Power grid project supply chain demand prediction and optimization system based on big data analysis

Abstract

The invention discloses a power grid project supply chain demand prediction and optimization system based on big data analysis. The system comprises a data acquisition module, a data preprocessing module, a dynamic demand prediction module, a supply chain optimization decision module, a real-time cooperative control module and a risk early warning and self-healing module. The data acquisition module is connected with an internet of things sensor, an engineering management system API and an external data interface. According to the invention, a full-chain intelligent system of data acquisition, demand prediction, dynamic optimization and risk management and control is constructed, so that the power grid project supply chain management efficiency is remarkably improved. The system fuses multi-source heterogeneous data, and adopts a machine learning model to adaptively correct prediction deviation, so that the material demand prediction precision is improved; and a multi-objective optimization decision is driven based on a dynamic prediction result, a global optimal scheme of purchase, storage and logistics is generated, redundant inventory is effectively reduced, and a material allocation period is shortened.

Description

Technical Field

[0001] The present application relates to the field of power grid engineering technology, and specifically to a power grid engineering supply chain demand forecasting and optimization system based on big data analysis. Background Art

[0002] Current power grid construction projects face challenges such as delayed supply chain response and imbalanced material supply and demand. Traditional forecasting methods, relying on historical experience and static models, struggle to adapt to the dynamic demands of power grid projects, often leading to inventory backlogs or supply shortages.

[0003] As smart grid construction expands, the variety of materials and complexity of supply chains surge. Big data technologies are urgently needed to integrate multi-source, heterogeneous data (such as project progress, weather conditions, and equipment failure records) to achieve accurate demand forecasting and dynamic resource allocation. Existing systems lack the ability to coordinate and optimize the entire supply chain in real time, making them unable to effectively address sudden project changes or supply chain disruptions. Summary of the Invention

[0004] To this end, this application provides a power grid engineering supply chain demand forecasting and optimization system based on big data analysis to solve the problem that the existing system lacks the ability to coordinate and optimize the entire supply chain in real time and cannot effectively respond to sudden engineering changes or supply chain interruption risks.

[0005] In order to achieve the above objectives, this application provides the following technical solutions:

[0006] The power grid engineering supply chain demand forecasting and optimization system based on big data analysis includes data acquisition module, data preprocessing module, dynamic demand forecasting module, supply chain optimization decision module, real-time collaborative control module and risk warning and self-healing module.

[0007] The data acquisition module uses IoT sensors, engineering management system APIs, and external data interfaces to obtain real-time data on power grid project material consumption, construction progress, meteorological and environmental data, supplier production capacity, and logistics status.

[0008] The data preprocessing module cleans, denoises, aligns time and space, and extracts features from multi-source heterogeneous data to generate a standardized supply chain analysis dataset.

[0009] The dynamic demand forecasting module forms a forecasting model based on the fusion of long short-term memory network and gradient boosting decision tree. The dynamic demand forecasting module associates the standardized supply chain analysis data set and outputs the forecast results of material demand;

[0010] The supply chain optimization decision module uses the forecast results as input and combines data on inventory costs, transportation constraints, and supplier ratings to generate procurement plans, warehouse allocation, and logistics routing solutions;

[0011] The real-time collaborative control module is used to visualize the status of the entire supply chain, receive manual correction instructions, and dynamically adjust optimization strategies;

[0012] The risk warning and self-healing module is used to monitor abnormal events in the supply chain. When an abnormal event occurs to a supplier in the supply chain, the preset emergency plan is automatically triggered and the alarm information is pushed to the related terminal.

[0013] The data acquisition module includes an RFID material tracking unit, a construction machinery status monitoring unit, and a meteorological microstation deployed at the construction site to collect material usage locations, equipment operating conditions, and local environmental parameters in real time.

[0014] The data preprocessing module includes an abnormal data adaptive filtering submodule, which uses the isolation forest algorithm to identify outliers in the collected data and automatically repair missing values ​​based on the distribution characteristics of historical data.

