Intelligent power grid optimal scheduling method and system based on multi-element energy storage cooperative scheduling

Abstract

The invention discloses an intelligent power grid optimal scheduling method and system based on multivariate energy storage cooperative scheduling, and relates to the technical field of power grid optimal scheduling, and the method comprises the following steps: building a prediction model based on first data, generating prediction data, coupling energy storage characteristic parameters of different types of energy storage equipment with the prediction data, and obtaining a prediction model; establishing a multi-energy collaborative scheduling model; dynamically screening the energy storage scheduling strategy set based on a preset real-time performance evaluation index to generate an optimal strategy subset; according to the optimal strategy subset, performing differentiated charging and discharging control instructions on the energy storage equipment cluster; and collecting second data in the charge and discharge control process, calculating a deviation value between the second data and the prediction data, converting the deviation value into a feature vector, inputting the feature vector into a preset incremental learning algorithm, and optimizing parameters of the multi-energy collaborative scheduling model. Layered screening is implemented in combination with real-time performance evaluation indexes, and it is ensured that the optimal scheduling scheme can be rapidly selected in different time periods and under the uncertain disturbance condition.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid optimization and dispatching, and more specifically, to a smart grid optimization and dispatching method and system based on multi-energy storage coordinated dispatching. Background Art

[0002] As the proportion of renewable energy continues to increase and grid load characteristics become increasingly complex, traditional power system dispatch methods face unprecedented challenges. Multi-element energy storage systems (including electrochemical energy storage, flywheel energy storage, pumped hydro, hydrogen energy storage, and thermal energy storage) offer advantages across diverse timescales, energy densities, and response speeds, making them a crucial enabler for enhancing grid flexibility and resilience. Smart grid optimization methods based on the coordinated dispatch of multi-element energy storage are becoming a key area of ​​research and engineering application in smart grids.

[0003] In practical applications, a single type of energy storage often struggles to meet the dual requirements of large-scale energy regulation and rapid dynamic response. Therefore, the complementary collaboration of multiple energy storage resources, coupled with layered, time-sharing, and zoned scheduling based on their respective characteristics, can effectively improve overall system performance. At the same time, the output of renewable energy sources such as wind and solar power is highly uncertain, requiring scheduling strategies to possess robustness and adaptability. Furthermore, achieving real-time scheduling typically requires the integration of edge computing, 5G communications, and IoT technologies to build a cloud-edge-end collaborative control system.

[0004] Existing multi-element energy storage collaborative optimization scheduling methods primarily utilize multi-objective optimization, rolling horizon control, distributed optimization algorithms (such as ADMM), and reinforcement learning techniques, taking into account multiple indicators such as economy, reliability, and environmental friendliness. At the same time, emerging technologies such as digital twins and artificial intelligence are being widely used in energy storage system state prediction, fault warning, and adaptive optimization, improving the intelligence of scheduling decisions.

[0005] For example, the invention patent announcement with announcement number: CN118539521B discloses a source-grid-load-storage coordination method and system, which relates to the field of data processing technology. A preliminary data set is formed by collecting power generation data of renewable energy power stations, real-time operation status data of power grids, and power load demand data of end users from multiple sources in real time; a prediction model is obtained based on data set training to output prediction results of renewable energy power generation and user load change trends in a future period of time; the prediction results are combined with the real-time operation status data of the power grid to construct a multi-objective optimization model, which includes multiple control strategies; the multi-objective optimization model is solved according to the multiple control strategies to obtain the optimal scheduling strategy to adjust the operation status of the power grid. After collecting necessary data from multiple sources in real time, the multi-objective optimization framework is introduced to comprehensively consider various constraints and objective functions to achieve the global optimal scheduling strategy.

[0006] For example, the invention patent announcement with announcement number: CN118263908A discloses a method and system for improving energy storage efficiency combined with energy management, which relates to the technical field of energy storage systems, including: reading the installed capacity and grid-connected status of the energy storage system in the preset control area, and monitoring the energy storage distribution status; making production capacity forecasts on the energy side and absorption forecasts on the demand side, performing forecast time series fitting, and determining energy storage trends; constructing an energy storage management module, including an energy management block and a battery management block of a staggered network structure, which can interact with data laterally and is configured with a penalty function; making electrochemical conversion decisions and energy scheduling decisions, and determining an energy storage scheduling plan; transmitting to the energy management system to assist the energy storage converter in performing energy storage management based on energy scheduling. The present invention solves the technical problem that traditional energy storage system management often lacks comprehensive data support and intelligent decision-making, and cannot accurately predict the supply and demand of energy, resulting in insufficient flexibility and intelligence in energy storage management.

[0007] The aforementioned technical solutions present at least the following technical issues: Most microgrid projects often employ a simple configuration approach based on maximum load demand or peak power consumption when configuring energy storage capacity. This approach ignores the synergistic effects between various renewable energy sources (such as wind power and photovoltaics) and energy storage systems, and fails to consider the efficiency of energy storage equipment. This results in excessive energy storage capacity, increased investment costs, and low energy storage utilization.

