Intelligent charging and discharging control method and system for energy storage equipment

Abstract

The present invention belongs to the field of intelligent control, and provides an intelligent charge and discharge control method and system for energy storage equipment, including: predicting power generation and power consumption; determining charge and discharge capacity; determining redundant power; obtaining the prior distribution of each battery; calculating the likelihood function of each state variable of each battery; determining the posterior probability of each state variable of each battery according to the likelihood function; determining the charge and discharge capacity and charge and discharge priority of redundant batteries and non-redundant batteries according to the charge and discharge capacity and redundant power and the posterior probability of each state variable of each battery; and performing charge and discharge control according to the charge and discharge capacity and charge and discharge priority of the non-redundant battery. The above scheme can accurately perform intelligent charge and discharge control of energy storage equipment.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent control, and in particular relates to an intelligent charging and discharging control method and system for energy storage equipment. Background Art

[0002] Energy storage equipment plays a vital role in modern power systems. Energy storage equipment mainly includes battery packs and supercapacitors, which can store electricity during the low electricity demand period and release electricity during the peak electricity demand period, thereby balancing the grid load and improving energy efficiency. Intelligent charge and discharge control uses advanced algorithms and control strategies to achieve efficient charging and discharging of energy storage equipment, optimize system performance, extend battery life, and ensure stable operation of the power grid.

[0003] Existing charging and discharging control methods usually predict power generation and power consumption, and adjust the charging and discharging strategy based on the predicted values. However, there are errors in the charging and discharging predictions and battery status measurements. The actual power consumption and power generation data are not completely consistent with the actual battery status, which results in the charging and discharging strategy being unable to achieve the optimal effect. Summary of the invention

[0004] In order to solve the problems in the prior art, the present invention provides an intelligent charging and discharging control method for energy storage equipment, the method comprising the following steps: in each cycle, using a long short-term memory network to predict power generation and power consumption; determining the charging and discharging amount according to the difference between the power generation and the power consumption; estimating the prediction error of power generation and power consumption according to historical data, and determining the redundant power according to the prediction error; obtaining the historical state data of each battery, and determining the prior distribution of each state variable of each battery according to the historical state data; calculating the likelihood function of each state variable of each battery according to the prior distribution and the current battery state data; determining the posterior probability of each state variable of each battery according to the likelihood function; determining the charging and discharging amount and charging and discharging priority of redundant batteries and non-redundant batteries according to the charging and discharging amount and the redundant power with the posterior probability of the variable of each state of each battery; and performing charging and discharging control according to the charging and discharging amount and the charging and discharging priority of the non-redundant battery.

[0005] Further, determining the redundant power according to the prediction error includes: using historical data and a prediction model to calculate a prediction error between actual power generation and predicted power generation in each time period;

[0006] Summarize the forecast errors of all time periods and calculate the standard deviation of the forecast errors. Determine the redundant power by multiples of the standard deviation. Add the standard deviations of the forecast errors and multiply them by the preset safety factor to get the redundant power that needs to be reserved.

[0007] Furthermore, obtaining historical status data of each battery, and determining the prior distribution of each state variable of each battery based on the historical status data includes: obtaining operating data of each battery, including voltage, current, temperature, health status, and remaining capacity; performing statistical analysis on the collected historical status data to determine the prior distribution of each state variable, wherein the prior distribution is described by a mean and a standard deviation.

[0008] Further, determining the charge and discharge amount and charge and discharge priority of redundant batteries and non-redundant batteries according to the charge and discharge amount and redundant power with the posterior probability of the variables of each state of each battery includes: comprehensively scoring each battery, the comprehensive score being composed of the posterior probability and weight of each state variable; selecting batteries with low comprehensive scores and low remaining capacity as charging redundant batteries according to the comprehensive score and remaining capacity, and ensuring that the total capacity meets the redundant power demand; selecting batteries with low comprehensive scores and low health status as discharging redundant batteries according to the comprehensive score and health status, and ensuring that the total capacity meets the redundant power demand; determining the charge amount of each non-redundant battery according to the remaining capacity and charging demand; determining the discharge amount of each non-redundant battery according to the health status and discharging demand: determining the charge and discharge priority of non-redundant batteries according to the comprehensive score, and batteries with high priority will be charged or discharged first; selecting batteries with low remaining capacity for charging first, and selecting batteries with high health status for discharging first.

