Photovoltaic- energy storage power supply control system based on shared energy storage power station

Abstract

This invention discloses a photovoltaic-energy storage power supply control system based on a shared energy storage power station, belonging to the field of energy storage control technology. The invention acquires raw datasets through a data acquisition module and performs preprocessing operations on the acquired raw datasets to obtain a standard dataset. A power output prediction module inputs the standard datasets into a pre-constructed photovoltaic power output prediction model to obtain a power output prediction sequence, and then obtains the net load demand curves for each user based on the power output prediction sequence. A status monitoring module processes the standard datasets to obtain real-time energy storage status information. A capacity allocation module generates an energy storage capacity allocation scheme based on the net load demand curves and the real-time energy storage status information. Finally, a grid connection control module executes power transmission control according to the energy storage capacity allocation scheme, realizing refined collaborative control of dynamic energy storage capacity allocation and photovoltaic-energy storage combined power supply in a shared energy storage power station under multiple user and time scale conditions.

Description

Technical Field

[0001] This invention relates to the field of energy storage control technology, specifically a photovoltaic-energy storage power supply control system based on a shared energy storage power station. Background Technology

[0002] Against the backdrop of large-scale development of photovoltaic power generation, the inherent intermittency and volatility of photovoltaic output lead to a significant time mismatch between it and the electricity load on the user side. This not only reduces the local absorption capacity of photovoltaic power generation, but also has an adverse impact on the voltage stability and power supply reliability of the distribution network.

[0003] However, existing shared energy storage power station technical solutions mostly focus on the static allocation of energy storage capacity and simple time-sharing scheduling based on reservations. They lack integrated consideration of multiple factors such as the uncertainty of photovoltaic output, the dynamic of differentiated loads of multiple users, the grid time-sharing electricity pricing mechanism, and the operating status of energy storage batteries. It is difficult to achieve dynamic optimization of energy storage capacity under multiple users and multiple time scales, and it is also impossible to perform fine control of power distribution in the process of photovoltaic-energy storage joint power supply.

[0004] How to construct a dynamic optimization configuration model for energy storage capacity under multiple time scales, and realize the fine scheduling and coordinated control of charging and discharging power in the process of photovoltaic-energy storage joint power supply, so as to maximize the local photovoltaic absorption rate, reduce the electricity cost of users and improve the stability of distribution network operation, is a problem we need to solve. To this end, we now provide a photovoltaic-energy storage power supply control system based on shared energy storage power station. Summary of the Invention

[0005] The purpose of this invention is to provide a photovoltaic-energy storage power supply control system based on a shared energy storage power station.

[0006] The objective of this invention can be achieved through the following technical solution: a photovoltaic-energy storage power supply control system based on a shared energy storage power station, comprising the following:

[0007] The data acquisition module is used to acquire raw datasets and perform preprocessing operations on the acquired raw datasets to obtain standard datasets;

[0008] The output prediction module is used to input the standard dataset into the pre-built photovoltaic output prediction model to obtain the output prediction sequence, and then obtain the net load demand curve based on the output prediction sequence.

[0009] The status monitoring module is used to process standard datasets to obtain real-time energy storage status information;

[0010] The capacity allocation module is used to generate an energy storage capacity allocation scheme based on the net load demand curve and real-time energy storage status information.

[0011] The grid-connected control module is used to perform power transmission control according to the energy storage capacity allocation scheme.

[0012] Furthermore, the process by which the data acquisition module acquires the raw dataset includes:

[0013] The data acquisition module establishes a data communication link with the photovoltaic inverters of each photovoltaic power generation user and acquires the power generation operation data corresponding to each photovoltaic power generation system at a first acquisition frequency.

[0014] Meteorological environmental data of the area is acquired at a second acquisition frequency by meteorological monitoring devices deployed in the areas where each photovoltaic power generation user is located.

[0015] By establishing a data communication link with the user-side energy management system of each photovoltaic power generation user, the electricity load data of each photovoltaic power generation user is acquired at a third acquisition frequency. The electricity load data includes a load power time series recorded at a preset time interval.

[0016] The voltage and current signals of the grid connection point are acquired at the fourth acquisition frequency by using voltage and current transformers deployed at the grid connection point of the shared energy storage power station.

