Energy storage energy management system for improving demand protection algorithm

Abstract

The invention discloses an energy storage energy management system for improving a demand protection algorithm, and the system comprises a data collection module which is used for obtaining energy storage system load data, distributed energy operation data, and energy storage system operation data; the correlation feature analysis module is used for acquiring a correlation feature set between the load and the energy supply; the control strategy generation module is used for adjusting a charging and discharging threshold value and formulating a charging and discharging control strategy; the control strategy execution module is used for formulating a charging and discharging control strategy and adjusting the working mode of the energy storage converter; the cooperative execution module is used for dynamically optimizing a power distribution strategy by dispatching distributed energy and power grid cooperative work; and the dynamic power adjustment strategy adjustment module is used for adjusting the charging and discharging protection threshold value and the power adjustment strategy through a feedback mechanism. Through an intelligent control and scheduling mechanism, the charging and discharging efficiency of the energy storage system is improved, the stability and flexibility of the power grid are enhanced, and effective guarantee is provided for safe operation of the power grid.

Description

Technical Field

[0001] The present invention relates to the field of power load control, and in particular to an energy storage management system with an improved demand protection algorithm. Background Art

[0002] With the transformation of the global energy structure and the widespread application of renewable energy, the stability of the power system faces new challenges. Renewable energy sources such as wind and solar energy are intermittent and volatile. The instability of power supply makes the load fluctuation of the power grid more intense, resulting in greater pressure on the load balance and voltage stability of the power grid. In this context, traditional power grid dispatching and energy storage management methods have exposed many shortcomings in dealing with these challenges. First, the management strategy of traditional energy storage systems is too static and lacks flexibility. Existing energy storage systems usually rely on fixed charging and discharging strategies, and fail to make efficient adjustments in real time when facing rapidly changing power grid loads and energy supply. The limitations of this fixed strategy make it impossible for energy storage systems to provide timely and effective power regulation when the power grid load fluctuates or the energy supply changes, thus failing to meet the requirements of the power grid for stability and flexibility. Secondly, with the widespread deployment of distributed energy, the load and energy supply in the power grid have become more complex and dynamic. The power generation capacity of distributed energy is affected by weather and seasonal changes, and often shows greater volatility. This volatility exacerbates the fluctuation of power grid load, making it difficult for traditional power grid dispatching methods to adapt. Although some advanced power dispatching methods attempt to integrate distributed energy data, in actual operation, how to efficiently dispatch these distributed energy resources and coordinate with energy storage systems is still an urgent problem to be solved. In addition, existing energy storage systems usually lack accurate protection mechanisms. When the load rises or falls sharply or the energy supply fluctuates suddenly, the energy storage system needs to be able to respond quickly and adjust the charging and discharging strategy. However, traditional protection strategies and control mechanisms fail to dynamically optimize the real-time changing power grid conditions, so that when emergencies occur, the response capability and protection strategy of the energy storage system cannot effectively guarantee the safe operation of the power grid. For example, when the load of the power grid suddenly increases or there is a sudden load fluctuation, the energy storage system may not be able to provide sufficient power support in a short time, resulting in the inability of the power grid to maintain balance, which may cause safety hazards such as voltage fluctuations or power grid overload. The existence of these problems not only affects the operating efficiency of the energy storage system, but also restricts the stable operation of the power grid under high load and high fluctuation environments. Therefore, how to improve the dynamic response capability of the energy storage system, optimize the load forecasting and dispatching strategy, and improve the adaptive control mechanism of the energy storage system has become a key challenge in the current power system management and energy storage technology development. Summary of the invention

[0003] The present invention solves the problems existing in the above-mentioned prior art and provides an energy storage energy management system with an improved demand protection algorithm, which mainly includes:

[0004] The data acquisition module is used to obtain the energy storage system load data, distributed energy operation data and energy storage system operation data based on the load monitoring equipment and smart meters, and to clean, pre-process, standardize and calibrate the acquired data;

[0005] The correlation feature analysis module is used to combine historical load and energy supply data, use long-term short-term memory networks to make long-term trend predictions, and use support vector machines to analyze the correlation between load fluctuations and energy supply fluctuations to obtain the correlation feature set and time delay information between load and energy supply;

[0006] The control strategy generation module is used to extract demand forecasting features using the chi-square test, combine the ARIMA model and the long short-term memory network model to forecast demand, and adjust the charge and discharge thresholds and formulate the charge and discharge control strategy according to the grid voltage and the charge state of the energy storage system;

[0007] The control strategy execution module is used to configure the charging, discharging and standby modes through the signal generator, adjust the control signal parameters and obtain the load data in real time, transmit the control instructions through the communication interface, formulate the charging and discharging control strategy and adjust the working mode of the energy storage converter;

[0008] The collaborative execution module is used to continuously monitor voltage and current data and adjust the converter working mode according to the grid load demand, distributed energy generation capacity and the charging and discharging status of the energy storage system, formulate dynamic power adjustment strategies, and dynamically optimize the power allocation strategy by dispatching distributed energy to work in coordination with the grid;

[0009] The dynamic power adjustment strategy adjustment module is used to dynamically update the hybrid prediction model parameters according to the real-time acquired grid load, energy storage system status and distributed energy data, and adjust the charge and discharge protection thresholds and power adjustment strategies through a feedback mechanism.

[0010] Furthermore, the data acquisition module is used to obtain energy storage system load data, distributed energy operation data and energy storage system operation data according to the load monitoring equipment and the smart meter, and perform data cleaning, preprocessing, standardization and calibration on the acquired data, including:

[0011] According to the load monitoring equipment and smart meters, the load data of the energy storage system is obtained, including current, voltage, instantaneous power change and power factor, where the smart meter records the load type information; by integrating with the distributed energy control system, the distributed energy operation data is obtained, including real-time output power, power change rate and energy quality; through the interface with the energy storage device, the energy storage system operation data is obtained, including voltage, current, charge state, health state and converter state; according to the obtained load data, distributed energy operation data and energy storage system operation data; using data synchronization technology, the load data, distributed energy operation data and energy storage system operation data are unified Integrate in the time dimension; use data cleaning and preprocessing algorithms to filter the integrated data, including setting range thresholds and filtering algorithms to eliminate noise and outliers in the data acquisition process, interpolate data, and handle null values ​​caused by equipment transmission interruptions; use data standardization technology to standardize the preprocessed data and convert them to the same dimension; set a number of data sampling points, and combine with known standard data sources to perform point-by-point calibration, and use the weighted average method to smooth the errors of the acquired data. The standard data source is a known load characteristic; transmit the calibrated data to the central monitoring system through the communication interface.