[0009] Further, the charge and discharge control according to the charge and discharge amount and charge and discharge priority of the non-redundant battery includes: charging the non-redundant batteries in sequence according to the charging priority order until the total charging demand is met; starting charging from the battery with the highest priority, performing a charging operation on each battery, the charging amount is the calculated charging amount, and gradually reducing the total charging demand until the charging demand is zero or all the non-redundant batteries are fully charged; discharging the non-redundant batteries in sequence according to the discharge priority order until the total discharge demand is met; starting discharging from the battery with the highest priority, performing a discharge operation on each battery, the discharge amount is the calculated discharge amount, and gradually reducing the total discharge demand until the discharge demand is zero or all the non-redundant batteries are discharged.

[0010] The present invention also provides an intelligent charging and discharging control system for energy storage equipment, comprising the following modules:

[0011] The prediction module is used to predict the power generation and power consumption in each cycle using the long short-term memory network;

[0012] The first calculation module is used to determine the charge and discharge amount according to the difference between the power generation amount and the power consumption amount;

[0013] A second calculation module is used to estimate the power generation and power consumption prediction errors according to historical data, and determine the redundant power according to the prediction errors;

[0014] A third calculation module is used to obtain historical state data of each battery; and determine the prior distribution of each state variable of each battery according to the historical state data;

[0015] A fourth calculation module, used for calculating the likelihood function of each state variable of each battery according to the prior distribution and the current battery state data;

[0016] a fifth calculation module, configured to determine a posterior probability of each state variable of each battery according to the likelihood function;

[0017] A determination module, configured to determine the charge and discharge capacity and charge and discharge priority of redundant batteries and non-redundant batteries according to the charge and discharge capacity and the redundant power with the posterior probability of the variable of each state of each battery;

[0018] The control module is used to control the charge and discharge according to the charge and discharge amount and the charge and discharge priority of the non-redundant battery.

[0019] The present invention can produce the following beneficial effects through the above technical solution:

[0020] In order to cope with prediction errors and emergencies, a certain amount of redundant power is reserved in the system. The redundant power is calculated based on historical data and the prediction error is determined by statistical methods. By collecting historical data on photovoltaic power generation, wind power generation and power consumption, the difference between the actual value and the predicted value for each time period is calculated, the mean and standard deviation of these prediction errors are counted, and the required redundant power is determined based on the standard deviation. By dynamically selecting redundant batteries and non-redundant batteries and determining the redundant power based on the prediction error for each cycle, it is possible to effectively cope with power load fluctuations and emergencies and ensure the stable operation of the system.

[0021] The historical status data of each battery is obtained, and the prior distribution of each state variable of each battery (such as voltage, current, temperature, state of health (SoH), remaining capacity (SoC), etc.) is determined based on this data. These prior distributions are combined with the current battery status data, and the likelihood function and posterior probability of each battery state variable are calculated through Bayesian inference, so as to more accurately evaluate the battery status. By accurately evaluating the battery status, batteries with better health status are selected for discharge operations, and batteries with poor health status are avoided from being frequently used, thereby extending the overall service life of the battery.

[0022] The comprehensive score of each battery is calculated based on the posterior probability and fixed weight parameters of each battery. The comprehensive score is used to evaluate the status of the battery. Batteries with lower comprehensive scores and lower remaining capacity are selected as charging redundant batteries, and batteries with lower comprehensive scores and lower health status are selected as discharging redundant batteries to ensure that the total capacity meets the redundant power demand. Intelligent regulation is carried out based on real-time prediction data and historical data to avoid overcharging or discharging, reduce energy waste, and improve energy utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0024] Figure 1 is a flow chart of the method of the present invention. DETAILED DESCRIPTION

[0025] Below, the invention is preferably described in conjunction with the accompanying drawings and specific implementation methods.

[0026] This embodiment solves the above problem through the following steps:

[0027] In one embodiment, reference Figure 1 The present invention provides an intelligent charging and discharging control method for energy storage equipment.

