Bus power balance adjusting method

Abstract

The invention belongs to the technical field of power balance adjustment methods, and particularly relates to a bus power balance adjustment method, which predicts and optimizes a power grid load, guides the adjustment of output power of a generator, realizes the balance of bus power, introduces energy storage equipment, and adopts a distributed control strategy for a power system. A power system is divided into a plurality of small areas, a localized control and optimization strategy is adopted in each small area, energy storage equipment is arranged in each small area, energy storage and release are performed to maintain bus power balance adjustment, and power balance control of a whole power grid is realized through cross-area coordination. Therefore, intelligent operation is realized, and the stability and balance capability of the system are improved.

Description

Technical Field

[0001] The invention belongs to the technical field of power balance adjustment methods, and in particular relates to a bus power balance adjustment method. Background Art

[0002] Bus power balancing regulation refers to the coordinated control of the output power of each generator in the power system to maintain the power balance and stable operation of the entire power grid. It mainly includes the following aspects: monitoring the load changes of the power system, and collecting the output power data of each generator in real time, processing and analyzing the real-time collected data, establishing a dynamic power balance model, and calculating the optimal output power of each generator. Sending instructions to each generator through the control system, adjusting the output power of the generator in real time, and intelligently optimizing and controlling the power system, including using technologies such as artificial intelligence to optimize the output power of the generator and improve the effect of bus power balancing, establishing a monitoring and alarm mechanism, and promptly discovering and solving bus power imbalance problems to ensure the safe and stable operation of the power system.

[0003] However, the bus power balancing regulation in the prior art still has the problem of being unable to conveniently control the output power of each generator and unable to achieve a better balanced distribution. Secondly, there is no energy storage device to assist. Therefore, a bus power balancing regulation method is designed to solve the above problems. Summary of the invention

[0004] In order to overcome the problems in the prior art, the present invention proposes a bus power balance adjustment method.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] The present invention provides a bus power balance adjustment method, comprising the following steps:

[0007] Step 1: Predict the grid load and optimize it to guide the power output adjustment of the generator;

[0008] Step 2: Introduce energy storage equipment and adopt a distributed control strategy for the power system, which is used to ensure that the power system can flexibly respond to load changes;

[0009] Step 3: Divide the power system into multiple small areas, and adopt a localized control and optimization strategy in each small area, wherein the localized control and optimization strategy is used to ensure power balance and independent regulation capability in each area;

[0010] Step 4: Store and release energy in the energy storage devices within each small area to maintain bus power balance regulation within the area and provide support for global regulation;

[0011] Step 5: Through cross-regional coordination, based on steps 3 and 4, the power balance conditions of each small area are integrated to achieve global power balance control of the entire power grid.

[0012] Furthermore, in step 1, the grid load is predicted and optimized, including the following steps:

[0013] S11: Collecting data in the power system, including real-time monitoring of power grid operation status, load conditions, temperature, humidity, and power;

[0014] S12: Build models to describe the behavior of the power system:

[0015] The model includes a load curve, a generator model, a transmission line model, a renewable energy model and a control model;

[0016] Among them, the load curve is used to determine the maximum load per day, week or month;

[0017] The generator model is used to describe the performance of the generator regulator at different load levels and determine what power output can be supported;

[0018] The transmission line model is used to describe the flow of electrical energy in the transmission line, including voltage reduction and common AC transmission effects;

[0019] The renewable energy model is used to understand the timing and power generation information captured by it;

[0020] The control model is used to describe the various devices that monitor network conditions to measure status and change settings to calibrate the flow of energy;

[0021] S13: Determine an objective function, wherein the objective function is to maximize profit or minimize cost under a given load, wherein the load level is a variable, and the operating cost and revenue need to be maximized or minimized;

[0022] Establishing constraints, wherein the constraints include minimum / maximum output limits of each generator, transmission line capacity constraints, and fuel consumption;

[0023] Select an optimization algorithm, which includes a nonlinear programming algorithm;

[0024] Conduct simulations and improve model accuracy: Conduct simulations based on data history and continuously check the model’s validity and accuracy;

[0025] Output results and optimize adjustments: adjust the output power by controlling the equipment at the power grid hub to keep all generators at their optimal output state;

[0026] S14: According to the calculated optimal output power, a command is sent to the generator control system to control the output power of the generator.

[0027] Furthermore, it also includes: monitoring and optimizing the system, and continuously adjusting the algorithm parameters and model parameters to adapt to the changes and optimization requirements of the power system; specifically including the following steps:

[0028] S151: Collect historical operation data and real-time data of the power system and store them in a database;

[0029] S152: Based on the data collection results, use machine learning and data analysis techniques to build a power system monitoring and optimization model;

[0030] S153: According to the established model, by testing different algorithm options and parameter values, the best parameter combination is selected to optimize the control strategy of the power system;

[0031] S154: As time goes by and data accumulates, the system will continue to learn and evolve;

[0032] S155: Regularly verify and evaluate the model through backtesting or real-time simulation testing;

[0033] S156: The optimized algorithms and models can be automatically deployed into the actual power system for intelligent monitoring and optimization functions.

