracted, and the characteristic data are subjected to correlation analysis to identify the characteristics of the main factors affecting the demand, and these characteristics are integrated to obtain the time series data of the material demand, and a material demand forecasting model is constructed based on the time series data using the ARIMA model to forecast the demand for various types of materials in the future, and the seasonal decomposition method is used to adjust the forecasting model; according to the results of the forecasting model, the demand frequency of various types of materials in the future time period is analyzed, the quantity demand distribution of high-frequency demand materials and low-frequency demand materials is statistically analyzed, and the quantity threshold is set according to the demand frequency and quantity demand to distinguish between large quantities of materials and small quantities of materials. ; Combine high-frequency and high-demand materials into one package, or combine interrelated or complementary materials into one package, and combine the real-time demand data of small-batch materials into one package; obtain the status information of power grid materials, use geographic information system technology to build a network diagram and generate the best distribution path, solve the minimum transportation energy consumption through linear programming algorithm, find the optimal distribution plan, and optimize the distribution strategy of large-volume materials based on this; extract the demand data of small-batch materials in the past period of time, calculate the demand mean and demand standard deviation, determine the reorder cycle, and calculate the safety stock to ensure that the inventory is neither too much nor too little. When the actual inventory is lower than the safety stock, issue an early warning message to remind you to replenish the inventory, and dynamically adjust the distribution plan according to the inventory level and demand frequency, respond to changes in demand in a timely manner, set priorities according to the urgency of demand, give priority to emergency needs, set delivery time windows, ensure that materials arrive on time, and use the path optimization algorithm to plan the optimal distribution path, make real-time adjustments to respond to sudden demands, monitor the time progress of the distribution process in real time, and ensure that the delivery time matches the demand time.

[0160] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A power material distribution management method based on big data, characterized in that: The following steps are involved: S1: Obtain the power material data of the power grid project and establish a unified power material database through integration; Classify the materials in the power material database; S2: Obtain historical demand data of various materials in power grid projects within a certain period of time in the past, and build a material demand prediction model; S3: According to the prediction results of the prediction model, the quantity demand distribution of high-frequency demand materials and low-frequency demand materials is obtained; Set quantity thresholds based on quantity demand distribution to distinguish between large and small quantities of materials; Combined with the demand frequency, large-volume materials and small-volume materials are divided into standard packages to obtain large-volume material standard packages and small-volume material standard packages; S4: Optimize the distribution strategy of bulk materials based on bulk material standard packages; Based on small batch material standard packages, the small batch material distribution strategy is optimized.

2. According to the big data-based power material distribution management method of claim 1, it is characterized in that: Power material data includes material name, material description information and supplier information; Material description information includes material model and parameters; Supplier information includes remaining quantity and geographic location of supplies.

3. According to the big data-based power material distribution management method of claim 1, it is characterized in that: Based on the power material database, the materials are preliminarily classified according to their names, and the materials are further classified according to their description information.

4. The method for managing power material distribution based on big data according to claim 1 is characterized in that: The specific process of building a material demand forecasting model is as follows: Obtain historical demand record data for various materials within a certain period of time in the past; The past certain period of time includes, but is not limited to, the past year and three years; Use time series analysis methods to identify demand trends in historical demand record data for various materials, and extract demand-related feature data, including date, time, material category, and demand quantity; All characteristic data extracted from the power material database are integrated and processed in chronological order to obtain the time series data of material demand; Build a material demand forecasting model based on the time series data of material demand; Based on the material demand forecasting model, the seasonal decomposition method is used to decompose the time series into three parts: trend, seasonality and residual; Adjust material demand forecast models based on seasonal components.

5. The method for managing power material distribution based on big data according to claim 4 is characterized in that: Based on the forecast results of the material demand forecasting model, the demand frequency and quantity requirements of various materials are obtained; Based on the demand frequency of materials, materials are divided into high-frequency demand materials and low-frequency demand materials; Conduct statistical analysis on the demand for high-frequency and low-frequency demand materials to obtain the demand distribution of different materials.

6. The method for managing power material distribution based on big data according to claim 1, characterized in that: The process of dividing large and small batches of materials into standard packages is as follows: Based on large quantities of materials, materials of the same or similar types are combined into one package, which is recorded as a large quantity material package; Combine related or complementary materials to form a package, which is recorded as a large-volume material package; Based on small batch materials, collect real-time demand data of small batch materials, extract the current actual demand situation, and obtain the actual demand quantity; According to the actual demand, small batches of materials are combined into a standard package, which is recorded as a small batch material standard package.

7. The method for managing power material distribution based on big data according to claim 6 is characterized in that: Based on large quantities of material packages, obtain the status information of power grid materials and historical distribution routes and time information; Among them, the status information of power grid materials includes warehouse location information, distribution point location information and remaining quantity information; Using GIS technology, a network map of all distribution points and warehouses was constructed; Generate paths from each warehouse to each distribution point based on the network graph; The path from each warehouse to each distribution point is calculated to obtain the total transportation energy consumption from each warehouse to each distribution point; The total transport energy consumption is analyzed and calculated to minimize the energy consumption, and the optimal route with the minimum total transport energy consumption is generated.

8. The power material distribution management method based on big data according to claim 7 is characterized in that: Based on the small batch material package, obtain the historical demand records of the small batch materials and obtain the demand data in the past period of time, where the past period of time includes but is not limited to one year and two years; Identify and extract the demand data of small batch materials in the past period of time to obtain the daily demand and total number of days in the past period of time; Based on the daily demand and total number of days for small batches of materials in the past period, the average demand for small batches of materials in the past period is calculated; The standard deviation of demand for small batches of materials over a period of time in the past is calculated through the demand mean; Based on the historical demand records of small batch materials, the demand fluctuations of small batch materials are extracted to obtain the reorder cycle of small batch materials; Based on the demand standard deviation and reorder cycle of small batch materials, the safety stock of small batch materials is calculated.

9. The power material distribution management method based on big data according to claim 8 is characterized in that: Based on the safety stock of small batch materials, obtain the actual inventory of small batch materials; Compare actual inventory of small-volume supplies with safety stock; When the actual inventory of small-batch materials is lower than the safety inventory, an early warning message is issued to remind you to replenish the inventory; When the actual inventory of small-batch materials is higher than or equal to the safety stock, the distribution plan is dynamically adjusted according to the actual inventory level and demand frequency, and the priority is set according to the urgency of demand.

10. A power material distribution management system based on big data, characterized in that: It includes data collection module, data processing module, model building module and material package division and strategy formulation module; The data collection module is used to obtain the power material data of the power grid project and establish a unified power material database through integration; The data processing module is used to classify the materials in the power material database; The model building module is used to obtain the historical demand data of various materials in the power grid project within a certain period of time in the past, and build a material demand prediction model; According to the prediction results of the prediction model, the quantity demand distribution of high-frequency demand materials and low-frequency demand materials is obtained; The material package division and strategy formulation module is used to set quantity thresholds according to quantity demand distribution and distinguish between large and small quantities of materials; Combined with the demand frequency, large-volume materials and small-volume materials are divided into standard packages to obtain large-volume material standard packages and small-volume material standard packages; Optimize the distribution strategy of bulk materials based on bulk material standard packages; Based on small batch material standard packages, the small batch material distribution strategy is optimized.

Images