[0028] In this embodiment, the energy storage device refers to a device for storing electric energy, which generally includes a battery pack, a supercapacitor, etc. The energy storage device can store electric energy during the low electricity demand period and release electric energy during the peak electricity demand period to balance the load of the power grid and improve energy utilization efficiency.

[0029] Intelligent charging and discharging control refers to the use of advanced algorithms and control strategies to achieve efficient charging and discharging of energy storage devices.

[0030] In this embodiment, the intelligent control can dynamically adjust the charging and discharging strategy to optimize system performance, extend battery life, and ensure stable operation of the power grid.

[0031] In each cycle, the long short-term memory network is used to predict the power generation and power consumption.

[0032] In power systems, operations are usually performed at a fixed time interval, and data acquisition, processing and control operations are repeated within this time interval. Common cycles can be minutes, hours or days, etc.

[0033] Long Short-Term Memory (LSTM) is a special recurrent neural network (RNN) designed to solve the long-term dependency problem in traditional RNN. LSTM can effectively capture and maintain the long-term and short-term dependencies in time series data by introducing memory units and gating mechanisms, and is widely used in time series prediction tasks.

[0034] Electricity consumption and power generation are usually time series related to time fluctuations, so in this embodiment, historical data and models are used to estimate the power generation of renewable energy (such as photovoltaic and wind power) in the future time period. Forecasting power generation helps to plan the dispatch and management of power resources in advance. Use historical data and models to estimate power demand in the future time period. Electricity consumption forecasting can help optimize power grid operation and ensure supply and demand balance.

[0035] Using LSTM network to perform data prediction is a conventional technical means in this field. Those skilled in the art can refer to the existing technology to train LSTM network and use it to achieve prediction.

[0036] The charge and discharge amount is determined based on the difference between power generation and power consumption.

[0037] At the beginning of each cycle, the forecast values ​​of photovoltaic power generation, wind power generation and power consumption for the next cycle are obtained from the forecast model. These forecast values ​​are based on current meteorological data, historical power generation data and power consumption data.

[0038] Add the photovoltaic power generation and wind power generation to get the total power generation forecast for the next cycle. For example, if the predicted photovoltaic power generation is 50 kWh and the wind power generation is 30 kWh, the total power generation is 80 kWh.

[0039] Calculate the difference between total power generation and power consumption. For example, if the predicted power consumption is 60 kWh, then the difference between power generation and power consumption is 20 kWh (80 kWh minus 60 kWh).

[0040] If the amount of electricity generated is greater than the amount of electricity consumed, the difference is positive, indicating that there is excess electricity, and the energy storage device needs to be charged. The amount of charge is equal to the difference between the amount of electricity generated and the amount of electricity consumed.

[0041] If the power generation is less than the power consumption, the difference is negative, indicating that the power is insufficient, and the energy storage device needs to discharge. The discharge amount is equal to the absolute value of the difference between the power generation and the power consumption.

[0042] If the power generation is equal to the power consumption, the difference is zero, which means that the electric energy is just balanced. At this time, the energy storage device does not need to be charged or discharged.

[0043] The power generation and power consumption prediction errors are estimated based on historical data, and the redundant power is determined based on the prediction errors.

[0044] In this system, redundant power is the additional energy storage capacity reserved to cope with forecast errors and emergencies. Redundant power ensures that the system can still operate normally when the actual power generation or power consumption is inconsistent with the forecast value.

[0045] Collect and organize the photovoltaic power generation, wind power generation and power consumption data over a period of time. These data are usually recorded at hourly, daily or other appropriate time intervals to form time series data. The collection of historical data is very important because it provides a basis for subsequent forecast error analysis and redundant power calculation.

[0046] Using historical data and forecasting models, the difference between actual power generation and forecasted power generation for each time period is calculated. These differences are called forecast errors. Specifically, for each time period, the difference between actual photovoltaic power generation and forecasted photovoltaic power generation, and the difference between actual wind power generation and forecasted wind power generation are calculated.

[0047] Similarly, the difference between the actual electricity usage and the predicted electricity usage is calculated for each time period. These differences are also called forecast errors. For each time period, the difference between the actual electricity usage and the predicted electricity usage is calculated.