[0034] Furthermore, the step 2 includes the following steps:

[0035] S21: During off-peak electricity price periods, energy storage devices are charged by purchasing low-priced electricity;

[0036] S22: During peak electricity price periods, the energy storage device discharges by supplying power to the grid;

[0037] S23: The energy storage device is adjusted in response to changes in grid load, i.e. it is fully charged before the peak period of grid demand and supplies power during the peak period.

[0038] Furthermore, the step 4 includes the following steps:

[0039] S41: connecting the energy storage device to a local area of ​​the power grid and coordinating their operation;

[0040] S42: Monitor and dispatch the entire power system through a centralized controller;

[0041] S43: Introduce a market-based mechanism to treat energy storage equipment as a dispatchable resource that participates in market transactions and is bought and sold according to changes in market supply and demand;

[0042] S44: Optimize the charging and discharging strategies and control strategies of energy storage equipment by analyzing historical data and real-time operating status and combining artificial intelligence algorithms.

[0043] Furthermore, the step five comprises the following steps:

[0044] S51: Establish a cross-regional communication network;

[0045] S52: Establish a cross-regional dispatch center to monitor and dispatch the entire network;

[0046] S53: Introduce market-based mechanisms, establish cross-regional electricity markets, and trade electricity as a commodity.

[0047] Furthermore, in S53, artificial intelligence and big data analysis technologies are specifically used to analyze historical data and real-time operating status to optimize the cross-regional power balance control strategy.

[0048] Furthermore, the optimization algorithm of the artificial intelligence technology is used to control the output power of each generator, wherein the algorithm includes genetic algorithm, neural network, fuzzy control, ant colony algorithm and optimal control.

[0049] Compared with the prior art, the present invention has the following technical effects:

[0050] The present invention provides a bus power balancing and regulating method, which predicts and optimizes the power grid load, guides the regulation of the generator output power, realizes the bus power balance, introduces energy storage equipment, and adopts a distributed control strategy for the power system, divides the power system into multiple small areas, adopts a localized control and optimization strategy within each small area, and sets energy storage equipment for each small area to store and release energy to maintain the bus power balance regulation. At the same time, the power balance control of the entire power grid is realized through cross-regional coordination, thereby realizing more intelligent operation and a more stable balancing function. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0052] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0053] In order to further explain the technical means and effects taken by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the technical solutions proposed by the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments. The specific features, structures or characteristics in one or more embodiments may be combined in any suitable form. Unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by technicians in the technical field of the present invention.

[0054] In one embodiment of the present invention, referring to Figure 1 , provides a bus power balance adjustment method, comprising the following steps:

[0055] Step 1: Predict the grid load and optimize it to guide the power output adjustment of the generator;

[0056] Step 2: Introduce energy storage equipment and adopt a distributed control strategy for the power system, which is used to ensure that the power system can flexibly respond to load changes;

[0057] Step 3: Divide the power system into multiple small areas, and adopt a localized control and optimization strategy in each small area, wherein the localized control and optimization strategy is used to ensure power balance and independent regulation capability in each area;

[0058] Step 4: Store and release energy in the energy storage devices within each small area to maintain bus power balance regulation within the area and provide support for global regulation;

[0059] Step 5: Through cross-regional coordination, based on steps 3 and 4, the power balance conditions of each small area are integrated to achieve global power balance control of the entire power grid.

[0060] The following is a detailed explanation of each of the above steps:

[0061] Step 1: Predict the grid load and optimize it to guide the power output adjustment of the generator and achieve bus power balance.

[0062] As an example, this step may include the following steps:

[0063] S11: Collect data in the power system;

[0064] Data in the power system is collected through sensors. Various sensors are installed in power transmission and distribution transformers, switchgear, and various applications. The sensors monitor the grid operation status, load conditions, temperature, humidity, power consumption, etc. in real time. The grid operation status includes: voltage stability, frequency stability, flow distribution, power balance, equipment operation status, fault and abnormality detection, power quality, etc.

[0065] S12: Build models to describe the behavior of power systems;

[0066] Build a model to describe the behavior of the power system, specifically using the load curve: The load curve is an important model to describe the characteristics of power demand and can be used to determine the maximum load per day, week or month. The load curve has the following characteristics:

[0067] Changes in daily load: During the day, industrial and commercial electricity consumption is relatively high, so the daytime load is relatively high; at night, people rest, household electricity consumption is relatively reduced, and the load decreases.

[0068] Periodic changes: There are differences in load demand between weekdays and weekends. The load on weekends and holidays is usually lower than on weekdays.

[0069] Seasonal changes: Temperature, weather conditions, and people’s production and living habits in different seasons will affect electricity demand, resulting in different load curves.

[0070] Impact of special events: Large-scale events, holiday lighting, and emergencies may have a temporary impact on the load curve of the power system.