[0008] In view of the above problems, the present invention proposes a solution. Summary of the Invention

[0009] To overcome the above-mentioned shortcomings of the prior art, embodiments of the present invention provide a smart grid optimized scheduling method and system based on multi-element energy storage coordinated scheduling. By deeply integrating the dynamic characteristics of energy storage equipment with load forecasts and renewable energy output forecasts, a coordinated scheduling model for fluctuations at different time scales is constructed to address the problems of neglecting the synergistic effects between multiple renewable energy sources (such as wind power and photovoltaics) and energy storage systems, and failing to consider the utilization efficiency of energy storage equipment, resulting in excessive energy storage capacity, increased investment costs, and low energy storage utilization.

[0010] To achieve the above object, the present invention provides the following technical solutions: The smart grid optimization scheduling method based on multi-energy storage collaborative scheduling includes the following steps: constructing a prediction model based on first data to generate prediction data, wherein the first data includes historical load data of grid nodes, meteorological characteristic parameters and physical parameters of renewable energy power generation equipment; coupling the energy storage characteristic parameters of different types of energy storage equipment with the predicted data, establishing a multi-energy collaborative scheduling model, and outputting several levels of energy storage scheduling strategy sets; dynamically screening the energy storage scheduling strategy set based on preset real-time performance evaluation indicators to generate an optimal strategy subset; based on the optimal strategy subset, performing differentiated charging and discharging control instructions on the energy storage device cluster; collecting second data during the charging and discharging control process, calculating the deviation value between the second data and the predicted data, converting the deviation value into a feature vector and inputting it into a preset incremental learning algorithm to optimize the parameters of the multi-energy collaborative scheduling model.

[0011] In a preferred embodiment, the coupling of energy storage characteristic parameters of different types of energy storage devices with prediction data is specifically as follows: establishing corresponding parameter models according to the types of energy storage devices; The parameter model is integrated with the forecast data to establish differentiated scheduling constraints for different types of energy storage equipment; the forecast data includes load forecast data and renewable energy output forecast data; based on the fusion results, the operating capacity boundaries and spatiotemporal distribution of available resources of various types of energy storage equipment in the future scheduling cycle are analyzed; according to the equipment type and its current state of charge and health status parameters, the load fluctuation characteristics at different time scales are matched to determine the scope of action and scheduling priority of various types of energy storage equipment in the multi-level scheduling system.

[0012] In a preferred embodiment, the energy storage characteristic parameters of different types of energy storage devices are coupled with the prediction data to establish a multi-energy collaborative scheduling model, and output several levels of energy storage scheduling strategy sets, specifically: obtaining load forecast data and renewable energy output forecast data within the target scheduling period; establishing parameterized characteristic models of each type of energy storage device according to the type of energy storage device; performing time-synchronized discrete processing on the parameterized characteristic model and the prediction data to generate the dynamic operating boundary of each energy storage unit in each scheduling period; constructing a collaborative scheduling constraint set based on the coupling relationship between the power balance constraints of the power grid, the state transition constraints of the energy storage device and the response capabilities of multiple devices; with the goal of minimizing the comprehensive operating cost, integrating the dynamic operating boundary and the coordinated scheduling constraint set to establish a multi-energy collaborative scheduling model; calling a preset optimization algorithm to solve the scheduling optimization model, outputting the optimal charging and discharging power instructions of each energy storage device in each scheduling period, and forming a hierarchical energy storage scheduling strategy set.

[0013] In a preferred embodiment, the preset optimization algorithm is called to solve the scheduling optimization model, output the optimal charging and discharging power instructions of each energy storage device in each scheduling period, and form a hierarchical energy storage scheduling strategy set, specifically: based on the multi-energy collaborative scheduling model, an optimization objective function is constructed; a sub-target optimization method is adopted to solve the multi-energy collaborative scheduling model and generate a scheduling strategy set; in view of the uncertainty of the prediction data, several disturbance scenarios based on the historical error distribution are constructed, and a multi-scenario performance simulation evaluation is performed on each scheduling strategy; the strategy score is calculated according to the real-time performance evaluation index, and the optimal energy storage scheduling strategy at the current moment is selected and executed according to the score sorting.

[0014] In a preferred embodiment, the dynamic screening of the energy storage scheduling strategy set based on preset real-time performance evaluation indicators is specifically as follows: constructing a performance evaluation indicator system that includes safety indicators, economic indicators, and equipment health indicators: first-level evaluation: hard-constrained filtering of the scheduling strategy set based on grid constraint parameters; second-level evaluation: for the remaining strategies that pass the first-level evaluation, calculating the weighted comprehensive score of their performance evaluation indicators; selecting the top k strategies with the highest comprehensive scores, using the optimal strategy as the execution plan, and storing the remaining strategies in the backup strategy pool.