[0017] Meanwhile, the data acquisition module obtains the energy storage battery operating status data of each energy storage battery unit through the battery management system of each energy storage battery unit in each shared energy storage power station. The energy storage battery operating status data includes state of charge, rated capacity, health status, cumulative number of charge and discharge cycles, battery cell temperature, current maximum rechargeable power and current maximum dischargeable power.

[0018] The original dataset is obtained by integrating power generation operation data, meteorological environment data, power load data, voltage and current signals at the grid connection point, and energy storage battery operation status data of each energy storage battery unit.

[0019] Furthermore, the process by which the data acquisition module performs preprocessing operations on the acquired raw dataset to obtain a standard dataset includes:

[0020] Data feature items are extracted from the original dataset, including power generation data feature items, meteorological data feature items, electricity load data feature items, grid connection point voltage and current signal feature items, and energy storage battery operating status feature items.

[0021] Set reasonable value ranges for each data feature item. If the value of a data feature item exceeds the corresponding reasonable value range, the data point is marked as an outlier and removed.

[0022] For gaps caused by outlier removal or missing data, linear interpolation is used to fill them with adjacent valid data points. If the gap is located at the beginning or end of the sequence, the nearest neighbor filling method is used.

[0023] After outlier removal and missing value imputation, the data features are normalized by min-max and max-min, and then integrated into a standard dataset.

[0024] Furthermore, the process of inputting the standard dataset into the pre-built photovoltaic power output prediction model to obtain the power output prediction sequence includes:

[0025] Collect sample data, which is a historical standard dataset, and divide the obtained sample data into a training sample dataset and a validation sample dataset according to a preset ratio;

[0026] A deep learning-based time series prediction model is constructed, and the time series prediction model is iteratively trained using the training sample dataset. The time series prediction model is then validated using the validation sample dataset after each iteration.

[0027] Record the average loss value on the training sample dataset and the average loss value on the validation sample dataset for each iteration. When the average loss value on the validation sample dataset does not decrease in multiple consecutive iterations and the average loss value on the training sample dataset continues to decrease, trigger the early stopping mechanism, end the model training, and save the model parameters of the current iteration as the optimal model parameters to complete the training of the time series prediction model and obtain the photovoltaic power output prediction model.

[0028] The real-time acquired standard dataset is input into the photovoltaic power output prediction model, which outputs the power output prediction sequence for each photovoltaic power generation user within a preset scheduling period.

[0029] Furthermore, the process by which the output prediction module obtains the net load demand curves for each photovoltaic power generation user includes:

[0030] Extract electricity load data for each photovoltaic power generation user from the standard dataset;

[0031] Based on the output prediction sequence of each photovoltaic power generation user, the output prediction sequence is aligned with the load power time sequence of the corresponding photovoltaic power generation user according to the unified time label of the standard dataset. At each time point after alignment, the net load power value at that time point is obtained by subtracting the output prediction power value at the same time point from the historical power load power value at that time point.

[0032] The net load power values ​​of each photovoltaic power generation user at all time points within the preset scheduling cycle are arranged in chronological order to generate the net load demand curve for each photovoltaic power generation user.

[0033] Furthermore, the process by which the status monitoring module processes the standard dataset to obtain real-time energy storage status information includes:

[0034] Extract the operating status data of each energy storage battery unit from the standard dataset, and add up the rated capacity of each energy storage battery unit to obtain the total energy storage capacity of the shared energy storage power station.

[0035] The total rechargeable power of the shared energy storage power station is obtained by adding up the current maximum rechargeable power of each energy storage battery unit.

[0036] The total dischargeable power of the shared energy storage power station is obtained by adding up the current maximum dischargeable power of each energy storage battery unit.

[0037] The state of charge, state of health, cell temperature, and cumulative charge-discharge cycle count of each energy storage battery cell are used as the differentiated state information of each energy storage battery cell.

[0038] The total energy storage capacity, total rechargeable power, total dischargeable power, and differentiated status information of each energy storage battery unit of the shared energy storage power station are summarized and integrated to obtain real-time energy storage status information.

[0039] Furthermore, the process by which the capacity allocation module generates an energy storage capacity allocation scheme based on the net load demand curve and real-time energy storage status information includes:

[0040] Using a preset scheduling cycle as the allocation time unit, at the beginning of each scheduling cycle, the system receives the output prediction sequence, net load demand curve, and real-time energy storage status information.