[0071] In the power grid load forecasting, a neural network model is used to forecast the power grid load, and the neural network model used is a BP (back propagation) neural network. Through the error back propagation algorithm for training, complex nonlinear mapping relationships can be approximated. In the process of power grid load forecasting, the data used is the collected historical power grid load data, where the collected historical power grid load data includes factors that may affect the load, such as load value, timestamp, weather conditions, holiday information, etc. The collected historical power grid load data is preprocessed, including cleaning, denoising, normalization, etc., to improve the training efficiency and prediction accuracy of the neural network.

[0072] Generator Models: Generator models can be used to describe the performance of generator regulators at different load levels and determine what power output they can support.

[0073] Transmission Line Models: Transmission line models can be used to describe how power flows in transmission lines, including voltage drops and common AC transmission effects such as inductive and capacitive coupling.

[0074] Renewable energy models: Renewable energy sources are highly stochastic, so they need to be analyzed in detail to understand the details such as the timing and amount of power generation they capture.

[0075] Control Model: The control model describes how various devices monitor network conditions to make state measurements and change settings to calibrate the flow of energy.

[0076] S13: Calculate the optimal output power of each generator using an optimization algorithm;

[0077] Determine the objective function: The objective function can be to maximize profit or minimize cost at a given load. This can be formulated as a linear programming problem where the load level is the variable and the operating cost and revenue need to be maximized or minimized.

[0078] Establishing constraints: Consider many constraints such as minimum / maximum output limits of each generator, transmission line capacity constraints, fuel consumption, etc.

[0079] Select an appropriate optimization algorithm: Since the problem contains multiple constraints, nonlinear programming algorithms such as Interior Point Method or Sequential Quadratic Programming Method can be used to solve the problem.

[0080] Conduct simulations and improve model accuracy: Conduct simulations based on data history and continuously check the model’s validity and accuracy. Models can be further improved in accuracy and predictability through machine learning algorithms.

[0081] Output results and optimize adjustments: Adjust the output power by controlling the equipment at the power grid hub to keep all generators in their optimal output state and meet load requirements. Continuously optimize the model and control strategy to achieve better power output effects.

[0082] S14: sending instructions to the generator control system according to the calculated optimal output power to control the output power of the generator;

[0083] S15: Monitor and optimize the system, and continuously adjust algorithm parameters and model parameters to adapt to changes and optimization requirements of the power system.

[0084] Preferably, S15 also includes the following steps:

[0085] S151: Collect historical operation data and real-time data of the power system and store them in a database for subsequent analysis and model optimization;

[0086] Among them, historical operation data and real-time data include historical and real-time voltage, frequency, power flow distribution, power balance, equipment operation status, fault and abnormality detection, power quality, etc., which are stored at regular intervals and the time interval can be set, the default is 1 minute.

[0087] S152: Based on the data collection results, a power system monitoring and optimization model is constructed using machine learning and data analysis techniques. The specific modeling process is as follows:

[0088] S1521: Data collection and preprocessing;

[0089] The data collection includes: determining the data source, data cleaning, and data storage and management. Among them, determining the data source means determining the type of data that needs to be collected, including power grid load data, power generation equipment operation data, user power consumption data, weather data, etc. Real-time collection is carried out through sensors and monitoring equipment, and historical data is obtained through database query and data import. Data cleaning removes noise, outliers and missing values ​​in the collected raw data to ensure data consistency and integrity. Use data mining and machine learning methods to detect data anomalies and complete data to improve data quality data storage and management, that is, select data storage technology to achieve efficient storage and management of massive data. Establish a data backup and recovery mechanism to ensure data security and reliability.

[0090] The data preprocessing includes data standardization and feature extraction and selection. Data standardization: normalize and standardize the data to make the data comparable and interpretable; feature extraction and selection: extract features related to the operation of the power system from the data, such as time features, seasonal features, holiday features, etc. Select the most representative and important features for subsequent analysis.

[0091] S1522: Build power system monitoring and optimization models based on machine learning and data analysis techniques, including:

[0092] Select appropriate machine learning algorithms: According to the specific needs of power system monitoring and optimization, select appropriate machine learning algorithms, such as support vector machine (SVM), decision tree, artificial neural network, etc.

[0093] Model training: Use the preprocessed data to train the selected machine learning algorithm to build a power system monitoring and optimization model. During the training process, the model parameters are continuously adjusted to improve the accuracy and generalization ability of the model.

[0094] S1523: Model validation and optimization.

[0095] Model validation: Use the test data set to validate the trained model and evaluate the prediction accuracy and generalization ability of the model. Use methods such as cross-validation to ensure the stability and reliability of the model on different data sets.

[0096] Model optimization: Optimize and adjust the model according to the verification results, such as adjusting model parameters, selecting more appropriate algorithms, etc. Introduce model integration methods, such as weighted average and stacked average, to further improve the prediction accuracy of the model.

