Data processing method for multi-energy power grid

Abstract

The invention discloses a data processing method for a multi-energy power grid, and belongs to the technical field of multi-energy power grids, and the method comprises the following steps: S1, data collection, S2, data cleaning, S3, data storage, S4, data processing and S5, data visualization, the data acquisition further comprises hardware equipment selection, communication technology and protocol conversion. According to the data processing method of the multi-energy power grid, the energy utilization efficiency is improved, and through historical data analysis and a machine learning model, the power load demand can be predicted in advance, the power generation plan can be optimized, and the energy waste can be reduced, for example, the power generation redundancy rate is reduced by 15% through data-driven load prediction of a certain power grid, and power, heat energy and natural gas energy data are integrated; cooperative scheduling of cross-energy systems is achieved, for example, in the heating peak period in winter, power generation waste heat is used for heating through a combined heat and power generation system, and the comprehensive utilization rate of energy is increased by 20% or above.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-energy power grids, and in particular to a data processing method for a multi-energy power grid. Background Art

[0002] A multi-energy grid refers to a power system that can integrate multiple types of energy resources (such as fossil energy, renewable energy, nuclear energy, etc.) and energy carriers (such as electricity, heat, natural gas, etc.), and achieve efficient energy utilization, flexible allocation and supply and demand balance through advanced energy conversion, storage and transmission technologies. For example, in the prior art 1 (application number CN202210770589.8, Chinese patent application date 2022-06-30), a photovoltaic power prediction method for a multi-energy distribution network is used. By normalizing the data preprocessing, the preprocessed data is used to calculate the correlation coefficient of each influencing factor using the grey correlation degree, and the correlation coefficient of each influencing factor is calculated by combining the relevant Connection number, extract similar sample data that need to be input into the model, and improve the position update formula and the flame number update formula of the moth optimization algorithm to reduce the situation of falling into local optimality, and improve the convergence speed and convergence accuracy of the algorithm. Prior art 2 (application number CN202011182483.3, Chinese patent application date 2020-10-29) is a method for online analysis and processing of power grid big data, which achieves the goal of facilitating the power department to efficiently obtain power output analysis data, making it easier for the power department to transmit and distribute electric energy and adjust the voltage according to the actual power consumption of the power output terminal.

[0003] Although existing technologies can improve the data processing efficiency of power grids and thus enhance energy utilization efficiency, they only use simple threshold detection (e.g., “alarm when pressure > 1.2 MPa”) when processing data in multi-energy power grids. They are unable to identify complex abnormal patterns (e.g., a continuous slight drop in pressure caused by a slow leak). The missed detection rate is as high as 20%, resulting in a single abnormal data processing mechanism. Therefore, a data processing method for multi-energy power grids is proposed to address the above-mentioned issues. Summary of the Invention

[0004] The purpose of the present invention is to provide a data processing method for a multi-energy grid to solve the problem proposed in the above background technology that the current data processing methods on the market cannot identify complex abnormal patterns, the missed detection rate is as high as 20%, and the abnormal data processing mechanism is single.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solutions: a data processing method for a multi-energy grid, comprising the following steps; S1, data collection; S2, data cleaning; S3, data storage; S4, data processing; S5, data visualization; Data cleaning also includes deduplication, missing value completion, noise reduction and standardization.

[0006] Preferably, the data acquisition also includes hardware equipment selection, communication technology and protocol conversion; Hardware equipment selection, sensors: Use current sensors, voltage sensors, temperature sensors, etc. to monitor the operating parameters of power equipment in real time; Smart meters: Install smart meters to collect real-time power consumption data, including active power, reactive power, peak and valley power, etc. Data acquisition terminal: responsible for collecting data from multiple smart meters or sensors, performing pre-processing (such as data filtering, format conversion, and outlier processing), and supporting multiple communication protocols (such as Modbus and TCP / IP); Communication technologies mainly include wired communication, wireless communication, and protocol conversion. Since different devices use different communication protocols (such as Modbus, DL / T645, and IEC104), the acquisition gateway needs to perform protocol conversion to ensure that the data can be uniformly managed by the upper-level system.

[0007] Preferably, the deduplication in the data cleaning is to detect and delete duplicate data records through an algorithm to ensure the uniqueness of the data, and the missing values ​​are filled using statistical methods (such as mean, median) and model prediction to improve the integrity of the data.