[0041] For each photovoltaic power generation user at each point in time within the aforementioned scheduling cycle, the determination is made based on the net load power value at that point in time:

[0042] If the net load power value is positive, it means that the user has a power supply gap at the corresponding time point, which needs to be supplemented by the discharge of the shared energy storage station.

[0043] If the net load power value is negative, it means that the user has a power generation surplus at the corresponding time point, which can be used to charge the shared energy storage station.

[0044] The total net load demand within the scheduling cycle is obtained based on the net load power value. Each photovoltaic power generation user is sorted from largest to smallest according to the value of its total net load demand to determine the priority of each user's net load demand. The larger the total net load demand value, the higher the priority. Energy storage discharge power is allocated to each photovoltaic power generation user one by one according to the order of net load demand priority from high to low.

[0045] The allocated energy storage discharge power value for photovoltaic power generation users is obtained based on the total dischargeable power and net load power value. For the time point when the net load power value is negative, that is, the time point when there is a charging demand, the absolute value of the net load power value at that time point is the power value that the user can charge the energy storage at that time point, which is recorded as the power generation surplus of the user. According to the power generation surplus of each user at that time point, the energy storage charging power is allocated to each user.

[0046] The allocated energy storage charging power value for photovoltaic power generation users is obtained based on the total rechargeable power and net load power value.

[0047] After completing the discharge power allocation and charging power allocation at all time points, the allocated energy storage discharge power value and allocated energy storage charging power value of each photovoltaic power generation user at each time point in the current scheduling cycle are arranged in chronological order to form the discharge power allocation time sequence and charging power allocation time sequence of each user.

[0048] By summing up the discharge power allocation time series and charging power allocation time series of each user, the energy storage capacity allocation scheme of each photovoltaic power generation user in the current scheduling cycle can be obtained.

[0049] Furthermore, the process by which the grid-connected control module performs power transmission control according to the energy storage capacity allocation scheme includes:

[0050] Receive the energy storage capacity allocation plan and extract the allocated energy storage discharge power value and allocated energy storage charging power value of each photovoltaic power generation user at each time point;

[0051] When energy storage discharge is required, a discharge control command is sent to the corresponding photovoltaic power generation user side to control the shared energy storage power station to deliver electrical energy to the user according to the allocated energy storage discharge power value.

[0052] When energy storage charging is required, a charging control command is sent to the corresponding photovoltaic power generation user to control the user to transmit power to the shared energy storage station according to the allocated energy storage charging power value.

[0053] Based on the voltage and current signals at the grid connection point of the shared energy storage power station, the real-time active power and real-time reactive power at the grid connection point are obtained.

[0054] The obtained real-time active power and real-time reactive power are compared with the preset grid-connected power limit to obtain the grid-connected active power deviation and grid-connected reactive power deviation.

[0055] When the deviation of grid-connected active power or grid-connected reactive power exceeds the preset safety threshold range, the grid-connected control module adjusts the discharge power and charging power of the shared energy storage station to restore the grid-connected interactive power to the safety threshold range.

[0056] Compared with the prior art, the beneficial effects of the present invention are:

[0057] This invention integrates multi-source heterogeneous data into a standard dataset through a unified data acquisition and preprocessing mechanism, eliminating the problems of inconsistent formats and timestamps between different data sources, and providing a high-quality data foundation with a unified timestamp alignment relationship for subsequent functional modules;

[0058] The output prediction sequence is obtained based on the photovoltaic output prediction model, and then the net load demand curve of each user is obtained based on the output prediction sequence. This realizes the feedforward compensation of photovoltaic output uncertainty, improves the accuracy and adaptability of power supply plan, and obtains real-time energy storage status information by processing the standard dataset through the status monitoring module. This realizes the real-time perception of relevant data of shared energy storage power station and provides accurate state boundary conditions for the dynamic allocation of energy storage capacity.

[0059] Based on the priority of net load demand and constrained by the total available energy storage capacity of the shared energy storage power station, the system realizes the discharge power allocation value and charging power allocation value of each user at each time point through clear allocation rules. This avoids the computational efficiency problem of traditional optimization models that rely on complex iterative solutions, and realizes the fast, fair and executable dynamic allocation of energy storage capacity under multiple users and multiple time scales.