[0097] S153: According to the established model, by testing different algorithm options and parameter values, the best parameter combination is selected to optimize the control strategy of the power system;

[0098] Different machine learning algorithm options are paralleled in the optimization process, which means that when optimizing power system control strategies, multiple machine learning or optimization algorithms can be considered simultaneously, such as neural networks, genetic algorithms, particle swarm optimization algorithms, simulated annealing algorithms, etc. Each algorithm has its own unique advantages and scope of application, and the most suitable algorithm for the current problem can be selected by comparing their performance.

[0099] Parameter values ​​are specific values ​​that need to be set during the operation of the algorithm, which directly affect the performance and results of the algorithm. In power system optimization control, parameter values ​​may include but are not limited to the following categories:

[0100] Neural network parameters, including learning rate, number of hidden layers and nodes, and activation function. The learning rate controls the step size of weight adjustment, and its value range is usually a decimal between 0 and 1, but it can also be other values. The number of hidden layers and nodes determines the complexity and learning ability of the network. The value range depends on the specific problem and there is no fixed limit. Activation functions such as Sigmoid, ReLU, etc., the choice of which activation function is also part of the parameter setting. Although they are not numerical parameters, they affect network performance.

[0101] Genetic algorithm parameters include population size, crossover probability, mutation probability, and number of iterations. Population size determines the initial size of the algorithm's search space, and its value range depends on the problem size and computing resources. Crossover probability and mutation probability control the crossover and mutation frequencies in genetic operations, and their value range is usually a decimal between 0 and 1. Number of iterations: The total number of rounds the algorithm runs, and its value range depends on the convergence speed and computing resources.

[0102] Other optimization algorithm parameters: such as the number of particles, inertia weight, acceleration constant, etc. in the particle swarm optimization algorithm; the initial temperature, cooling rate, etc. in the simulated annealing algorithm. The specific value range of these parameters depends on the specific implementation of the algorithm, the complexity of the problem, and the limitation of computing resources. In practical applications, it is usually necessary to determine the best parameter combination through experiments and debugging.

[0103] Each algorithm has its own unique parameter settings, which have a significant impact on the performance and results of the algorithm. The parameter values ​​between different algorithms are not comparable because they serve different algorithm logics and optimization goals. Therefore, when optimizing the power system control strategy, it is necessary to perform parameter tuning for each algorithm separately to find the parameter combination that best suits the current problem.

[0104] Verification evaluation criteria generally include the following:

[0105] Historical data fit, strategy stability, parameter sensitivity, real-time prediction accuracy, system response speed, fault simulation and response capabilities, resource utilization efficiency, and user satisfaction

[0106] Specific verification and evaluation standards should be formulated and adjusted according to the actual situation of the power system and the specific requirements of the model. At the same time, with the continuous development of technology and the continuous changes in the power system, the verification and evaluation standards also need to be continuously updated and improved.

[0107] S154: As time goes by and data accumulates, the system continuously improves its prediction and optimization capabilities by iteratively updating models and algorithms;

[0108] S155: Regularly verify and evaluate the model through backtesting or real-time simulation testing;

[0109] Verification evaluation criteria generally include the following:

[0110] Historical data fit, strategy stability, parameter sensitivity, real-time prediction accuracy, system response speed, fault simulation and response capabilities, resource utilization efficiency, and user satisfaction

[0111] Specific verification and evaluation standards should be formulated and adjusted according to the actual situation of the power system and the specific requirements of the model. At the same time, with the continuous development of technology and the continuous changes in the power system, the verification and evaluation standards also need to be continuously updated and improved.

[0112] S156: The optimized algorithms and models can be automatically deployed into the actual power system for intelligent monitoring and optimization functions.

[0113] Step 2: Introduce energy storage equipment and adopt a distributed control strategy for the power system to ensure that the power system can flexibly respond to load changes. As an example, this step may include the following steps:

[0114] S21: During off-peak electricity price periods, energy storage devices can be charged by purchasing low-priced electricity;

[0115] S22: During peak electricity price periods, the energy storage device can discharge by supplying power to the grid;

[0116] S23: The energy storage device is adjusted in response to changes in grid load, i.e. it is fully charged before the peak period of grid demand and then supplies power during the peak period.

[0117] Step 3: Divide the power system into multiple small areas, and adopt localized control and optimization strategies within each small area to ensure power balance and independent regulation capabilities within each area.

[0118] As an example, this step may include the following steps:

[0119] The power system is divided into multiple small areas, and localized control and optimization strategies are adopted in each small area to expose specific local information: in each small area, a part of the local information belonging to the small area is exposed, such as grid topology, load data, generator output, etc.

[0120] It is worth explaining the meaning of exposing local information:

[0121] Internal visibility: This information is exposed within the small area, which means that the control system, optimization algorithm, monitoring equipment, etc. within the small area can directly access and use this information.

[0122] Specificity: The exposed local information is closely related to the community, such as grid topology (referring only to the connection status within a small area), load data (power consumption of users in a small area), generator output (operating status and output of power generation equipment in the community), etc.

[0123] Restrict access: This information is usually not exposed unconditionally to all external systems or users, but is restricted based on security, privacy, and efficiency needs.

[0124] By establishing a distributed communication network in a small area, the required information can be shared among cells.