[0012] Furthermore, the correlation feature analysis module is used to combine historical load and energy supply data, use long short-term memory network to perform long-term trend prediction, and use support vector machine to analyze the correlation between load fluctuation and energy supply fluctuation to obtain the correlation feature set and time delay information between load and energy supply, including:

[0013] The calibrated data is converted from the time domain to the frequency domain using the Fourier transform algorithm, and the instantaneous frequency changes of the signal are extracted through wavelet transform to identify high-frequency fluctuations of different loads or supply changes; the convolutional neural network is used to extract features of short-term fluctuations to obtain the short-term fluctuation trend of load and energy supply, including its high-frequency fluctuations, instantaneous change frequency and amplitude information. The short-term fluctuation characteristics include frequency distribution, amplitude and fluctuation intensity; the historical load data, historical energy supply data, and the corresponding timestamp data are obtained through the central monitoring system, and the long short-term memory network is used for model training to predict the long-term trend changes and seasonal fluctuations of load and energy supply; according to the short-term fluctuation trend and long-term trend of load and energy supply, the support vector machine algorithm is used to judge the correlation between load fluctuations and energy supply fluctuations, and the correlation feature set between load and energy supply is obtained through regression analysis and Pearson correlation coefficient calculation method, including time delay, time delay correlation and time delay correlation intensity.

[0014] Furthermore, the control strategy generation module is used to extract demand forecasting features using a chi-square test, combine the ARIMA model and the long short-term memory network model to perform demand forecasting, and adjust the charge and discharge thresholds and formulate a charge and discharge control strategy according to the grid voltage and the state of charge of the energy storage system, including:

[0015] Obtain historical load data, load type, energy supply data, weather data, and distributed energy generation data, and use the chi-square test to extract demand forecasting features related to demand forecasting, including load type, seasonal pattern, load fluctuation, and energy supply change. The seasonal pattern is the correlation between loads of different load types and different seasons and energy supply. Load types include industrial loads and residential loads. Based on the historical data of load demand forecasting features, a hybrid forecasting model combining the ARIMA model and the long-short-term memory network model is used to forecast demand, including using the ARIMA model to identify the linear law and seasonal pattern of demand, combining the long-short-term memory network model to identify nonlinear fluctuations, and predicting demand within a preset time period in the future. Obtain the current state of charge and power of the energy storage system. The charging and discharging demand protection threshold adjustment unit is used to adjust the charging and discharging demand protection threshold based on the grid voltage data, and a charging and discharging control strategy for the energy storage system is formulated; if the difference between the grid voltage and the preset voltage upper limit threshold is less than the preset difference threshold, the charging demand protection threshold is adjusted; if the current state of charge is lower than the preset state of charge threshold, the discharge demand protection threshold is lowered; the sudden load change or energy supply fluctuation information in the demand forecast result is obtained to determine whether there is a risk of extreme fluctuation; if the demand increase rate is higher than the preset increase rate threshold or the energy supply decrease speed is higher than the preset speed threshold, a rapid response control strategy is generated based on real-time data and preset emergency response rules, including emergency adjustment of the charging and discharging power of the energy storage system, starting the backup energy storage resources, and emergency power interaction and coordination with the grid.

[0016] It also includes a charge and discharge demand protection threshold adjustment unit, which is used to obtain the current charge state of the energy storage system and the grid voltage data, adjust the charge and discharge demand protection threshold, and formulate the charge and discharge control strategy of the energy storage system, which specifically includes:

[0017] The state of charge of the energy storage system and the voltage of the power grid are obtained based on real-time data to determine whether the charge and discharge thresholds need to be adjusted; the state of charge of the energy storage system and the voltage of the power grid are obtained based on real-time data to calculate the grid voltage deviation and the state of charge deviation; according to the grid voltage deviation and the state of charge deviation, the charging demand protection threshold adjustment formula C1=C-(ΔV×α) is used to determine the dynamically adjusted charging demand protection threshold C1, where C is the original charging threshold, ΔV is the grid voltage deviation, and α is the sensitivity of the voltage deviation to the charging threshold. V maxis the upper voltage limit of the power grid, V is the current voltage value of the power grid, P is the current power grid load power, P max is the maximum load power of the power grid, γ is a constant, which represents the voltage stability requirement of the power grid and is obtained by fitting historical data; according to the voltage deviation and state of charge deviation of the power grid, the discharge demand protection threshold adjustment formula D1=D+(ΔS×β) is used to determine the dynamically adjusted discharge demand protection threshold D1, where D is the original discharge threshold, ΔS is the state of charge deviation, and β is the sensitivity of the state of charge deviation to the discharge threshold. is the preset target state of charge, S is the current state of charge of the energy storage system, and λ is a constant that indicates the degree of influence of battery health and grid load on the adjustment of the discharge threshold. According to the adjusted charging threshold and discharging threshold, the charging and discharging control strategy of the energy storage system is formulated.

[0018] Furthermore, the control strategy execution module is used to configure the charging, discharging and standby modes through a signal generator, adjust the control signal parameters and obtain load data in real time, transmit control instructions through a communication interface, formulate a charging and discharging control strategy and adjust the working mode of the energy storage converter, including:

[0019] According to the charge and discharge control strategy and the operation requirements of the energy storage system, different working modes are configured through the signal generator, and the corresponding control signal parameters are determined. The working modes include charging mode, discharging mode, and standby mode. The control signal parameters include but are limited to power size and charge and discharge rate; the current load data of the energy storage system, the energy storage system status, and the grid demand data are obtained, the adaptive communication protocol is configured, and the control instructions are transmitted to the energy storage device in real time through the communication interface, and the communication interface includes CAN bus or Ethernet; by calculating the current load demand, the charge state of the energy storage system, and the grid load fluctuation, the power and time period required for charging and discharging are determined, and control strategy instructions are formulated, including control parameters of charging power, discharging power, and duration; by sending the strategy instructions to the energy storage converter control unit, the execution unit controls the charging and discharging process of the energy storage device according to the instructions; the real-time status information of the energy storage converter control unit is obtained, including the current charging and discharging state, converter working mode, and output power, wherein the converter working mode includes bidirectional conversion mode and constant current mode; through the energy storage converter control unit, the output of the energy storage converter is adjusted according to the set power range to perform charging and discharging operations.

[0020] Furthermore, the collaborative execution module is used to continuously monitor voltage and current data and adjust the converter working mode according to the grid load demand, distributed energy generation capacity and energy storage system charging and discharging status, formulate dynamic power adjustment strategy, and dynamically optimize power allocation strategy by scheduling distributed energy and grid to work together, including:

[0021] Continuously monitor the real-time data of voltage and current during the charging and discharging process, and adjust the converter working mode in real time according to the monitoring information; obtain the total demand information of the power grid, the real-time power generation capacity of distributed energy and the charging and discharging status of the energy storage system, determine the load demand information of the power grid and the energy supply capacity of each party of distributed energy, and dynamically adjust the power generation power of distributed energy by dispatching distributed energy and the power grid to work together, supplement the lack of power in the energy storage system or adjust the excess power; obtain the real-time load information of the power grid, the power generation data of distributed energy, and the charging and discharging status of the energy storage system, adopt the dynamic programming algorithm, and formulate a dynamic power adjustment strategy based on the power allocation optimization target setting. The power allocation optimization target setting includes the load balance of the power grid, the effective utilization of the energy storage system and the maximum efficiency output of the distributed energy, while avoiding overload or unstable fluctuations. The dynamic adjustment strategy includes increasing the discharge power of the energy storage system when the load fluctuation of the power grid is greater than the preset fluctuation threshold, giving priority to the use of the energy storage system to provide electricity when the power grid voltage is lower than the preset voltage threshold, and giving priority to the dispatch of the distributed energy when the power generation capacity of the distributed energy is higher than the preset capacity threshold, and reducing the load of the energy storage system according to the needs of the power grid.

