Energy storage combined power supply system with priority management and scheduling method thereof

Abstract

The invention discloses an energy storage combined power supply system with priority management and a scheduling method thereof, and belongs to the technical field of power systems, and the system comprises a data collection module which is used for collecting energy storage combined power supply real-time data; the data processing module is used for processing the energy storage combined power supply real-time data and determining energy storage combined power supply characteristic data; the analysis and prediction module is used for analyzing the energy storage combined power supply characteristic data and predicting the load demand of energy storage combined power supply; and the scheduling management module is used for performing scheduling management on combined power supply of the energy storage equipment and meeting load requirements according to a priority sequence. According to the method, the problems that energy storage combined power supply cannot be effectively scheduled and managed according to a priority sequence and complicated and changeable power utilization requirements are difficult to adapt in the prior art are solved. According to the invention, effective scheduling management can be carried out on energy storage combined power supply according to a priority sequence, intelligent scheduling can be realized to optimize energy distribution, and complex and changeable power utilization requirements can be met.

Description

Technical Field

[0001] This invention relates to the field of power system technology, specifically to an energy storage combined power supply system with priority management and its scheduling method. Background Technology

[0002] With the growth of global energy demand and the rapid development of renewable energy, the power system faces enormous challenges in balancing supply and demand and maintaining stability. Among these challenges, energy storage technology serves as a key means to balance grid load fluctuations, smooth renewable energy output, and improve power supply reliability and economy.

[0003] Existing technologies cannot effectively schedule and manage energy storage combined with power supply according to priority order, resulting in a lack of dynamic management of load priorities for energy storage devices. This makes it impossible to achieve intelligent scheduling to optimize energy distribution and adapt to complex and ever-changing electricity demands. Summary of the Invention

[0004] The purpose of this invention is to provide an energy storage combined power supply system with priority management and its scheduling method, which can effectively schedule and manage the energy storage combined power supply according to the priority order, so that the energy storage device does not lack dynamic management of load priority, can realize intelligent scheduling to optimize energy distribution, and can adapt to complex and ever-changing electricity demand, thus solving the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: An energy storage combined power supply system with priority management includes: The data acquisition module is used to monitor load and power supply data in real time and collect real-time data of energy storage combined power supply in the power system. The data processing module is used to process real-time data of energy storage combined power supply, extract key features, and determine characteristic data of energy storage combined power supply. The analysis and prediction module is used to build a load prediction model for energy storage combined power supply, analyze the characteristic data of energy storage combined power supply, and predict the load demand of energy storage combined power supply. The scheduling and management module is used to schedule and manage the joint power supply of energy storage devices based on load demand forecasts, and to meet load demands in a priority order.

[0006] Preferably, the characteristic data of energy storage combined power supply are analyzed, and the load demand of energy storage combined power supply is predicted, by performing the following operations: Deploy the energy storage combined power supply load forecasting model in the energy storage combined power supply load forecasting environment; The characteristic data of energy storage combined power supply is input into the load prediction model of energy storage combined power supply. The characteristic data of energy storage combined power supply is analyzed according to the load prediction model of energy storage combined power supply, and the load demand of energy storage combined power supply is predicted to determine the load prediction result of energy storage combined power supply. The load forecast results of energy storage combined power supply are compared and analyzed with those of distributed power sources in the power system to determine whether distributed power sources in the power system can meet the load demand of energy storage combined power supply. When distributed power energy storage in the power system meets the load demand for combined energy storage power supply, energy storage devices capable of storing electrical energy are dispatched and used to store excess distributed power energy. When distributed power sources in the power system cannot meet the load demand for combined energy storage power supply, energy storage devices that can release electrical energy are dispatched, and the electrical energy released by the energy storage devices is used to meet the load demand in order of priority.

[0007] Preferably, energy storage devices are used to release electrical energy and meet load demands according to priority, and the following operations are performed: The multiple loads on the load side are classified into critical loads, normal loads and interruptible loads, and different power supply priorities are assigned to critical loads, normal loads and interruptible loads. Among them, the power supply priority of critical loads is higher than that of ordinary loads, and the power supply priority of ordinary loads is higher than that of interruptible loads. When distributed power sources in the power system cannot meet the load demand for combined power supply from energy storage, the electrical energy released by energy storage devices should be used to prioritize meeting the load demand of critical loads. When there is still residual energy after the energy storage device releases electricity to meet the load demand of critical loads, the residual energy is used to meet the load demand of ordinary loads first. When there is still residual energy after the energy storage device releases electricity to meet the load requirements of critical and ordinary loads, the residual energy is used to meet the load requirements of interruptible loads.