[0125] Localized controller design: A local controller is set up in each cell to use local information to optimize and control the local system.

[0126] For example, in a small area, the output power of the generator can be adjusted and energy can be saved according to the local weather conditions and predicted load. This includes data collection, data processing and analysis, predictive calculation and decision making, and system integration and feedback.

[0127] The data collection includes weather data collection and load forecast data collection. Weather data collection includes collection through sensor networks and external data sources. Sensor network collection: deploying meteorological sensors in or around the community, such as temperature, humidity, wind speed, light intensity and other sensors, to collect local meteorological data in real time; external data sources: using meteorological data services on the Internet, such as the API of the Meteorological Bureau, professional meteorological websites, etc., to obtain more comprehensive weather forecast information.

[0128] Load forecasting data collection includes historical load data and real-time load monitoring data. Historical load data: collects power load data in the community over the past period of time, including daily, weekly, and monthly peak and valley periods; real-time load monitoring data: obtains the current power load situation of the community in real time through smart meters or other power monitoring equipment. User behavior analysis: analyzes users' power usage habits and predicts possible load changes in the future.

[0129] The data processing and analysis include data cleaning and feature extraction. Data cleaning: pre-processing the collected data to remove noise and outliers to ensure the accuracy and reliability of the data; feature extraction: extracting key features from weather data and load data, such as temperature, humidity, light intensity, historical load trends, etc.

[0130] Prediction model establishment: Use machine learning or statistical methods (such as time series analysis, regression analysis, neural networks, etc.) to establish a prediction model to predict future load changes based on historical data and current weather conditions.

[0131] Prediction calculation and decision making include: real-time prediction, optimization algorithm and dynamic adjustment. Real-time prediction: Combine real-time weather data and load monitoring data to use prediction models to make real-time load predictions; Optimization algorithm: Based on the prediction results, use optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.) to calculate the optimal output power of the generator to achieve maximum energy savings while meeting load requirements; Dynamic adjustment: According to real-time changes in weather and load, dynamically adjust the output power of the generator to ensure system stability and efficiency.

[0132] System integration and feedback,System integration: Integrate the above links into the localized controller to realize automatic data collection, processing, analysis and decision-making. Feedback mechanism: Establish a feedback mechanism to continuously improve and optimize the prediction model and optimization algorithm based on the actual operation effect of the system.

[0133] Through the above steps, the localized controller can optimize and control the local system using local weather conditions and predicted load information, thereby achieving output power adjustment of the generator and effective energy conservation.

[0134] Multi-agent coordination: Multiple controllers need to coordinate operations to avoid interference and conflict. Therefore, an agent needs to be set up between the controllers of each cell to coordinate the energy flow and interaction in the entire cell to achieve more efficient operation management and corresponding optimization goals.

[0135] Sustainable production management: For the management of renewable energy, models based on weather forecasts and historical data are used for operation and management within small areas. For example, in a solar community, the battery storage capacity can be adjusted according to local weather conditions and historical data to ensure energy balance.

[0136] In actual operation, for different power systems, it is necessary to adopt corresponding local control and optimization strategies based on specific circumstances to bring a more stable and efficient operation state to the entire power system.

[0137] An intelligent agent is set between the controllers of each small area to coordinate the energy flow and interaction in the entire small area to achieve more efficient operation management and optimization goals. The specific implementation measures include the following aspects:

[0138] 1. Agent Architecture Design

[0139] Distributed architecture: With a distributed architecture, intelligent nodes are deployed on the controller of each small area. These nodes can process local data autonomously and communicate with adjacent nodes or the central control center to achieve information sharing and collaboration.

[0140] Modular design: The intelligent agent system adopts modular design, and each module is responsible for specific functions, such as data collection, processing, prediction, scheduling, etc. This design makes the system more flexible, easy to expand and maintain.

[0141] 2. Data Collection and Monitoring

[0142] Real-time data collection: Utilize sensor networks, smart meters and other equipment to collect real-time data on energy production (such as solar energy, wind energy, etc.), consumption (such as residential electricity, commercial electricity, etc.) and status data of energy storage equipment within the community to ensure the accuracy and real-time nature of the data.

[0143] Data transmission and storage: Establish an efficient data transmission network and transmit the collected data to the intelligent node by wired or wireless means. At the same time, establish a large-capacity, high-reliability data storage system for the preservation and analysis of historical data.

[0144] 3. Intelligent Forecasting and Scheduling

[0145] Load forecasting: Based on historical load data, weather forecasts, holiday schedules and other factors, the forecasting algorithm is used to forecast the future power load of the community. The forecast results will serve as an important basis for energy scheduling.

[0146] Energy production forecast: For renewable energy power generation equipment in the community, power generation forecasts are made based on historical power generation data, weather forecasts, and other information. The forecast results will be used to optimize energy production and distribution.

