A micro-grid automatic power regulation method

Abstract

The application discloses a micro-grid automatic power regulation method and relates to the technical field of power regulation and resource management of micro-grid. In order to solve the defects of poor new energy consumption capacity, high difficulty of distributed resource management, poor grid regulation flexibility, poor resource coordination effect and poor power and load prediction accuracy of the existing micro-grid power regulation method, the application realizes flexible switching of grid connection and off-grid of A and B power grids by constructing an integrated construction architecture. Multi-source data is integrated, new energy power generation and load demand are intelligently predicted through data management of a data center, power scheduling is determined by a power dispatching integrated system when the power grids are connected to the grid, and a scheduling plan is formulated by a B micro-grid management system. When the power grids are off-grid, power supply is provided according to user priority. In the grid connection and off-grid mode, power distribution, load regulation, resource optimization configuration are carried out, distributed resources are real-timely regulated through a coordination controller, and stable operation of the power grid is ensured. The application is mainly used for flexible regulation of the micro-grid.

Description

Technical Field

[0001] This invention relates to the field of power regulation and resource management technology for microgrids, and in particular to an automatic power regulation method for microgrids based on multi-source data fusion and intelligent prediction. Background Technology

[0002] Amid the global trend of actively addressing climate change and vigorously promoting energy transition, many regions are using islands as important testing grounds for achieving large-scale application of renewable energy and reducing carbon emissions, striving to build them into "zero-carbon islands," "low-carbon islands," and even "negative-carbon islands."

[0003] With continuous economic development and social progress, Islands A and B have been designated as demonstration areas for development. The load demand on Island B is experiencing rapid growth, and its load characteristics are undergoing significant changes. Currently, the power grid on Island B consists of only two 35kV lines erected on the same tower. This line structure results in severely insufficient transmission capacity, a relatively weak overall grid structure, and is a purely AC network. Although the plan includes a series of improvement measures such as upgrading the line voltage and constructing interconnection channels, the construction period for these projects is relatively long. During this long transition period, the reliability of the power grid still faces significant risks.

[0004] Overall, current technology has significant shortcomings in several aspects.

[0005] First, the capacity for renewable energy absorption is limited. In grid-connected mode, when renewable energy generates a large amount of electricity, problems such as power backflow and voltage exceeding the upper limit are likely to occur. In off-grid mode, it is also often faced with the dilemma that the generated electricity cannot be fully absorbed or is insufficient to support the load. This leads to frequent wind and solar curtailment, which seriously reduces energy utilization efficiency and wastes energy resources.

[0006] Secondly, distributed resource management is quite challenging. Island B contains a wide variety of distributed resources, each with different communication methods and parameter configurations. Currently, there is a lack of effective means to integrate and manage these independent distributed resources in a unified manner.

[0007] Third, the flexibility of power grid control is insufficient. Faced with the complex and diverse operating scenarios of Island B, including different operating modes such as grid connection, off-grid, and black start, as well as different working conditions such as normal, heavy overload, fault, and maintenance, the existing technology has significant deficiencies in the adaptability and flexibility of control strategies.

[0008] Fourth, energy storage's synergy with other resources is inadequate. Energy storage falls short in its ability to optimize synergy with new energy sources, traditional power sources, and loads. On a temporal scale, it cannot fully leverage the power generation and consumption characteristics and response speeds of different controllable resources to achieve precise charging and discharging control at different time stages, such as day-ahead, intraday, and real-time, making it difficult to maximize benefits. On a spatial scale, its zoning and layering synergy with other resources is insufficient, failing to fully utilize energy storage's crucial role in balancing local power and improving overall power supply reliability.

[0009] Fifth, the accuracy of forecasts needs further improvement. Although artificial intelligence technologies and related algorithms have been introduced to forecast renewable energy generation and load demand, current forecasting methods still struggle to achieve a high level of accuracy due to the complex influence of meteorological and other factors on renewable energy generation, and the constraints of user behavior and other uncertainties on load demand. The existence of forecast errors may lead to inaccurate formulation of control strategies, thereby affecting the economy and reliability of power grid operation.

[0010] Therefore, there is a need for an automatic power control method for microgrids that can improve the flexibility of microgrid regulation, enhance the absorption of new energy sources, optimize distributed resource management, and improve energy utilization efficiency. Summary of the Invention

[0011] To address the shortcomings of existing microgrid power regulation methods, such as poor renewable energy absorption capacity, high difficulty in distributed resource management, poor grid regulation flexibility, poor resource coordination effect, and poor accuracy in power and load prediction, this invention provides an automatic power regulation method for microgrids that can improve the regulation flexibility of microgrids, enhance renewable energy absorption, optimize distributed resource management, and improve energy utilization efficiency.

[0012] The automatic power regulation method for microgrids according to the present invention includes the following steps:

[0013] S1. Construct an integrated construction architecture to realize the switching and coordinated control of power grids A and B in grid-connected and off-grid modes;

[0014] S2. Collect, process, and analyze multi-source data, and integrate multi-source data through a data platform to obtain historical and real-time data;

[0015] S3. Based on the historical and real-time data, intelligently predict the power generation of new energy power generation equipment and the load demand of energy storage system on a time scale.

[0016] S4. In grid-connected mode, the integrated distribution and dispatching system determines power dispatch based on the optimal target of the entire network. The B microgrid management system formulates a source-grid-load-storage dispatch plan based on power generation forecast and energy storage load forecast data.

[0017] S5. In off-grid mode, power supply management is carried out step by step according to user priority based on power supply and load conditions to ensure frequency and voltage stability.

[0018] S6. In both grid-connected and off-grid modes, power allocation, load regulation, resource optimization and scheduling are performed based on load demand and power generation conditions, and distributed resources are controlled in real time through a coordination controller.

[0019] Furthermore, in S6, the specific adjustment method for dynamically adjusting resources in the time and space dimensions is as follows: resources are adjusted on a time scale based on the power generation and consumption characteristics and response speed of different controllable resources, and dynamic adjustments are made according to the time scale and resource status during the adjustment process; in the spatial dimension, the principle of local, nearby, and same voltage level balance and absorption is followed to realize the zoning and hierarchical coordination of resources; wherein, in the grid-connected mode, the source, grid, load and storage are optimized and coordinated based on the whole grid, and in the off-grid mode, the B microgrid management system independently controls the resources within the island as a microgrid management system.

[0020] Furthermore, in S6, the real-time control of distributed resources through the coordination controller specifically includes the following steps:

[0021] S61. During the load regulation process, the optimal decision is made based on the load characteristics and the grid status to guide the load to be adjusted at different times and assist the power balance of the microgrid.

[0022] S62. Allocate power according to the power generation of new energy power generation equipment and the load demand of energy storage system, explore the grid support capacity of flexible and controllable load, and dynamically adjust resources in time and space based on the power generation and consumption characteristics of controllable resources to achieve regional coordination and optimized allocation of resources.

[0023] The dynamic resource adjustment includes energy storage and hydrogen production during periods of high renewable energy generation, as well as charging operations using the power grid during periods of low load.

[0024] S63. The coordination controller performs distributed resource coordination control in off-grid mode through GOOSE communication and automatically adjusts the microgrid frequency and voltage to ensure quasi-synchronous operation of the power grid.

[0025] S64. The energy management system interacts with different types of distributed resources using different methods and then converts them into a unified way to interact with the upper-level dispatcher. Based on the dynamic topology of the power grid, it aggregates independent distributed resource individuals into virtual resource aggregates to achieve monitoring and management of distributed resources. In the process of distributed resource management, it optimizes resource aggregation and dispatch strategies.

[0026] Furthermore, in S63, the distributed resource coordination control includes grid-connected power coordination control, off-grid voltage and frequency control, off-grid power coordination control, grid-connected / off-grid switching control, energy storage economic operation control, and black start control.

[0027] Furthermore, in S63, the switching and control process during the grid-connected to off-grid and off-grid to grid-connected transitions specifically includes the following steps:

[0028] S631. Initialization Phase: Set the initial control parameters of the adaptive control algorithm in the microgrid coordinating controller and establish a reference model to describe the ideal values ​​of the microgrid under different operating modes: In grid-connected mode, the voltage and frequency in the reference model are consistent with the main grid; in off-grid mode, the reference model is based on the local load demand and maintains a stable voltage and frequency range and power distribution method.

[0029] The initial control parameters include the initial charge and discharge power coefficient of the energy storage system, the initial output power ratio of the distributed power source, and the initial gain of frequency and voltage regulation.

[0030] S632, Monitoring and Data Acquisition Phase: The microgrid coordinating controller acquires real-time operational data from the microgrid and the main grid; it also acquires status information of distributed resources within the microgrid.

[0031] S633, Error Calculation Stage: The real-time monitored microgrid data is compared with the ideal value in the reference model to calculate the corresponding error; for power coordination control, the deviation between the actual power and the reference power is calculated, and the error is used to reflect the degree of deviation between the current operating state of the microgrid and the ideal state.