[0008] Preferably, the step of classifying the multiple loads at the load end into critical loads, normal loads, and interruptible loads includes: Based on relevant information about the load, determine the life safety risks, public order impact, amount of loss, number of system crashes, and interruption tolerance of the load after a power outage, and determine the scenario type of the load. The risk to life safety, impact on public order, amount of loss, number of system crashes, interruption tolerance, and scenario type are normalized, and the criticality value of the load end is calculated based on the normalization result. Based on the criticality value, the initial load type of the load end is determined; Real-time monitoring of the power supply status of the load end under the initial load type, including user complaint rate, user complaint interval time, and user appeal rate, to calculate the feedback value of the load end under the initial load type; When the feedback value is greater than a preset threshold, the load type of the load end is increased; otherwise, the initial load type of the load end remains unchanged.

[0009] Preferably, real-time data on energy storage combined with power supply in the power system is collected, and the following operations are performed: Real-time monitoring of voltage, current, power, energy consumption, equipment operating status, and environmental parameters at the load end of the power system, and collection of real-time data at the load end; Real-time monitoring of voltage, frequency, power output, energy type, and energy storage device status at the power supply end of the power system, and collection of real-time data from the power supply end; Real-time data on combined energy storage and power supply in the power system is generated based on real-time data from the load side and the power supply side.

[0010] Preferably, the real-time data of the combined energy storage and power supply is processed by performing the following operations: The real-time data of energy storage combined power supply is cleaned to remove noise and outliers that are not useful for the priority management of energy storage combined power supply, and outliers that are useful for the priority management of energy storage combined power supply are replaced. Standardize the real-time data of energy storage combined power supply, unify the format of the real-time data of energy storage combined power supply, eliminate the differences in the units of measurement in the real-time data of energy storage combined power supply, and form standardized real-time data of energy storage combined power supply.

[0011] Preferably, the real-time data of the combined energy storage and power supply is processed by performing the following operations: The real-time data of energy storage combined with power supply is organized and put into a data view. The integrity of the real-time data of energy storage combined with power supply in the data view is evaluated. After the data evaluation, the real-time data of energy storage combined with power supply in the data view is stored and backed up. Feature extraction is performed on real-time data of energy storage combined power supply. Features related to the priority management of energy storage combined power supply are extracted from the real-time data of energy storage combined power supply to determine the feature data of energy storage combined power supply.

[0012] Preferably, an energy storage-integrated power supply load forecasting model is constructed, and the following operations are performed: Historical data on the combined power supply of energy storage in the power system are collected, and the collected historical data on the combined power supply of energy storage in the power system are divided into training set and test set; The machine learning model is trained using a training set, enabling it to autonomously learn the load prediction behavior of energy storage combined power supply from the training set and predict the load demand of energy storage combined power supply, thus determining the load prediction model of energy storage combined power supply. The test set was used to test the load prediction model of energy storage combined power supply, evaluate the generalization ability of the energy storage combined power supply load prediction model, and determine whether the energy storage combined power supply load prediction model can achieve the expected effect of predicting the load demand of energy storage combined power supply. If the constructed energy storage combined power supply load prediction model does not meet the expected requirements, the parameters of the energy storage combined power supply load prediction model will be adjusted and optimized to make the constructed energy storage combined power supply load prediction model meet the expected requirements, thereby determining the energy storage combined power supply load prediction model that meets the expected requirements.

[0013] Preferably, when the constructed energy storage combined power supply load prediction model does not meet the expected requirements, the parameters of the energy storage combined power supply load prediction model are adjusted and optimized, including: The time period error, scenario error, time scale error, and feature contribution are obtained from the test results of the energy storage combined power supply load prediction model on the test set. The target time period with a value greater than a preset time period error threshold is obtained from the time period error; the target scene with a value greater than a preset scene error threshold is obtained from the scene error; the target time scale with a value greater than a preset time scale error threshold is obtained from the time scale error; and the target feature with a value less than a preset feature contribution threshold is obtained from the feature contribution. Based on the target time period and target scenario, determine the additional time period data and additional scenario data; based on the target features, determine the additional feature type; based on the target time scale, determine the data time series length. The training set is augmented by adding time period data, scene data, feature types, and data time series length to obtain the target training set. Based on the difference between the feature contribution of the target feature and the preset feature contribution threshold, the weighting weight of the target feature is determined, and an attention layer is added to the power supply load prediction model based on the weighting weight. Based on the test results of the energy storage combined power supply load prediction model on the test set and the initial hyperparameter combination of the energy storage combined power supply load prediction model, multiple hyperparameter combinations are adaptively established. Based on the target training set and attention layer, the energy storage combined power supply load prediction model is optimized, and multiple hyperparameter combinations are iterated based on the Bayesian optimization algorithm to obtain the optimal hyperparameter combination, thus achieving the optimization of the energy storage combined power supply load prediction model.