[0015] In a preferred embodiment, the dynamic screening of the energy storage scheduling strategy set based on the preset real-time performance evaluation index also includes: when a power instruction logic conflict is detected between the strategies in the backup strategy pool, executing: calling the strategy fusion algorithm to identify the conflict segment set, calculating the conflict power difference, constructing the conflict matrix, and constructing the compromise scheduling power instruction.

[0016] In a preferred embodiment, the parameters of the optimized multi-energy collaborative scheduling model also include the judgment of the optimization results, specifically: obtaining the voltage value and system frequency of each grid node in real time; calculating the system frequency deviation, and dividing the disturbance level according to the preset frequency threshold; for the key node set, calculating the voltage fluctuation range and instantaneous voltage offset: if the instantaneous voltage offset is greater than the preset voltage offset threshold, it is marked as an unstable node, and a storage priority scheduling instruction is generated; when the deviation is greater than the preset second deviation threshold and there are multiple nodes with instantaneous voltage offsets greater than the preset voltage offset threshold, it is determined that the system has a large power disturbance or a renewable energy mutation.

[0017] In a preferred embodiment, the deviation value is converted into a eigenvector, specifically as follows: based on a preset time sliding window, the actual load value and the predicted load value at each moment in the window, the actual output and the predicted output data of renewable energy are obtained, and a multidimensional disturbance observation matrix is ​​constructed; the real-time operating status parameters of the energy storage device cluster are integrated into the observation matrix to form an extended state matrix; the covariance matrix of the extended state matrix is ​​calculated, and the eigenvalue decomposition of the covariance matrix is ​​performed to obtain a set of eigenvectors arranged in descending order of eigenvalue size; a principal component retention threshold is set, the first g eigenvectors whose cumulative contribution rate meets the threshold are selected, and the key disturbance response eigenvectors are generated by splicing.

[0018] In a preferred embodiment, the conversion of the deviation value into a feature vector and inputting it into a preset incremental learning algorithm is specifically as follows: constructing a strategy parameter learning model based on an incremental extreme learning machine, using the current power system collaborative dispatch parameter set as the output layer target value of the model to complete model initialization; inputting the key disturbance response feature vector into the initialized strategy parameter learning model, and constructing a nonlinear mapping relationship between the disturbance characteristics and the dispatch parameters in combination with the feedback data; the feedback data includes the charging and discharging efficiency of the energy storage system, the grid frequency recovery time, and the voltage stability coefficient; calculating the deviation value between the model output parameter and the current operating strategy parameter; and iteratively updating the hidden layer-output layer weight matrix of the strategy parameter learning model according to a preset learning rate.

[0019] The smart grid optimization scheduling system based on multi-energy storage collaborative scheduling includes the following modules: a prediction modeling module: used to obtain first data and build a prediction model to generate prediction data; a scheduling modeling module: used to couple the energy storage characteristics of different types of energy storage equipment with the prediction data, establish a multi-energy collaborative scheduling model, and output several levels of energy storage scheduling strategy sets; a strategy screening module: used to dynamically screen the energy storage scheduling strategy set based on real-time performance evaluation indicators; an execution control module: used to perform differentiated charging and discharging control of the energy storage device cluster based on the screened energy storage scheduling strategy; an adaptive optimization module: used to optimize the collaborative scheduling parameters according to the second data in the charging and discharging control process, and the optimization is performed by converting the deviation value between the actual data and the predicted data into a feature vector and inputting it into a preset incremental learning algorithm for optimization.

[0020] The technical effects and advantages of the smart grid optimization scheduling method and system based on multi-energy storage coordinated scheduling of the present invention are as follows: 1. This invention fully considers the differences in response rate, efficiency, and capacity decay characteristics among energy storage types such as lithium-ion batteries, supercapacitors, and flow batteries, thereby constructing differentiated parameter models and scheduling constraints. By deeply integrating the dynamic characteristics of energy storage equipment with load forecasts and renewable energy output forecasts, a collaborative scheduling model for fluctuations on different time scales is constructed. This allows fast-acting energy storage to quickly respond to short-term disturbances, while energy-based energy storage takes on the task of medium- and long-term energy balance. This significantly improves the system's overall responsiveness and scheduling accuracy, avoiding the waste of resources that comes with a "one-size-fits-all" approach.