[0032] S634, Adjustment Stage: Based on the calculated error and the pre-set algorithm rules, the control parameters are automatically adjusted; the algorithm rules are based on the rule of dynamically changing the control parameters according to the error: in off-grid mode, the voltage error is positive, that is, the measured voltage is higher than the reference voltage, and the frequency error is also positive, that is, the measured frequency is higher than the reference frequency.

[0033] S635, Function Implementation Stage: When the microgrid coordinating controller executes control functions, it uses the control parameters adjusted by the algorithm to realize functions such as grid-connected power coordination control, off-grid voltage and frequency control, and off-grid power coordination control. In off-grid voltage and frequency control, the voltage and frequency of the microgrid are stabilized within the range specified by the reference model by adjusting the charging and discharging power of energy storage and the output power of distributed power sources. In the process of grid-connected to off-grid switching control, the algorithm dynamically adjusts the control parameters according to the data monitored during the switching process to realize the switching and control during the grid-connected to off-grid and off-grid to grid-connected processes.

[0034] Further: In S64, the optimized resource aggregation and scheduling strategy specifically includes the following steps:

[0035] S641. In the initial stage, the virtual "explorer" representing resource allocation attempts will randomly try on each resource connection path and leave a corresponding degree of "mark" on the path based on the resource allocation effect of each attempt; the degree of "mark" left on the path is calculated based on the resource allocation effect.

[0036] S642. Over time and through repeated exploration, the path selection and "imprint" level are continuously optimized, resulting in continuous improvement of resource aggregation and scheduling strategies.

[0037] S643. Based on the real-time dynamic topology of the power grid, dispersed distributed resources are effectively aggregated into a virtual resource aggregate.

[0038] Furthermore, in S3, the intelligent prediction specifically includes the following steps:

[0039] S31. Preprocess the historical data and real-time data to construct an algorithm network model;

[0040] S32. Divide the preprocessed data into training data group and test data group to train the algorithm network model;

[0041] S33. Update the model weights using backpropagation, and iterate repeatedly until the model converges.

[0042] Further: In S4, the integrated distribution and dispatch system in grid-connected mode specifically includes the following steps:

[0043] S41. Construct a microgrid model and use the power generation forecast of new energy power generation equipment and the load forecast data of energy storage system to predict the power trend in the future period. Taking into account the optimal operation target of the whole network and the power forecast situation inside the microgrid, determine the optimal setpoint sequence of tie line exchange power in the future preset time period by solving an optimization problem in each control cycle, so as to achieve the operation target under the preset constraints.

[0044] S42. Based on power forecast information, plan the charging and discharging power of energy storage at different times in advance. According to the predicted peak and off-peak periods of new energy power generation, allow energy storage to charge when there is excess power generation and allow energy storage to discharge when there is insufficient power generation or peak load. At the same time, continuously optimize the scheduling plan, recalculate and adjust the tie-line exchange power target and energy storage scheduling strategy to ensure that the microgrid always operates in the direction of maximizing green energy consumption and frequency and voltage stability.

[0045] Furthermore: In S5, the power supply management for users in off-grid mode specifically includes the following steps:

[0046] S51. Balance the base power and frequency voltage, take grid-type energy storage and diesel generator power equipment as key control objects, and automatically adjust the output power according to the real-time power supply and demand of the microgrid to keep the frequency and voltage of the microgrid within the preset range.

[0047] S52. Power supply management is based on user priority. Users in the microgrid are classified according to priority, and users are divided into important users and multi-level priority users. When the power output meets the power demand of important users, the power supply of important users is guaranteed in a timely manner. When there is still power remaining after the power output meets the needs of important users, the power supply of users with different priorities is gradually restored according to the pre-set priority sequence.

[0048] The power supply management is achieved by monitoring and controlling the power supply equipment and loads of each user through a microgrid coordinating controller.

[0049] Further: In S51, the specific steps of automatically adjusting the output power are as follows: if the frequency of the microgrid shows a downward trend, indicating insufficient power generation, the diesel generator will quickly increase its output and the grid-type energy storage will release electrical energy in a timely manner according to the preset rules to support the frequency to return to the normal level; conversely, if the frequency increases, the power generation output will be reduced accordingly.

[0050] The beneficial effects of this invention are:

[0051] This invention sets A Island and B Island as the construction targets for relevant demonstration areas, and builds an energy supply system that can ensure both efficient power supply and reliable and clean power supply. The system is based on the integrated architecture of the power grids of A Island and B Island, and is a comprehensive power supply and demand control platform composed of a flexible grid-connected and off-grid switching mechanism, a data platform, a microgrid coordination controller, and multi-source collaborative regulation, which aims to adapt to the needs of the times for energy transformation and low-carbon development.

[0052] The microgrid automatic power regulation method described in this invention, in terms of renewable energy consumption, can accurately address the volatility of renewable energy generation through the effective combination of intelligent prediction and coordinated regulation strategies, significantly improving consumption capacity and reducing wind and solar curtailment rates; it also improves energy utilization efficiency, promotes the development of clean energy, and assists in the optimization and transformation of the energy structure; in terms of energy storage, the optimized charging and discharging strategies and coordinated control not only extend the life of energy storage but also enhance its efficiency in balancing power, providing backup power, and black start capability, ensuring the stable operation of the microgrid; the grid regulation flexibility is significantly enhanced, adapting to various modes and operating conditions, ensuring smooth switching and stable control between grid connection and disconnection, and effectively guaranteeing power supply reliability; distributed resource management realizes unified monitoring and optimized configuration, fully tapping load potential, assisting in grid power balance and improving overall economic efficiency; multi-source data fusion provides accurate basis for regulation and enhances the scientific nature of decision-making; the platform has good scalability, compatibility, and efficiency, supporting the platform function expansion and algorithm iteration upgrade of the integrated distribution and dispatch system and the microgrid management system, facilitating integration and updates, adapting to the development needs of microgrids, and thus powerfully promoting energy transformation and sustainable development. Attached Figure Description

[0053] Figure 1 This is a framework diagram of the integrated allocation and dispatch system;

[0054] Figure 2 yes Figure 1 The framework diagram of the platform on which the integrated distribution and dispatch system and the B microgrid management system are based;

[0055] Figure 3 This is a flowchart;

[0056] Figure 4 This is a simplified diagram of a microgrid controller;

[0057] Figure 5 This is a connection diagram for the microgrid controller;

[0058] Figure 6 This is a functional diagram of the microgrid controller;

[0059] Figure 7 This is a schematic diagram of microgrid operation. Detailed Implementation

[0060] The following are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. The embodiments described below are only for explaining the present invention and should not be construed as limiting the present invention. The scope of protection of the present invention should be determined by the scope of the claims. The embodiments of the present invention are described in detail below. In order to facilitate the description of the present invention and simplify the description, the technical terms used in the specification of the present invention should be interpreted broadly, including but not limited to conventional alternatives not mentioned in this application, and including both direct and indirect implementation methods.

[0061] Example 1

[0062] Combination Figure 1 Figure 7 This embodiment describes an automatic power regulation method for microgrids based on multi-source data fusion and intelligent prediction, comprising the following steps:

[0063] Reference Appendix Figure 1 and Figure 2 The project constructs an integrated architecture to lay the foundation for regulation: focusing on creating a highly adaptable, flexible, and efficient power grid architecture. By constructing an integrated dual-active architecture for the A and B power grids, and a unified construction plan for Zone III, flexible switching and coordinated operation between grid-connected and off-grid modes are achieved. During grid connection, deep optimization and coordination of power generation, grid, load, and storage are conducted from a grid-wide perspective, promoting synergistic development among various elements and improving overall efficiency. In off-grid mode, the B microgrid management system, as a microgrid management system, can independently manage resources within the island, ensuring the autonomy and stability of power supply, thus laying a solid foundation for subsequent precise regulation.

[0064] Integrating diverse data to provide decision-making support: Data is extensively collected from various professional systems (such as EMS: Energy Management System, DMS: Distribution Management System, AMI: Advanced Metering Infrastructure, PMS: Power Production Management System, etc.), integrating information from meteorology, power generation, and load to ensure data richness and accuracy. The data acquisition and processing processes maintain data accuracy and timeliness. Simultaneously, the data platform plays a crucial role, integrating multi-source data to achieve unified management and sharing. It supports the interaction of data including, but not limited to, distributed resource information, user information, and maintenance plans to assist power regulation decisions. The data platform has data preprocessing capabilities to provide standard data formats for subsequent algorithms. Integrating and managing this massive and complex data not only covers detailed information on distributed resources and users but also incorporates important data such as maintenance plans. After preprocessing, the data is transformed into a standard format suitable for subsequent complex algorithms, providing comprehensive and accurate data support for power regulation decisions.

[0065] Utilizing intelligent forecasting to gain insights into power grid trends: This approach leverages advanced artificial intelligence technology to fully uncover potential patterns in historical and real-time data. The algorithm constructs a network model based on carefully prepared data (incorporating multiple factors such as meteorology). Through rigorous training (data grouping, loss function selection, and multiple iterations until convergence), the algorithm achieves accurate and intelligent forecasting of renewable energy generation power and load demand at different time scales (day-ahead, intraday, and real-time). The forecast results are used to formulate power regulation strategies. The algorithm is further developed, trained, and optimized to improve forecast accuracy. This provides crucial forward-looking information for the timely and effective formulation of regulation strategies.