[0147] Intelligent dispatching algorithm: The intelligent agent uses optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.) to formulate energy dispatching plans based on load forecasts and energy production forecasts. This includes determining the output of each power generation equipment, the charging and discharging strategy of energy storage equipment, and the power flow distribution of the power grid. The dispatching algorithm should be able to adjust in real time to adapt to changes in the system's operating status.

[0148] 4. Collaborative Interaction and Optimization

[0149] Local collaboration: Intelligent nodes exchange information and work together through communication protocols. They can share local data, scheduling plans, and optimization results to achieve energy balance and efficient utilization in a small area.

[0150] Global optimization: In some cases, the intelligent system can also communicate with a higher-level control center to achieve cross-cell energy interaction and optimization. Through the global optimization algorithm, the efficiency and stability of the entire energy network can be further improved.

[0151] Dynamic adjustment: The intelligent system should have the ability to dynamically adjust and automatically adjust the scheduling plan and optimization strategy according to the real-time monitoring data and changes in the system operation status. This helps to deal with emergencies and uncertain factors and ensure the stable operation of the system.

[0152] V. Security and Privacy Protection

[0153] System security: Establish a comprehensive security protection mechanism to ensure the stable operation of the intelligent system and data security. Use encryption technology to protect sensitive information during data transmission to prevent illegal access and tampering.

[0154] Privacy protection: When collecting and using user data, privacy protection laws and regulations should be strictly followed. Sensitive data should be desensitized or anonymized before storage and analysis to ensure that user privacy is not violated.

[0155] 6. Continuous optimization and upgrading

[0156] Data analysis and mining: Regularly analyze and mine the stored historical data to discover potential operating rules and optimization points. Continuously improve the energy prediction model and scheduling algorithm through data analysis results to improve the performance and efficiency of the system.

[0157] Technology upgrade: Pay attention to the latest developments in energy technology and information technology, and apply new technologies to the intelligent system in a timely manner. Maintain the advancement and competitiveness of the system through technology upgrades and updates.

[0158] Step 4: Store and release energy in the energy storage devices within each small area to maintain bus power balance regulation within the area and provide support for global regulation;

[0159] As an example, this step may include the following steps:

[0160] S41: connecting the energy storage device to a local area of ​​the power grid and coordinating their operation;

[0161] S42: Monitor and dispatch the entire power system through a centralized controller;

[0162] In the power system, the centralized controller plays a vital role. It is responsible for monitoring and dispatching the entire power system to ensure the safe, stable and efficient operation of the system. The specific steps of dispatching usually include the following aspects:

[0163] 1. Data Collection and Monitoring

[0164] Data collection: The centralized controller collects various operating parameters of the power system in real time through various sensors and monitoring equipment, such as voltage, current, power, frequency, load, etc. These parameters are the basis for dispatching decisions.

[0165] Monitoring and analysis: Real-time monitoring and analysis of collected data to evaluate the operating status and trends of the power system. Use technologies such as big data and artificial intelligence to process massive amounts of data and extract valuable information.

[0166] 2. Load Forecasting and Generation Plan Formulation

[0167] Load forecasting: Based on historical load data, weather forecasts, holiday schedules and other factors, the forecasting model is used to forecast the load for a period of time in the future. The accuracy of load forecasting is crucial to the formulation of power generation plans.

[0168] Power generation plan formulation: formulate a reasonable power generation plan based on load forecast results and power generation resources (such as hydropower, thermal power, wind power, photovoltaic power, etc.). The power generation plan needs to take into account multiple factors such as the economy, environmental protection and safety of the system.

[0169] 3. Scheduling Decisions and Instructions

[0170] Dispatch decision: Based on load forecast, power generation plan and real-time operation status of power system, the centralized controller makes dispatch decision. Dispatch decision includes determining the output of each generator set, power flow distribution of power grid, voltage and reactive power compensation, etc.

[0171] Instruction issuance: The dispatch decision results are converted into specific operation instructions, and issued to various power plants, substations and other execution units through the communication system. The instructions include the start and stop of the generator set, output adjustment, and the switch operation of the power grid.

[0172] 4. Execution and Feedback

[0173] Execution: After receiving the instructions, each power plant, substation and other execution units will operate according to the requirements. For example, adjust the output of the generator set, switch the operation mode of the power grid, etc.

[0174] Feedback: The execution unit feeds back the operation results to the centralized controller. The centralized controller verifies and processes the feedback information to ensure the correct execution of the dispatch instructions. At the same time, the dispatch plan is adjusted and optimized as necessary based on the feedback information.

[0175] 5. Safety verification and adjustment

[0176] Safety verification: During the dispatch decision-making process, the centralized controller needs to conduct safety verification to ensure that the dispatch plan will not cause instability or safety problems in the power system. Safety verification includes static safety analysis, transient stability analysis, etc.

[0177] Adjustment: If the safety check finds problems with the dispatch plan, the centralized controller will make automatic or manual adjustments to ensure the safe and stable operation of the power system.

[0178] S43: Introduce a market-based mechanism to treat energy storage equipment as a dispatchable resource that participates in market transactions and is bought and sold according to changes in market supply and demand;

[0179] S44: Optimize the charging and discharging strategies and control strategies of energy storage equipment by analyzing historical data and real-time operating status and combining artificial intelligence algorithms.