[0021] 2. This invention constructs a multi-objective collaborative optimization model to generate three types of strategies: economic dispatch, emergency response, and balance regulation. It then implements hierarchical screening based on real-time performance evaluation indicators, ensuring rapid selection of the optimal dispatch solution under different time periods and uncertain disturbance scenarios. This mechanism not only avoids the local optimality issues associated with a single objective function but also enhances the robustness of strategy switching through a pool of backup strategies and a conflict fusion mechanism, thereby achieving global optimization of dispatch economy and equipment lifespan while balancing grid security constraints. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of the process of the smart grid optimization scheduling method based on multi-energy storage coordinated scheduling of the present invention; Figure 2 This is a structural diagram of the smart grid optimization scheduling system based on multi-energy storage coordinated scheduling of the present invention. DETAILED DESCRIPTION

[0023] The following will provide a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0024] Example 1, Figure 1 The present invention provides a smart grid optimization scheduling method based on multi-energy storage coordinated scheduling, which includes the following steps: S1, building a prediction model based on the first data to generate prediction data; The first data includes historical load data of power grid nodes, meteorological characteristic parameters and physical parameters of renewable energy power generation equipment.

[0025] The forecast data includes the electricity load forecast value and the renewable energy power generation forecast interval.

[0026] Obtain historical power load data of key grid nodes in the target microgrid system, with time scales including daily, hourly, and minute levels; Simultaneously collect meteorological characteristic data in the target area, including but not limited to ambient temperature, wind speed, wind direction, solar irradiance, humidity, etc.; Obtain the physical and operational parameters of deployed renewable energy equipment such as photovoltaic modules and wind power generation equipment, such as photovoltaic panel area, conversion efficiency, wind turbine blade diameter, and rated power.

[0027] The prediction model is constructed based on the first data to generate prediction data, specifically: Obtaining the first data, and extracting the long-term fluctuation trend and short-term change characteristics based on a multi-model structure combining long-term and short-term; The long-term forecast part uses time series modeling methods to capture seasonal and trend characteristics; The short-term forecasting part introduces machine learning or deep learning models to predict short-term load fluctuations and renewable energy output changes within several time steps in the future; Output the power load forecast value sequence in the future scheduling period, including the forecast value and confidence interval; Outputs power generation forecast intervals for renewable energy sources such as photovoltaic and wind power, including minimum-maximum ranges or quantile forecast results.

[0028] S2, coupling the energy storage characteristic parameters of different types of energy storage devices with the predicted data, establishing a multi-energy coordinated scheduling model, and outputting a set of energy storage scheduling strategies at several levels; The energy storage devices include lithium-ion batteries, supercapacitors and flow batteries; The coupling of energy storage characteristic parameters of different types of energy storage devices with prediction data is specifically as follows: Establish corresponding parameter models according to the type of energy storage device, including but not limited to charge and discharge efficiency, capacity attenuation coefficient, state response time, maximum and minimum charge and discharge power limits, current state of charge, and health status; Integrate the parameter model with load forecast data and renewable energy output forecast data, and establish differentiated dispatch constraints for different energy storage types; Based on the fusion results of the parameter model and the predicted data, the operating capacity boundaries and available resource distribution of various energy storage devices in the future scheduling cycle are analyzed; Furthermore, the load fluctuation characteristics within different time scales are matched according to the device type (such as fast response type and energy type) and its current state parameters (such as SOC, SOH, response delay), and the scope of action and scheduling priority of various types of energy storage devices in the multi-level scheduling system are determined.

[0029] Among them, fast-response energy storage devices are matched to suppress short-term power disturbances with high-frequency fluctuations, and energy-type energy storage devices are used to support medium- and long-term energy balance control.

[0030] The parameter model is integrated with the load forecast data and the renewable energy output forecast data, specifically: Obtain the grid load forecast values ​​and renewable energy output forecast values ​​for multiple future time periods generated by the forecast model to form a scheduling input data set; According to each type of energy storage device, calling its corresponding parameterized mathematical model, the parameterized mathematical model includes a charge and discharge efficiency function, a health status update model, and a response rate model; Align the predicted data with the energy storage model's state parameters (such as SOC, SOH, and available capacity) by timestamp to achieve consistency in the time dimension. Based on the aligned data and combined with the models of various devices, the maximum charge and discharge power range, efficiency change trends, and health status boundaries that can be supported in each time period are calculated to extract and form differentiated device operating capacity constraints. The predicted values ​​are integrated with the equipment capability constraints to construct a structured multi-source collaborative scheduling constraint matrix.

[0031] The corresponding parameter models are respectively established, specifically: Nonlinear charge and discharge efficiency function: For lithium batteries, considering the efficiency change under the influence of SOC, the model is:

[0032] Equipment health status update model: Considering the influence of temperature, DOD and cycle number, the SOH update expression is:

[0033] Supercapacitor response rate modeling: Define the maximum response slope:

[0034] in, is the charge and discharge efficiency at different powers, is the rated efficiency, is the current power, is the current state of charge, is the power attenuation coefficient, is the SOC offset penalty coefficient, is the health status of the device at time t+1, is the health status of the device at time t, is the battery aging loss function, with inputs of depth of discharge (DOD), temperature (T) and number of cycles (N). is the maximum response rate of the supercapacitor, is the supercapacitor power change rate, is the current output of the supercapacitor.