[0048] Summarize the forecast errors of all time periods and calculate the statistical indicators of the forecast errors, such as the mean and standard deviation. The standard deviation is an important indicator to measure the volatility of the forecast error, which indicates the average difference between the forecast value and the actual value. By calculating the standard deviation of the forecast errors of photovoltaic power generation, wind power generation and electricity consumption, we can understand the accuracy and stability of the forecast model.

[0049] The redundant power required by the system is determined based on the statistical indicators of the prediction error. The redundant power is the energy storage capacity reserved to cope with possible prediction errors. Specifically, the redundant power can be determined by multiples of the standard deviation. An appropriate safety factor can be selected, the standard deviation of the prediction errors of photovoltaic power generation, wind power generation and power consumption are added together, and then multiplied by this safety factor to obtain the redundant power that needs to be reserved.

[0050] Acquire historical state data of each battery; and determine the prior distribution of each state variable of each battery according to the historical state data.

[0051] In this embodiment, the historical status data refers to the operating data of each battery recorded in the past period of time, including voltage, current, temperature, state of health (SoH), remaining capacity (SoC), etc. The historical status data can reflect the long-term operation and performance changes of the battery.

[0052] State variables refer to key parameters used to describe the battery status, such as battery voltage, current, temperature, state of health (SoH), and remaining capacity (SoC), etc. These variables together characterize the current state and health of the battery.

[0053] Prior distribution refers to the probability distribution of parameters based on previous knowledge or historical data before considering new data. Prior distribution combined with observed data can be updated to posterior distribution through Bayes' theorem.

[0054] First, it is necessary to obtain the historical status data of each battery from the battery management system (BMS). Including:

[0055] Voltage: reflects the voltage level of the battery, measured in volts (V).

[0056] Current: reflects the charging and discharging current of the battery, measured in ampere (A).

[0057] Temperature: reflects the operating temperature of the battery, in degrees Celsius (℃).

[0058] State of Health (SoH): Indicates the health of the battery, relative to the performance ratio of a new battery, usually expressed as a percentage.

[0059] Remaining capacity (SoC, State of Charge): Indicates the current charging state of the battery, usually expressed as a percentage.

[0060] These data are usually recorded at time intervals of hours or minutes to form detailed time series data. Through long-term monitoring, the state changes of the battery under various operating conditions can be obtained.

[0061] The collected historical state data are statistically analyzed to determine the prior distribution of each state variable.

[0062] Collect the voltage data of the battery at different time points and calculate the mean and standard deviation of the voltage. Assuming that the voltage data follows a normal distribution, its prior distribution can be described by the mean and standard deviation.

[0063] For example, if the mean voltage data of a battery is 3.7V and the standard deviation is 0.1V, its prior distribution can be expressed as N(3.7,0.1 2 ).

[0064] Similarly, the current data and temperature data of the battery are collected to calculate their mean and standard deviation, and it is assumed that the current data follows a normal distribution.

[0065] For example, if the mean current data of a battery is 1.5A and the standard deviation is 0.2A, its prior distribution can be expressed as N(1.5,0.2 2).

[0066] Collect the battery temperature data, calculate its mean and standard deviation, and assume that the temperature data follows a normal distribution.

[0067] For example, if the mean of the temperature data of a battery is 25°C and the standard deviation is 2°C, then its prior distribution can be expressed as N(25,2 2 ).

[0068] Collect the SoH data of the batteries, calculate their mean and standard deviation, and assume that the SoH data follows a normal distribution.

[0069] For example, the mean of the SoH data of a battery is 90% and the standard deviation is 5%, so its prior distribution can be expressed as N(90,5 2 ).

[0070] Collect the battery SoC data, calculate its mean and standard deviation, and assume that the SoC data follows a normal distribution.

[0071] For example, if the mean of the SoC data of a battery is 50% and the standard deviation is 10%, its prior distribution can be expressed as N(50,10 2 ).

[0072] Through the above steps, the historical status data of each battery is obtained, and the prior distribution of each state variable of each battery is determined based on these data, which provides a basis for subsequent inference. This process helps to more accurately evaluate the battery status and optimize the charging and discharging strategy of the energy storage system.

[0073] The likelihood function of each state variable of each battery is calculated according to the prior distribution and the current battery state data.