[0066] The specific process is as follows: First, in the data preparation stage, comprehensive historical and real-time data are collected. The data covers, but is not limited to, key information such as renewable energy power generation and load demand, and incorporates meteorological data. Preprocessing is then performed on the collected data, adjusting it through methods including, but not limited to, normalization and standardization. Let the historical data sequence of renewable energy power generation be... ;

[0067] in, This represents a power adjustment factor, where t is the current time and n is the length of the historical data. This refers to the original renewable energy power generation capacity, and the historical load demand data sequence is as follows: Where m is the length of historical data and d represents the load change rate. Representing load demand, meteorological data sequences are collected and preprocessed using a normalization formula:

[0068]

[0069] in, This represents the normalized power generation of new energy sources, with values ​​between [0,1], which facilitates its use in algorithms (such as machine learning, control optimization, etc.). It is the original new energy power generation capacity; This represents the maximum power generation from historical data. This represents the minimum power generation capacity in historical data.

[0070] The value of n represents the length of the historical time window used to predict the power of new energy sources. It is usually set according to the needs of the prediction model and the sampling period. For example, if the sampling frequency is 15 minutes, n=96 means reviewing the data of the previous 24 hours.

[0071] The value of m represents the length of the historical data sequence used to predict load demand. It is usually consistent with n, but can also be set separately according to the load change characteristics. For example, m=48 represents the past 12 hours.

[0072] The meaning of t: It is the current prediction time point, usually the current sampling time, which can be dynamically advanced by the scheduling cycle (such as 5 minutes, 15 minutes).

[0073] The load demand and meteorological data are processed in a similar manner; then the data are organized in an orderly manner according to the time steps corresponding to different time scales, including day-ahead, intraday and real-time, and the real-time forecast is set according to the actual demand by setting the time interval.

[0074] Next, the model building phase begins, constructing the network model for the algorithm. The number of input layer neurons is determined based on the feature dimension of the input data, while the number of hidden layer neurons is set according to the problem complexity and data volume. The number of output layer neurons is then determined according to the prediction target. Furthermore, appropriate time steps and sequence lengths are set for different time scales to lay the foundation for accurate predictions. Let the feature dimension of the input data be k, including but not limited to renewable energy power generation, load demand, and meteorological data features; then the number of input layer neurons is set to k. Considering the problem complexity and data volume, the number of hidden layer neurons is set to h. The prediction targets are renewable energy power generation and load demand; the number of output layer neurons is set to 2, corresponding to the predicted values ​​of power generation and load demand, respectively. Appropriate time steps and sequence lengths are set for different time scales. Indicates the current step length. Indicates the intraday step length. Indicates the real-time step size. Indicates the length of the day-ahead sequence. Indicates the length of the intraday sequence. Indicates the length of the real-time sequence;

[0075] The preprocessed data is then divided into training and testing data groups according to a certain ratio: the majority of the data is used as the training data group, and the remaining portion is used as the testing data group; let the proportion of training data be α, then the number of training data groups is... Q represents the total number of data sets, and the number of test data sets is... Based on this, select loss functions including but not limited to mean squared error, and input the training data into the algorithm network model. Let the input data be... The weights of the network model are , bias is Then the forward propagation formula is: f is the activation function; calculate the loss value. , To predict the output vector, Given the true value vector, the mean squared error (MSE) is selected as the loss function:

[0076]

[0077] p represents the number of samples, and i represents the current index, from 1 to p, used to traverse all data points;

[0078] The model weights are updated using backpropagation, and the process is repeated until the model converges.

[0079] Reference Appendix Figure 3 To ensure stable operation, precise strategies are formulated: Optimized coordination under grid-connected mode: The integrated distribution and dispatch system determines the tie-line exchange power based on the optimal target for the entire grid, while the B microgrid management system plans a reasonable SOC level for energy storage based on new energy sources and load forecasts, and formulates a detailed island-wide source-grid-load-storage dispatch plan. In this process, by constructing a microgrid model and utilizing algorithms to comprehensively consider multiple factors (equipment characteristics, power forecasts, etc.), an optimal balance between generation costs and green energy consumption is achieved under constraints. Simultaneously, the plan is continuously optimized based on new data to ensure stable and efficient operation of the microgrid during grid connection, continuously moving towards maximizing the utilization of green energy and maintaining frequency and voltage stability.

[0080] Specifically, the implementation involves: constructing a microgrid model that encompasses, but is not limited to, the operational characteristics and interrelationships of renewable energy generation equipment, energy storage systems, loads, and tie-line components. Based on this model, the algorithm uses forecast data from renewable energy sources and loads to predict future power trends. For determining the tie-line power exchange target, the algorithm comprehensively considers the overall optimal operating target of the entire network and the power forecast within the microgrid. Within each control cycle, it solves an optimization problem to determine the optimal setpoint sequence for the tie-line power exchange over a future period, thereby achieving the lowest generation cost and the maximum green energy consumption while satisfying constraints including, but not limited to, tie-line power limitations and equipment operation limitations. The algorithm aims for optimal operation across the entire network. In determining the State of Charge (SOC) level and formulating an island-wide power generation, grid, load, and storage scheduling plan, it also pre-plans the charging and discharging power of energy storage at different times based on power prediction information. It schedules energy storage charging during periods of overcapacity to boost power generation, and allows energy storage to discharge during periods of undercapacity or peak load to meet load demand and track upper-level dispatch instructions. Simultaneously, it continuously optimizes this scheduling plan, recalculating and adjusting the tie-line exchange power target and energy storage scheduling strategy as time progresses and new prediction data becomes available. This ensures the microgrid always operates towards maximizing green energy consumption and maintaining stable frequency and voltage, achieving efficient and reliable grid-connected power regulation.

[0081] Reference Appendix Figure 4 - Figure 7Emergency power supply in off-grid mode: Based on actual power supply and load conditions, algorithms are used to combine grid-connected energy storage, diesel generators, and other equipment to prioritize power supply to important users. Regarding basic power balance and frequency and voltage stability, power output is adjusted in real time; for load management, priority-based orderly power supply is implemented, with precise monitoring and control through a microgrid coordinating controller to achieve reliable power supply to all loads on the island and effective utilization of green energy. Whether facing sudden off-grid situations or long-term independent operation, the microgrid can ensure stable and reliable power service to users.

[0082] Figure 5 In Chinese, EMS stands for Energy Management System.

[0083] DMS: Distribution Management System;

[0084] AMI: Advanced Metering Infrastructure;

[0085] PMS: Power Management System;

[0086] SOC: State of Charge;

[0087] MGCC (Microgrid Central Controller).

[0088] MGLC (Microgrid Local Controller);

[0089] PQ control (Power-Reactive Power Control).

[0090] VF control (Voltage-Frequency Control);

[0091] Specifically, the first step of control focuses on basic power balance and frequency and voltage stability. This involves using key power supply equipment, including but not limited to grid-connected energy storage and diesel generators, as the primary control targets. Utilizing their rapid power regulation, the output power is automatically adjusted based on the real-time power supply and demand of the microgrid to maintain the microgrid's frequency and voltage within a stable range. If a downward trend in the microgrid frequency is detected, indicating insufficient power generation, the diesel generator will rapidly increase its output, and the grid-connected energy storage will release energy according to preset rules to support the frequency recovery to normal levels. Conversely, if the frequency increases, the power output will be reduced accordingly. The second step of control focuses on orderly power supply management based on load priority. This involves detailed classification of various loads within the microgrid and clearly distinguishing between important users and low-priority users. When the power output meets the electricity needs of important users, priority is given to ensuring the continuous power supply to important users, including but not limited to hospitals, critical infrastructure, and important industrial production, to ensure that these users with extremely high requirements for power reliability are not affected by power fluctuations. When the power supply has a margin after meeting the needs of important users, the power supply to low-priority loads is gradually restored according to a pre-set priority sequence. The entire process is carried out by the microgrid coordinator to accurately monitor and control each power supply device and load, so that the microgrid can prioritize the key power demand in the off-grid state, and meet the power demand of more users to the greatest extent possible when conditions permit. This maintains the stability of the microgrid frequency and voltage, realizes the reliable power supply of the entire island's loads and the effective utilization of green energy, and provides a solid guarantee for the stable and efficient operation of the off-grid microgrid.

[0092] Coordinating diverse resources to enhance power supply efficiency: Focusing on the synergistic optimization of resources between new energy sources and various power sources and loads. Through intelligent algorithms, power is dynamically allocated based on new energy output and load demand, effectively suppressing new energy fluctuations, enhancing power supply reliability, and achieving optimal coordination between power sources. Simultaneously, the potential of flexible and controllable loads is explored, guiding their rational electricity consumption through demand response, enabling them to provide strong support for microgrid power balance at different times. In this process, the algorithm makes intelligent decisions based on load characteristics and real-time grid status, ensuring a significant improvement in the power supply efficiency of the entire microgrid automatic power regulation system (i.e., including power sources, energy storage, loads, distribution networks, and control systems within the microgrid), resulting in more rational and efficient resource utilization.