[0035] The establishment of differentiated dispatch constraints for different energy storage types is specifically as follows: SOC update equation:

[0036] State-dependent dispatchable power range:

[0037] Frequency support equipment dispatch response trigger conditions:

[0038] in, is the state of charge of the energy storage device at time t+1, is the state of charge of the energy storage device at time t, and is the charging efficiency and discharging efficiency at different powers, is the charging power of the energy storage device at time t, is the discharge power of the energy storage device at time t, is the rated capacity of the energy storage device, is the scheduling time step, is the maximum discharge power limit of the i-th device, is the frequency response threshold, is the response power of the supercapacitor at the tth moment, is the grid frequency deviation at time t, The response power is proportional to the frequency deviation.

[0039] The multi-energy coordinated scheduling model is established as follows: Obtain load forecast data and renewable energy output forecast data within the target scheduling period; Establish parameterized characteristic models for each type of energy storage device, including charge and discharge efficiency functions, capacity decay functions, and response rate limiting conditions; The parameterized characteristic model and the predicted data are discretized in time synchronization to obtain the dynamic operating boundary of each energy storage unit in each scheduling period; Construct a collaborative scheduling constraint set based on the coupling relationship between power balance, state constraints and device response capabilities; Based on dynamic operating boundaries and constraint sets, a multi-energy collaborative scheduling model is constructed with the goal of minimizing comprehensive operating costs. The operating costs include the grid purchase cost, energy storage aging cost, and scheduling ramp-up cost. The preset optimization algorithm is called to solve the multi-energy coordinated scheduling model, and the optimal charging and discharging power instructions of each energy storage device in each scheduling period are output to form a hierarchical energy storage scheduling strategy set.

[0040] The multi-energy coordinated scheduling model is specifically as follows:

[0041] in, is the total scheduling period, is the cost of purchasing electricity from the main grid at time t, is the aging cost weight coefficient of the i-th equipment, is the aging loss cost of the i-th energy storage device at time t, is the ramp cost weight coefficient of the jth device, is the ramp-up cost of the j-th energy storage device at time t. The equipment aging cost weight coefficient and the ramp-up cost weight coefficient are obtained by the hierarchical analysis method.

[0042] The output of the energy storage scheduling strategy set at several levels is specifically: Based on the multi-energy coordinated dispatch model, optimization objective functions are constructed, corresponding to the economic optimization objectives, safety and stability objectives, and balance objectives respectively; Using a sub-objective optimization method to generate a dispatching strategy; the dispatching strategy includes an economic dispatching strategy, an emergency response strategy, and a balance adjustment strategy; In view of the uncertainty of the predicted data, several disturbance scenarios are constructed, and the effectiveness of the scheduling strategy is simulated and evaluated under various scenarios; Based on the set real-time performance evaluation indicators, multiple scheduling strategies are scored and ranked, and the optimal energy storage scheduling strategy at the current moment is selected for actual execution.

[0043] Emergency response strategy: When a node voltage deviation exceeds 7% or a grid frequency fluctuation exceeds 0.5Hz, the system will use a preset safe charging and discharging template to ensure system stability. Economic dispatch strategy: Based on the trough and peak periods of electricity prices, a charging and discharging plan is formulated that prioritizes the use of low-cost electricity. Equipment aging loss models are also introduced for cost calculation. Balancing regulation strategy: Based on minute-level photovoltaic or wind power output fluctuation detection (amplitude exceeding ±10%), a smooth scheduling algorithm is triggered to suppress the impact of drastic power changes on the system.

[0044] The real-time performance evaluation indicators are specifically: S3, dynamically screening the energy storage scheduling strategy set based on preset real-time performance evaluation indicators to generate the optimal strategy subset; The energy storage scheduling strategy set is dynamically screened based on the preset real-time performance evaluation indicators to generate the optimal strategy subset, specifically: Constructing performance indicators that include safety indicators, economic indicators, and equipment health indicators; the safety indicators include voltage stability coefficient; the economic indicators include peak-valley arbitrage net income and equipment loss cost; the equipment health indicators include the standard deviation of energy storage unit aging distribution; The first level of evaluation eliminates strategies that violate safety constraints during the simulated scheduling process based on the multi-energy coordinated scheduling model and grid constraint parameters (voltage upper and lower limits, frequency deviation thresholds, and power balance conditions). The second level of evaluation calculates a weighted comprehensive score based on the effectiveness evaluation indicators of the remaining strategies; The top k strategies with the highest comprehensive scores are selected as the execution plan for the current cycle, and the suboptimal strategies are stored in the backup strategy pool for subsequent switching; When there is a logical conflict between backup strategies (e.g., simultaneous charging and discharging in the same period), the strategy fusion algorithm is called to calculate the conflicting power difference and construct a compromise scheduling power. Actual operating data is continuously collected during the strategy execution cycle, and the safety, economy, and energy storage utilization scores are updated in real time. When the current strategy effectiveness deviates from the predicted score by more than the preset threshold, the strategy switching mechanism is triggered, and the suboptimal solution in the backup strategy pool is replaced first.