[0074] In statistics, the likelihood function is the probability of observation data occurring given a parameter. In Bayesian inference, the likelihood function is used to quantify the relationship between observation data and parameter values. In an embodiment, the likelihood function reflects the possibility of the current observation data under the assumed state variable parameter value.

[0075] In the intelligent charging and discharging control method of energy storage equipment, it is necessary to calculate the likelihood function of each state variable of each battery based on the prior distribution and the current battery state data. This process is a key step in Bayesian inference, which can help us combine prior knowledge with current observation data and update our understanding of the battery state.

[0076] Obtain battery status data at the current time point from the battery management system (BMS). This data includes the voltage, current, temperature, state of health (SoH) and remaining capacity (SoC) of each battery.

[0077] Based on the current state data and prior distribution, the likelihood function of each state variable is constructed. The likelihood function describes the probability of observing the current data when the parameter value is given.

[0078] The likelihood function usually assumes that the observed data follows a certain probability distribution, such as a normal distribution. Assuming that the voltage data follows a normal distribution, the likelihood function can be expressed as:

[0079]

[0080] Among them, D is the current observation data, μ and σ are the mean and standard deviation of the prior distribution, and θ is the parameter of the state variable.

[0081] By substituting the current observed data into the likelihood function, the likelihood function value of each state variable of each battery is calculated. The likelihood function value represents the probability of observing the current data under the assumed prior distribution parameters.

[0082] Exemplarily

[0083] Current voltage D = 3.8, prior distribution N (3.7, 0.1 2 )

[0084] Likelihood function:

[0085]

[0086] The posterior probability of each state variable of each battery is determined according to the likelihood function.

[0087] The posterior probability is the probability distribution of the parameter obtained by combining the prior distribution and the observed data. It represents the updated cognition of the parameter after observing the current data. In this embodiment, the posterior probability represents the updated cognition of the battery state variable after combining the prior distribution and the current battery state data.

[0088] The prior distribution reflects the perception of the parameters in the absence of current data, while the likelihood function reflects the likelihood of the current data occurring given the parameter value.

[0089] Bayes’ theorem is used to calculate the posterior probability, and the formula is as follows:

[0090]

[0091] Where: P(θ|D) is the posterior probability, which represents the probability of parameter θ after observing data D.

[0092] P(D|θ) is the likelihood function, which represents the probability of observing data D given the parameters θ.

[0093] P(θ) is the prior distribution, which represents the initial understanding of the parameter θ when there is no observed data.

[0094] P(D) is the marginal likelihood of observing the data and can be calculated by integrating or summing over all possible parameter values.

[0095] For example, the voltage posterior probability is:

[0096] Prior distribution P(θ voltage )=N(3.7,0.1 2 )

[0097] Current observed voltage D = 3.8

[0098] Likelihood function P(D|θ voltage )=L(θ voltage |3.8)

[0099] Posterior probability

[0100]

[0101] The charge and discharge amounts and charge and discharge priorities of the redundant batteries and the non-redundant batteries are determined according to the charge and discharge amounts and the redundant power with the posterior probability of the variables of each state of each battery.

[0102] In the energy storage system, a portion of battery capacity is specially reserved to cope with load fluctuations and emergencies. These batteries are mainly used to ensure system stability and are called redundant batteries. In this embodiment, a certain number of redundant batteries are selected in each cycle. The redundant batteries in different cycles may be different and are dynamically updated in each cycle to achieve system optimization.

[0103] Non-redundant batteries are the battery capacity used for charging and discharging operations in the energy storage system, excluding the redundant parts used to deal with emergencies. Non-redundant batteries will also be updated according to the status of the battery.

[0104] Specifically

[0105] The comprehensive score of each battery is used to evaluate its status. The comprehensive score consists of the posterior probability of each state variable and a fixed weight. The formula is as follows:

[0106] Score i =w1·P(SoH i )+w2·P(SoC i )-w3·P(T i )+w4·P(DOD i )

[0107] in

[0108] Score i : The overall rating of Battery i.

[0109] w1,w2,w3,w4: weight parameters, reflecting the importance of each state variable.

[0110] P(SoH i ): the posterior probability of the health state of battery i.

[0111] P(SoC i ): the posterior probability of the remaining capacity of battery i.

[0112] P(T i ): posterior probability of battery i temperature.