[0060] Finally, the grid-connected control module executes power transmission control according to the energy storage capacity allocation scheme, realizing the dynamic allocation of energy storage capacity and the refined collaborative control of photovoltaic-energy storage power supply in multi-user and multi-time scale shared energy storage power stations. This effectively improves the local photovoltaic absorption rate, reduces users' electricity costs, and improves the stability of the distribution network operation. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0062] Figure 1 This is a schematic diagram of the present invention. Detailed Implementation

[0063] like Figure 1 As shown, a photovoltaic-energy storage power supply control system based on a shared energy storage power station includes a data acquisition module, an output prediction module, a status monitoring module, a capacity allocation module, and a grid connection control module.

[0064] The data acquisition module is used to acquire raw datasets and perform preprocessing operations on the acquired raw datasets to obtain standard datasets;

[0065] The output prediction module is used to input the standard dataset into the pre-built photovoltaic output prediction model to obtain the output prediction sequence, and then obtain the net load demand curve based on the output prediction sequence.

[0066] The status monitoring module is used to process the standard dataset to obtain real-time energy storage status information;

[0067] The capacity allocation module is used to generate an energy storage capacity allocation scheme based on the net load demand curve and real-time energy storage status information.

[0068] The grid-connected control module is used to perform power transmission control according to the energy storage capacity allocation scheme.

[0069] It should be further explained that, in the specific implementation process, the process of the data acquisition module acquiring the raw dataset includes:

[0070] The data acquisition module establishes a data communication link with the photovoltaic inverters of each photovoltaic power generation user and acquires the power generation operation data corresponding to each photovoltaic power generation system at a first acquisition frequency. The power generation operation data includes real-time output power, DC side voltage, DC side current, AC side voltage, AC side current and inverter operating status parameters.

[0071] Meteorological environmental data of the area is acquired by meteorological monitoring devices deployed in the areas where photovoltaic power generation users are located at a second acquisition frequency. The meteorological environmental data includes total horizontal irradiance, ambient temperature, relative humidity, wind speed and cloud cover.

[0072] By establishing a data communication link with the user-side energy management system of each photovoltaic power generation user, the electricity load data of each photovoltaic power generation user is obtained at the third acquisition frequency.

[0073] The power load data includes a load power time series recorded at preset time intervals, and the third sampling frequency has the same period as the second sampling frequency.

[0074] The voltage and current signals of the grid connection point are acquired at the fourth acquisition frequency by using voltage and current transformers deployed at the grid connection point of the shared energy storage power station.

[0075] Meanwhile, the data acquisition module obtains the energy storage battery operating status data of each energy storage battery unit through the battery management system of each energy storage battery unit in each shared energy storage power station. The energy storage battery operating status data includes state of charge, rated capacity, health status, cumulative number of charge and discharge cycles, battery cell temperature, current maximum rechargeable power and current maximum dischargeable power.

[0076] The original dataset is obtained by integrating power generation operation data, meteorological environment data, power load data, voltage and current signals at the grid connection point, and energy storage battery operation status data of each energy storage battery unit.

[0077] It should be further explained that, in the specific implementation process, the process by which the data acquisition module performs preprocessing operations on the acquired raw dataset to obtain a standard dataset includes:

[0078] Data feature items are extracted from the original dataset, including power generation data feature items, meteorological data feature items, electricity load data feature items, grid connection point voltage and current signal feature items, and energy storage battery operating status feature items.

[0079] Set reasonable value ranges for each data feature item. If the value of a data feature item exceeds the corresponding reasonable value range, the data point is marked as an outlier and removed.

[0080] For gaps caused by outlier removal or missing data, linear interpolation is used to fill them with adjacent valid data points. If the gap is located at the beginning or end of the sequence, the nearest neighbor filling method is used.

[0081] After outlier removal and missing value imputation, each data feature is uniformly mapped to the [0,1] interval using the min-max normalization method. The normalized value for each data feature is denoted as... ,Right now:

[0082] ;

[0083] in, and These represent the minimum and maximum values ​​of the corresponding data feature items throughout the entire data collection period;

[0084] The normalized power generation data features, meteorological data features, electricity load data features, grid connection point voltage and current signal features, and energy storage battery operating status features are integrated into a standard dataset.

[0085] The standard dataset uses a unified time label for data alignment to ensure a one-to-one timestamp relationship between power generation data features, meteorological data features, and energy storage battery operating status features.

[0086] It should be further explained that, in the specific implementation process, the process of inputting the standard dataset into the pre-built photovoltaic power output prediction model to obtain the power output prediction sequence includes:

[0087] Collect sample data, which is a historical standard dataset;

[0088] The historical standard dataset includes historical power generation data features, historical meteorological data features, historical electricity load data features, historical grid connection point voltage and current signal features, and historical energy storage battery operating status features.