[0015] The dynamic demand forecasting module is integrated with a project stage identifier, which analyzes construction drawings and progress images through a convolutional neural network, automatically divides the project into stages and associates them with a stage-dependent material demand model.

[0016] The supply chain optimization decision module includes a multi-level inventory collaborative optimization sub-module, which is used to establish a three-level inventory network model of central warehouse-regional warehouse-construction site temporary warehouse, and dynamically calculate the optimal replenishment strategy with the goal of minimizing transportation costs and out-of-stock losses.

[0017] The real-time collaborative control module can generate a digital twin view of the supply chain, which maps the physical supply chain status to a virtual model. The digital twin view can adjust and optimize the replenishment strategy by dragging and dropping and simulate the replenishment results in real time.

[0018] The risk warning and self-healing module includes a supplier risk rating library, which includes supplier industrial and commercial credit data, historical delivery records and public opinion data to generate a dynamic credit score for suppliers and automatically trigger a dual-source procurement strategy for low-scoring suppliers.

[0019] It also includes a low-carbon optimization constraint module, which embeds a carbon emission calculation model in supply chain decisions, converts transportation methods, packaging materials, and warehousing energy consumption into carbon footprint data, and generates the optimal solution that meets the carbon neutrality goal.

[0020] Compared with the prior art, this application has at least the following beneficial effects:

[0021] This invention significantly improves the efficiency of power grid project supply chain management by building a comprehensive intelligent system encompassing data collection, demand forecasting, dynamic optimization, and risk management. The system integrates heterogeneous data from multiple sources and uses machine learning models to adaptively correct forecast deviations, improving the accuracy of material demand forecasts. Dynamic forecast results drive multi-objective optimization decisions, generating globally optimal solutions for procurement, warehousing, and logistics, effectively reducing excess inventory and shortening material allocation cycles.

[0022] Real-time monitoring of the supply chain status is achieved through the digital twin visualization platform. When project changes, logistics disruptions or supplier risks are detected, emergency plans are automatically triggered and resource allocation is dynamically adjusted, greatly reducing the response time to emergencies and significantly enhancing the risk resistance and continuity of power grid project construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a module diagram of the power grid engineering supply chain demand forecasting and optimization system based on big data analysis in this application. DETAILED DESCRIPTION

[0024] The present application will be further described below in detail through specific embodiments in conjunction with the accompanying drawings.

[0025] like Figure 1 As shown, this application discloses a power grid engineering supply chain demand forecasting and optimization system based on big data analysis, including a data acquisition module, a data preprocessing module, a dynamic demand forecasting module, a supply chain optimization decision module, a real-time collaborative control module, and a risk warning and self-healing module.

[0026] The data acquisition module uses IoT sensors, engineering management system APIs, and external data interfaces to obtain real-time data on power grid project material consumption, construction progress, meteorological and environmental data, supplier production capacity, and logistics status, solving the problem of a single data source in traditional systems.

[0027] The data preprocessing module cleans, denoises, aligns time and space, and extracts features from multi-source heterogeneous data to generate a standardized supply chain analysis dataset.

[0028] The dynamic demand forecasting module forms a forecasting model based on the fusion of long short-term memory networks and gradient boosting decision trees. The module associates standardized supply chain analysis datasets and outputs forecast results for material demand. The module combines time series models with feature engineering to adapt to changes in nonlinear demand for power grid projects.

[0029] The supply chain optimization decision module uses forecast results as input and combines data on inventory costs, transportation constraints, and supplier ratings to generate procurement plans, warehouse allocation, and logistics routing solutions. It converts forecast results into executable plans to ensure the global optimization of supply chain costs and efficiency.

[0030] The real-time collaborative control module is used to visualize the status of the entire supply chain, receive manual correction instructions, and dynamically adjust optimization strategies;

[0031] The risk warning and self-healing module is used to monitor abnormal events in the supply chain. When an abnormal event occurs to a supplier in the supply chain, the preset emergency plan is automatically triggered and the alarm information is pushed to the related terminal.