[0022] Furthermore, the dynamic power adjustment strategy adjustment module is used to dynamically update the hybrid prediction model parameters according to the real-time acquired grid load, energy storage system status and distributed energy data, and adjust the charge and discharge protection threshold and power adjustment strategy through a feedback mechanism, including:

[0023] Obtain grid load, energy storage system status and distributed energy data in real time through the communication interface and transmit it to the central monitoring system; summarize the real-time grid load, energy storage system status and distributed energy data to the human-computer interaction interface, and use the human-computer interaction interface to display the operating status information of the energy storage system, load and distributed energy, the real-time forecast value of demand, the dynamically changing protection threshold and potential demand risk warning information; use real-time data to update the parameters of the hybrid prediction model, and predict the demand within a preset time period in the future. According to the prediction results, the current state of charge value of the energy storage system and the grid voltage data, dynamically adjust the charging demand protection threshold and the discharging demand protection threshold; during the operation of the energy storage system, obtain feedback data in real time and record it All relevant control instructions and execution results, identify the error between demand forecast and actual demand, determine the responsiveness of the protection strategy, and transmit feedback data to the central monitoring system and the hybrid prediction model; use the error correction algorithm to optimize the hybrid prediction model, re-run the optimized hybrid prediction model for demand forecast, and evaluate the actual operating effect of the optimized hybrid prediction model; adjust the parameters of the dynamic power adjustment strategy according to the optimized demand forecast value; compare the actual operating status with the forecast value through real-time monitoring and feedback mechanism, judge the real-time effect of the hybrid prediction model and the dynamic power adjustment strategy, and adjust the parameters of the hybrid prediction model and the dynamic power adjustment strategy through the automatic feedback system and the real-time monitoring data.

[0024] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0025] The present invention provides an energy storage energy management system with an improved demand protection algorithm. The present invention adjusts the charge and discharge mode of the energy storage system in real time to cope with the load fluctuation of the power grid and the power generation fluctuation of distributed energy, thereby effectively balancing the load and voltage of the power grid, reducing power waste, and improving the stability and reliability of the power grid. By dynamically adjusting the charge and discharge control strategy of the energy storage system, it can respond quickly when the load of the power grid is at a peak or when there is a sudden load fluctuation, ensuring that the power grid will not fail due to excessive load or unstable voltage. The present invention can also perceive and respond to changes in the load of the power grid and fluctuations in energy supply in real time by continuously monitoring the load of the power grid, the state of the energy storage system, and the operation data of distributed energy. Combined with real-time data and dynamic scheduling strategies, the present invention can optimize power distribution, ensure that the energy storage system and distributed energy can play the maximum efficiency in the optimal period, thereby reducing dependence on traditional power grids and improving the demand response capability of the power grid. The present invention can accurately adjust the charge and discharge thresholds and protection strategies under different load demands and power grid voltage conditions to ensure that the energy storage system maintains the best working state under various working conditions. Through efficient operation strategies, the present invention can improve the adaptability of the energy storage system, so that it can better adapt to different load fluctuations and energy supply conditions, and improve the overall stability and safety of the power grid. Therefore, the present invention realizes the efficient operation of the energy storage system through the intelligent control and scheduling mechanism, which not only improves the charging and discharging efficiency of the energy storage system, but also enhances the stability and flexibility of the power grid, providing effective guarantee for the safe operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 A flow chart of an energy storage energy management system with an improved demand protection algorithm of the present invention;

[0027] Figure 2 A schematic diagram of an energy storage energy management system with an improved demand protection algorithm of the present invention;

[0028] Figure 3 It is another schematic diagram of an energy storage energy management system with an improved demand protection algorithm according to the present invention. DETAILED DESCRIPTION

[0029] In order to make the objectives, technical strategies and advantages of the present invention more clear, the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0030] like Figure 1-3 In this embodiment, an energy storage energy management system with an improved demand protection algorithm may specifically include:

[0031] Step S101, a data acquisition module is used to obtain energy storage system load data, distributed energy operation data and energy storage system operation data according to load monitoring equipment and smart meters, and perform data cleaning, preprocessing, standardization and calibration on the obtained data.

[0032] According to the load monitoring equipment and smart meters, the load data of the energy storage system is obtained, including current, voltage, instantaneous power change and power factor, where the smart meter records the load type information. By integrating with the distributed energy control system, the distributed energy operation data is obtained, including real-time output power, power change rate and energy quality. Through the interface with the energy storage device, the energy storage system operation data is obtained, including voltage, current, state of charge, health state and converter state. According to the acquired load data, distributed energy operation data and energy storage system operation data. Using data synchronization technology, the load data, distributed energy operation data and energy storage system operation data are integrated on a unified time dimension. According to the integrated data, the data cleaning and preprocessing algorithms are used to filter the data, including setting range thresholds and filtering algorithms to eliminate noise and outliers in the data acquisition process, interpolating the data, and processing null values ​​caused by device transmission interruption. Using data standardization technology, the preprocessed data is standardized and converted to the same dimension. By setting several data sampling points and combining with known standard data sources, point-by-point calibration is performed, and the error of the acquired data is smoothly corrected using the weighted average method. The standard data source is the known load characteristics. The calibrated data is transmitted to the central monitoring system through the communication interface.

[0033] For example, there is an energy storage system connected to a residential load, multiple distributed energy units, and a central monitoring system. The energy storage system includes a smart meter for monitoring the current, voltage, instantaneous power change, and power factor of the load. At the same time, the smart meter also records the load type information. The loads in the residential area include air conditioners, electric water heaters, and lighting equipment, while the distributed energy system consists of several solar photovoltaic panels and a small wind turbine. According to the operating information of these devices, the data is transmitted to the central monitoring system through a high-speed communication interface. The load data, distributed energy data, and energy storage system operation data are integrated in a unified time dimension through data synchronization technology. Within a sampling period of 10 minutes, the load data of the energy storage system is obtained through the load monitoring device and the smart meter, including a current of 10A, a voltage of 220V, a power of 2200W, and a power factor of 0.95. By integrating with the distributed energy control system, the distributed energy operation data is obtained, including the output power of the solar photovoltaic panel is 1200W, the power change rate is 3%, and the output power of the wind turbine is 800W, and the power change rate is 1%. Through the interface with the energy storage device, the energy storage system operation data is obtained, including the voltage of 48V, the current of 5A, the state of charge SOC of 60%, the energy storage system health status is good, and the converter is working normally. The load data, distributed energy data and energy storage system operation data are integrated in a unified time dimension through data synchronization technology. The data is cleaned by setting range thresholds and filtering algorithms. If the range of the collected voltage and power data changes within the normal working range within this 10 minutes, does not exceed the preset voltage range of ±10V, and there is no abnormal power fluctuation, the data will be considered reliable and will not be filtered. If some measured values ​​in the data are found to be abnormal, such as the current value is higher than the maximum safe working current or the power factor is not within the predetermined effective range, these unreasonable data are automatically cleared through the set filtering algorithm. After completing the data cleaning, the pre-processed data needs to be unified into a standard dimension, such as the current unit needs to be converted into a standardized current value, and the voltage is converted into a standard voltage value. At this time, all voltage, current and power data will be converted into standard data units through standardized formulas. Point-by-point calibration is performed based on known standard data sources. For example, if the standard current of an air conditioner under a certain load is known to be 9A, and the monitored current is 10A, the error of this data is smoothed and corrected based on the weighted average method to obtain a more accurate current value. These calibrated standardized data will provide a basis for subsequent load forecasting, energy storage scheduling, and grid coordination. After data cleaning, standardization, calibration, and other processing, the data will be input into the central monitoring system.