[0014] According to another aspect of the present invention, a method for joint power supply scheduling of energy storage with priority management is provided, implemented based on the above-described joint power supply system of energy storage with priority management, comprising: Collect and process real-time data on energy storage combined power supply in the power system, determine the characteristic data of energy storage combined power supply, analyze the characteristic data of energy storage combined power supply, and predict the load demand of energy storage combined power supply. When distributed power energy storage meets the load demand of energy storage combined power supply, energy storage devices that can store electrical energy are dispatched and used to store excess distributed power energy. When distributed power sources cannot meet the load demand for combined energy storage power supply, energy storage devices that can release electrical energy are dispatched, and the electrical energy released by the energy storage devices is used to meet the load demand in order of priority.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention collects real-time data on energy storage combined power supply in the power system by monitoring data at the load end and the power supply end in real time. By processing the real-time data on energy storage combined power supply and extracting key features, the characteristic data of energy storage combined power supply is determined. By constructing a load prediction model for energy storage combined power supply and analyzing the characteristic data of energy storage combined power supply, the load demand of energy storage combined power supply is predicted, the load prediction result of energy storage combined power supply is determined, and the combined power supply of energy storage devices is scheduled and managed according to the load demand prediction, and the load demand is met in a priority order.

[0016] 2. When the distributed power energy storage of this invention meets the load demand of the energy storage combined power supply, it dispatches energy storage devices that can store electrical energy and uses these devices to store excess distributed power energy. When the distributed power energy storage cannot meet the load demand of the energy storage combined power supply, it dispatches energy storage devices that can release electrical energy and uses the released electrical energy to meet the load demand according to priority. It can effectively dispatch and manage the energy storage combined power supply according to priority, so that the energy storage devices do not lack dynamic management of load priority. It can realize intelligent dispatch to optimize energy distribution and adapt to complex and ever-changing electricity demand. Attached Figure Description

[0017] Figure 1 This is a block diagram of the energy storage combined power supply system with priority management according to the present invention; Figure 2 This is a flowchart of the energy storage combined power supply system with priority management according to the present invention; Figure 3 This is a flowchart of the energy storage joint power supply scheduling method with priority management according to the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] To address the current limitations of effective scheduling and management of energy storage-based power supply based on priority, which results in energy storage devices lacking dynamic management of load priorities and failing to achieve intelligent scheduling to optimize energy distribution, thus hindering their ability to adapt to complex and ever-changing electricity demands, please refer to [link to relevant documentation]. Figures 1-3 This embodiment provides the following technical solution: An energy storage combined power supply system with priority management includes: a data acquisition module, a data processing module, an analysis and prediction module, and a scheduling management module.

[0020] Specifically, through the interaction between the data acquisition module, data processing module, analysis and prediction module, and scheduling management module, the energy storage combined power supply can be effectively scheduled and managed according to the priority order. This ensures that the energy storage equipment does not lack dynamic management of load priorities, and enables intelligent scheduling to optimize energy distribution, thus adapting to complex and ever-changing electricity demands.

[0021] The data acquisition module is used to monitor data at the load end and the power supply end, and to collect real-time data on energy storage combined with power supply in the power system.

[0022] In this embodiment, real-time data of energy storage combined with power supply in the power system is collected, and the following operations are performed: Real-time monitoring of voltage, current, power, energy consumption, equipment operating status, and environmental parameters at the load end of the power system, and collection of real-time data at the load end; Real-time monitoring of voltage, frequency, power output, energy type, and energy storage device status at the power supply end of the power system, and collection of real-time data from the power supply end; Real-time data on combined energy storage and power supply in the power system is generated based on real-time data from the load side and the power supply side.

[0023] It should be noted that by monitoring data from both the load and power supply sides in real time, real-time data on energy storage combined with power supply in the power system is collected, providing a data foundation for subsequent energy storage combined with power supply.