[0180] Optimization of charging and discharging strategies: Develop a reasonable charging and discharging plan based on load forecasts and renewable energy generation forecasts. Charge during low-load periods and discharge during peak load periods to ease the contradiction between supply and demand in the power system. Optimize charging and discharging power and rate to improve the utilization efficiency and life of energy storage equipment.

[0181] Control strategy optimization: Optimize the control algorithm of energy storage equipment so that it can more accurately control the charging and discharging process and reduce energy loss and battery loss.

[0182] Advanced control strategies, such as fuzzy control and adaptive control, are introduced to improve the robustness and stability of the system.

[0183] Implement optimization strategies: Apply the optimized charging and discharging strategies and control strategies to the actual power system for trial operation and verification.

[0184] Feedback and adjustment: Make necessary adjustments and improvements to the optimization strategy based on the trial operation results and real-time operation status feedback. Continuously collect new operation data for model update and iteration to continuously improve the operation efficiency and performance of energy storage equipment.

[0185] Step 5: Through cross-regional coordination, based on steps 3 and 4, the power balance conditions of each small area are integrated to achieve global power balance control of the entire power grid.

[0186] As an example, this step may include the following steps:

[0187] S51: Establish a cross-regional communication network;

[0188] S52: Establish a cross-regional dispatch center to monitor and dispatch the entire network;

[0189] S53: Introduce market-based mechanisms, establish cross-regional electricity markets, and trade electricity as a commodity.

[0190] Preferably, in S53, artificial intelligence and big data analysis technology can be used to analyze historical data and real-time operating status, optimize cross-regional power balance control strategies, and improve the reliability and economic benefits of the power system. The following are specific processing, usage methods, and specific analysis steps for historical data and real-time operating status:

[0191] 1. Historical Data Processing

[0192] Data collection: Collect historical data from the power systems in various regions, including power generation, power consumption, load curves, energy storage equipment status, grid topology, etc. Ensure the integrity and accuracy of the data, and conduct preliminary screening and cleaning of the data.

[0193] Data storage: Use distributed databases or big data storage platforms (such as Hadoop, Spark, etc.) to store massive amounts of historical data. Classify and index the data for quick retrieval and analysis.

[0194] Data preprocessing: denoising, filling missing values, standardizing or normalizing the data. Extracting key features, such as seasonal load changes, holiday load patterns, and the impact of weather on load.

[0195] Feature engineering: Build and select useful feature sets based on business needs and analysis objectives. This may include time series features, statistical features, and transformed features (such as differential, cumulative, etc.).

[0196] 2. Real-time operation status processing

[0197] Real-time data collection: Use sensors, smart meters, SCADA systems, etc. to collect real-time data on the operation status of the power system to ensure the real-time and accuracy of the data.

[0198] Data cleaning and verification: Perform preliminary cleaning and verification of real-time data to remove abnormal values ​​and erroneous data. Perform synchronization and consistency checks on real-time data.

[0199] Condition monitoring and assessment: Real-time monitoring of key parameters of the power system, such as voltage, current, frequency, load, etc. Evaluate the stability and safety of the system and predict potential risks and failures.

[0200] 3. Specific analysis steps

[0201] Historical data analysis: Pattern recognition: Use cluster analysis, time series analysis and other methods to identify load patterns, power generation patterns, etc. in historical data. Association analysis: Analyze the correlation between different variables, such as weather and load, electricity price and electricity consumption, etc. Trend prediction: Use machine learning algorithms (such as LSTM, ARIMA, etc.) to predict future loads, power generation, etc.

[0202] Real-time data analysis and fusion: Real-time prediction: Combine real-time data and historical models to make rolling predictions for short-term load, power generation, etc. State assessment: Real-time assessment of the operating status of the power system, including supply and demand balance, energy storage equipment status, etc. Anomaly detection: Use anomaly detection algorithms to identify abnormal points or abnormal situations in real-time data.

[0203] Cross-regional power balance optimization: Multi-regional coordination: Consider the power supply and demand relationship between different regions and establish a cross-regional coordination model. Optimization algorithm: Use optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, reinforcement learning, etc.) to formulate cross-regional power balance control strategies.

[0204] Strategy implementation: Based on the optimization results, adjust the power generation plan, load management, and energy storage equipment charging and discharging strategies of each region.

[0205] Effect evaluation and feedback: Effect evaluation: Evaluate the effect after implementation, including the stability of the power system, economic benefits, environmental benefits, etc. Feedback and adjustment: According to the evaluation results, make necessary adjustments and improvements to the optimization strategy to adapt to changes in the power system.

[0206] Continuous learning and optimization: Model update: Regularly update and optimize the model with new historical data and real-time data. Strategy iteration: Continuously iterate and optimize the cross-regional power balance control strategy as the power system develops and changes.

[0207] Through the above steps, the cross-regional power balance control strategy can be optimized and the reliability and economic benefits of the power system can be improved.