[0008] Preferably, the noise reduction in the data cleaning is to remove noise and impurities in the data through filtering algorithms or other noise reduction technologies to ensure the authenticity and reliability of the data. Standardization is to unify data units and formats to ensure that all data are recorded with the same standards to facilitate subsequent comparison and analysis.

[0009] Preferably, the data storage includes a distributed database, cloud storage and a data warehouse; Among them, the distributed database stores data in multiple servers to achieve high availability and high throughput of data; Cloud storage is a storage service provided by cloud computing platforms that flexibly expands storage capacity to meet the needs of large-scale data storage; A data warehouse is a data storage system specifically used for analysis. It provides efficient data query and analysis capabilities through structured storage and indexing.

[0010] Preferably, the data processing includes data mining, multi-energy flow collaborative analysis, multi-objective optimization scheduling and multi-source data fusion; Data mining is the use of machine learning algorithms (such as regression analysis, time series analysis, and cluster analysis) to discover valuable information from massive data, such as power load forecasting, fault detection, and user behavior analysis.

[0011] Preferably, the multi-energy flow collaborative analysis utilizes edge computing technology to process and analyze data locally in real time, reduce dependence on upper-level systems, improve response speed, build an energy hub model, analyze the electricity-gas-heat coupling relationship (such as the waste heat generated by gas turbine power generation can be used for heating), and identify system bottlenecks through flow calculations.

[0012] Preferably, the multi-objective optimization scheduling includes an objective function, constraints and an algorithm; The objective function is to minimize comprehensive energy consumption costs, carbon emissions, and equipment wear, while maximizing renewable energy consumption; Constraints: power balance, gas pipeline flow limit, and thermal network temperature threshold; Algorithms: Particle swarm optimization and distributed optimization to solve nonlinear optimization problems of multi-energy coupling systems.

[0013] Preferably, the multi-source data fusion is to integrate information from different data sources and improve the accuracy and comprehensiveness of the data through data fusion technology.

[0014] Preferably, the data visualization includes chart presentation, dashboard and interactive analysis; The chart display uses pie charts, bar charts, line charts, and heat map charts to intuitively show the trends and patterns of the data; The dashboard is a real-time dashboard that displays the real-time status and key indicators of the power grid, helping managers quickly understand the operation status of the power grid; Interactive analysis provides functions such as drag-and-drop, data slicing, and drilling, supporting users to conduct in-depth data analysis.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) Improve energy utilization efficiency. Through historical data analysis and machine learning models, power load demand can be predicted in advance, power generation plans can be optimized, and energy waste can be reduced. The power grid has reduced the power generation redundancy rate by 15% through data-driven load forecasting, integrated energy data such as electricity, heat, and natural gas, and achieved coordinated scheduling across energy systems. During the winter heating peak, the waste heat from power generation can be used for heating through the cogeneration system, and the comprehensive energy utilization rate has increased by more than 20%.

[0016] (2) Enhance the reliability and stability of the power grid. By using sensor data and real-time analysis technology, abnormal conditions in the power grid (such as voltage fluctuations and equipment overheating) can be quickly detected, and the fault point can be accurately located. Through real-time data analysis, the power grid can shorten the fault response time from hours to minutes. Through data-driven distributed energy resource scheduling, the fluctuations of intermittent energy such as distributed photovoltaic and wind power can be balanced, reducing the impact on the main power grid. Through distributed energy management, the power grid has reduced the wind and solar power curtailment rate by 30%.

[0017] (3) Promote the consumption of renewable energy. Based on meteorological data and historical power generation data, a renewable energy power generation forecast model is established to optimize the scheduling strategy. Wind farms use data-driven power generation forecasts to increase the accuracy of power generation plans to over 90%, significantly improving wind power consumption capacity. Through data analysis, the charging and discharging strategies of energy storage systems are optimized to balance the intermittent nature of renewable energy and the stability of the power grid. For example, a power grid has increased the proportion of renewable energy connected to the grid by 25% through energy storage system optimization.

[0018] (4) Reduce operating costs. Through sensor data and machine learning models, equipment status can be monitored in real time, equipment failures can be predicted, and preventive maintenance can be achieved. The power grid has reduced equipment failure rates by 40% and maintenance costs by 30% through equipment status monitoring. Through market data analysis and optimization algorithms, the optimal energy trading strategy is formulated to reduce electricity purchase costs. Power companies have reduced electricity purchase costs by 10% through data-driven energy trading.