[0032] The data acquisition module includes an RFID material tracking unit, a construction machinery status monitoring unit, and a meteorological microstation deployed at the construction site to collect real-time information on the location of material use, equipment operating conditions, and local environmental parameters. The RFID unit enables full traceability of material flow, avoiding errors in traditional manual recording; machinery status monitoring correlates construction efficiency with material consumption rate, improving the timeliness of demand forecasting; and the meteorological microstation captures local environmental changes on the construction site (such as temperature, humidity, and wind speed) to provide data support for supply chain adjustments in severe weather. The three work together to form a high-precision data perception network, resolving the problem that macro data cannot reflect local dynamics. For example, in power transmission projects, by deploying RFID, the movement trajectory of cable reels can be tracked to facilitate cable management.

[0033] The data preprocessing module includes an abnormal data adaptive filtering submodule, which uses the isolation forest algorithm to identify outliers in the collected data and automatically repairs missing values ​​based on the distribution characteristics of historical data. This enables the solution to identify abnormal values ​​caused by equipment failure or human error, avoiding abnormal data contamination of the prediction model; at the same time, it combines the missing value repair mechanism of historical distribution to ensure data continuity.

[0034] The dynamic demand forecasting module integrates a project phase identifier. Using a convolutional neural network to analyze construction drawings and progress images, it automatically divides power grid projects into phases and associates them with a phase-dependent material demand model. By analyzing construction images, it identifies the current phase (e.g., civil engineering, installation, commissioning) and correlates material consumption patterns across different phases. When a phase transition is detected, the system automatically switches forecasting model parameters. In power grid projects where project progress isn't easily observable, such as tunnel construction, the module uses the shield machine's tunneling speed to determine the project's progress and predict grid resource demand for deployment.

[0035] The supply chain optimization decision module includes a multi-level inventory collaborative optimization submodule, which is used to establish a three-level inventory network model consisting of a central warehouse, a regional warehouse, and a temporary construction site warehouse. This submodule dynamically calculates the optimal replenishment strategy with the goal of minimizing transportation costs and out-of-stock losses. By establishing a mathematical model for this three-level inventory network, it comprehensively considers storage costs, transportation timeliness, and fluctuations in construction site demand, minimizing total costs while ensuring construction continuity.

[0036] The real-time collaborative control module can generate a digital twin view of the supply chain, which maps the physical supply chain status to a virtual model. The digital twin view can adjust and optimize the replenishment strategy by dragging and dropping and simulate the replenishment results in real time.

[0037] Users can adjust parameters within the virtual environment (e.g., adding suppliers, modifying transportation routes), and the system calculates the resulting cost and delivery cycle changes in real time, presenting risk points in the form of a heat map. For example, the digital twin view can simulate a typhoon-induced route disruption, automatically highlighting affected projects and alternative options. This lowers the operational decision threshold for the solution and enhances managers' emergency response capabilities.

[0038] The risk warning and self-healing module includes a supplier risk rating library, which includes supplier industrial and commercial credit data, historical delivery records, and public opinion data to generate dynamic credit scores for suppliers, automatically triggering a dual-source procurement strategy for low-scoring suppliers. Through multi-dimensional data analysis, the reliability of suppliers (such as fine records, negative news) is evaluated in real time. When the credit score is lower than the threshold, the system automatically splits the order to backup suppliers and uses blockchain to certify the breach of contract, thereby avoiding the risk of construction suspension due to supply chain problems. For some higher-priced equipment, such as transformers, this solution can retrieve the supplier's court default record to ensure the security of procurement.

[0039] It also includes a low-carbon optimization constraint module, which embeds a carbon emission calculation model in supply chain decisions, converts transportation methods, packaging materials, and warehousing energy consumption into carbon footprint data, and generates the optimal solution that meets the carbon neutrality goal.