[0045] The voltage stability coefficient is specifically:

[0046] The conflicting power difference is specifically:

[0047] The compromise scheduling power is:

[0048] The weighted comprehensive score is specifically:

[0049]

[0050]

[0051]

[0052] in, is the voltage stability coefficient, is the frequency recovery time, is the voltage deviation, For normal voltage, is the conflict power difference, is the expected power of the i-th scheduling object, is the average power that can be allocated to the current system, is the total number of scheduled objects, To compromise the dispatch power, is the priority weight of the i-th scheduling object, is the total power that the system can provide, is the preset adjustment coefficient, is the weighted comprehensive score, Score safety. For economic rating, Score your health. 、 、 The weight coefficients corresponding to the three indicators are set based on historical experience. Net income = peak-valley arbitrage income - equipment loss cost, For the maximum net profit, is the minimum net profit, is the standard deviation of the energy storage unit aging distribution, the smaller it is, the healthier it is. is the maximum standard deviation, is the minimum standard deviation.

[0053] S4, based on the optimal strategy subset, provides differentiated charge and discharge control instructions to the energy storage device cluster; The differentiated charge and discharge control instructions for the energy storage device cluster based on the optimal strategy subset are specifically: For lithium battery devices, a nonlinear power control function is called based on the current state of charge, health status, and the target power value set in the strategy to calculate the actual execution power, ensuring that overcharging and over-discharging are avoided and life loss is suppressed; For supercapacitor devices, based on the frequency response control logic, the fast response mode is triggered when the grid frequency deviation exceeds the threshold, and its maximum output power is set according to the current SOC and the upper limit of the response rate; A scheduling execution window is set for each type of energy storage device, and the power instructions are dynamically updated within each window period based on the time priority label in the scheduling strategy. During the entire control cycle, the operating status data of each device in the energy storage cluster is continuously monitored. When parameter abnormalities (such as temperature exceeding the limit or voltage exceeding the limit) are detected, the charging and discharging power of the device is automatically adjusted or participation in the control of this cycle is suspended.

[0054] S5, collects the second data during the charge and discharge control process, calculates the deviation value between the second data and the predicted data, converts the deviation value into a feature vector and inputs it into a preset incremental learning algorithm to optimize the parameters of the multi-energy collaborative scheduling model.

[0055] The second data includes grid frequency deviation and node voltage fluctuation; The optimization of the parameters of the multi-energy coordinated scheduling model also includes the judgment of the optimization results, specifically: Obtain the voltage value and system frequency of each grid node in real time; Calculate the deviation between the system frequency and the reference frequency, and determine the frequency disturbance level; For each key node, the voltage fluctuation range is counted and the instantaneous voltage offset is calculated: If the instantaneous voltage deviation is greater than the preset voltage deviation, it is marked as an unstable node and priority is given to energy storage support. When it is detected at the same time that the deviation is greater than the preset second deviation threshold and there are multiple nodes whose instantaneous voltage deviations are greater than the preset voltage deviation, it is determined that there is a large power disturbance or a sudden change in renewable energy in the system.

[0056] The frequency disturbance level is determined as follows: If the deviation is less than the preset first deviation threshold, it is considered a normal fluctuation; If the deviation is greater than a preset first deviation threshold and less than a preset second deviation threshold, a light adjustment strategy is triggered; If the deviation is greater than a preset second deviation threshold, the emergency strategy is activated and the output power of the energy storage rapid response device is adjusted.

[0057] The conversion of the deviation value into a feature vector is specifically as follows: Based on the preset time sliding window, the actual load value and predicted load value at the corresponding moment, the actual output and predicted output of renewable energy are obtained, and a multi-dimensional disturbance observation matrix is ​​constructed; Based on the observation matrix, the real-time operating status parameters of the current energy storage device cluster are integrated to form an extended state matrix; the real-time operating status parameters include average state of charge, health state, frequency deviation and node voltage variance; Calculate the covariance matrix of the extended state matrix and perform eigenvalue decomposition to obtain a set of eigenvectors and corresponding eigenvalues; The principal component retention threshold is set, and the first g eigenvalues ​​that meet the cumulative contribution rate are concatenated to obtain the key disturbance response eigenvector.

[0058] The input preset incremental learning algorithm is used to optimize the parameters of the multi-energy collaborative scheduling model, specifically: Initialize the policy parameter learning model based on the incremental extreme learning machine and use the current collaborative scheduling parameter set as the output target; Inputting key disturbance response feature vectors into a strategy parameter learning model and building a nonlinear mapping relationship between disturbances and scheduling parameters based on feedback data, including the energy storage system's charge and discharge efficiency, grid frequency recovery time, and voltage stability coefficient. Calculate the deviation between the model output and the current running policy parameters and update the parameters at a preset learning rate.