[0024] The data processing module is used to process real-time data of energy storage combined power supply, extract key features, and determine characteristic data of energy storage combined power supply.

[0025] In this embodiment, the real-time data of the combined energy storage and power supply is processed by performing the following operations: Cleaning the real-time data of energy storage combined power supply removes noise and outliers that are not useful for the priority management of energy storage combined power supply, and replaces outliers that are useful for the priority management of energy storage combined power supply, which can improve the data quality of the real-time data of energy storage combined power supply. Standardize the real-time data of energy storage combined power supply to unify the format of the data, eliminate the differences in dimensions, and form standardized real-time data of energy storage combined power supply, which will facilitate better analysis of the data.

[0026] In this embodiment, the real-time data of the combined energy storage and power supply is processed by performing the following operations: The real-time data of energy storage combined with power supply is organized and put into a data view. The integrity of the real-time data of energy storage combined with power supply in the data view is evaluated. After the data evaluation, the real-time data of energy storage combined with power supply in the data view is stored and backed up. Feature extraction is performed on real-time data of energy storage combined power supply. Features related to the priority management of energy storage combined power supply are extracted from the real-time data of energy storage combined power supply to determine the feature data of energy storage combined power supply.

[0027] It should be noted that by extracting features from real-time data of energy storage combined power supply and extracting key features, the characteristic data of energy storage combined power supply can be determined, which facilitates better prediction of load demand for energy storage combined power supply in the future.

[0028] The analysis and prediction module is used to construct a load prediction model for energy storage combined power supply, analyze the characteristic data of energy storage combined power supply, and predict the load demand of energy storage combined power supply.

[0029] In this embodiment, an energy storage-integrated power supply load prediction model is constructed, and the following operations are performed: Historical data on the combined power supply of energy storage in the power system are collected, and the collected historical data on the combined power supply of energy storage in the power system are divided into training set and test set; The machine learning model is trained using a training set, enabling it to autonomously learn the load prediction behavior of energy storage combined power supply from the training set and predict the load demand of energy storage combined power supply, thus determining the load prediction model of energy storage combined power supply. The test set was used to test the load prediction model of energy storage combined power supply, evaluate the generalization ability of the energy storage combined power supply load prediction model, and determine whether the energy storage combined power supply load prediction model can achieve the expected effect of predicting the load demand of energy storage combined power supply. If the constructed energy storage combined power supply load prediction model does not meet the expected requirements, the parameters of the energy storage combined power supply load prediction model will be adjusted and optimized to make the constructed energy storage combined power supply load prediction model meet the expected requirements, thereby determining the energy storage combined power supply load prediction model that meets the expected requirements.

[0030] In one embodiment, when the constructed energy storage combined power supply load forecasting model does not meet the expected requirements, the parameters of the energy storage combined power supply load forecasting model are adjusted and optimized, including: The time period error, scenario error, time scale error, and feature contribution are obtained from the test results of the energy storage combined power supply load prediction model on the test set. The target time period with a value greater than a preset time period error threshold is obtained from the time period error; the target scene with a value greater than a preset scene error threshold is obtained from the scene error; the target time scale with a value greater than a preset time scale error threshold is obtained from the time scale error; and the target feature with a value less than a preset feature contribution threshold is obtained from the feature contribution. Based on the target time period and target scenario, determine the additional time period data and additional scenario data; based on the target features, determine the additional feature type; based on the target time scale, determine the data time series length. The training set is augmented by adding time period data, scene data, feature types, and data time series length to obtain the target training set. Based on the difference between the feature contribution of the target feature and the preset feature contribution threshold, the weighting weight of the target feature is determined, and an attention layer is added to the power supply load prediction model based on the weighting weight. Based on the test results of the energy storage combined power supply load prediction model on the test set and the initial hyperparameter combination of the energy storage combined power supply load prediction model, multiple hyperparameter combinations are adaptively established. Based on the target training set and attention layer, the energy storage combined power supply load prediction model is optimized, and multiple hyperparameter combinations are iterated based on the Bayesian optimization algorithm to obtain the optimal hyperparameter combination, thus achieving the optimization of the energy storage combined power supply load prediction model.

[0031] In this embodiment, the iteration of multiple hyperparameter combinations based on the Bayesian optimization algorithm typically uses Gaussian processes or random forests to model the performance of known hyperparameter combinations. Based on the alternative model, the acquisition value of each hyperparameter combination is calculated to guide the next sampling, continuously updating the alternative model and gradually approaching the global optimum.