[0089] The obtained sample data is divided into training sample dataset and validation sample dataset according to a preset ratio;

[0090] A deep learning-based time series prediction model is constructed, and the time series prediction model is iteratively trained using the training sample dataset. The time series prediction model is then validated using the validation sample dataset after each iteration.

[0091] Record the average loss value on the training sample dataset and the average loss value on the validation sample dataset for each iteration. When the average loss value on the validation sample dataset does not decrease in multiple consecutive iterations and the average loss value on the training sample dataset continues to decrease, trigger the early stopping mechanism, end the model training, and save the model parameters of the current iteration as the optimal model parameters to complete the training of the time series prediction model and obtain the photovoltaic power output prediction model.

[0092] The real-time acquired standard dataset is input into the photovoltaic power output prediction model, which outputs the power output prediction sequence for each photovoltaic power generation user within a preset scheduling period.

[0093] It should be further explained that, in the specific implementation process, the process by which the output prediction module obtains the net load demand curve of each photovoltaic power generation user includes:

[0094] Extract electricity load data for each photovoltaic power generation user from the standard dataset;

[0095] Based on the power output prediction sequences of each photovoltaic (PV) power generation user, the power output prediction sequences are aligned with the corresponding PV power load time series according to the unified time label of the standard dataset. At each aligned time point, the net load power value at that time point is obtained by subtracting the corresponding power output prediction power value from the historical power load value at that time point, denoted as . ,Right now:

[0096] ;

[0097] in, This represents the net load power value. This represents the historical power load value. This indicates the predicted power output value;

[0098] The net load power values ​​of each photovoltaic power generation user at all time points within the preset scheduling cycle are arranged in chronological order to generate the net load demand curve for each photovoltaic power generation user.

[0099] It should be further explained that, in the specific implementation process, the process by which the status monitoring module processes the standard dataset to obtain real-time energy storage status information includes:

[0100] Extract the energy storage battery operating status data of each energy storage battery unit from the standard dataset;

[0101] The energy storage battery operation status data of each energy storage battery unit are summarized and integrated to obtain real-time energy storage status information. The real-time energy storage status information includes the total energy storage capacity of the shared energy storage power station, the total rechargeable power, the total dischargeable power, and the differentiated status information of each energy storage battery unit.

[0102] Specifically:

[0103] The total energy storage capacity of the shared energy storage power station is obtained by adding the rated capacities of each energy storage battery unit, denoted as . ,Right now:

[0104] ;

[0105] in, For the first The rated capacity of each energy storage battery cell The total number of energy storage battery units within a shared energy storage power station;

[0106] The total rechargeable power of the shared energy storage power station is obtained by summing the current maximum rechargeable power of each energy storage battery unit, denoted as . ,Right now:

[0107] ;

[0108] in, This represents the current maximum rechargeable power of the r-th energy storage battery cell;

[0109] The total dischargeable power of the shared energy storage power station is obtained by summing the current maximum dischargeable power of each energy storage battery unit. ,Right now:

[0110] ;

[0111] in, This represents the current maximum discharge power of the r-th energy storage battery cell;

[0112] The state of charge, state of health, cell temperature, and cumulative charge-discharge cycle count of each energy storage battery cell are used as the differentiated state information of each energy storage battery cell.

[0113] The total energy storage capacity, total rechargeable power, total dischargeable power, and differentiated status information of each energy storage battery unit of the shared energy storage power station are summarized and integrated to obtain real-time energy storage status information.

[0114] It should be further explained that, in the specific implementation process, the process by which the capacity allocation module generates an energy storage capacity allocation scheme based on the net load demand curve and real-time energy storage status information includes:

[0115] Using a preset scheduling cycle as the allocation time unit, at the beginning of each scheduling cycle, the system receives the output prediction sequence, net load demand curve, and real-time energy storage status information.

[0116] For each photovoltaic power generation user at each point in time within the aforementioned scheduling cycle, the determination is made based on the net load power value at that point in time:

[0117] If the net load power value is positive, it means that the user has a power supply gap at the corresponding time point, which needs to be supplemented by the discharge of the shared energy storage station.