[0093] The resource collaborative optimization steps are as follows: First, achieve collaborative optimization between renewable energy sources (including but not limited to wind, solar, and biomass power generation) and power sources such as energy storage and diesel generators. Second, rationally allocate power supply based on renewable energy output and load demand using an algorithm consistent with the S4 control strategy formulation steps. Third, smooth out renewable energy fluctuations and improve power supply reliability, ensuring optimal coordination among power sources. Fourth, explore the grid support capacity of flexible and controllable loads (including but not limited to charging piles, V2G, shore power, agricultural production, and air conditioning). Fifth, guide loads to use electricity rationally at different times through demand response and other means, assisting in microgrid power balance. Sixth, during load control, the algorithm makes optimal decisions based on load characteristics and grid conditions.

[0094] Spatiotemporal integrated regulation to achieve optimal resource allocation: Dynamic adjustment in the time dimension: Fully considering the power generation and consumption characteristics and response speed of different resources, resources are flexibly adjusted on time scales such as day-ahead, intraday, and real-time. For example, energy storage or hydrogen production can be carried out in a timely manner during periods of high renewable energy generation, and the power grid can be used to charge energy storage during periods of low load. The algorithm optimizes and adjusts in real time according to time changes and resource status to ensure that the rational allocation of resources on the time axis maximizes benefits.

[0095] The specific spatiotemporal coordinated control steps are as follows: On a time scale, based on the power generation and consumption characteristics and response speed of different controllable resources, resources are rationally controlled at the day-ahead, intraday, and real-time time scales to maximize benefits. This includes, but is not limited to, energy storage and hydrogen production during periods of high new energy generation, and charging operations using the main grid power supply during off-peak periods. During the control process, an algorithm consistent with the S4 control strategy formulation steps is used for dynamic adjustments based on the time scale and resource status. On a spatial dimension, resource zoning and hierarchical coordination are achieved by following the principles of local, proximity, and voltage level balance and absorption. This includes hierarchical aggregation of adjustable capabilities between high-voltage transmission layers, medium-voltage distribution networks, and low-voltage distribution areas; hierarchical decomposition of control targets; and autonomous and mutually supportive zoning within the same layer. The periods of high new energy generation are typically concentrated between noon and afternoon, with specific times varying depending on regional policies.

[0096] Spatial hierarchical coordination: Following the principles of localization, proximity, and balance and absorption at the same voltage level, resources are coordinated in a zoned and hierarchical manner at the high-voltage transmission layer, medium-voltage distribution network layer, and low-voltage distribution area layer. Through hierarchical aggregation of adjustable capabilities, hierarchical decomposition of adjustment targets, and autonomous and mutually supportive zones within the same layer, the complexity of the microgrid control system is reduced, the adaptability between different levels and the stability of the overall control strategy are improved, thereby achieving optimal resource allocation in space and ensuring the efficient and stable operation of the microgrid.

[0097] Strengthening microgrid coordination to ensure stable operation: Leveraging the master-slave configuration and high-speed communication capabilities of the microgrid coordinator controller, high-speed real-time coordination of distributed resources in off-grid mode is achieved. Key control parameters are set and a reference model is established during the initialization phase, providing a basic framework for subsequent regulation. Real-time monitoring of grid operation data and distributed resource status calculation errors is performed, and control parameters are dynamically adjusted based on algorithmic rules, bringing the actual operation of the microgrid closer to the ideal state. In terms of functional implementation, various control functions (grid-connected / off-grid power coordination, voltage and frequency control, economical operation of energy storage, black start, etc.) are achieved by adjusting energy storage, distributed power sources, and other equipment, ensuring stable operation of the microgrid under various operating conditions. Whether in normal operation or during switching processes, a smooth transition provides a solid guarantee for the reliable operation of the microgrid.

[0098] The microgrid coordination control steps are as follows: This step is implemented based on a microgrid coordination controller, using a master-slave configuration. The master controller and slave controller communicate via GOOSE high-speed communication to achieve high-speed real-time coordination control of regional distributed resources in off-grid mode. It has processing capabilities including but not limited to grid-connected power coordination control, off-grid voltage and frequency control, off-grid power coordination control, grid-connected / off-grid switching control, energy storage economic operation control, and black start control. The microgrid coordination controller works in collaboration with the overall regulation algorithm to execute control functions, accepting decision instructions from the algorithm and executing corresponding operations. Based on the voltage and frequency monitoring of the microgrid and the main grid, the microgrid coordination controller automatically adjusts the frequency and voltage on the microgrid side to achieve quasi-synchronization requirements, realizing smooth switching and stable control during grid-connected to off-grid and off-grid to grid-connected processes. During the switching process, the algorithm adjusts the control parameters based on monitoring data to make the switching process stable and reliable.

[0099] Specifically, in the S7-1 initialization phase: the initial control parameters of the adaptive control algorithm are set in the microgrid coordinator controller. These parameters include, but are not limited to, the initial charge and discharge power coefficient of the energy storage system, the initial output power ratio of the distributed power source, and the initial gain of frequency and voltage regulation. At the same time, a reference model is established, which describes the ideal voltage, frequency, and power state of the microgrid under different operating modes: in grid-connected mode, the voltage and frequency in the reference model are consistent with the main grid; in off-grid mode, the reference model is based on local load demand and maintains a stable voltage and frequency range and power allocation method.

[0100] S7-2 Monitoring and Data Acquisition Phase: The microgrid coordinating controller acquires key operational data, including but not limited to voltage and frequency, of the microgrid and the main grid in real time; it also acquires the status information of distributed resources within the microgrid, including but not limited to the SOC of energy storage and the output power of distributed power sources; these data will be used as inputs to the algorithm for subsequent calculations and decisions.

[0101] S7-3 Error Calculation Stage: The real-time monitored microgrid voltage, frequency, and power data are compared with the ideal values ​​in the reference model to calculate the corresponding errors. For power coordination control, the deviation between the actual power and the reference power is calculated. The error reflects the degree of deviation between the current operating state of the microgrid and the ideal state, and is an important basis for the algorithm to adjust the control parameters.

[0102] S7-4 Adjustment Phase: Based on the calculated error and the pre-set algorithm rules, the control parameters are automatically adjusted. The algorithm rules are a set of rules that dynamically change the control parameters according to the error: In off-grid mode, the voltage error is positive (i.e., the measured voltage is higher than the reference voltage) and the frequency error is also positive (i.e., the measured frequency is higher than the reference frequency). The algorithm then reduces the charging power of the energy storage and the output power of the distributed power source to reduce the voltage and frequency. By continuously adjusting the control parameters according to the error, the actual operating state of the microgrid is made closer to the reference model.

[0103] S7-5 Function Implementation Phase: When the microgrid coordinator executes control functions, it utilizes the algorithm-adjusted control parameters to achieve functions including but not limited to grid-connected power coordination control, off-grid voltage and frequency control, and off-grid power coordination control. In off-grid voltage and frequency control, the algorithm adjusts the charging and discharging power of energy storage and the output power of distributed power sources to stabilize the voltage and frequency of the microgrid within the range specified by the reference model. During grid-connected / off-grid switching control, the algorithm dynamically adjusts control parameters based on data monitored during the switching process, ensuring smooth switching and stable control during grid-connected to off-grid and off-grid to grid-connected processes. For energy storage economic operation control, the algorithm dynamically adjusts the charging and discharging strategy of energy storage based on the SOC of energy storage, electricity price information, and the power demand of the microgrid to maximize economic benefits while meeting the operational requirements of the microgrid. In black-start control, the algorithm gradually adjusts the power output of the starting equipment based on the state changes during the microgrid recovery process, ensuring a smooth black-start process.

[0104] Optimizing distributed resource management to contribute to grid stability: The microgrid automatic power regulation system adopts differentiated interaction methods for different distributed resources and uniformly interfaces with the upper-level dispatch center. Based on the dynamic topology of the power grid, resources are aggregated into virtual aggregates for efficient monitoring and management. An innovative algorithm simulating biological community collaboration optimizes resource aggregation and scheduling strategies. This algorithm continuously improves resource allocation through repeated attempts by virtual "explorers" on resource connection paths and path selection optimization based on "imprints." Simultaneously, it simplifies resource equipment modeling, enhances overall operational economy, ensures stable service to the power grid by distributed resources, and assists in power regulation. The algorithm flexibly optimizes management strategies based on resource characteristics and grid demands, providing strong support for the stable operation of the microgrid.