[0059] Example 2, Figure 2 The present invention provides a smart grid optimization scheduling system based on multi-energy storage coordinated scheduling, which is characterized by including the following modules: Prediction modeling module: used to build a prediction model based on first data to generate prediction data, wherein the first data includes historical load data of power grid nodes, meteorological characteristic parameters and physical parameters of renewable energy power generation equipment; Scheduling modeling module: This module is used to couple the energy storage characteristic parameters of different types of energy storage devices with forecast data, establish a multi-energy collaborative scheduling model, and output a set of energy storage scheduling strategies at several levels; Strategy screening module: used to dynamically screen the energy storage scheduling strategy set based on preset real-time performance evaluation indicators and generate the optimal strategy subset; Execution control module: used to provide differentiated charge and discharge control instructions for the energy storage device cluster based on the optimal strategy subset; Adaptive optimization module: used to collect the second data during the charging and discharging control process, calculate the deviation value between the second data and the predicted data, convert the deviation value into a feature vector and input it into the preset incremental learning algorithm to optimize the parameters of the multi-energy collaborative scheduling model.

[0060] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0061] The above embodiments may be implemented in whole or in part through software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0062] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0063] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0064] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

[0065] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A smart grid optimization scheduling method based on multi-energy storage coordinated scheduling, characterized by: The following steps are involved: Building a prediction model based on first data to generate prediction data, wherein the first data includes historical load data of power grid nodes, meteorological characteristic parameters, and physical parameters of renewable energy power generation equipment; The energy storage characteristic parameters of different types of energy storage devices are coupled with the predicted data to establish a multi-energy coordinated scheduling model and output a set of energy storage scheduling strategies at several levels; Dynamically screen the energy storage scheduling strategy set based on preset real-time performance evaluation indicators to generate the optimal strategy subset; Based on the optimal strategy subset, differentiated charging and discharging control instructions are issued to the energy storage device cluster; Collect the second data during the charge and discharge control process, calculate the deviation between the second data and the predicted data, convert the deviation into a feature vector and input it into the preset incremental learning algorithm to optimize the parameters of the multi-energy collaborative scheduling model.

2. The smart grid optimization scheduling method based on multi-energy storage coordinated scheduling according to claim 1 is characterized in that: The coupling of energy storage characteristic parameters of different types of energy storage devices with prediction data is specifically as follows: Establish corresponding parameter models according to the type of energy storage equipment; The parameter model is integrated with the forecast data to establish differentiated scheduling constraints for different types of energy storage equipment; the forecast data includes load forecast data and renewable energy output forecast data; Based on the fusion results, the operating capacity boundaries and spatiotemporal distribution of available resources of various energy storage devices in the future scheduling cycle are analyzed; According to the equipment type and its current state of charge and health status parameters, the load fluctuation characteristics of different time scales are matched to determine the scope of action and scheduling priority of various types of energy storage equipment in the multi-level scheduling system.

3. The smart grid optimization scheduling method based on multi-energy storage coordinated scheduling according to claim 2 is characterized in that: The energy storage characteristic parameters of different types of energy storage devices are coupled with the prediction data to establish a multi-energy coordinated scheduling model, and output a set of energy storage scheduling strategies at several levels, specifically: Obtain load forecast data and renewable energy output forecast data within the target scheduling period; Establish parameterized characteristic models for each type of energy storage device according to the type of energy storage device; The parameterized characteristic model and the predicted data are discretized in time synchronization to generate the dynamic operating boundaries of each energy storage unit in each scheduling period; Based on the coupling relationship between grid power balance constraints, energy storage device state transition constraints, and multi-device response capabilities, a collaborative scheduling constraint set is constructed; With the goal of minimizing comprehensive operating costs, a multi-energy coordinated scheduling model is established by integrating dynamic operating boundaries and coordinated scheduling constraints. The preset optimization algorithm is called to solve the scheduling optimization model, and the optimal charging and discharging power instructions of each energy storage device in each scheduling period are output to form a hierarchical energy storage scheduling strategy set.

4. The smart grid optimization scheduling method based on multi-energy storage coordinated scheduling according to claim 3 is characterized in that: The preset optimization algorithm is called to solve the scheduling optimization model, and the optimal charge and discharge power instructions of each energy storage device in each scheduling period are output to form a hierarchical energy storage scheduling strategy set, specifically: Based on the multi-energy coordinated scheduling model, the optimization objective function is constructed; Adopting the sub-objective optimization method, we solve the multi-energy coordinated scheduling model and generate a set of scheduling strategies; To address the uncertainty of forecast data, several disturbance scenarios based on historical error distribution are constructed, and multi-scenario performance simulation evaluation is performed on each scheduling strategy. The strategy score is calculated based on the real-time performance evaluation indicators, and the optimal energy storage scheduling strategy at the current moment is selected and executed based on the score sorting.