[0118] If the net load power value is negative, it means that the user has a power generation surplus at the corresponding time point, which can be used to charge the shared energy storage station.

[0119] Calculate the total net load demand of each photovoltaic power generation user during the dispatch period, denoted as . ,in For photovoltaic power generation users' numbers, The total number of photovoltaic power generation users, i.e.:

[0120] ;

[0121] in, For the j-th photovoltaic power generation user Net load power value at any given time. The total number of time points within the scheduling period. The time interval between adjacent time points;

[0122] The photovoltaic power generation users are sorted from largest to smallest according to their total net load demand to determine the priority of each user's net load demand. The larger the total net load demand value, the higher the priority.

[0123] According to the priority of net load demand from high to low, the energy storage discharge power is allocated to each photovoltaic power generation user one by one.

[0124] The first A photovoltaic power generation user The allocated energy storage discharge power value at each moment is denoted as ,Right now:

[0125] ;

[0126] in, This refers to the remaining available discharge capacity of the shared energy storage power station, which is the remaining value of the total dischargeable power after the allocated discharge power. This refers to the set of photovoltaic power generation users whose allocation has not yet been completed. For the first individual users at any time Net load power value;

[0127] For the time point when the net load power value is negative, that is, the time point when there is a charging demand, the absolute value of the net load power value at that time point is the power value that the user can charge the energy storage at that time point, which is recorded as the user's power generation surplus. According to the power generation surplus of each user at that time point, the energy storage charging power is allocated to each user.

[0128] Let the allocated energy storage charging power value of the j-th photovoltaic power generation user at time t be denoted as . ,Right now:

[0129] ;

[0130] in, This refers to the currently available charging capacity of a shared energy storage power station, i.e., the remaining value of the total rechargeable power after the allocated charging power. This refers to the collection of photovoltaic power generation users who currently have surplus power generation capacity.

[0131] After completing the discharge power allocation and charging power allocation at all time points, the allocated energy storage discharge power value and allocated energy storage charging power value of each photovoltaic power generation user at each time point in the current scheduling cycle are arranged in chronological order to form the discharge power allocation time sequence and charging power allocation time sequence of each user.

[0132] By summing up the discharge power allocation time series and charging power allocation time series of each user, the energy storage capacity allocation scheme of each photovoltaic power generation user in the current scheduling cycle can be obtained.

[0133] It should be further explained that, in the specific implementation process, the process by which the grid-connected control module performs power transmission control according to the energy storage capacity allocation scheme includes:

[0134] Receive the energy storage capacity allocation plan and extract the allocated energy storage discharge power value and allocated energy storage charging power value of each photovoltaic power generation user at each time point;

[0135] When energy storage discharge is required, a discharge control command is sent to the corresponding photovoltaic power generation user side to control the shared energy storage power station to deliver electrical energy to the user according to the allocated energy storage discharge power value.

[0136] When energy storage charging is required, a charging control command is sent to the corresponding photovoltaic power generation user to control the user to transmit power to the shared energy storage station according to the allocated energy storage charging power value.

[0137] Based on the voltage and current signals at the grid connection point of the shared energy storage power station, the real-time active power and real-time reactive power at the grid connection point are calculated and denoted as follows: , ,Right now:

[0138] ;

[0139] ;

[0140] in, This represents the instantaneous real-time voltage value at the grid connection point. This represents the instantaneous value of the real-time current at the grid connection point. It is the power frequency cycle;

[0141] The calculated real-time active power and real-time reactive power are compared with the preset grid-connected power limits to obtain the grid-connected active power deviation and grid-connected reactive power deviation, where the grid-connected active power deviation is denoted as... The grid-connected reactive power deviation is recorded as ,Right now:

[0142] ;

[0143] ;

[0144] in, The preset grid-connected active power limit, The preset grid-connected reactive power limit;

[0145] When the deviation of grid-connected active power or grid-connected reactive power exceeds the preset safety threshold range, the grid-connected control module adjusts the discharge power and charging power of the shared energy storage station to restore the grid-connected interactive power to the safety threshold range.

[0146] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications or equivalent substitutions made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A photovoltaic-energy storage power supply control system based on a shared energy storage power station, characterized in that, include: The data acquisition module is used to acquire raw datasets and perform preprocessing operations on the acquired raw datasets to obtain standard datasets; The output prediction module is used to input the standard dataset into the pre-built photovoltaic output prediction model to obtain the output prediction sequence, and then obtain the net load demand curve based on the output prediction sequence. The status monitoring module is used to process standard datasets to obtain real-time energy storage status information; The capacity allocation module is used to generate an energy storage capacity allocation scheme based on the net load demand curve and real-time energy storage status information. The grid-connected control module is used to perform power transmission control according to the energy storage capacity allocation scheme.