[0019] (5) Promote the intelligent and digital transformation of energy systems, and provide data-driven decision support for grid operations through big data analysis and visualization technology. Through real-time dashboards, managers can quickly understand the grid operation status and make scientific decisions. Through data sharing and collaboration, the construction of the energy Internet is promoted to achieve global optimal energy allocation. Regional energy Internet realizes cross-system collaboration of electricity, heat and natural gas through data sharing, and energy utilization efficiency is improved by 25%. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a data processing flow chart of the present invention; Figure 2 This is a data collection flow chart of the present invention; Figure 3 This is a data cleaning flow chart of the present invention; Figure 4 This is a data storage flow chart of the present invention; Figure 5 This is a data processing flow chart of the present invention; Figure 6 This is a data visualization flow chart of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0022] The present invention provides the following technical solution, a data processing method for a multi-energy grid: Example 1

[0023] The following steps are included: S1, data collection; S2, data cleaning; S3, data storage; S4, data processing; S5, data visualization; Data cleaning also includes deduplication, missing value completion, noise reduction and standardization.

[0024] Data collection also includes hardware equipment selection, communication technology and protocol conversion; Hardware equipment selection, sensors: Use current sensors, voltage sensors, temperature sensors, etc. to monitor the operating parameters of power equipment in real time; Smart meters: Install smart meters to collect real-time power consumption data, including active power, reactive power, peak and valley power, etc. Data acquisition terminal: responsible for collecting data from multiple smart meters or sensors, performing pre-processing (such as data filtering, format conversion, and outlier processing), and supporting multiple communication protocols (such as Modbus and TCP / IP); Communication technologies mainly include wired communication, wireless communication, and protocol conversion. Since different devices use different communication protocols (such as Modbus, DL / T645, and IEC104), the acquisition gateway needs to perform protocol conversion to ensure that the data can be uniformly managed by the upper-level system.

[0025] Deduplication in data cleaning is to detect and delete duplicate data records through algorithms to ensure the uniqueness of the data. Missing values ​​are filled using statistical methods (such as mean, median) and model predictions to improve data integrity.

[0026] Noise reduction in data cleaning is to remove noise and impurities in the data through filtering algorithms or other noise reduction technologies to ensure the authenticity and reliability of the data. Standardization is to unify data units and formats to ensure that all data are recorded with the same standards to facilitate subsequent comparison and analysis.

[0027] Data storage includes distributed databases, cloud storage, and data warehouses; Among them, the distributed database stores data in multiple servers to achieve high availability and high throughput of data; Cloud storage is a storage service provided by cloud computing platforms that flexibly expands storage capacity to meet the needs of large-scale data storage; A data warehouse is a data storage system specifically used for analysis. It provides efficient data query and analysis capabilities through structured storage and indexing.

[0028] Data processing includes data mining, multi-energy flow collaborative analysis, multi-objective optimization scheduling and multi-source data fusion; Data mining is the use of machine learning algorithms (such as regression analysis, time series analysis, and cluster analysis) to discover valuable information from massive data, such as power load forecasting, fault detection, and user behavior analysis.

[0029] Collaborative analysis of multiple energy flows utilizes edge computing technology to process and analyze data locally in real time, reducing dependence on upper-level systems and improving response speed. This allows the construction of energy hub models, analysis of the electricity-gas-heat coupling relationship (e.g., waste heat generated by gas turbines during power generation can be used for heating), and identification of system bottlenecks through flow calculations.

[0030] Multi-objective optimization scheduling includes objective functions, constraints and algorithms; The objective function is to minimize comprehensive energy consumption costs, carbon emissions, and equipment wear, while maximizing renewable energy consumption; Constraints: power balance, gas pipeline flow limit, and thermal network temperature threshold; Algorithms: Particle swarm optimization and distributed optimization to solve nonlinear optimization problems of multi-energy coupling systems.

[0031] Multi-source data fusion is the process of integrating information from different data sources and improving the accuracy and comprehensiveness of data through data fusion technology.

[0032] Data visualization includes chart presentation, dashboards, and interactive analysis; The chart display uses pie charts, bar charts, line charts, and heat map charts to intuitively show the trends and patterns of the data; The dashboard is a real-time dashboard that displays the real-time status and key indicators of the power grid, helping managers quickly understand the operation status of the power grid; Interactive analysis provides functions such as drag-and-drop, data slicing, and drilling, supporting users to conduct in-depth data analysis.