[0032] In this embodiment, the time period error is, for example, a sudden increase in error at 18:00 every day, which may not capture the characteristics of the evening peak. The corresponding increase in time period data is the addition of data at 18:00.

[0033] In this embodiment, the scene error is, for example, that the prediction error during the Spring Festival is three times that of usual, and the corresponding additional scene data is the scene data during the Spring Festival.

[0034] In this embodiment, the feature contribution is, for example, the temperature feature is found to be only 5% important, and the corresponding feature type is added as temperature feature.

[0035] In this embodiment, the time scale error is, for example, a short-term prediction error of <5% but a long-term prediction error of >20%, and the corresponding data time series length is determined based on the long-term prediction.

[0036] In this embodiment, the attention layer added to the power supply load prediction model based on the weighted weights specifically adds corresponding weighted weights to each input feature.

[0037] The beneficial effects of the above design scheme are as follows: By extracting time period error, scene error, time scale error, and feature contribution, it directly identifies the model's deficiencies in specific dimensions. The quantized error-based localization method avoids the inefficient traditional approach of blindly adjusting parameters, making the optimization direction clearer and improving optimization efficiency. Furthermore, by adding time period data and scene data for the identified target time period, scene, and time scale, data augmentation can directly strengthen the model's training in weak areas. Adding an attention layer allows the model to actively focus on previously overlooked key features, solving the problem of insufficient feature learning and enhancing its overall strength. The model's feature mining capabilities were enhanced, reducing prediction bias caused by insufficient feature information. Based on test results and optimized training data, a Bayesian optimization algorithm was used to iteratively search for the optimal hyperparameter combination. This improved prediction accuracy while avoiding overfitting. The precise optimization of hyperparameters ensured that the model could fully learn data patterns after adjustment and maintain its generalization ability to new data. Ultimately, this improved the accuracy, robustness, adaptability, and optimization efficiency of the energy storage combined power supply load prediction model, providing a reliable decision-making basis for the combined power supply scheduling and load priority management of energy storage systems, and ensuring the stability and economy of the power supply system.

[0038] In this embodiment, the characteristic data of energy storage combined power supply are analyzed, and the load demand of energy storage combined power supply is predicted. The following operations are performed: Deploy the energy storage combined power supply load forecasting model in the energy storage combined power supply load forecasting environment; The characteristic data of energy storage combined power supply is input into the load prediction model of energy storage combined power supply. The characteristic data of energy storage combined power supply is analyzed according to the load prediction model of energy storage combined power supply, and the load demand of energy storage combined power supply is predicted to determine the load prediction result of energy storage combined power supply. The load forecast results of energy storage combined power supply are compared and analyzed with those of distributed power sources in the power system to determine whether distributed power sources in the power system can meet the load demand of energy storage combined power supply. When distributed power energy storage in the power system meets the load demand for combined energy storage power supply, energy storage devices capable of storing electrical energy are dispatched and used to store excess distributed power energy. When distributed power sources in the power system cannot meet the load demand for combined energy storage power supply, energy storage devices that can release electrical energy are dispatched, and the electrical energy released by the energy storage devices is used to meet the load demand in order of priority.

[0039] The scheduling and management module is used to schedule and manage the joint power supply of energy storage devices based on load demand forecasts, and to meet load demands in a priority order.

[0040] In this embodiment, energy storage devices are used to release electrical energy and meet load demands according to priority order, performing the following operations: The multiple loads on the load side are classified into critical loads, normal loads and interruptible loads, and different power supply priorities are assigned to critical loads, normal loads and interruptible loads. Among them, the power supply priority of critical loads is higher than that of ordinary loads, and the power supply priority of ordinary loads is higher than that of interruptible loads. When distributed power sources in the power system cannot meet the load demand for combined power supply from energy storage, the electrical energy released by energy storage devices should be used to prioritize meeting the load demand of critical loads. When there is still residual energy after the energy storage device releases electricity to meet the load demand of critical loads, the residual energy is used to meet the load demand of ordinary loads first. When there is still residual energy after the energy storage device releases electricity to meet the load requirements of critical and ordinary loads, the residual energy is used to meet the load requirements of interruptible loads.