[0113] P(DOD i ): the posterior probability of the depth of discharge of battery i.

[0114] According to the comprehensive score and remaining capacity (SoC), select batteries with lower comprehensive scores and lower remaining capacity as charging redundant batteries to ensure that the total capacity meets the redundant power requirements

[0115]

[0116] Redundant charge : Charge the redundant battery set.

[0117] Capacity j : Capacity of battery j.

[0118] E redundant : Redundant power demand.

[0119] According to the comprehensive score and the state of health (SoH), batteries with lower comprehensive scores and lower state of health are selected as discharge redundant batteries to ensure that the total capacity meets the redundant power demand.

[0120]

[0121] Redundant discharge : Discharge redundant battery set

[0122] According to the current cycle charge P charge and discharge capacity P discharge , and the remaining power of the non-redundant battery, determine the charge and discharge capacity of each non-redundant battery.

[0123] According to the remaining capacity (SoC) and charging demand P charge , determine the charge of each non-redundant battery:

[0124]

[0125] P charge,i : Charge level of battery i.

[0126] Non-Redundant: Non-redundant battery collection.

[0127] According to the state of health (SoH) and discharge demand P discharge , determine the discharge of each non-redundant battery:

[0128]

[0129] P discharge,i : Discharge capacity of battery i.

[0130] The charging and discharging priority of the non-redundant batteries is determined based on the comprehensive score, and the batteries with higher priority will be charged or discharged first.

[0131] Select the battery with lower remaining capacity (SoC) to charge first:

[0132] Priority charge,i =argmin(SoC i )

[0133] Prioririty charge,i : Charging priority of battery i.

[0134] Discharge priority:

[0135] Select batteries with higher state of health (SoH) to be discharged first:

[0136] Priority discharge,i =argmax(SoH i )

[0137] Prioririty discharge,i : Discharge priority of battery i.

[0138] Through the above steps, the charge and discharge capacity and priority of redundant batteries and non-redundant batteries can be determined in each cycle according to the charge and discharge capacity, redundant power and the posterior probability of each battery state variable. This ensures that the energy storage system maintains the necessary redundant capacity while meeting daily power demand to cope with uncertainty and improve the reliability and stability of the system.

[0139] The charging and discharging control is performed according to the charging and discharging amount and the charging and discharging priority of the non-redundant battery.

[0140] The non-redundant batteries are charged in order of charging priority until the total charging demand P is met. charge .

[0141] Start charging from the battery with the highest priority. Perform charging operation on each battery, and the charging capacity is calculated

[0142] P charge,i Gradually reduce the total charging demand P charge , until the charging demand is zero or all non-redundant batteries are fully charged.

[0143] Discharge the non-redundant batteries in order of discharge priority until the total discharge demand P is met. discharge .

[0144] Start discharging from the battery with the highest priority. Perform a discharge operation on each battery, and the discharge amount is the calculated P discharge,i . Gradually reduce the total discharge demand P discharge , until the discharge demand is zero or all non-redundant batteries are discharged.

[0145] On the other hand, the present invention also provides an intelligent charging and discharging control system for energy storage equipment, comprising:

[0146] The prediction module is used to predict the power generation and power consumption in each cycle using the long short-term memory network;

[0147] The first calculation module is used to determine the charge and discharge amount according to the difference between the power generation amount and the power consumption amount;

[0148] A second calculation module is used to estimate the power generation and power consumption prediction errors according to historical data, and determine the redundant power according to the prediction errors;

[0149] A third calculation module is used to obtain historical state data of each battery; and determine the prior distribution of each state variable of each battery according to the historical state data;

[0150] A fourth calculation module, used for calculating the likelihood function of each state variable of each battery according to the prior distribution and the current battery state data;

[0151] a fifth calculation module, configured to determine a posterior probability of each state variable of each battery according to the likelihood function;

[0152] A determination module, configured to determine the charge and discharge capacity and charge and discharge priority of redundant batteries and non-redundant batteries according to the charge and discharge capacity and the redundant power with the posterior probability of the variable of each state of each battery;

[0153] The control module is used to control the charge and discharge according to the charge and discharge amount and the charge and discharge priority of the non-redundant battery.