[0034] Step S102, a correlation feature analysis module is used to combine historical load and energy supply data, use a long short-term memory network to perform long-term trend prediction, and use a support vector machine to analyze the correlation between load fluctuations and energy supply fluctuations to obtain a correlation feature set and time delay information between load and energy supply.

[0035] The Fourier transform algorithm is used to convert the calibrated data from the time domain to the frequency domain. The instantaneous frequency changes of the signal are extracted by wavelet transform to identify high-frequency fluctuations of different load or supply changes. The convolutional neural network is used to extract features of short-term fluctuations to obtain the short-term fluctuation trend of load and energy supply, including its high-frequency fluctuations, instantaneous change frequency and amplitude information. The short-term fluctuation characteristics include frequency distribution, amplitude and fluctuation intensity. The historical load data, historical energy supply data and corresponding timestamp data are obtained through the central monitoring system. The long short-term memory network is used for model training to predict the long-term trend changes and seasonal fluctuations of load and energy supply. According to the short-term fluctuation trend and long-term trend of load and energy supply, the support vector machine algorithm is used to judge the correlation between load fluctuation and energy supply fluctuation. Through regression analysis and Pearson correlation coefficient calculation method, the correlation feature set between load and energy supply is obtained, including time delay, time delay correlation and time delay correlation intensity.

[0036] For example, there is a smart grid system including multiple load devices and a group of distributed energy systems. The multiple load devices include air conditioners, lighting equipment and electric water heaters, and the distributed energy systems include solar photovoltaic and wind turbines. The load and energy supply data are recorded in real time through smart meters and monitoring equipment, including current, voltage, power factor, etc. in time series, and each data point contains a timestamp. For the calibrated load and energy supply data, such as in a certain period of time, the time series of load power is [2200, 2250, 2300, 2250, 2200] W, these data are converted from the time domain to the frequency domain by Fourier transform algorithm, and frequency domain data are obtained to identify higher frequency fluctuations in the data, such as rapid fluctuations in power demand. Through Fourier transform, these instantaneous fluctuations are extracted as different frequency components, and possible high-frequency noise or load fluctuations are identified. The wavelet transform algorithm is used to further process the data, extract the instantaneous frequency changes of the signal, and identify short-term high-frequency changes in load or supply fluctuations. For example, within an hour, the instantaneous frequency of the load power is found to fluctuate frequently between 10-20 minutes, which may be caused by the start-up or shutdown of the air conditioning equipment. These instantaneous frequency changes are extracted through the wavelet transform algorithm, and the frequency, amplitude and change intensity of each fluctuation can be accurately captured. If the high-frequency fluctuation data of the load power is extracted within a certain period of time, including its frequency of 10Hz, amplitude of 50W and fluctuation intensity of 30%, the convolutional neural network is used to extract the features of the short-term fluctuations of the load and energy supply, and generate a feature set of short-term fluctuations, including the frequency distribution and change trend of these fluctuations. The historical load data and historical energy supply data are obtained through the central monitoring system. The historical load data includes the load power data of the past 6 months, as well as the corresponding seasonal data, such as higher load in summer and lower load in winter. The long short-term memory network is used for model training to learn the long-term trend changes and seasonal fluctuations of the load. The long short-term memory network model finds that in summer, the load power is 2500W on average, while in winter it is 2000W, and there are annual periodic fluctuations. According to the short-term fluctuation trend and long-term trend of load and energy supply, the support vector machine algorithm is used to judge the correlation between load fluctuation and energy supply fluctuation. For example, the SVM algorithm finds that there is a certain correlation between load fluctuation and energy supply fluctuation, especially during the peak load period in summer, when the output power of solar photovoltaic power generation is high, and the fluctuations of the two show synchronous changes. This correlation is further quantified through regression analysis and Pearson correlation coefficient calculation method, and the result is that the time delay is 2 hours and the correlation strength is 0.85, indicating that in certain periods of time, the time delay and correlation between load fluctuation and energy supply fluctuation are very strong.

[0037] Step S103, a control strategy generation module is used to extract demand forecasting features using a chi-square test, perform demand forecasting in combination with an ARIMA model and a long short-term memory network model, and adjust charge and discharge thresholds and formulate charge and discharge control strategies according to grid voltage and the state of charge of the energy storage system.

[0038] Obtain historical load data, load type, energy supply data, weather data, and distributed energy generation data, and use the chi-square test to extract demand forecasting features related to demand forecasting, including load type, seasonal pattern, load fluctuation, and energy supply change. The seasonal pattern is the correlation feature between load and energy supply in different seasons of different load types. Load types include industrial loads and residential loads. According to the historical data of load demand forecasting features, a hybrid forecasting model combining the ARIMA model and the long short-term memory network model is used to forecast demand, including using the ARIMA model to identify the linear law and seasonal pattern of demand, combining the long short-term memory network model to identify nonlinear fluctuations, and predicting demand in the future preset time period. Obtain the current state of charge and grid voltage data of the energy storage system, use the charge and discharge demand protection threshold adjustment unit to adjust the charge and discharge demand protection threshold, and formulate the charge and discharge control strategy of the energy storage system. If the difference between the grid voltage and the preset voltage upper limit threshold is less than the preset difference threshold, the charge demand protection threshold is adjusted. If the current state of charge is lower than the preset state of charge threshold, the discharge demand protection threshold is lowered. Obtain information on sudden load changes or energy supply fluctuations in demand forecast results to determine whether there is a risk of extreme fluctuations. If the demand increase rate is higher than the preset increase rate threshold or the energy supply decrease rate is higher than the preset speed threshold, a rapid response control strategy is generated based on real-time data and preset emergency response rules, including emergency adjustment of the charging and discharging power of the energy storage system, starting backup energy storage resources, and emergency power interaction and coordination with the power grid.