[0040] The technical features of the above embodiments can be combined arbitrarily (as long as there is no contradiction in the combination of these technical features). In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described; these embodiments that are not explicitly written should also be considered to be within the scope of this specification.

Claims

1. A power grid engineering supply chain demand forecasting and optimization system based on big data analysis, characterized by: It includes data acquisition module, data preprocessing module, dynamic demand forecasting module, supply chain optimization decision module, real-time collaborative control module and risk warning and self-healing module; The data acquisition module uses IoT sensors, engineering management system APIs, and external data interfaces to obtain real-time data on power grid project material consumption, construction progress, meteorological and environmental data, supplier production capacity, and logistics status. The data preprocessing module cleans, denoises, aligns time and space, and extracts features from multi-source heterogeneous data to generate a standardized supply chain analysis dataset. The dynamic demand forecasting module forms a forecasting model based on the fusion of long short-term memory network and gradient boosting decision tree. The dynamic demand forecasting module associates the standardized supply chain analysis data set and outputs the forecast results of material demand; The supply chain optimization decision module uses the forecast results as input and combines data on inventory costs, transportation constraints, and supplier ratings to generate procurement plans, warehouse allocation, and logistics routing solutions; The real-time collaborative control module is used to visualize the status of the entire supply chain, receive manual correction instructions, and dynamically adjust optimization strategies; The risk warning and self-healing module is used to monitor abnormal events in the supply chain. When an abnormal event occurs to a supplier in the supply chain, the preset emergency plan is automatically triggered and the alarm information is pushed to the related terminal.

2. The power grid engineering supply chain demand forecasting and optimization system based on big data analysis according to claim 1 is characterized in that: The data acquisition module includes an RFID material tracking unit, a construction machinery status monitoring unit, and a meteorological microstation deployed at the construction site to collect material usage locations, equipment operating conditions, and local environmental parameters in real time.

3. The power grid engineering supply chain demand forecasting and optimization system based on big data analysis according to claim 1 is characterized in that: The data preprocessing module includes an abnormal data adaptive filtering submodule, which uses the isolation forest algorithm to identify outliers in the collected data and automatically repair missing values ​​based on the distribution characteristics of historical data.

4. The power grid engineering supply chain demand forecasting and optimization system based on big data analysis according to claim 1 is characterized in that: The dynamic demand forecasting module is integrated with a project stage identifier, which analyzes construction drawings and progress images through a convolutional neural network, automatically divides the project into stages and associates them with a stage-dependent material demand model.

5. The power grid engineering supply chain demand forecasting and optimization system based on big data analysis according to claim 1 is characterized in that: The supply chain optimization decision module includes a multi-level inventory collaborative optimization sub-module, which is used to establish a three-level inventory network model of central warehouse-regional warehouse-construction site temporary warehouse, and dynamically calculate the optimal replenishment strategy with the goal of minimizing transportation costs and out-of-stock losses.

6. The power grid engineering supply chain demand forecasting and optimization system based on big data analysis according to claim 1 is characterized in that: The real-time collaborative control module can generate a digital twin view of the supply chain, which maps the physical supply chain status to a virtual model. The digital twin view can adjust and optimize the replenishment strategy by dragging and dropping and simulate the replenishment results in real time.

7. The power grid engineering supply chain demand forecasting and optimization system based on big data analysis according to claim 1 is characterized in that: The risk warning and self-healing module includes a supplier risk rating library, which includes supplier industrial and commercial credit data, historical delivery records and public opinion data to generate a dynamic credit score for suppliers and automatically trigger a dual-source procurement strategy for low-scoring suppliers.

8. The power grid engineering supply chain demand forecasting and optimization system based on big data analysis according to claim 1 is characterized in that: It also includes a low-carbon optimization constraint module, which embeds a carbon emission calculation model in supply chain decisions, converts transportation methods, packaging materials, and warehousing energy consumption into carbon footprint data, and generates the optimal solution that meets the carbon neutrality goal.

Images