[0154] Furthermore, the specific implementation methods of the above-mentioned intelligent charging and discharging control system for energy storage equipment are the same as an intelligent charging and discharging control method for energy storage equipment, and all further technical solutions in an intelligent charging and discharging control method for energy storage equipment are fully introduced into an intelligent charging and discharging control system for energy storage equipment.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

[0156] The present invention does not particularly specify the structure of some modules, which shall be subject to the contents recorded in the prior art. The prior art mentioned in the above background technology section and the specific embodiment section of the present invention can be used as part of the present invention to understand the meaning of some technical features or parameters. The scope of protection of the present invention shall be subject to the contents actually recorded in the claims.

Claims

1. An intelligent charging and discharging control method for energy storage equipment, characterized in that The method comprises the following steps: In each cycle, the long short-term memory network is used to predict the power generation and power consumption; Determine the charge and discharge amount based on the difference between power generation and power consumption; Estimate power generation and power consumption prediction errors based on historical data, and determine redundant power based on the prediction errors, wherein the historical data are historical power generation data and power consumption data; Acquire historical state data of each battery, and determine a priori distribution of each state variable of each battery according to the historical state data; Calculate the likelihood function of each state variable of each battery according to the prior distribution and the current battery state data; determining a posterior probability of each state variable of each battery according to the likelihood function; Determine the charge and discharge capacity and charge and discharge priority of redundant batteries and non-redundant batteries according to the charge and discharge capacity, redundant power and the posterior probability of the variable of each state of each battery; and perform charge and discharge control according to the charge and discharge capacity and charge and discharge priority of the non-redundant battery.

2. According to claim 1, a method for intelligent charging and discharging control of energy storage equipment is characterized in that Estimating power generation and power consumption prediction errors based on historical data, and determining redundant power based on the prediction errors include: Using historical data and forecasting models, calculate the forecast error between actual power generation and forecasted power generation for each time period; Summarize the forecast errors for all time periods and calculate the standard deviation of the forecast errors; The redundant power is determined by multiples of the standard deviation. The predicted error standard deviation is added and then multiplied by the preset safety factor to obtain the redundant power that needs to be reserved.

3. The intelligent charging and discharging control method of energy storage equipment according to claim 1 is characterized in that Acquiring historical state data of each battery, and determining the prior distribution of each state variable of each battery according to the historical state data includes: Obtain the operating data of each battery, including voltage, current, temperature, health status, and remaining capacity; Performing statistical analysis on the collected historical state data to determine a prior distribution of each state variable, wherein the prior distribution is described by a mean and a standard deviation; The historical status data refers to the operating data of each battery recorded in the past period of time.

4. The intelligent charging and discharging control method for energy storage equipment according to claim 1 is characterized in that Determining the charge and discharge amount and charge and discharge priority of the redundant battery and the non-redundant battery according to the charge and discharge amount, the redundant power, and the posterior probability of the variable of each state of each battery includes: A comprehensive score is given to each battery, which is composed of the posterior probability and weight of each state variable; According to the comprehensive score and remaining capacity, select batteries with low comprehensive scores and low remaining capacity as charging redundant batteries, and ensure that the total capacity meets the redundant power requirements; According to the comprehensive score and health status, select batteries with low comprehensive scores and low health status as discharge redundant batteries, and ensure that the total capacity meets the redundant power requirements; Determine the charge amount of each non-redundant battery based on the remaining capacity and the charging demand; Determine the discharge amount of each non-redundant battery based on the health status and discharge requirements: Determine the charging and discharging priority of non-redundant batteries based on the comprehensive score, and batteries with higher priority will be charged or discharged first; Batteries with lower remaining capacity should be charged first, and batteries with higher health status should be discharged first.

5. The intelligent charging and discharging control method of energy storage equipment according to claim 1 is characterized in that The charging and discharging control according to the charging and discharging amount and the charging and discharging priority of the non-redundant battery includes: The non-redundant batteries are charged in order of charging priority until the total charging demand is met; Starting from the battery with the highest priority, charging is performed on each battery, and the charging amount is the calculated charging amount, gradually reducing the total charging demand until the charging demand is zero or all non-redundant batteries are fully charged; Discharge the non-redundant batteries in order of discharge priority until the total discharge demand is met; Discharging starts from the battery with the highest priority, and performs a discharge operation on each battery with a discharge amount equal to the calculated discharge amount, gradually reducing the total discharge demand until the discharge demand is zero or all non-redundant batteries are discharged.