[0033] Example 2: Optimal Scheduling of Multi-Energy Microgrids in Industrial Parks 1. Data acquisition and preprocessing: Sensors collect real-time operating data from various energy devices, such as the output power of photovoltaic panels, the speed and power generation of wind turbines, the gas consumption and power generation of gas turbines, and the charge and discharge status and power of energy storage systems. Edge computing nodes are used to perform preliminary cleaning of the raw data to remove outliers and noise. For example, a sliding average filter algorithm is used to process fan vibration data to smooth out abnormal fluctuations in vibration amplitude, ensuring that the data accurately reflects the actual operating status of the equipment and reducing misjudgments caused by sensor interference. 2. Load Forecasting and Energy Synergy Analysis: Long-short-term memory networks are used in conjunction with historical load data and meteorological data (such as sunlight intensity, wind speed, and temperature) to predict the park's electrical and thermal loads. Based on the prediction results, an electricity-gas-heat coupling model is established to analyze the conversion and synergy between different energy sources. For example, when there is sufficient sunlight and strong wind power generation, but the electrical load is low, excess electricity can be stored as thermal energy through power-to-heat equipment or used to charge the energy storage system. When the gas supply is sufficient and the electricity price is high, gas turbines are prioritized for power generation. 3. Optimizing scheduling decisions: With the lowest overall operating cost and carbon emissions as the objective function, and considering equipment operating constraints (such as the upper and lower power limits of power generation equipment and the charge and discharge rate limits of energy storage systems), a particle swarm optimization algorithm was used to solve the problem. After optimized scheduling, the industrial park's comprehensive energy utilization efficiency increased from 65% to 80%, annual operating costs decreased by 15%, and carbon emissions were reduced by 20%.

[0034] Example 3: Cross-system fault diagnosis in a city energy data center 1. Multi-source data integration and synchronization: Utilize unified data interface protocols such as OPC UA to integrate SCADA data and metering data from different systems. High-precision time synchronization technology ensures the temporal consistency of data from each energy system, with errors controlled to the microsecond level. For example, precise time alignment can be achieved between the voltage and current data of the power grid and the pressure and flow data of the gas grid to facilitate subsequent cross-system analysis.

[0035] 2. Fault feature extraction and diagnostic model construction: For different energy system fault types, corresponding feature quantities are extracted. For example, in power grid fault diagnosis, the harmonic characteristics of voltage and current are analyzed through Fourier transform. In gas grid fault diagnosis, the dynamic time warping algorithm is used to compare the difference between the real-time pressure curve and the normal operating condition template. A multi-classification fault diagnosis model based on support vector machines is constructed. The extracted features are used as model input to identify and locate the fault type.

[0036] 3. Fault diagnosis and emergency response: When an energy supply anomaly occurs in a certain area of ​​the city, the data center quickly calls the fault diagnosis model for analysis. In a gas pipeline leak incident, the abnormal downward trend of gas pressure was promptly detected through data processing. Combined with the power fluctuations of surrounding power grid nodes (because some power plants that rely on gas-fired power generation were affected), the fault location was quickly located. The system immediately activated the emergency response mechanism, notifying the gas repair department while coordinating the power grid to adjust the power supply strategy to ensure the power supply of critical loads. Compared with traditional single-system fault diagnosis, the fault location time was shortened from an average of 15 minutes to less than 5 minutes, greatly improving the reliability and safety of the city's energy supply.

[0037] Example 4: Energy management and energy saving optimization for residential users 1. User energy consumption data collection and profile construction: Collect residents' daily, weekly, and monthly energy consumption data, clean and organize the data, and use the K-means clustering algorithm to classify residential users into different categories based on energy consumption levels, peak electricity and gas consumption periods, such as high-energy-consuming users, energy-saving users, and peak-hour electricity-intensive users, to construct detailed user energy profiles.

[0038] 2. Energy-saving strategy formulation and promotion: Develop personalized energy-saving strategies for different types of users. For high-energy-consuming users, analyze the reasons for their high energy consumption (such as long-term use of high-power appliances and low energy equipment efficiency), and promote energy-saving transformation suggestions, such as replacing energy-saving lamps and upgrading high-efficiency water heaters, and provide relevant product coupons. For users with concentrated electricity consumption during peak hours, promote peak and valley electricity price information and peak-shifting electricity consumption suggestions through the electricity price incentive mechanism, guiding users to use adjustable load equipment (such as washing machines and electric vehicle charging) during off-peak hours.