[0041] In one embodiment, classifying the multiple loads on the load side into critical loads, normal loads, and interruptible loads includes: Based on relevant information about the load, determine the life safety risks, public order impact, amount of loss, number of system crashes, and interruption tolerance of the load after a power outage, and determine the scenario type of the load. The risk to life safety, impact on public order, amount of loss, number of system crashes, interruption tolerance, and scenario type are normalized, and the criticality value of the load end is calculated based on the normalization result. in, This indicates the criticality value of the load end. This indicates that the interruption affects the weight. Indicates the risk value to life safety. Indicates the impact value on public order. Indicates the amount of loss. This indicates the number of system crashes. Indicates the level of tolerance for interruption. This indicates the urgency value of the scenario type on the load side. Indicates the weight of electricity consumption behavior. Indicates the weight of electricity tolerance. Indicates the weight of electricity consumption scenarios; Based on the criticality value, the initial load type of the load end is determined; Real-time monitoring of the power supply status of the load end under the initial load type, including user complaint rate, user complaint interval time, and user appeal rate, to calculate the feedback value of the load end under the initial load type; in, This indicates the feedback value from the load side under the initial load type. Indicates the interval between user complaints. This indicates the reference interval time under the initial load type. Indicates the user complaint rate. Indicates the user complaint rate; When the feedback value is greater than a preset threshold, the load type of the load end is increased; otherwise, the initial load type of the load end remains unchanged.

[0042] In this embodiment, the larger the feedback value of the load side under the initial load type, the lower the user satisfaction, indicating that the load type setting is unreasonable.

[0043] In this embodiment, upgrading the load type on the load side specifically involves upgrading a normal load to a critical load and upgrading an interruptible load to a normal load.

[0044] In this embodiment, the values ​​for life safety risk, public order impact, amount of loss, number of system crashes, interruption tolerance, and scenario type are all normalized values.

[0045] In this embodiment, the scenario types on the load side include, in ascending order of urgency value, daily scenarios, power grid stress scenarios, and disaster emergency scenarios.

[0046] In this embodiment, the higher the criticality value of the load side, the higher the priority of the initial load type.

[0047] In this embodiment, the user appeal rate is the number of user appeals regarding load type on the APP.

[0048] In this embodiment, the user complaint interval refers to the time interval between the power outage and the user complaint.

[0049] In this embodiment, the weights for interruption impact, electricity consumption behavior, electricity tolerance, and electricity consumption scenarios can be determined based on actual needs or through a weight allocation model based on historical data.

[0050] The beneficial effects of the above design scheme are as follows: By using quantifiable indicators such as life safety risks, impact on public order, amount of loss, number of system crashes, and interruption tolerance, combined with the urgency of the scenario type, a criticality value is calculated. It integrates weights for interruption impact, electricity consumption behavior, and electricity consumption scenarios, avoiding subjective assumptions and making the classification criteria clearer and more convincing, thus improving the accuracy of classification from the source. Furthermore, by introducing a user feedback adjustment mechanism, feedback values ​​are calculated through real-time monitoring of user complaint rates, complaint intervals, and user appeal rates. When the feedback value exceeds a threshold, the load type is automatically upgraded. This design breaks the limitation of fixed initial classifications, enhances the system's adaptability to complex and variable electricity consumption scenarios, and achieves both scientific and objective classification while enabling flexible responses to actual needs. It provides accurate and reliable load classification criteria for the priority scheduling of energy storage combined power supply systems, thereby improving the stability, economy, and user acceptance of the entire power supply system.

[0051] To better demonstrate the priority-management-based energy storage joint power supply scheduling process, this embodiment provides a priority-management-based energy storage joint power supply scheduling method, implemented based on the aforementioned priority-management-based energy storage joint power supply system, including: Collect and process real-time data on energy storage combined power supply in the power system, determine the characteristic data of energy storage combined power supply, analyze the characteristic data of energy storage combined power supply, and predict the load demand of energy storage combined power supply. When distributed power energy storage meets the load demand of energy storage combined power supply, energy storage devices that can store electrical energy are dispatched and used to store excess distributed power energy. When distributed power sources cannot meet the load demand for combined energy storage power supply, energy storage devices that can release electrical energy are dispatched, and the electrical energy released by the energy storage devices is used to meet the load demand in order of priority.

[0052] In summary, by monitoring load and power supply data in real time, real-time data of energy storage combined power supply in the power system is collected. This data is then processed, key features are extracted, and characteristic data of energy storage combined power supply is determined. By constructing a load prediction model for energy storage combined power supply and analyzing the characteristic data, the load demand of energy storage combined power supply is predicted, and the load prediction results are determined. Based on the load demand prediction, the combined power supply of energy storage devices is scheduled and managed, and load demand is met according to priority. This allows for effective scheduling and management of energy storage combined power supply according to priority, enabling intelligent scheduling to optimize energy allocation and adapt to complex and ever-changing electricity demands.