[0039] For example, in a certain power management system, load data, load type, distributed energy generation data, weather data and grid data are obtained in real time through smart meters, energy storage systems and distributed energy devices. The load data includes current, voltage and power factor. The load type includes residential load and industrial load. The distributed energy generation data includes the real-time power of solar and wind power generation. The weather data includes temperature and humidity. The grid data includes voltage and frequency. Through these data, the chi-square test method is used to extract the features related to demand forecasting. If in the past 6 months of data, it is found that the residential load shows a higher volatility in the winter from December to February and the summer from June to August, while the industrial load fluctuates less in the spring and autumn seasons, according to these seasonal patterns, the correlation characteristics between loads and energy supply in different seasons can be confirmed, especially the correlation between residential loads and temperature changes. For example, when the temperature is higher than 30°C, the demand for residential electricity increases, especially in the use of air-conditioning equipment. Based on these load demand characteristics, a hybrid forecasting model combining the ARIMA model and the long short-term memory network model is used to forecast demand, including identifying the linear law and seasonal pattern of demand through the ARIMA model, predicting the load demand for the next week, and predicting that the load demand in July will reach 3000kW through the ARIMA model, and the peak of the load will appear between 2 pm and 5 pm during warm weather. At the same time, the nonlinear fluctuations contained in the load demand are further identified through the long short-term memory network model, and it is predicted that the load demand fluctuates greatly during these periods, especially in extreme weather. Therefore, the long short-term memory network model gives a more accurate demand forecast value, and the load demand may reach 3300kW during these periods. Obtain real-time data of the energy storage system, including the current state of charge of the energy storage system is 40%, the grid voltage is 240V, the preset voltage upper limit threshold is 250V, the voltage difference is 10V, and the preset voltage difference threshold is 5V. In this case, according to the system settings, the difference of 10V between the grid voltage and the voltage upper limit threshold is greater than the preset difference threshold of 5V, so the charging demand protection threshold needs to be adjusted to prevent further charging from causing the grid voltage to be too high. The state of charge of the energy storage system is lower than the preset threshold of 50%, so the discharge demand protection threshold will be lowered to avoid damage to the energy storage equipment due to excessive discharge. Obtain information about load changes and energy supply fluctuations in the demand forecast results. If in a certain period of time, the load demand is monitored to increase from 2500kW to 2875kW within 1 hour, that is, the increase rate of the load demand reaches 15% of the preset increase rate threshold. At the same time, the system detects that the output power of solar power generation is showing a downward trend. Solar power generation decreases from 500kW to 410kW in the same time period, that is, the decline rate reaches 18% of the preset speed threshold, which indicates that the system faces the risk of sudden load increase and insufficient energy supply.Based on these real-time data, an emergency response control strategy is generated through preset emergency response rules, including emergency adjustment of the charging and discharging power of the energy storage system, adjusting from discharge mode to charging mode, and starting backup energy storage resources. At the same time, emergency power interaction is carried out with the power grid to stabilize the grid load and respond to sudden fluctuations.

[0040] Among them, the charge and discharge demand protection threshold adjustment unit is used to obtain the current charge state and grid voltage data of the energy storage system, adjust the charge and discharge demand protection threshold, and formulate the charge and discharge control strategy of the energy storage system.

[0041] Obtain the state of charge of the energy storage system and the voltage of the grid based on real-time data to determine whether the charge and discharge thresholds need to be adjusted. Obtain the state of charge of the energy storage system and the voltage of the grid based on real-time data, and calculate the grid voltage deviation and state of charge deviation. According to the grid voltage deviation and state of charge deviation, use the charging demand protection threshold adjustment formula C1=C-(ΔV×α) to determine the dynamically adjusted charging demand protection threshold C1, where C is the original charging threshold, ΔV is the grid voltage deviation, and α is the sensitivity of the voltage deviation to the charging threshold. V max is the upper voltage limit of the power grid, V is the current voltage value of the power grid, P is the current power grid load power, P max is the maximum load power of the power grid, γ is a constant, which represents the voltage stability requirement of the power grid, and is obtained by fitting historical data. According to the voltage deviation and state of charge deviation of the power grid, the discharge demand protection threshold adjustment formula D1=D+(ΔS×β) is used to determine the dynamically adjusted discharge demand protection threshold D1, where D is the original discharge threshold, ΔS is the state of charge deviation, and β is the sensitivity of the state of charge deviation to the discharge threshold. is the preset target state of charge, S is the current state of charge of the energy storage system, and λ is a constant that indicates the degree of influence of battery health and grid load on the adjustment of the discharge threshold. According to the adjusted charging threshold and discharging threshold, the charging and discharging control strategy of the energy storage system is formulated.

[0042] For example, in an actual application scenario, if the original charging threshold C of the energy storage system is 500kW, the upper limit Vmax of the grid voltage is set to 1.1kV, the current grid voltage V is 1.05kV, the grid load power P is 800MW, and the maximum load power Pmax of the grid is 1000MW. The sensitivity α of the voltage deviation to the charging threshold is given by the formula The result is 10, which means that every 1% voltage deviation will cause the charging threshold to change by 10kW, where γ is a constant, which represents the voltage stability requirement of the power grid, obtained by fitting historical data. The grid voltage deviation is calculated to be 0.05kV. Using the charging demand protection threshold adjustment formula C1=C-(ΔV×α), it is determined that the dynamically adjusted charging demand protection threshold C1 is 550kV. Therefore, the dynamically adjusted charging demand protection threshold is 550kW. If the original discharge threshold D of the energy storage system is 300kW, the target state of charge is 80%, and the current state of charge S of the energy storage system is 60%. The sensitivity of the state of charge deviation to the discharge threshold β is expressed by the formula The result is 20, which means that every 1% state of charge deviation will cause the discharge threshold to change by 20kW, where λ is a constant that indicates the degree of influence of battery health and grid load on the adjustment of the discharge threshold. The state of charge of the energy storage system is obtained through real-time data, and the state of charge deviation is calculated to be 20%. The discharge demand protection threshold adjustment formula D1=D+(ΔS×β) is used to calculate the dynamically adjusted discharge demand protection threshold D1 to be 620kW. Through the dynamically adjusted charging and discharging thresholds, the energy storage system adjusts the charging and discharging strategy according to the current grid voltage deviation and the state of charge of the energy storage system, including when the grid voltage deviation is large, the charging threshold is increased, and the energy storage system is charged to support grid stability. When the state of charge of the energy storage system is low, the discharge threshold is increased, and the energy storage system can discharge faster to meet the load demand.

[0043] Step S104, a control strategy execution module is used to configure charging, discharging and standby modes through a signal generator, adjust control signal parameters and obtain load data in real time, transmit control instructions through a communication interface, formulate charging and discharging control strategies and adjust the working mode of the energy storage converter.

[0044] According to the charge and discharge control strategy and the operation requirements of the energy storage system, different working modes are configured through the signal generator, and the corresponding control signal parameters are determined. The working modes include charging mode, discharging mode, and standby mode. The control signal parameters include but are limited to power size and charge and discharge rate. The current load data, energy storage system status, and grid demand data of the energy storage system are obtained, and the adaptive communication protocol is configured. The control instructions are transmitted to the energy storage device in real time through the communication interface, which includes CAN bus or Ethernet. By calculating the current load demand, the charge state of the energy storage system, and the grid load fluctuation, the power and time period required for charging and discharging are determined, and control strategy instructions are formulated, including control parameters of charging power, discharging power, and duration. By sending the strategy instructions to the energy storage converter control unit, the execution unit controls the charging and discharging process of the energy storage device according to the instructions. The real-time status information of the energy storage converter control unit is obtained, including the current charging and discharging state, converter working mode, and output power, where the converter working mode includes bidirectional conversion mode and constant current mode. Through the energy storage converter control unit, the output of the energy storage converter is adjusted according to the set power range to perform charging and discharging operations.