2. The photovoltaic-energy storage power supply control system based on a shared energy storage power station according to claim 1, characterized in that, The process by which the data acquisition module acquires the raw dataset includes: The data acquisition module establishes a data communication link with the photovoltaic inverters of each photovoltaic power generation user and acquires the power generation operation data corresponding to each photovoltaic power generation system at a first acquisition frequency. Meteorological environmental data of the area is acquired at a second acquisition frequency by meteorological monitoring devices deployed in the areas where each photovoltaic power generation user is located. By establishing a data communication link with the user-side energy management system of each photovoltaic power generation user, the electricity load data of each photovoltaic power generation user is acquired at a third acquisition frequency. The electricity load data includes a load power time series recorded at a preset time interval. The voltage and current signals of the grid connection point are acquired at the fourth acquisition frequency by using voltage and current transformers deployed at the grid connection point of the shared energy storage power station. Meanwhile, the data acquisition module obtains the energy storage battery operating status data of each energy storage battery unit through the battery management system of each energy storage battery unit in each shared energy storage power station. The energy storage battery operating status data includes state of charge, rated capacity, health status, cumulative number of charge and discharge cycles, battery cell temperature, current maximum rechargeable power and current maximum dischargeable power. The original dataset is obtained by integrating power generation operation data, meteorological environment data, power load data, voltage and current signals at the grid connection point, and energy storage battery operation status data of each energy storage battery unit.

3. A photovoltaic-energy storage power supply control system based on a shared energy storage power station according to claim 2, characterized in that, The process by which the data acquisition module performs preprocessing operations on the acquired raw dataset to obtain a standard dataset includes: Data feature items are extracted from the original dataset, including power generation data feature items, meteorological data feature items, electricity load data feature items, grid connection point voltage and current signal feature items, and energy storage battery operating status feature items. Set reasonable value ranges for each data feature item. If the value of a data feature item exceeds the corresponding reasonable value range, the data point is marked as an outlier and removed. For gaps caused by outlier removal or missing data, linear interpolation is used to fill them with adjacent valid data points. If the gap is located at the beginning or end of the sequence, the nearest neighbor filling method is used. After outlier removal and missing value imputation, the data features are normalized by min-max and max-min, and then integrated into a standard dataset.

4. A photovoltaic-energy storage power supply control system based on a shared energy storage power station according to claim 3, characterized in that, The process of inputting a standard dataset into a pre-built photovoltaic power output prediction model to obtain a power output prediction sequence includes: Collect sample data, which is a historical standard dataset, and divide the obtained sample data into a training sample dataset and a validation sample dataset according to a preset ratio; A deep learning-based time series prediction model is constructed, and the time series prediction model is iteratively trained using the training sample dataset. The time series prediction model is then validated using the validation sample dataset after each iteration. Record the average loss value on the training sample dataset and the average loss value on the validation sample dataset for each iteration. When the average loss value on the validation sample dataset does not decrease in multiple consecutive iterations and the average loss value on the training sample dataset continues to decrease, trigger the early stopping mechanism, end the model training, and save the model parameters of the current iteration as the optimal model parameters to complete the training of the time series prediction model and obtain the photovoltaic power output prediction model. The real-time acquired standard dataset is input into the photovoltaic power output prediction model, which outputs the power output prediction sequence for each photovoltaic power generation user within a preset scheduling period.

5. A photovoltaic-energy storage power supply control system based on a shared energy storage power station according to claim 4, characterized in that, The process by which the power output prediction module obtains the net load demand curves for each photovoltaic power generation user includes: Extract electricity load data for each photovoltaic power generation user from the standard dataset; Based on the output prediction sequence of each photovoltaic power generation user, the output prediction sequence is aligned with the load power time sequence of the corresponding photovoltaic power generation user according to the unified time label of the standard dataset. At each time point after alignment, the net load power value at that time point is obtained by subtracting the output prediction power value at the same time point from the historical power load power value at that time point. The net load power values ​​of each photovoltaic power generation user at all time points within the preset scheduling cycle are arranged in chronological order to generate the net load demand curve for each photovoltaic power generation user.