5. The smart grid optimization scheduling method based on multi-energy storage coordinated scheduling according to claim 4 is characterized in that: The dynamic screening of the energy storage scheduling strategy set based on the preset real-time performance evaluation indicators is specifically as follows: Construct an efficiency evaluation index system that includes safety indicators, economic indicators, and equipment health indicators: First-level evaluation: hard constraint filtering of the dispatch strategy set based on grid constraint parameters; Second-level evaluation: For the remaining strategies that have passed the first-level evaluation, calculate the weighted comprehensive scores of their effectiveness evaluation indicators; The top k strategies with the highest comprehensive scores are selected, the optimal strategy is used as the execution plan, and the remaining strategies are stored in the backup strategy pool.

6. The smart grid optimization scheduling method based on multi-element energy storage coordinated scheduling according to claim 5 is characterized in that: The dynamic screening of the energy storage scheduling strategy set based on the preset real-time performance evaluation index also includes: When a power instruction logic conflict is detected between the strategies in the backup strategy pool, the following steps are executed: the strategy fusion algorithm is called to identify the conflict segment set, the conflict power difference is calculated, the conflict matrix is ​​constructed, and a compromise scheduling power instruction is constructed.

7. The smart grid optimization scheduling method based on multi-energy storage coordinated scheduling according to claim 6 is characterized in that: The optimization of the parameters of the multi-energy coordinated scheduling model also includes the judgment of the optimization results, specifically: Obtain the voltage value and system frequency of each grid node in real time; Calculate the system frequency deviation and classify the disturbance level according to the preset frequency threshold; For the set of key nodes, calculate the voltage fluctuation range and instantaneous voltage offset: If the instantaneous voltage deviation is greater than the preset voltage deviation threshold, it is marked as an unstable node and an energy storage priority dispatch instruction is generated; When the deviation is greater than a preset second deviation threshold and there are multiple nodes whose instantaneous voltage deviations are greater than a preset voltage deviation threshold, it is determined that there is a large power disturbance or a sudden change in renewable energy in the system.

8. The smart grid optimization scheduling method based on multi-element energy storage coordinated scheduling according to claim 7 is characterized in that: The conversion of the deviation value into a feature vector is specifically as follows: Based on a preset time sliding window, the actual load value and predicted load value at each moment in the window, the actual output and predicted output data of renewable energy are obtained, and a multi-dimensional disturbance observation matrix is ​​constructed; Integrating the real-time operating status parameters of the energy storage device cluster into the observation matrix to form an extended state matrix; Calculate the covariance matrix of the expanded state matrix and perform eigenvalue decomposition on the covariance matrix to obtain a set of eigenvectors arranged in descending order of eigenvalue size; The principal component retention threshold is set, the first g eigenvectors whose cumulative contribution rate meets the threshold are selected, and the key disturbance response eigenvectors are generated by splicing.

9. The smart grid optimization scheduling method based on multi-energy storage coordinated scheduling according to claim 7 is characterized in that: The deviation value is converted into a feature vector and input into a preset incremental learning algorithm, specifically: A strategy parameter learning model is constructed based on an incremental extreme learning machine, with the current power system coordinated dispatch parameter set as the output layer target value of the model to complete model initialization. Inputting the key disturbance response feature vector into the initialized strategy parameter learning model, and combining it with feedback data to construct a nonlinear mapping relationship between the disturbance feature and the scheduling parameters; the feedback data includes the energy storage system's charging and discharging efficiency, grid frequency recovery time, and voltage stability coefficient; Calculate the deviation between the model output parameters and the current operation strategy parameters; The hidden layer-output layer weight matrix of the strategy parameter learning model is iteratively updated according to the preset learning rate.

10. A system using the smart grid optimization scheduling method based on multi-energy storage coordinated scheduling according to any one of claims 1 to 9, characterized in that: Includes the following modules: Prediction modeling module: used to obtain first data and build a prediction model to generate prediction data; Scheduling modeling module: This module is used to couple the energy storage characteristics of different types of energy storage devices with forecast data, establish a multi-energy collaborative scheduling model, and output a set of energy storage scheduling strategies at several levels. Strategy screening module: used to dynamically screen energy storage scheduling strategy sets based on real-time performance evaluation indicators; Execution control module: used to perform differentiated charging and discharging control of energy storage device clusters based on the selected energy storage scheduling strategy; Adaptive optimization module: used to optimize the collaborative scheduling parameters according to the second data in the charge and discharge control process. The optimization is performed by converting the deviation value between the actual data and the predicted data into a feature vector and inputting it into a preset incremental learning algorithm for optimization.

Images