[0045] For example, in a specific application scenario, the current state of charge of the energy storage system is 70%, the grid load demand is 1200kW, and the maximum charge and discharge power of the energy storage system is 500kW. According to the real-time load data and grid demand, the energy storage system needs to adjust the working mode to meet the grid demand and ensure the stable operation of the energy storage system. If the current load demand is 100kW more than the expected load demand, and the grid load is expected to continue to rise in the next hour, the energy storage system needs to respond quickly to this increase in demand. According to the current state of charge and grid demand data, configure the adapted working mode. Due to the large fluctuation of the grid load, the energy storage system is set to the discharge mode, and the discharge power is determined to be 400kW, and the remaining 100kW is used for the standby load regulation of the energy storage system. The system calculates that the discharge time should be 1 hour to help balance the grid load. At this time, the charge and discharge rate of the energy storage system is set to 400kW, and a control signal is sent to the energy storage converter control unit through a high-speed communication interface such as CAN bus or Ethernet, instructing the energy storage converter to start discharging at this rate. During this process, the working status of the energy storage equipment is continuously monitored to obtain the real-time output power and working mode of the converter. If the working mode of the converter is abnormal or the power output does not reach the predetermined target, the control signal is automatically adjusted to ensure the smooth progress of the charging and discharging process. If the state of charge of the energy storage system drops to 40%, it is determined whether the working mode needs to be adjusted. If the grid load continues to rise, the energy storage system may need to increase the discharge power or switch to the charging mode for reverse compensation. Otherwise, when the grid load returns to normal levels, the energy storage system will be adjusted to standby mode and remain in standby status to cope with future load fluctuations.

[0046] Step S105, the collaborative execution module is used to continuously monitor the voltage and current data and adjust the converter working mode according to the grid load demand, distributed energy generation capacity and the charging and discharging status of the energy storage system, formulate a dynamic power adjustment strategy, and dynamically optimize the power allocation strategy by scheduling distributed energy to work in coordination with the grid.

[0047] Continuously monitor the real-time data of voltage and current during the charging and discharging process, and adjust the converter working mode in real time according to the monitoring information. Obtain the total demand information of the power grid, the real-time power generation capacity of distributed energy, and the charging and discharging status of the energy storage system, determine the load demand information of the power grid and the energy supply capacity of each party of distributed energy, and dynamically adjust the power generation power of distributed energy by dispatching distributed energy and the power grid to work together, supplement the lack of power in the energy storage system or adjust the excess power. Obtain the real-time load information of the power grid, the power generation data of distributed energy, and the charging and discharging status of the energy storage system, adopt the dynamic programming algorithm, and formulate a dynamic power adjustment strategy based on the power distribution optimization target setting. The power distribution optimization target setting includes the load balance of the power grid, the effective use of the energy storage system, and the maximum efficiency output of distributed energy, while avoiding overload or unstable fluctuations. The dynamic adjustment strategy includes increasing the discharge power of the energy storage system when the load fluctuation of the power grid is greater than the preset fluctuation threshold, giving priority to the use of the energy storage system to provide electricity when the power grid voltage is lower than the preset voltage threshold, and giving priority to the distributed energy when the power generation capacity of the distributed energy is higher than the preset capacity threshold, and reducing the load of the energy storage system according to the needs of the power grid.

[0048] For example, in a typical energy management system, the real-time load demand of the power grid is 3000kW, the real-time power generation capacity of solar energy and wind energy is 1000kW, the current state of charge of the energy storage system is 60%, and its charging power limit is 500kW and the discharge power limit is 400kW. Real-time acquisition of power grid load information, including the load demand of the power grid is 3000kW, and the power generation capacity of distributed energy is 1000kW, which means that the remaining load demand is 2000kW. Since the energy storage system is currently in a state of charge of 60%, the power demand of the power grid is judged based on this information, and the system can provide a maximum discharge power of 400kW. Applying a feedback mechanism, by monitoring the real-time data of the grid voltage and current, ensure that the working mode of the converter is stable during the charging and discharging process. If it is detected that the grid load fluctuates greatly and the voltage drops below the set voltage threshold of 210V, the energy storage system is used first to provide power to prevent the grid voltage from further dropping and ensure the stability of the grid. If the grid load fluctuation exceeds the preset fluctuation threshold of 500kW, the dynamic programming algorithm is used to optimize the power allocation and adjust it according to the predetermined power allocation target. Since the grid load demand is 2000kW, which is still high, and the distributed energy generation capacity is 1000kW, the discharge power of the energy storage system is increased according to the dynamic power adjustment strategy, such as adjusting the energy storage system to a discharge power of 350kW to meet the remaining power demand and avoid excessive discharge of the energy storage system. If the solar power generation capacity exceeds 800kW at this time and the wind power output is 200kW, the total power generation of 1000kW has reached the maximum efficiency output of distributed energy. In order to optimize energy use, this part of distributed energy is dispatched first to reduce the load demand of the energy storage system, thereby maintaining the effective utilization rate of the energy storage system and avoiding excessive discharge. The discharge power of the energy storage system is adjusted to 150kW to ensure that it does not exceed its maximum discharge power limit, and by effectively coordinating the output of distributed energy, the grid load is balanced and the grid voltage is kept in a stable range.

[0049] Step S106, a dynamic power adjustment strategy adjustment module is used to dynamically update the hybrid prediction model parameters according to the real-time acquired grid load, energy storage system status and distributed energy data, and adjust the charge and discharge protection threshold and power adjustment strategy through a feedback mechanism.

[0050] The grid load, energy storage system status and distributed energy data are acquired in real time through the communication interface and transmitted to the central monitoring system. The real-time grid load, energy storage system status and distributed energy data are aggregated to the human-computer interaction interface, and the operation status information of the energy storage system, load and distributed energy, the real-time forecast value of demand, the dynamically changing protection threshold and the potential demand risk warning information are displayed on the human-computer interaction interface. The parameters of the hybrid prediction model are updated using real-time data, and the demand in the future preset time period is predicted. The charging demand protection threshold and the discharging demand protection threshold are dynamically adjusted according to the prediction results, the current state of charge value of the energy storage system and the grid voltage data. During the operation of the energy storage system, feedback data is acquired in real time, and all relevant control instructions and execution results are recorded, the error between the demand forecast and the actual demand is identified, and the responsiveness of the protection strategy is determined, and the feedback data is transmitted to the central monitoring system and the hybrid prediction model. The hybrid prediction model is optimized by the error correction algorithm, and the optimized hybrid prediction model is re-run for demand forecasting, and the actual operation effect of the optimized hybrid prediction model is evaluated. According to the optimized demand forecast value, the parameters of the dynamic power adjustment strategy are adjusted. Through real-time monitoring and feedback mechanisms, the actual operating status is compared with the predicted value to determine the real-time effects of the hybrid prediction model and the dynamic power adjustment strategy. The parameters of the hybrid prediction model and the dynamic power adjustment strategy are adjusted using the real-time monitoring data through an automatic feedback system.