6. An intelligent charging and discharging control system for energy storage equipment, characterized in that The system includes the following modules: The prediction module is used to predict the power generation and power consumption in each cycle using the long short-term memory network; The first calculation module is used to determine the charge and discharge amount according to the difference between the power generation amount and the power consumption amount; A second calculation module is used to estimate power generation and power consumption prediction errors based on historical data, and determine redundant power based on the prediction errors, wherein the historical data are historical power generation data and power consumption data; A third calculation module is used to obtain historical state data of each battery; and determine the prior distribution of each state variable of each battery according to the historical state data; a fourth calculation module, used for calculating the likelihood function of each state variable of each battery according to the prior distribution and the current battery state data; a fifth calculation module, configured to determine a posterior probability of each state variable of each battery according to the likelihood function; A determination module, configured to determine the charge and discharge capacity and charge and discharge priority of redundant batteries and non-redundant batteries according to the charge and discharge capacity, redundant power, and the posterior probability of the variable of each state of each battery; The control module is used to control the charge and discharge according to the charge and discharge amount and the charge and discharge priority of the non-redundant battery.

7. An intelligent charging and discharging control system for energy storage equipment according to claim 6, characterized in that Estimating power generation and power consumption prediction errors based on historical data, and determining redundant power based on the prediction errors include: Using historical data and forecasting models, calculate the forecast error between actual power generation and forecasted power generation for each time period; Summarize the forecast errors for all time periods and calculate the standard deviation of the forecast errors; The redundant power is determined by multiples of the standard deviation. The predicted error standard deviation is added and then multiplied by the preset safety factor to obtain the redundant power that needs to be reserved.

8. An intelligent charging and discharging control system for energy storage equipment according to claim 6, characterized in that Acquiring historical state data of each battery, and determining the prior distribution of each state variable of each battery according to the historical state data includes: Obtain the operating data of each battery, including voltage, current, temperature, health status, and remaining capacity; Performing statistical analysis on the collected historical state data to determine a prior distribution of each state variable, wherein the prior distribution is described by a mean and a standard deviation; The historical status data refers to the operating data of each battery recorded in the past period of time.

9. An intelligent charge and discharge control system for energy storage equipment according to claim 6, characterized in that Determining the charge and discharge amount and charge and discharge priority of the redundant battery and the non-redundant battery according to the charge and discharge amount, the redundant power, and the posterior probability of the variable of each state of each battery includes: A comprehensive score is given to each battery, which is composed of the posterior probability and weight of each state variable; According to the comprehensive score and remaining capacity, select batteries with low comprehensive scores and low remaining capacity as charging redundant batteries, and ensure that the total capacity meets the redundant power requirements; According to the comprehensive score and health status, select batteries with low comprehensive scores and low health status as discharge redundant batteries, and ensure that the total capacity meets the redundant power requirements; Determine the charge amount of each non-redundant battery based on the remaining capacity and the charging demand; Determine the discharge amount of each non-redundant battery based on the health status and discharge requirements: Determine the charging and discharging priority of non-redundant batteries based on the comprehensive score, and batteries with higher priority will be charged or discharged first; Batteries with lower remaining capacity should be charged first, and batteries with higher health status should be discharged first.

10. An intelligent charge and discharge control system for energy storage equipment according to claim 6, characterized in that The charging and discharging control according to the charging and discharging amount and the charging and discharging priority of the non-redundant battery includes: The non-redundant batteries are charged in order of charging priority until the total charging demand is met; Starting from the battery with the highest priority, charging is performed on each battery, and the charging amount is the calculated charging amount, gradually reducing the total charging demand until the charging demand is zero or all non-redundant batteries are fully charged; Discharge the non-redundant batteries in order of discharge priority until the total discharge demand is met; Discharging starts from the battery with the highest priority, and performs a discharge operation on each battery with a discharge amount equal to the calculated discharge amount, gradually reducing the total discharge demand until the discharge demand is zero or all non-redundant batteries are discharged.

Images