[0053] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0054] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A combined energy storage and power supply system with priority management, characterized in that, include: The data acquisition module is used to monitor load and power supply data in real time and collect real-time data of energy storage combined power supply in the power system. The data processing module is used to process real-time data of energy storage combined power supply, extract key features, and determine characteristic data of energy storage combined power supply. The analysis and prediction module is used to build a load prediction model for energy storage combined power supply, analyze the characteristic data of energy storage combined power supply, and predict the load demand of energy storage combined power supply. The scheduling and management module is used to schedule and manage the joint power supply of energy storage devices based on load demand forecasts, and to meet load demands in a priority order.

2. The energy storage combined power supply system with priority management according to claim 1, characterized in that, Analyze the characteristic data of energy storage combined power supply, predict the load demand of energy storage combined power supply, and perform the following operations: Deploy the energy storage combined power supply load forecasting model in the energy storage combined power supply load forecasting environment; The characteristic data of energy storage combined power supply is input into the load prediction model of energy storage combined power supply. The characteristic data of energy storage combined power supply is analyzed according to the load prediction model of energy storage combined power supply, and the load demand of energy storage combined power supply is predicted to determine the load prediction result of energy storage combined power supply. The load forecast results of energy storage combined power supply are compared and analyzed with those of distributed power sources in the power system to determine whether distributed power sources in the power system can meet the load demand of energy storage combined power supply. When distributed power energy storage in the power system meets the load demand for combined energy storage power supply, energy storage devices capable of storing electrical energy are dispatched and used to store excess distributed power energy. When distributed power sources in the power system cannot meet the load demand for combined energy storage power supply, energy storage devices that can release electrical energy are dispatched, and the electrical energy released by the energy storage devices is used to meet the load demand in order of priority.

3. The energy storage combined power supply system with priority management according to claim 2, characterized in that, By using energy storage devices to release electrical energy and prioritizing it to meet load demands, the following operations are performed: The multiple loads on the load side are classified into critical loads, normal loads and interruptible loads, and different power supply priorities are assigned to critical loads, normal loads and interruptible loads. Among them, the power supply priority of critical loads is higher than that of ordinary loads, and the power supply priority of ordinary loads is higher than that of interruptible loads. When distributed power sources in the power system cannot meet the load demand for combined power supply from energy storage, the electrical energy released by energy storage devices should be used to prioritize meeting the load demand of critical loads. When there is still residual energy after the energy storage device releases electricity to meet the load demand of critical loads, the residual energy is used to meet the load demand of ordinary loads first. When there is still residual energy after the energy storage device releases electricity to meet the load requirements of critical and ordinary loads, the residual energy is used to meet the load requirements of interruptible loads.

4. The energy storage combined power supply system with priority management according to claim 3, characterized in that, The classification of multiple loads at the load end into critical loads, normal loads, and interruptible loads includes: Based on relevant information about the load, determine the life safety risks, public order impact, amount of loss, number of system crashes, and interruption tolerance of the load after a power outage, and determine the scenario type of the load. The risk to life safety, impact on public order, amount of loss, number of system crashes, interruption tolerance, and scenario type are normalized, and the criticality value of the load end is calculated based on the normalization result. Based on the criticality value, the initial load type of the load end is determined; Real-time monitoring of the power supply status of the load end under the initial load type, including user complaint rate, user complaint interval time, and user appeal rate, to calculate the feedback value of the load end under the initial load type; When the feedback value is greater than a preset threshold, the load type of the load end is increased; otherwise, the initial load type of the load end remains unchanged.

5. A combined energy storage and power supply system with priority management according to claim 3, characterized in that, Collect real-time data on energy storage combined with power supply in the power system and perform the following operations: Real-time monitoring of voltage, current, power, energy consumption, equipment operating status, and environmental parameters at the load end of the power system, and collection of real-time data at the load end; Real-time monitoring of voltage, frequency, power output, energy type, and energy storage device status at the power supply end of the power system, and collection of real-time data from the power supply end; Real-time data on combined energy storage and power supply in the power system is generated based on real-time data from the load side and the power supply side.