[0051] For example, in a smart grid system, the real-time grid load is 3200kW, the current state of charge of the energy storage system is 75%, the upper limit of the charging power of the energy storage device is 500kW, the upper limit of the discharge power is 400kW, and the real-time power generation capacity of distributed energy, including solar energy and wind energy, is 1500kW. These data are transmitted to the central monitoring system through a high-speed communication interface, and the operating status of the energy storage system, load, and distributed energy, as well as the real-time forecast value of the demand and the dynamically changing protection threshold are displayed on the interface. At this time, the demand value in the next hour is predicted to be 3500kW by the hybrid prediction model. By comparing with the real-time data, it is identified that there is an error of 300kW between the demand forecast value and the actual load demand, indicating that the error of the prediction model is large. Based on this information, all relevant control instructions are recorded, and the actual demand data is transmitted to the central monitoring system and the hybrid prediction model through a feedback mechanism to update the model parameters and perform error correction. In this process, the error correction algorithm is used to optimize the hybrid prediction model. The optimized prediction results show that the demand in the next hour is 3400kW, which is more accurate than the initial prediction results. Through the optimized demand forecast value, the charging demand protection threshold and the discharging demand protection threshold are dynamically adjusted, such as dynamically adjusting the charging demand protection threshold from the original 2500kW to 2700kW, and adjusting the discharging demand protection threshold from 2000kW to 2100kW to cope with the upcoming load peak. The power allocation strategy is dynamically adjusted according to the new demand forecast value and the charge state of the energy storage system of 75% and the grid voltage of 220V. In order to balance the grid load and the energy use of the energy storage system, the energy storage system is used to provide 300kW of discharge power, and the stability of the grid is ensured according to the grid voltage and load fluctuations. When the grid load is greater than 3000kW, the output of distributed energy is dispatched first, and when the grid voltage shows a downward trend, such as the voltage is lower than 215V, the power of the energy storage system is used first. Through real-time monitoring, the system compares the actual operating status with the predicted value. If a new prediction error is found, the automatic feedback mechanism is activated. Based on the new error, the parameters of the hybrid prediction model and dynamic power adjustment strategy are automatically adjusted to further optimize the grid load scheduling and the charging and discharging process of the energy storage system, ensuring the reliability and stability of the grid power supply.

[0052] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the concept of the present application. For example, the above features are replaced with the technical features with similar functions disclosed in the present application (but not limited to) to form a technical solution.

Claims

1. An energy storage energy management system with an improved demand protection algorithm, characterized in that: The system comprises: The data acquisition module is used to obtain energy storage system load data, distributed energy operation data, and energy storage system operation data based on load monitoring equipment and smart meters, and to clean, preprocess, standardize, and calibrate the acquired data; The correlation feature analysis module is used to combine historical load and energy supply data, use long-term short-term memory networks to perform long-term trend forecasts, and use support vector machines to analyze the correlation between load fluctuations and energy supply fluctuations to obtain the correlation feature set and time delay information between load and energy supply; The control strategy generation module is used to extract demand forecast features using the chi-square test, combine the ARIMA model and the long-short-term memory network model to perform demand forecasting, and adjust the charge and discharge thresholds and formulate the charge and discharge control strategy based on the grid voltage and the state of charge of the energy storage system; The control strategy execution module is used to configure the charging, discharging, and standby modes through a signal generator, adjust control signal parameters and obtain load data in real time, transmit control instructions through a communication interface, formulate charge and discharge control strategies, and adjust the operating mode of the energy storage converter; The collaborative execution module is used to continuously monitor voltage and current data and adjust the converter operating mode according to the grid load demand, distributed energy generation capacity and the charge and discharge status of the energy storage system. It also formulates dynamic power adjustment strategies and dynamically optimizes the power allocation strategy by coordinating distributed energy with the grid. The dynamic power adjustment strategy adjustment module is used to dynamically update the hybrid prediction model parameters based on real-time grid load, energy storage system status and distributed energy data, and adjust the charge and discharge protection thresholds and power adjustment strategies through a feedback mechanism.

2. The system according to claim 1, wherein: The data acquisition module is used to obtain energy storage system load data, distributed energy operation data, and energy storage system operation data based on load monitoring equipment and smart meters, and to perform data cleaning, preprocessing, standardization, and calibration on the obtained data, including: According to the load monitoring equipment and smart meters, the load data of the energy storage system is obtained, including current, voltage, instantaneous power change and power factor, among which the smart meter records the load type information; by integrating with the distributed energy control system, the distributed energy operation data is obtained, including real-time output power, power change rate and energy quality; through the interface with the energy storage device, the energy storage system operation data is obtained, including voltage, current, charge state, health state and converter state; based on the obtained load data, distributed energy operation data and energy storage system operation data; using data synchronization technology, the load data, distributed energy operation data and energy storage system operation data are unified Integration is carried out in the time dimension; based on the integrated data, data cleaning and preprocessing algorithms are used to filter the data, including setting range thresholds and filtering algorithms to eliminate noise and outliers in the data acquisition process, interpolating the data, and processing null values ​​caused by equipment transmission interruptions; data standardization technology is used to standardize the preprocessed data and convert it to the same dimension; by setting several data sampling points and combining them with known standard data sources, point-by-point calibration is performed, and the weighted average method is used to smooth the errors of the acquired data. The standard data source is a known load characteristic; the calibrated data is transmitted to the central monitoring system through the communication interface.

3. The system according to claim 1, wherein: The correlation feature analysis module is used to combine historical load and energy supply data, use a long-term short-term memory network to perform long-term trend prediction, and use a support vector machine to analyze the correlation between load fluctuations and energy supply fluctuations to obtain a correlation feature set and time delay information between load and energy supply, including: The calibrated data is converted from the time domain to the frequency domain using the Fourier transform algorithm, and the instantaneous frequency changes of the signal are extracted through wavelet transform to identify high-frequency fluctuations of different load or supply changes; a convolutional neural network is used to extract features of short-term fluctuations to obtain the short-term fluctuation trends of load and energy supply, including their high-frequency fluctuations, instantaneous change frequency and amplitude information. The short-term fluctuation characteristics include frequency distribution, amplitude and fluctuation intensity; historical load data, historical energy supply data, and corresponding timestamp data are obtained through the central monitoring system, and a long short-term memory network is used for model training to predict long-term trend changes and seasonal fluctuations of load and energy supply; based on the short-term fluctuation trends and long-term trends of load and energy supply, the support vector machine algorithm is used to determine the correlation between load fluctuations and energy supply fluctuations, and through regression analysis and Pearson correlation coefficient calculation method, a set of correlation features between load and energy supply is obtained, including time delay, time delay correlation and time delay correlation intensity.

4. The system according to claim 1, wherein: The control strategy generation module is used to extract demand forecast features using the chi-square test, combine the ARIMA model and the long short-term memory network model to perform demand forecasting, and adjust the charge and discharge thresholds and formulate the charge and discharge control strategy based on the grid voltage and the energy storage system state of charge, including: Obtain historical load data, load type, energy supply data, weather data and distributed energy generation data, and use chi-square test to extract demand forecasting features related to demand forecasting, including load type, seasonal pattern, load fluctuation, and energy supply change. Seasonal pattern is the correlation feature between load and energy supply of different load types and seasons. Load types include industrial load and residential load. Based on the historical data of load demand forecasting features, a hybrid forecasting model combining ARIMA model and long short-term memory network model is used to forecast demand, including using ARIMA model to identify linear laws and seasonal patterns of demand, combining long short-term memory network model to identify nonlinear fluctuations, and predicting demand within a preset time period in the future. Obtain the current state of charge and power of the energy storage system. The system uses grid voltage data, uses the charge and discharge demand protection threshold adjustment unit to adjust the charge and discharge demand protection threshold, and formulates the charge and discharge control strategy of the energy storage system; if the difference between the grid voltage and the preset voltage upper limit threshold is less than the preset difference threshold, the charge demand protection threshold is adjusted; if the current state of charge is lower than the preset state of charge threshold, the discharge demand protection threshold is lowered; obtains sudden load changes or energy supply fluctuations in the demand forecast results to determine whether there is a risk of extreme fluctuations. If the demand increase rate is higher than the preset increase rate threshold or the energy supply decrease rate is higher than the preset speed threshold, a rapid response control strategy is generated based on real-time data and preset emergency response rules, including emergency adjustment of the charge and discharge power of the energy storage system, activation of backup energy storage resources, and emergency power interaction and coordination with the grid.