[0105] The specific steps of the distributed resource management are as follows: the microgrid automatic power regulation system interacts with different types of distributed resources using different methods and converts them into a unified way to interact with the upper-level dispatcher. Based on the dynamic topology of the power grid, independent distributed resources are aggregated into virtual resource aggregates to achieve effective monitoring and management of distributed resources. During the distributed resource management process, algorithms are used to optimize resource aggregation and dispatch strategies. The algorithm simulates the collaborative pattern of social organisms in nature when exploring survival resources. These organisms leave a special "mark" in their paths, and subsequent organisms of the same species are influenced by this "mark" and tend to follow paths with denser "marks." Mapping this pattern to the field of distributed resource management, the connection links between distributed resources are like the movement trajectories of these organisms. Key indicators, including but not limited to the efficiency, cost, and stability of resource transmission, are abstracted as the salience of this "mark." In the initial stage, virtual "explorers" representing resource allocation attempts randomly try different resource connection paths and leave a corresponding degree of "mark" on the path based on the resource allocation effect of each attempt. Let the set of distributed resources be... , m is the number of distributed resources, and the set of connection links between resources is . This includes, but is not limited to, a vector composed of key indicators such as resource transmission efficiency, cost, and stability. Define the significance function of "imprint". = The virtual "explorer" k randomly attempts resource connection paths, assuming it chooses a path... 'l' represents the path length, and the degree of 'mark' left on the path is calculated based on the resource allocation effect. = h is a function that calculates the degree of imprinting based on the effect of resource allocation;

[0106] in, Let i be the i-th resource in the distributed resource set R; Representing resources and The connection path between them; i is the index, indicating that the calculation is performed on the i-th sample; j is the index, indicating that the calculation is performed on the j-th sample; Let J be the j-th resource in the distributed resource set R; It is a resource and The transmission efficiency between them involves data transmission rate, bandwidth, etc. It is a resource and Transmission costs between them, such as energy consumption and other expenses; It is a resource and The stability between them, such as reliability metrics or packet loss rate; Resource connection path The degree of prominence of the "imprint"; A vector containing multiple metrics is defined as follows: This is used to comprehensively evaluate the performance of the connection path; It represents a power regulation factor that determines the power allocation priority; This represents the input resource of a certain scheduling node; k is the number of trials, that is, the number of times a certain resource path is explored; A function that calculates the degree of imprinting based on the effectiveness of resource allocation.

[0107] As time goes by and through repeated exploration, subsequent "explorers" are guided by the "imprints" accumulated in the early stages, increasing the probability of them choosing a path. Related to the salience of the "marks" on the path; let the path The sum of the salience of the "imprints" on the surface is Then select the path The probability formula is , For the set of all possible paths; when the "explorer" chooses a path Then, update the "mark" level on the path based on the resource allocation effect. = By continuously repeating the above exploration process and optimizing path selection and "imprint" updates, resource aggregation and scheduling strategies are continuously improved.

[0108] Ultimately, based on the real-time dynamic topology of the power grid, the dispersed distributed resources are effectively aggregated into virtual resource aggregates; let the set of virtual resource aggregates be denoted as . q represents the number of aggregates, determined by specific aggregation rules. , For aggregate functions, The optimized path aggregates relevant resources to achieve precise monitoring and rational management of distributed resources, providing strong support for the scientific utilization of distributed resources and the stable operation of the power grid. Through intelligent methods, the modeling of distributed resource devices is simplified, their internal details are hidden, and the management of distributed resources is optimized to improve the overall operating economy. This enables distributed resources to provide continuous and stable services to the power grid and assists in microgrid power regulation. The algorithm optimizes management strategies based on the characteristics of distributed resources and the needs of the power grid.

[0109] Example 2

[0110] This embodiment, in conjunction with Example 1, discloses an automatic power regulation method for microgrids based on multi-source data fusion and intelligent prediction. The grid-connected regulation during periods of high renewable energy generation specifically includes the following steps:

[0111] Step 1. Data Fusion and Utilization:

[0112] The microgrid automatic power regulation system collects real-time data on photovoltaic power generation, wind power generation, and load (including real-time power of various loads such as residential and commercial electricity consumption) and non-real-time data (such as historical load curves) from Island B through systems such as EMS, DMS, AMI, and PMS. It also integrates meteorological data (such as solar irradiance, wind speed, and temperature). The data platform integrates this data, for example, by linking the location and capacity information of distributed photovoltaic power stations with their real-time power generation data, while also combining user information (such as electricity consumption habits and peak / valley periods for users in different areas) and maintenance plans (such as recent maintenance arrangements for a transmission line). This provides a comprehensive and accurate data foundation for subsequent regulation, ensuring data accuracy and timeliness, and preprocessing the data to transform it into a format suitable for subsequent algorithm analysis.

[0113] Step 2. Intelligent Prediction

[0114] Based on collected historical data (such as renewable energy power generation and load demand data for the past month) and real-time data, artificial intelligence algorithms (such as Long Short-Term Memory (LSTM) networks in deep learning algorithms) are used to make day-ahead, intraday, and real-time predictions of renewable energy power generation and load demand. During the data preparation phase, key information including renewable energy power generation and load demand is comprehensively collected and integrated with meteorological data. The data is normalized and standardized, and organized into time series data according to day-ahead (24 hours, with a 1-hour time step), intraday (4 hours, with a 15-minute time step), and real-time (a number of time intervals after the current moment). When constructing the network model, the number of input layer neurons is determined based on the feature dimensions of the input data (such as the number of features in meteorological data, historical power generation data, etc.). The number of hidden layer neurons is set according to the complexity of the problem (such as the complexity of the B Island power grid, renewable energy access status, etc.) and the amount of data. The number of output layer neurons is set to correspond to the prediction target (such as predicting renewable energy power generation and load demand at each future time step). The system divides the data into training and testing sets (e.g., 80% for training and 20% for testing). Mean squared error is selected as the loss function. The training data is input into the algorithm's network model. After forward propagation to obtain the predicted output, the loss value is calculated. Backpropagation is then used to update the model weights, iterating repeatedly until the model converges. Finally, real-time renewable energy generation, load demand, and meteorological data are input into the trained model to predict future generation and load demand. For example, it might predict that renewable energy generation will gradually increase in the morning and peak at noon.

[0115] Step 3. Formulating Control Strategies

[0116] In grid-connected mode, the integrated dispatch system determines the tie-line exchange power target based on the optimal operation goal of the entire network (such as achieving the lowest power generation cost and the maximum green energy consumption under the premise of ensuring the safe and stable operation of the power grid), taking into account the forecast of peak renewable energy generation, and allows some renewable energy power generation to be transmitted externally. The B microgrid management system, based on renewable energy and load forecasts, determines the SOC level for energy storage. During peak renewable energy generation periods (such as 10:00 AM to 2:00 PM), it schedules energy storage charging or hydrogen production according to the predicted power generation and load demand, and plans the charging and discharging power of energy storage at different times in advance. For example, if it is predicted that renewable energy power generation will exceed local load demand at 11:00 AM, the energy storage will be charged at a certain power starting at 11:00 AM. Simultaneously, an island-wide source-grid-load-storage dispatch plan is formulated, tracking the upper-level dispatch demand. Through optimization algorithms, power is rationally allocated to increase the local consumption of renewable energy (such as allocating excess renewable energy power to adjustable industrial loads), reduce the output of traditional power sources (such as diesel generators), and maximize green energy consumption and frequency and voltage stability.

[0117] Step 4. Resource Coordination Optimization

[0118] During periods of high renewable energy generation, wind, solar, and other renewable energy sources operate in synergistic and optimized manner with energy storage, diesel generators, and other power sources. Based on renewable energy output (e.g., real-time power of wind and solar power) and load demand (e.g., actual power demand of various loads), optimization algorithms are used to rationally allocate power and smooth out renewable energy fluctuations. For example, when solar power generation suddenly increases, the algorithm controls energy storage charging power to increase accordingly, while diesel generator output is appropriately reduced to maintain microgrid power balance. Simultaneously, the grid's support capacity for flexible and adjustable loads (e.g., charging piles, air conditioners) is utilized, and demand response measures guide loads to use electricity rationally during periods of high renewable energy generation. For instance, control signals are sent to charging piles to encourage them to charge more during periods of high renewable energy generation; for air conditioning loads, temperature setpoints or operating times are adjusted to assist in microgrid power balance.

[0119] Step 5. Spatiotemporal Coordinated Regulation

[0120] On a time scale, based on the power generation and consumption characteristics and response speed of different controllable resources, day-ahead dispatching plans energy storage charging schedules and adjustable load power consumption arrangements according to the predicted surge in renewable energy generation. Intraday dispatching further optimizes and adjusts based on real-time monitoring of renewable energy power generation and load changes. Real-time control ensures that control strategies are accurately executed according to actual conditions at each time step. For example, during periods of high renewable energy generation, if real-time monitoring shows that photovoltaic power generation exceeds expectations, energy storage charging power and load power consumption strategies are adjusted promptly. On a spatial scale, following the principles of local, nearby, and voltage-level balance and absorption, local photovoltaic power generation is prioritized to supply nearby loads, with excess power stored in local energy storage devices or transmitted to adjacent areas for absorption via suitable transmission lines. For example, the power generated by distributed photovoltaic power stations in a certain area of ​​an island is prioritized to meet the residential and commercial loads of that area. If there is any surplus, it is transmitted to energy storage stations in adjacent areas for storage or directly supplied to the loads of that area via the regional distribution lines, achieving zoned and hierarchical resource coordination, hierarchical aggregation of adjustable capacity, hierarchical decomposition of adjustment targets, and autonomy and mutual assistance among zones at the same level.