6. A photovoltaic-energy storage power supply control system based on a shared energy storage power station according to claim 5, characterized in that, The process by which the status monitoring module processes the standard dataset to obtain real-time energy storage status information includes: Extract the operating status data of each energy storage battery unit from the standard dataset, and add up the rated capacity of each energy storage battery unit to obtain the total energy storage capacity of the shared energy storage power station. The total rechargeable power of the shared energy storage power station is obtained by adding up the current maximum rechargeable power of each energy storage battery unit. The total dischargeable power of the shared energy storage power station is obtained by adding up the current maximum dischargeable power of each energy storage battery unit. The state of charge, state of health, cell temperature, and cumulative charge-discharge cycle count of each energy storage battery cell are used as the differentiated state information of each energy storage battery cell. The total energy storage capacity, total rechargeable power, total dischargeable power, and differentiated status information of each energy storage battery unit of the shared energy storage power station are summarized and integrated to obtain real-time energy storage status information.

7. A photovoltaic-energy storage power supply control system based on a shared energy storage power station according to claim 6, characterized in that, The process by which the capacity allocation module generates an energy storage capacity allocation scheme based on the net load demand curve and real-time energy storage status information includes: Using a preset scheduling cycle as the allocation time unit, at the beginning of each scheduling cycle, the system receives the output prediction sequence, net load demand curve, and real-time energy storage status information. For each photovoltaic power generation user at each point in time within the aforementioned scheduling cycle, the determination is made based on the net load power value at that point in time: If the net load power value is positive, it means that the user has a power supply gap at the corresponding time point, which needs to be supplemented by the discharge of the shared energy storage station. If the net load power value is negative, it means that the user has a power generation surplus at the corresponding time point, which can be used to charge the shared energy storage station. The total net load demand within the scheduling cycle is obtained based on the net load power value. Each photovoltaic power generation user is sorted from largest to smallest according to the value of its total net load demand to determine the priority of each user's net load demand. The larger the total net load demand value, the higher the priority. Energy storage discharge power is allocated to each photovoltaic power generation user one by one according to the order of net load demand priority from high to low. The allocated energy storage discharge power value for photovoltaic power generation users is obtained based on the total dischargeable power and net load power value. For the time point when the net load power value is negative, that is, the time point when there is a charging demand, the absolute value of the net load power value at that time point is the power value that the user can use to charge the energy storage at that time point, which is recorded as the power generation surplus of the user. According to the power generation surplus of each user at that time point, the energy storage charging power is allocated to each user. The allocated energy storage charging power value for photovoltaic power generation users is obtained based on the total rechargeable power and net load power value. After completing the discharge power allocation and charging power allocation at all time points, the allocated energy storage discharge power value and allocated energy storage charging power value of each photovoltaic power generation user at each time point in the current scheduling cycle are arranged in chronological order to form the discharge power allocation time sequence and charging power allocation time sequence of each user. By summing up the discharge power allocation time series and charging power allocation time series of each user, the energy storage capacity allocation scheme of each photovoltaic power generation user in the current scheduling cycle can be obtained.

8. A photovoltaic-energy storage power supply control system based on a shared energy storage power station according to claim 7, characterized in that, The process by which the grid-connected control module performs power transmission control according to the energy storage capacity allocation scheme includes: Receive the energy storage capacity allocation plan and extract the allocated energy storage discharge power value and allocated energy storage charging power value of each photovoltaic power generation user at each time point; When energy storage discharge is required, a discharge control command is sent to the corresponding photovoltaic power generation user side to control the shared energy storage power station to deliver electrical energy to the user according to the allocated energy storage discharge power value. When energy storage charging is required, a charging control command is sent to the corresponding photovoltaic power generation user to control the user to transmit power to the shared energy storage station according to the allocated energy storage charging power value. Based on the voltage and current signals at the grid connection point of the shared energy storage power station, the real-time active power and real-time reactive power at the grid connection point are obtained. The obtained real-time active power and real-time reactive power are compared with the preset grid-connected power limit to obtain the grid-connected active power deviation and grid-connected reactive power deviation. When the deviation of grid-connected active power or grid-connected reactive power exceeds the preset safety threshold range, the grid-connected control module adjusts the discharge power and charging power of the shared energy storage station to restore the grid-connected interactive power to the safety threshold range.

Images