6. The energy storage combined power supply system with priority management according to claim 5, characterized in that, The real-time data of the combined energy storage and power supply is processed, and the following operations are performed: The real-time data of energy storage combined power supply is cleaned to remove noise and outliers that are not useful for the priority management of energy storage combined power supply, and outliers that are useful for the priority management of energy storage combined power supply are replaced. Standardize the real-time data of energy storage combined power supply, unify the format of the real-time data of energy storage combined power supply, eliminate the differences in the units of measurement in the real-time data of energy storage combined power supply, and form standardized real-time data of energy storage combined power supply.

7. A combined energy storage and power supply system with priority management according to claim 6, characterized in that, The real-time data of the combined energy storage and power supply is processed, and the following operations are performed: The real-time data of energy storage combined with power supply is organized and put into a data view. The integrity of the real-time data of energy storage combined with power supply in the data view is evaluated. After the data evaluation, the real-time data of energy storage combined with power supply in the data view is stored and backed up. Feature extraction is performed on real-time data of energy storage combined power supply. Features related to the priority management of energy storage combined power supply are extracted from the real-time data of energy storage combined power supply to determine the feature data of energy storage combined power supply.

8. A combined energy storage and power supply system with priority management according to claim 7, characterized in that, Construct an energy storage-integrated power supply load forecasting model and perform the following operations: Historical data on the combined power supply of energy storage in the power system are collected, and the collected historical data on the combined power supply of energy storage in the power system are divided into training set and test set; The machine learning model is trained using a training set, enabling it to autonomously learn the load prediction behavior of energy storage combined power supply from the training set and predict the load demand of energy storage combined power supply, thus determining the load prediction model of energy storage combined power supply. The test set was used to test the load prediction model of energy storage combined power supply, evaluate the generalization ability of the energy storage combined power supply load prediction model, and determine whether the energy storage combined power supply load prediction model can achieve the expected effect of predicting the load demand of energy storage combined power supply. If the constructed energy storage combined power supply load prediction model does not meet the expected requirements, the parameters of the energy storage combined power supply load prediction model will be adjusted and optimized to make the constructed energy storage combined power supply load prediction model meet the expected requirements, thereby determining the energy storage combined power supply load prediction model that meets the expected requirements.

9. A combined energy storage and power supply system with priority management according to claim 1, characterized in that, When the constructed energy storage combined power supply load forecasting model does not meet the expected requirements, the parameters of the energy storage combined power supply load forecasting model will be adjusted and optimized, including: The time period error, scenario error, time scale error, and feature contribution are obtained from the test results of the energy storage combined power supply load prediction model on the test set. The target time period with a value greater than a preset time period error threshold is obtained from the time period error; the target scene with a value greater than a preset scene error threshold is obtained from the scene error; the target time scale with a value greater than a preset time scale error threshold is obtained from the time scale error; and the target feature with a value less than a preset feature contribution threshold is obtained from the feature contribution. Based on the target time period and target scenario, determine the additional time period data and additional scenario data; based on the target features, determine the additional feature type; based on the target time scale, determine the data time series length. The training set is augmented by adding time period data, scene data, feature types, and data time series length to obtain the target training set. Based on the difference between the feature contribution of the target feature and the preset feature contribution threshold, the weighting weight of the target feature is determined, and an attention layer is added to the power supply load prediction model based on the weighting weight. Based on the test results of the energy storage combined power supply load prediction model on the test set and the initial hyperparameter combination of the energy storage combined power supply load prediction model, multiple hyperparameter combinations are adaptively established. Based on the target training set and attention layer, the energy storage combined power supply load prediction model is optimized, and multiple hyperparameter combinations are iterated based on the Bayesian optimization algorithm to obtain the optimal hyperparameter combination, thus achieving the optimization of the energy storage combined power supply load prediction model.

10. A method for joint power supply scheduling of energy storage with priority management, implemented based on the joint power supply system of energy storage with priority management as described in claim 9, characterized in that, include: Collect and process real-time data on energy storage combined power supply in the power system, determine the characteristic data of energy storage combined power supply, analyze the characteristic data of energy storage combined power supply, and predict the load demand of energy storage combined power supply. When distributed power energy storage meets the load demand of energy storage combined power supply, energy storage devices that can store electrical energy are dispatched and used to store excess distributed power energy. When distributed power sources cannot meet the load demand for combined energy storage power supply, energy storage devices that can release electrical energy are dispatched, and the electrical energy released by the energy storage devices is used to meet the load demand in order of priority.

Images