5. The system according to claim 4, wherein: The charge and discharge demand protection threshold adjustment unit is used to obtain the current state of charge of the energy storage system and the grid voltage data, adjust the charge and discharge demand protection threshold, and formulate the charge and discharge control strategy of the energy storage system, including: The state of charge of the energy storage system and the voltage of the grid are obtained based on real-time data to determine whether the charge and discharge thresholds need to be adjusted. The state of charge of the energy storage system and the voltage of the grid are obtained based on real-time data, and the grid voltage deviation and state of charge deviation are calculated. Based on the grid voltage deviation and state of charge deviation, the charging demand protection threshold adjustment formula C1 = C-(ΔV×α) is used to determine the dynamically adjusted charging demand protection threshold C1, where C is the original charging threshold, ΔV is the grid voltage deviation, and α is the sensitivity of the voltage deviation to the charging threshold. V max is the upper limit of the grid voltage, V is the current grid voltage, P is the current grid load power, P max is the maximum load power of the power grid, γ is a constant representing the voltage stability requirement of the power grid, obtained by fitting historical data; according to the voltage deviation and state of charge deviation of the power grid, the discharge demand protection threshold adjustment formula D1 = D + (ΔS × β) is used to determine the dynamically adjusted discharge demand protection threshold D1, where D is the original discharge threshold, ΔS is the state of charge deviation, and β is the sensitivity of the state of charge deviation to the discharge threshold. is the preset target state of charge, S is the current state of charge of the energy storage system, and λ is a constant that represents the degree of influence of battery health and grid load on the discharge threshold adjustment. Based on the adjusted charging and discharging thresholds, the charge and discharge control strategy of the energy storage system is formulated.

6. The system according to claim 1, wherein: The control strategy execution module is used to configure the charging, discharging and standby modes through a signal generator, adjust the control signal parameters and obtain load data in real time, transmit control instructions through a communication interface, formulate the charging and discharging control strategy and adjust the energy storage converter operating mode, including: According to the charge and discharge control strategy and the operating requirements of the energy storage system, different operating modes are configured through a signal generator and corresponding control signal parameters are determined. The operating modes include charging mode, discharging mode, and standby mode. The control signal parameters include but are limited to power size and charge and discharge rate. The current load data, energy storage system status, and grid demand data of the energy storage system are obtained, an adaptive communication protocol is configured, and control instructions are transmitted to the energy storage device in real time through a communication interface, which may include a CAN bus or Ethernet. By calculating the current load demand, the charge state of the energy storage system, and the grid load fluctuation, the power and time period required for charging and discharging are determined, and control strategy instructions are formulated, including control parameters such as charging power, discharging power, and duration. By sending the strategy instructions to the energy storage converter control unit, the execution unit controls the charging and discharging process of the energy storage device according to the instructions. Real-time status information of the energy storage converter control unit is obtained, including the current charge and discharge status, converter operating mode, and output power. The converter operating mode includes bidirectional conversion mode and constant current mode. The output of the energy storage converter is adjusted according to the set power range through the energy storage converter control unit to perform charging and discharging operations.

7. The system according to claim 1, wherein: The collaborative execution module is used to continuously monitor voltage and current data and adjust the converter operating mode according to the grid load demand, distributed energy generation capacity, and the charge and discharge status of the energy storage system, formulate a dynamic power adjustment strategy, and dynamically optimize the power allocation strategy by scheduling distributed energy and working in coordination with the grid, including: The system continuously monitors real-time voltage and current data during the charging and discharging process, and adjusts the converter operating mode in real time based on this monitoring information. It obtains information on the total grid demand, the real-time generation capacity of distributed energy resources, and the charge and discharge status of the energy storage system, determines the grid load demand information and the energy supply capacity of each distributed energy source, and dynamically adjusts the generation power of distributed energy resources by coordinating the distributed energy resources with the grid to supplement the energy storage system's energy shortage or regulate excess energy. It obtains real-time grid load information, distributed energy generation data, and the charge and discharge status of the energy storage system, and uses a dynamic programming algorithm to dynamically formulate a dynamic power adjustment strategy based on the power allocation optimization objective setting. The power allocation optimization objectives include grid load balancing, efficient utilization of the energy storage system, and maximum efficiency output of distributed energy resources, while avoiding overload or unstable fluctuations. The dynamic adjustment strategy includes increasing the discharge power of the energy storage system when the grid load fluctuation exceeds a preset fluctuation threshold, prioritizing the use of the energy storage system when the grid voltage is below a preset voltage threshold, and prioritizing the dispatch of distributed energy resources when their generation capacity exceeds a preset capacity threshold, while reducing the load on the energy storage system based on grid demand.

8. The system according to claim 1, wherein: The dynamic power adjustment strategy adjustment module is used to dynamically update the hybrid prediction model parameters based on the real-time grid load, energy storage system status and distributed energy data, and adjust the charge and discharge protection thresholds and power adjustment strategy through a feedback mechanism, including: The grid load, energy storage system status and distributed energy data are acquired in real time through the communication interface and transmitted to the central monitoring system; the real-time acquired grid load, energy storage system status and distributed energy data are aggregated into the human-computer interaction interface, and the human-computer interaction interface is used to display the operating status information of the energy storage system, load and distributed energy, the real-time forecast value of demand, the dynamically changing protection threshold and the potential demand risk warning information; the real-time data is used to update the parameters of the hybrid prediction model, and the demand within the preset time period in the future is predicted. According to the prediction results, the current state of charge value of the energy storage system and the grid voltage data, the charging demand protection threshold and the discharging demand protection threshold are dynamically adjusted; during the operation of the energy storage system, feedback data is acquired in real time and recorded All relevant control instructions and execution results are used to identify the error between demand forecast and actual demand, determine the responsiveness of the protection strategy, and transmit feedback data to the central monitoring system and the hybrid prediction model; the hybrid prediction model is optimized using an error correction algorithm, the optimized hybrid prediction model is re-run for demand forecast, and the actual operating effect of the optimized hybrid prediction model is evaluated; the parameters of the dynamic power adjustment strategy are adjusted according to the optimized demand forecast value; the actual operating status is compared with the forecast value through a real-time monitoring and feedback mechanism to determine the real-time effect of the hybrid prediction model and the dynamic power adjustment strategy, and the parameters of the hybrid prediction model and the dynamic power adjustment strategy are adjusted using the real-time monitoring data through an automatic feedback system.

Images