[0121] Step 6. Microgrid Coordination and Control

[0122] The microgrid coordinating controller adopts a master-slave configuration, with high-speed GOOSE communication between the master and slave controllers to monitor key operational data such as voltage and frequency of the microgrid and the main grid in real time, as well as status information such as the SOC of energy storage and the output power of distributed generation. During the initialization phase, initial control parameters for the adaptive control algorithm are set, such as the initial charge / discharge power coefficient of the energy storage system, the initial output power ratio of the distributed generation, and the initial gains for frequency and voltage regulation. Simultaneously, a reference model is established to describe the ideal voltage, frequency, and power state of the microgrid under grid-connected mode (consistent with the main grid). During the monitoring and data acquisition phase, real-time operational data is continuously acquired. In the error calculation phase, the real-time monitored microgrid voltage, frequency, and power data are compared with the ideal values ​​in the reference model to calculate the error. For power coordination control, the deviation between the actual power and the reference power is calculated. In the adjustment phase, based on the calculated error and pre-set algorithm rules (e.g., when the voltage error and frequency error are both positive, the charging power of the energy storage is reduced and the output power of the distributed generation is decreased), the control parameters are automatically adjusted to make the actual operating state of the microgrid approach the reference model. During the functional implementation phase, the adjusted control parameters are used to achieve functions such as grid-connected power coordination control and off-grid voltage and frequency control, ensuring that the power exchange between the microgrid and the main grid is reasonable when new energy sources are generated in large quantities, and that the voltage and frequency within the microgrid are stable within the specified range. For example, the voltage of the microgrid is stabilized within ±5% of the rated voltage range, and the frequency is stabilized within 50Hz ±0.2Hz.

[0123] Step 7. Distributed Resource Management

[0124] The microgrid automatic power regulation system interacts with different types of distributed resources (such as distributed photovoltaic power stations and decentralized wind farms) using various methods. For example, it communicates with distributed photovoltaic power stations through dedicated communication lines to obtain information such as their power generation and operating status, and sends control commands (such as adjusting power output). This is then converted into a unified method for interaction with the higher-level dispatch center. Based on the dynamic topology of the power grid (such as changes in the access points of distributed power sources and the online connection of new distributed resources), optimization algorithms are used to aggregate independent distributed resources into virtual resource aggregates. For instance, multiple geographically close distributed photovoltaic power stations with similar power generation characteristics can be aggregated into a single virtual power source for unified management. This enables effective monitoring and management of distributed resources, optimizes resource aggregation and dispatch strategies, improves overall operational economy, ensures that distributed resources provide continuous and stable services to the grid, and assists in microgrid power regulation, such as adjusting the output power of the virtual resource aggregate based on the power balance requirements of the microgrid.

[0125] Example 3

[0126] This embodiment, in conjunction with Example 1, discloses an automatic power regulation method for microgrids based on multi-source data fusion and intelligent prediction. The off-grid regulation during peak load periods and when renewable energy output is insufficient specifically includes the following steps:

[0127] 1. Data fusion and utilization

[0128] The microgrid automatic power regulation system utilizes various data sources, focusing on collecting load demand data during peak load periods (such as 7 PM to 10 PM) (including power demands from various loads such as peak residential electricity consumption and concentrated commercial electricity consumption periods), as well as data on lower renewable energy generation power, and integrating meteorological data (such as the impact of low nighttime wind speeds on wind power generation). The data platform integrates data, for example, linking load data from different regions with distributed resource information for those regions, while also combining user information (such as the distribution and electricity priority of important users) and maintenance plans (such as the recent maintenance schedule for a certain energy storage device), providing accurate data support for off-grid regulation, ensuring data accuracy and timeliness, and performing preprocessing to adapt to subsequent algorithm processing.

[0129] 2. Intelligent prediction

[0130] Based on historical peak load data (such as nighttime peak load data from the past week), real-time monitored load growth trends, and renewable energy power generation forecasts (such as predicting nighttime wind and solar power generation based on weather forecasts and current renewable energy equipment status), algorithms are used for accurate prediction. In the data preparation phase, key information including load demand and renewable energy power generation is collected and integrated with meteorological data. This data is then normalized and standardized, and organized into time series data according to a suitable time scale for peak load prediction (such as predicting nighttime load demand changes in 30-minute time steps). When constructing the network model, the number of input layer neurons is determined based on the feature dimensions of the input data (such as the number of features in historical load data and meteorological data). The number of hidden layer neurons is set based on the complexity of the problem (such as the complex electricity consumption situation during nighttime peak load on Island B) and the amount of data. The number of output layer neurons is also set to correspond to the prediction target (such as predicting load demand and renewable energy power generation at each time step during peak load periods). The model is divided into training and testing data groups, and a suitable loss function (such as mean absolute error) is selected. The training data is then input into the model for training. The model undergoes forward propagation, loss calculation, and backpropagation to update the model weights. This process is iterated repeatedly until the model converges. Finally, real-time load demand, renewable energy generation capacity, and meteorological data are input into the trained model to predict power deficits during peak load periods.

[0131] 3. Formulation of regulatory strategies

[0132] In off-grid mode, based on power supply and load conditions, algorithms are used to prioritize power supply to critical users through the joint control of grid-connected energy storage, diesel generators, and other equipment, before gradually restoring power to lower-priority loads. For example, first, continuous power supply to critical users such as hospitals and key infrastructure is ensured, and then lower-priority loads such as commercial loads are gradually restored based on remaining power capacity. Before peak load arrives, a reasonable discharge plan is formulated for energy storage based on forecast information, such as starting to discharge energy storage at a higher power level at 6:30 pm to cope with the upcoming peak load. At the same time, based on the insufficient power generation of new energy sources, the start-up and output of diesel generators are reasonably arranged to ensure power supply reliability. For example, if it is predicted that new energy generation cannot meet the power supply needs of critical users, diesel generators are started in advance and their output is adjusted.

[0133] 4. Resource Coordination Optimization

[0134] When renewable energy generation is insufficient, diesel generators start up and adjust their output according to load demand, while energy storage discharges to supplement the power deficit. The algorithm optimizes power allocation based on load characteristics (such as the importance and adjustability of different types of loads) and grid conditions. For example, for critical loads that cannot be interrupted (such as hospital equipment), priority is given to ensuring their power supply; for adjustable loads (such as some commercial air conditioning units), their operating status is controlled to reduce load demand. At the same time, the potential of flexible and adjustable loads is explored, such as by sending control signals to air conditioning loads to adjust their temperature setpoints or operating periods, thereby reducing power consumption, assisting in microgrid power balance, and improving power supply reliability.

[0135] 5. Spatiotemporal coordinated regulation

[0136] On a time scale, day-ahead dispatching plans strategies in advance for nighttime peak load periods, such as energy storage discharge plans and diesel generator backup arrangements. Intraday dispatching further optimizes these strategies based on real-time load changes and renewable energy generation. Real-time control ensures timely adjustments to resource allocation during peak load periods. For example, during peak load periods, if load demand exceeds expectations, diesel generator output can be increased or energy storage discharge strategies adjusted accordingly. Spatially, following the principles of local availability and proximity, priority is given to utilizing local energy storage and nearby startable diesel generators to meet load demands, achieving regional autonomy. If local resources are insufficient, support is obtained from other regions through hierarchical coordination (e.g., coordinating resource sharing between adjacent regions through a microgrid coordinator), ensuring balance and absorption at the same voltage level. For instance, during peak load periods in a certain region, priority is given to using local energy storage and diesel generators. If this still cannot meet the demand, some power support is obtained from microgrids in adjacent regions.

[0137] 6. Microgrid Coordination and Control

[0138] The microgrid coordinator monitors key operational data of the microgrid in real time, such as voltage and frequency, as well as status information like the SOC of energy storage and the output of diesel generators. During the initialization phase, suitable control parameters for off-grid operation are set, such as the charge / discharge power coefficient of the energy storage system and the initial output setting of the diesel generator. The monitoring and data acquisition phase continuously acquires real-time data. In the error calculation phase, the real-time monitored microgrid voltage, frequency, and power data are compared with ideal values ​​in the reference model (voltage and frequency ranges and power distribution methods based on local load demand and maintaining stable operation) to calculate the error. In the adjustment phase, control parameters are automatically adjusted according to the error and algorithm rules; for example, when the measured voltage is lower than the reference voltage and the frequency shows a downward trend, the energy storage discharge power is increased or the diesel generator output is increased. In the functional implementation phase, by controlling diesel generators, energy storage, and other equipment, off-grid voltage and frequency control, and power coordination control functions are achieved to ensure stable operation of the microgrid during peak load periods and when renewable energy output is insufficient. The voltage is stabilized within ±7% of the rated voltage range, and the frequency is stabilized within 50Hz ±0.5Hz, ensuring reliable power supply for important users.