[0208] Preferably, the optimization algorithm of the artificial intelligence technology is used to control the output power of each generator, wherein the algorithm includes genetic algorithm, neural network, fuzzy control, ant colony algorithm and optimal control.

[0209] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A bus power balance adjustment method, characterized in that: The steps include: Step 1: Predict the grid load and optimize it to guide the power output adjustment of the generator; Step 2: Introducing energy storage equipment and adopting a distributed control strategy for the power system, the distributed control strategy is used to ensure that the power system copes with load changes; Step 3: Divide the power system into multiple small areas, and adopt a localized control and optimization strategy in each small area, wherein the localized control and optimization strategy is used to ensure power balance and independent regulation capability in each area; Step 4: Store and release energy in the energy storage devices within each small area to maintain bus power balance regulation within the area and provide support for global regulation; Step 5: Through cross-regional coordination, based on steps 3 and 4, the power balance conditions of each small area are integrated to achieve global power balance control of the entire power grid.

2. A bus power balance adjustment method according to claim 1, characterized in that: In step 1, the grid load is predicted and optimized, including the following steps: S11: Collecting data in the power system, including real-time monitoring of power grid operation status, load conditions, temperature, humidity, and power; S12: Build models to describe the behavior of the power system: The model includes a load curve, a generator model, a transmission line model, a renewable energy model and a control model; Among them, the load curve is used to determine the maximum load per day, week or month; The generator model is used to describe the performance of the generator regulator at different load levels and determine what power output is supported; The transmission line model is used to describe the flow of electric energy in the transmission line, including voltage reduction and AC transmission effects; The renewable energy model is used to understand the timing and power generation information captured by it; The control model is used to describe the various devices that monitor network conditions to measure status and change settings to calibrate the flow of energy; S13: determining an objective function for maximizing profit or minimizing cost at a given load, where the load level is a variable and the operating cost and revenue need to be maximized or minimized; Establishing constraints, wherein the constraints include minimum / maximum output limits of each generator, transmission line capacity constraints, and fuel consumption; Select an optimization algorithm, which includes a nonlinear programming algorithm; Conduct simulations and improve model accuracy: Conduct simulations based on data history and continuously check the model’s validity and accuracy; Output results and optimize adjustments: adjust the output power by controlling the equipment at the power grid hub to keep all generators at their optimal output state; S14: According to the calculated optimal output power, a command is sent to the generator control system to control the output power of the generator.

3. A bus power balance adjustment method according to claim 2, characterized in that: It also includes: monitoring and optimizing the system, and constantly adjusting the algorithm parameters and model parameters to adapt to the changes and optimization requirements of the power system; specifically, it includes the following steps: S151: Collect historical operation data and real-time data of the power system and store them in a database; S152: Based on the data collection results, use machine learning and data analysis techniques to build a power system monitoring and optimization model; S153: According to the established model, by testing different algorithm options and parameter values, the best parameter combination is selected to optimize the control strategy of the power system; S154: As time goes by and data accumulates, the system continuously improves its prediction and optimization capabilities by iteratively updating models and algorithms; S155: Regularly verify and evaluate the model through backtesting or real-time simulation testing; S156: The optimized algorithms and models are automatically deployed to the actual power system for intelligent monitoring and optimization functions.

4. A bus power balance adjustment method according to claim 1, characterized in that: The distributed control strategy described in step 2 includes: S21: During off-peak electricity price periods, energy storage devices are charged by purchasing low-priced electricity; S22: During peak electricity price periods, the energy storage device discharges by supplying power to the grid; S23: The energy storage device is adjusted in response to changes in grid load, i.e. it is fully charged before the peak period of grid demand and supplies power during the peak period.

5. A bus power balance adjustment method according to claim 4, characterized in that: The step 4 comprises the following steps: S41: connecting the energy storage device to a local area of ​​the power grid and coordinating their operation; S42: Monitor and dispatch the entire power system through a centralized controller; S43: Introduce a market-based mechanism to treat energy storage equipment as a dispatchable resource that participates in market transactions and is bought and sold according to changes in market supply and demand; S44: Optimize the charging and discharging strategies and control strategies of energy storage equipment by analyzing historical data and real-time operating status and combining artificial intelligence algorithms.

6. A bus power balance adjustment method according to claim 5, characterized in that: The step five comprises the following steps: S51: Establish a cross-regional communication network; S52: Establish a cross-regional dispatch center to monitor and dispatch the entire network; S53: Introduce market-based mechanisms, establish cross-regional electricity markets, and trade electricity as a commodity.

7. A bus power balance adjustment method according to claim 6, characterized in that: In S53, artificial intelligence and big data analysis technologies are specifically used to analyze historical data and real-time operating status to optimize the cross-regional power balance control strategy.

8. A bus power balance adjustment method according to claim 7, characterized in that: The optimization algorithm of the artificial intelligence technology is used to control the output power of each generator, wherein the algorithm includes genetic algorithm, neural network, fuzzy control, ant colony algorithm and optimal control.

Images