[0039] 3. Energy-saving effect evaluation: After a period of implementation of the energy-saving strategy, the energy-saving effect was evaluated by comparing the energy consumption data of users before and after. The overall energy consumption of residents in the community was reduced by 12%, among which the average energy consumption of high-energy-consuming users decreased by 20%, and the electricity consumption of peak-hour concentrated users during peak hours decreased by 18%, effectively improving residents' energy-saving awareness and energy utilization efficiency.

[0040] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A data processing method for a multi-energy grid, comprising the following steps: S1, data collection; S2, data cleaning; S3, data storage; S4, data processing; S5, data visualization; Data cleaning also includes deduplication, missing value completion, noise reduction and standardization.

2. The data processing method of a multi-energy grid according to claim 1, characterized in that: Said data collection also includes hardware equipment selection, communication technology and protocol conversion; Hardware equipment selection, sensors: Use current sensors, voltage sensors, and temperature sensors to monitor the operating parameters of power equipment in real time; Smart meters: Install smart meters to collect real-time power consumption data, including active power, reactive power, and peak and valley power; Data acquisition terminal: responsible for collecting data from multiple smart meters or sensors, performing pre-processing, and supporting multiple communication protocols; Communication technologies mainly include wired communication, wireless communication, and protocol conversion. Since different devices use different communication protocols, the acquisition gateway needs to perform protocol conversion to ensure that the data can be uniformly managed by the upper-level system.

3. The data processing method of a multi-energy grid according to claim 1, characterized in that: The deduplication in the data cleaning is to detect and delete duplicate data records through algorithms to ensure the uniqueness of the data, and to fill in missing values ​​using statistical methods and model predictions to improve the integrity of the data.

4. The data processing method of a multi-energy grid according to claim 1, characterized in that: The noise reduction in the data cleaning is to remove noise and impurities in the data through filtering algorithms to ensure the authenticity and reliability of the data. Standardization is to unify data units and formats to ensure that all data are recorded with the same standards to facilitate subsequent comparison and analysis.

5. The data processing method of a multi-energy grid according to claim 1, characterized in that: The data storage includes distributed databases, cloud storage and data warehouses; Among them, the distributed database stores data in multiple servers to achieve high availability and high throughput of data; Cloud storage is a storage service provided by cloud computing platforms that flexibly expands storage capacity to meet the needs of large-scale data storage; A data warehouse is a data storage system specifically used for analysis. It provides efficient data query and analysis capabilities through structured storage and indexing.

6. The data processing method of a multi-energy grid according to claim 1, characterized in that: Said data processing includes data mining, multi-energy flow collaborative analysis, multi-objective optimization scheduling and multi-source data fusion; Data mining is the use of machine learning algorithms to discover valuable information from massive data, such as power load forecasting, fault detection, and user behavior analysis.

7. The data processing method of a multi-energy grid according to claim 6, characterized in that: The multi-energy flow collaborative analysis utilizes edge computing technology to process and analyze data locally in real time, reduce dependence on upper-level systems, improve response speed, build an energy hub model, analyze the electricity-gas-heat coupling relationship, and identify system bottlenecks through flow calculations.

8. The data processing method of a multi-energy grid according to claim 6, characterized in that: The multi-objective optimization scheduling includes objective functions, constraints and algorithms; The objective function is to minimize comprehensive energy consumption costs, carbon emissions, and equipment wear, while maximizing renewable energy consumption; Constraints: power balance, gas pipeline flow limit, and thermal network temperature threshold; Algorithm: Particle swarm optimization and distributed optimization to solve nonlinear optimization problems of multi-energy coupling systems.

9. The data processing method of a multi-energy grid according to claim 6, characterized in that: The multi-source data fusion is to integrate information from different data sources and improve the accuracy and comprehensiveness of the data through data fusion technology.

10. The data processing method of a multi-energy grid according to claim 1, characterized in that: The data visualization includes chart presentation, dashboard and interactive analysis; The chart display uses pie charts, bar charts, line charts, and heat map charts to intuitively show the trends and patterns of the data; The dashboard is a real-time dashboard that displays the real-time status and key indicators of the power grid, helping managers quickly understand the operation status of the power grid; Interactive analysis provides drag-and-drop operations, data slicing, and drilling functions, supporting users to conduct in-depth data analysis.

Images