[0139] 7. Distributed Resource Management

[0140] The microgrid automatic power regulation system interacts with distributed resources to obtain their status information and manage them uniformly. For example, it monitors the remaining power of distributed energy storage and the operating status of distributed power sources in real time. Based on the dynamic topology of the power grid (e.g., distributed resources in certain areas may become critical supports during peak load periods) and load demand, algorithms are used to optimize the scheduling of distributed resources. Distributed resources are aggregated into virtual resource aggregates, such as managing dispersed small energy storage devices and distributed power sources together. This improves the efficiency of monitoring and managing distributed resources, ensuring that they can effectively assist the microgrid in power regulation during peak load periods and provide stable services to the power grid. For example, the output power of the virtual resource aggregates can be adjusted according to load demand, prioritizing the power supply needs of important loads.

[0141] Example 2

[0142] This embodiment, described in conjunction with Example 1, discloses an automatic power regulation method for microgrids based on multi-source data fusion and intelligent prediction. This method addresses unplanned grid disconnection caused by extreme weather conditions and includes the following steps:

[0143] 1. Data fusion and utilization

[0144] When extreme weather events (such as strong typhoons and torrential rain) occur, the microgrid's automatic power regulation system rapidly collects grid operation data, including current load demand (e.g., loads may suddenly increase or decrease due to weather), renewable energy generation capacity (e.g., strong winds may affect wind power, and torrential rain may affect photovoltaic power), and energy storage status (e.g., remaining energy capacity and charging / discharging status). This data is combined with meteorological data (e.g., real-time wind speed, rainfall, and air pressure) and previously collected information on the B-island grid infrastructure (e.g., the wind resistance of transmission lines and the stability of towers). The data platform integrates this data, for example, linking load data from areas heavily affected by weather with the status of distributed resources and transmission lines in that area. It also combines user information (e.g., user distribution and identification of important users) and maintenance plans (e.g., preventative maintenance of equipment scheduled before extreme weather events) to provide crucial data support for regulation during unplanned grid disconnection, ensuring data accuracy and timeliness, and rapidly preprocessing data to meet the algorithmic processing needs of emergency situations.

[0145] 2. Intelligent prediction

[0146] Based on historical data from extreme weather events (such as grid operation data under similar weather conditions in the past) and real-time monitoring, algorithms are used to predict load demand changes and renewable energy generation during unplanned grid disconnection. In the data preparation phase, key information including load demand and renewable energy generation is collected and integrated with meteorological data. This data undergoes rapid normalization and standardization, and is then organized into time series data according to a time scale suitable for extreme weather conditions (e.g., predicting grid operation over the next few hours in 15-minute time steps). When constructing the network model, the number of input layer neurons is determined based on the feature dimensions of the input data (e.g., the number of features in meteorological data and grid operation data). The number of hidden layer neurons is set based on the complexity of the problem (e.g., the complex impact of extreme weather on the grid) and the amount of data. The number of output layer neurons is also set to correspond to the prediction objective (e.g., predicting load demand and renewable energy generation at each time step during unplanned grid disconnection). The model is divided into training and testing data groups, and a suitable loss function (such as root mean square error) is selected. The training data is then input into the model for rapid training (possibly using a simplified training algorithm or fine-tuning a previously trained model). The model undergoes forward propagation, loss calculation, and backpropagation to update the model weights, aiming for rapid convergence. Finally, real-time load demand, renewable energy generation capacity, and meteorological data are input into the trained model to predict power balance during unplanned grid disconnection periods. For example, it can predict whether the remaining grid can meet load demand and how much support renewable energy generation can provide after strong winds damage some transmission lines.

[0147] 3. Formulation of regulatory strategies

[0148] Following unplanned grid disconnection, the integrated distribution and dispatch system rapidly adjusts the operating mode of distributed resources. For example, based on predicted renewable energy generation and load demand, it adjusts the output settings of distributed power sources (e.g., reducing unstable renewable energy output to prevent grid impact). The microgrid management system, based on the current operating mode, quickly adjusts protection settings and load balancer status to ensure the safe and stable operation of the grid under unplanned disconnection conditions. Simultaneously, based on power supply and load conditions, it prioritizes power supply to important users, such as hospitals and emergency command centers, and formulates load switching plans for emergency situations, such as cutting off non-critical commercial loads to ensure power supply to critical loads. It utilizes grid-connected energy storage and diesel generators to establish microgrid voltage and frequency, and formulates energy storage charging and discharging strategies based on forecast information. For example, when renewable energy generation is unstable, it rationally controls the charging and discharging of energy storage to maintain microgrid power balance, gradually restoring power to low-priority loads. Simultaneously, it achieves optimized coordinated control of power generation, grid, load, and storage through the microgrid coordinating controller.

[0149] 4. Resource Coordination Optimization

[0150] During unplanned grid disconnection, renewable energy sources such as wind and solar power, along with energy storage and diesel generators, work together to address extreme weather challenges. Renewable energy sources generate electricity based on weather changes (e.g., wind power may fluctuate when wind speeds are unstable), energy storage rapidly adjusts its charging and discharging based on load demand and renewable energy generation, and diesel generators serve as backup power, activating promptly when renewable energy generation is insufficient and energy storage capacity is low. Algorithms optimize power allocation based on real-time resource status (e.g., remaining energy storage capacity and renewable energy power fluctuations) and load demand (e.g., power demands from critical users and remaining loads), ensuring stable grid operation during unplanned grid disconnection due to extreme weather, mitigating the impact of renewable energy fluctuations on the microgrid's automatic power regulation system during off-grid operation, and improving power supply reliability. For example, when wind power generation suddenly increases, the algorithm controls energy storage charging power to increase, while diesel generator output is appropriately reduced; when wind power generation suddenly decreases, energy storage discharges to compensate for the power shortfall, and diesel generator output responds quickly by increasing its output. At the same time, we can explore the grid support capacity of flexible and controllable loads (such as air conditioners and some industrial loads) and guide loads to use electricity rationally under extreme weather conditions through demand response measures. For example, we can send control signals to air conditioners to adjust their temperature setpoints or operating modes, reduce power consumption, and assist in the power balance of microgrids.

[0151] 5. Spatiotemporal coordinated regulation

[0152] On a temporal scale, due to the uncertainty of extreme weather, real-time control becomes crucial after unplanned grid disconnection. Algorithms are used to rapidly adjust control strategies based on real-time monitoring of resource status and load demand, such as reassessing renewable energy generation capacity and load demand every 15 minutes, dynamically adjusting the charging and discharging power of energy storage and the output of diesel generators. Spatially, following the principles of locality and proximity, priority is given to utilizing locally undamaged or minimally affected distributed resources (such as local distributed energy storage and still operational distributed power sources) to meet the needs of important local loads, achieving regional autonomy. For example, if a transmission line in a certain area is damaged due to extreme weather, distributed power sources and energy storage within that area form an independent power supply unit, prioritizing power supply to important loads such as hospitals and emergency shelters within the area. If local resources cannot meet all demands, a microgrid coordinator controller coordinates resource sharing between adjacent areas to ensure balance and absorption at the same voltage level, introducing some power support from microgrids in adjacent areas.

[0153] 6. Microgrid Coordination and Control

[0154] The microgrid coordinating controller plays a crucial role during unplanned grid disconnection. Its master-slave configuration uses GOOSE high-speed communication between the master and slave controllers to monitor key operational data of the microgrid in real time, such as voltage and frequency, as well as information on the SOC of energy storage, the output power of distributed power sources, and the operating status of the diesel generator. During the initialization phase, control parameters suitable for unplanned grid disconnection emergencies are quickly set, such as the emergency charging and discharging power coefficient of the energy storage system, and the rapid start-up and maximum power output settings of the diesel generator. During the monitoring and data acquisition phase, real-time data is continuously and rapidly acquired. In the error calculation phase, the real-time monitored microgrid voltage, frequency, and power data are compared with ideal values ​​in the reference model (voltage and frequency ranges and power distribution methods based on local critical load demand and stable operation under extreme weather conditions) to calculate the error. In the adjustment phase, based on the error and pre-set emergency algorithm rules (such as immediately starting the diesel generator and outputting maximum power when the voltage drops sharply and the frequency is unstable, while simultaneously adjusting the energy storage discharge strategy), the control parameters are automatically adjusted to bring the actual operating state of the microgrid closer to the reference model as quickly as possible. During the functional implementation phase, by controlling equipment such as diesel generators, energy storage, and distributed power sources, functions such as off-grid voltage and frequency control and power coordination control are realized to ensure the stable operation of the microgrid when unplanned off-grid situations occur due to extreme weather. The voltage is stabilized within a range as close as possible to the rated voltage (e.g., ±10% of the rated voltage), and the frequency is stabilized within an acceptable range (e.g., 50Hz ±1Hz), ensuring reliable power supply for important users.

[0155] 7. Distributed Resource Management

[0156] Off-grid microgrid automatic power regulation systems employ emergency interaction with distributed resources to ensure their status information is still available and effectively managed even under extreme weather conditions. For example, they maintain contact with distributed power sources and energy storage devices through backup communication channels (such as wireless communication) to obtain their operating status (e.g., fault status, remaining power), and send emergency control commands (e.g., adjusting power output, switching operating modes). Based on the dynamic topology of the power grid (e.g., partial line tripping due to extreme weather, islanded operation of distributed resources), optimization algorithms are used to aggregate distributed resources into virtual resource aggregates, improving the efficiency of monitoring and management. For instance, islanded distributed power sources and energy storage devices in a certain area due to transmission line faults can be aggregated into a single virtual power source, with unified power dispatch to prioritize the power supply needs of important loads in the area. This ensures that distributed resources can continuously provide stable service to the grid under extreme weather conditions, assisting the microgrid in achieving power balance and stable operation. Simultaneously, intelligent methods simplify the modeling of distributed resource devices, shielding their internal details and optimizing distributed resource management to improve overall operational economy. For example, they can quickly assess the availability and reliability of distributed resources under extreme weather conditions and rationally allocate their participation in microgrid power regulation.

Claims

1. A method for automatic power regulation in a microgrid, characterized in that, Includes the following steps: S1. Construct an integrated construction architecture to realize the switching and coordinated control of power grids A and B in grid-connected and off-grid modes; S2. Collect, process, and analyze multi-source data, and integrate multi-source data through a data platform to obtain historical and real-time data; S3. Based on the historical and real-time data, intelligently predict the power generation of new energy power generation equipment and the load demand of energy storage system on a time scale. S4. In grid-connected mode, the integrated distribution and dispatching system determines power dispatch based on the optimal target of the entire network. The B microgrid management system formulates a source-grid-load-storage dispatch plan based on power generation forecast and energy storage load forecast data. In S4, the integrated distribution and dispatch system in grid-connected mode specifically includes the following steps: S41. Construct a microgrid model and use the power generation forecast of new energy power generation equipment and the load forecast data of energy storage system to predict the power trend in the future period. Taking into account the optimal operation target of the whole network and the power forecast situation inside the microgrid, determine the optimal setpoint sequence of tie line exchange power in the future preset time period by solving an optimization problem in each control cycle, so as to achieve the operation target under the preset constraints. S42. Based on power forecast information, plan the charging and discharging power of energy storage at different times in advance. According to the predicted peak and off-peak periods of new energy power generation, allow energy storage to charge when there is excess power generation and allow energy storage to discharge when there is insufficient power generation or peak load. At the same time, continuously optimize the scheduling plan, recalculate and adjust the tie-line exchange power target and energy storage scheduling strategy to ensure that the microgrid always operates in the direction of maximizing green energy consumption and frequency and voltage stability. S5. In off-grid mode, power supply management is carried out step by step according to user priority based on power supply and load conditions to ensure frequency and voltage stability. S6. In both grid-connected and off-grid modes, power allocation, load regulation, resource optimization and scheduling are carried out according to load demand and power generation conditions, and distributed resources are controlled in real time through a coordination controller. In S6, the real-time control of distributed resources through the coordination controller specifically includes the following steps: S61. During the load regulation process, the optimal decision is made based on the load characteristics and the grid status to guide the load to be adjusted at different times and assist the power balance of the microgrid. S62. Allocate power according to the power generation of new energy power generation equipment and the load demand of energy storage system, explore the grid support capacity of flexible and adjustable load, and dynamically adjust resources in time and space based on the power generation and consumption characteristics of adjustable resources to achieve regional coordination and optimized allocation of resources. The specific adjustment method for dynamically adjusting resources in the time and space dimensions is as follows: resources are adjusted on a time scale based on the power generation and consumption characteristics and response speed of different controllable resources, and dynamic adjustments are made according to the time scale and resource status during the adjustment process; in the spatial dimension, the principle of local, nearby, and same voltage level balance and absorption is followed to achieve the zoning and hierarchical coordination of resources; wherein, in the grid-connected mode, the source, grid, load and storage are optimized and coordinated based on the whole grid, and in the off-grid mode, the B microgrid management system independently controls the resources within the island as a microgrid management system; The dynamic resource adjustment includes energy storage and hydrogen production during periods of high renewable energy generation, as well as charging operations using the power grid during periods of low load. S63. The coordination controller performs distributed resource coordination control in off-grid mode through GOOSE communication and automatically adjusts the microgrid frequency and voltage to ensure quasi-synchronous operation of the power grid. S64. The energy management system interacts with different types of distributed resources using different methods and converts them into a unified way to interact with the upper-level dispatcher. Based on the dynamic topology of the power grid, it aggregates independent distributed resource individuals into virtual resource aggregates to achieve monitoring and management of distributed resources. In the process of distributed resource management, it optimizes resource aggregation and dispatch strategies. In S63, the distributed resource coordination control includes grid-connected power coordination control, off-grid voltage and frequency control, off-grid power coordination control, grid-connected / off-grid switching control, energy storage economic operation control, and black start control processing.

2. The microgrid automatic power regulation method according to claim 1, characterized in that, In S63, the handover and control process during grid-connected to off-grid and off-grid to grid-connected transitions specifically includes the following steps: S631. Initialization Phase: Set the initial control parameters of the adaptive control algorithm in the microgrid coordinating controller and establish a reference model to describe the ideal values ​​of the microgrid under different operating modes: In grid-connected mode, the voltage and frequency in the reference model are consistent with the main grid; in off-grid mode, the reference model is based on the local load demand and maintains a stable voltage and frequency range and power distribution method. The initial control parameters include the initial charge and discharge power coefficient of the energy storage system, the initial output power ratio of the distributed power source, and the initial gain of frequency and voltage regulation. S632, Monitoring and Data Acquisition Phase: The microgrid coordinating controller acquires real-time operational data from the microgrid and the main grid; it also acquires status information of distributed resources within the microgrid. S633, Error Calculation Stage: The real-time monitored microgrid data is compared with the ideal value in the reference model to calculate the corresponding error; for power coordination control, the deviation between the actual power and the reference power is calculated, and the error is used to reflect the degree of deviation between the current operating state of the microgrid and the ideal state. S634, Adjustment Stage: Based on the calculated error and the pre-set algorithm rules, the control parameters are automatically adjusted; the algorithm rules are based on the rule of dynamically changing the control parameters according to the error: in off-grid mode, the voltage error is positive, that is, the measured voltage is higher than the reference voltage, and the frequency error is also positive, that is, the measured frequency is higher than the reference frequency. S635, Function Implementation Stage: When the microgrid coordinating controller executes control functions, it uses the adaptive control algorithm to adjust the control parameters to achieve functions such as grid-connected power coordination control, off-grid voltage and frequency control, and off-grid power coordination control. In off-grid voltage and frequency control, the voltage and frequency of the microgrid are stabilized within the range specified by the reference model by adjusting the charging and discharging power of energy storage and the output power of distributed power sources. In the process of grid-connected to off-grid switching control, the algorithm dynamically adjusts the control parameters according to the data monitored during the switching process to realize the switching and control during the grid-connected to off-grid and off-grid to grid-connected processes.

3. The microgrid automatic power regulation method according to claim 1, characterized in that, In S64, the optimized resource aggregation and scheduling strategy specifically includes the following steps: S641. In the initial stage, the virtual "explorer" representing resource allocation attempts will randomly try on each resource connection path and leave a corresponding degree of "mark" on the path based on the resource allocation effect of each attempt; the degree of "mark" left on the path is calculated based on the resource allocation effect. S642. Over time and through repeated exploration, the path selection and "imprint" level are continuously optimized, resulting in continuous improvement of resource aggregation and scheduling strategies. S643. Based on the real-time dynamic topology of the power grid, dispersed distributed resources are effectively aggregated into a virtual resource aggregate.

4. The microgrid automatic power regulation method according to claim 1, characterized in that, In S3, the intelligent prediction specifically includes the following steps: S31. Preprocess the historical data and real-time data to construct an algorithm network model; S32. Divide the preprocessed data into training data group and test data group to train the algorithm network model; S33. Update the weights of the algorithm network model using backpropagation, and iterate repeatedly until the algorithm network model converges.

5. The microgrid automatic power regulation method according to claim 1, characterized in that, In S5, the power supply management for users in off-grid mode specifically includes the following steps: S51. Balance the base power and frequency voltage, take grid-type energy storage and diesel generator power equipment as key control objects, and automatically adjust the output power according to the real-time power supply and demand of the microgrid to keep the frequency and voltage of the microgrid within the preset range. S52. Power supply management is based on user priority. Users in the microgrid are classified according to priority, and users are divided into important users and multi-level priority users. When the power output meets the power demand of important users, priority is given to ensuring the continuous power supply of important users. When the power output still has remaining power after meeting the needs of important users, power supply to users of different priorities is gradually restored according to a pre-set priority sequence. The power supply management is achieved by monitoring and controlling the power supply equipment and loads of each user through a microgrid coordinating controller.

6. The automatic power regulation method for a microgrid according to claim 5, characterized in that, In S51, the specific steps for automatically adjusting the output power are as follows: if the frequency of the microgrid shows a downward trend, indicating insufficient power generation, the diesel generator will quickly increase its output and the grid-type energy storage will release electrical energy in a timely manner according to the preset rules to support the frequency to return to the normal level; conversely, if the frequency increases, the power generation output will be reduced accordingly.

Images