Centralized control method and system for multi-source energy storage system
By predicting and planning the historical data of multi-source energy storage systems and optimizing energy storage strategies, the problems of low energy storage efficiency and energy waste caused by the lack of comprehensive considerations in the existing technology are solved, and more efficient energy storage management is achieved and the utilization of renewable energy.
Patent Information
- Application Number
- CN202411055873.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-08-02
AI Technical Summary
The existing multi-source energy storage systems lack comprehensive consideration of factors such as electricity price fluctuations, renewable energy generation situation and electricity load, resulting in low energy storage efficiency and possible energy waste.
By collecting historical data, data cleaning and prediction, including electricity price, power generation and load prediction, statistics of power storage and discharge sequences, building a centralized control method for multi-source energy storage systems, optimizing energy storage strategy planning, and considering the coordination of different energy storage devices. manage.
Improve the overall efficiency of the energy storage system, can better cope with the fluctuations in the power market and the uncertainty of renewable energy generation, balance the supply and demand of electricity, reduce wind and light abandonment, promote the sustainable development of renewable energy, and achieve Optimal allocation and efficient utilization of energy storage resources.
Smart Images

Figure CN118572702B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of multi-source energy storage systems, and in particular to a centralized control method and system for a multi-source energy storage system. Background Art
[0002] With the transformation of the global energy structure and the rapid development of renewable energy, multi-source energy storage systems have become an indispensable and important part of modern power systems. Multi-source energy storage systems integrate different types of energy storage devices, such as electrochemical energy storage, thermal energy storage, and supercapacitors, to achieve efficient storage and flexible use of energy, which is of great significance for balancing power supply and demand, improving grid stability, and promoting the consumption of renewable energy. However, the energy storage strategy planning of multi-source energy storage systems has always been a technical problem in the industry. The rationality and effectiveness of its planning method directly affect the economy and operation efficiency of the energy storage system.
[0003] At present, the configuration of energy storage capacity often relies on empirical estimates and lacks scientific analysis methods based on actual data and needs. This leads to the configuration of energy storage capacity that is often too large or too small, which not only increases investment costs, but may also fail to meet actual needs. The planning of energy storage strategies needs to consider a variety of factors, including electricity price fluctuations, renewable energy generation, electricity load, etc. However, current energy storage strategies often lack comprehensive consideration of these factors, resulting in low energy storage efficiency and even energy waste. Multi-source energy storage systems integrate multiple different types of energy storage devices, each of which has its own unique advantages and limitations. However, in the existing energy storage strategy planning, there is often a lack of consideration for the coordination between multiple energy storage devices, resulting in waste of energy storage resources and unstable system operation.
[0004] For example, a Chinese patent with authorization announcement number CN109861292B discloses a method for improving clean energy consumption based on a multi-energy storage system, including: collecting information at the current moment; calculating the output fluctuations of photovoltaic power stations and wind farms; and deciding to invest in heat storage or electricity storage batteries according to the specific circumstances of judgment conditions 1, 2, 3, and 4. The invention proposes that when the output of wind farms and photovoltaic power stations is greater than the load, corresponding energy storage and electricity storage equipment can be invested to consume the electric energy, solving the current power balance problem and ensuring the full consumption of clean energy. When the output of wind farms and photovoltaic power stations is insufficient, the batteries and heat storage equipment will output a certain power to ensure the demand for electricity and heating. The invention can increase the consumption of clean energy and ensure the stability of the power system. The multi-energy storage system that combines electricity storage and energy storage can improve the clean energy consumption capacity.
[0005] For example, the patent application with publication number CN116565905A discloses a multi-energy complementary water-gas co-storage energy system and energy storage method, which utilizes valley electricity and new energy electricity to compress air and water energy storage, utilizes the surplus electric energy with the help of air compressor to compress the air, stores the compressed air in the air storage tank, and at the same time, pumps the working fluid liquid in the liquid storage component to the liquid chamber to form a high head liquid level, so as to maximize the use of excess electricity. In the energy release stage, the compressed air pushes the working fluid liquid in the liquid chamber to flow to the hydro-turbine generator to generate electricity, and the tail water of the power generation can be discharged into the liquid storage component to keep the hydro-turbine generator in the optimal working condition under stable pressure. When the water pressure of the hydro-turbine generator is unstable or the liquid level in the liquid chamber is as low as the warning line, the liquid is replenished in time, and the pressure of the air chamber is adjusted by the cooling component, so that the water pressure of the hydro-turbine generator can be quickly adjusted.
[0006] The above patents all have the problems raised by this background technology: lack of comprehensive consideration of factors such as electricity price fluctuations, renewable energy generation, and electricity load, resulting in low energy storage efficiency and possible energy waste.
[0007] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to ordinary technicians in this field. Summary of the invention
[0008] The technical problem to be solved by the present invention is to overcome the defects of the prior art and provide a centralized control method and system for a multi-source energy storage system, so as to promote multi-source coordinated optimization of the multi-source energy storage system, enhance the stability of the multi-source energy storage system, and promote the consumption of renewable energy.
[0009] In order to solve the above technical problems, the present invention provides the following technical solutions:
[0010] In a first aspect, the present invention provides a centralized control method for a multi-source energy storage system, comprising the following steps:
[0011] S1: Collect historical data and perform data cleaning; the historical data includes historical electricity price data, historical power generation data, historical load data, historical power storage data, and historical discharge data;
[0012] S2: Performing power data forecasting based on the historical data, including power price forecasting, power generation forecasting, and load forecasting;
[0013] S3: Determine the amount of storage capacity based on the historical storage data and the result of the power data prediction;
[0014] S4: Counting the historical discharge data to generate a discharge sequence;
[0015] S5: Based on the storage amount and discharge sequence, perform energy storage strategy planning for the multi-source energy storage system.
[0016] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, the power generation prediction method is as follows:
[0017] Acquire historical meteorological data and perform data cleaning and normalization processing; the historical meteorological data includes historical light intensity data, historical daylight time data, historical wind speed data, historical temperature data, and historical humidity data;
[0018] The power generation potential index, including wind power potential index and photovoltaic potential index, is calculated as follows:
[0019] ;
[0020] in, It represents the wind power potential index; represents the wind power potential factor; v represents the daily average wind speed;
[0021] ;
[0022] in, It represents the photoelectric potential index; represents the photoelectric potential factor; G represents the average daily light intensity; T represents the average daily light time;
[0023] Arranging the historical power generation data into a historical power generation sequence in chronological order;
[0024] Construct a power generation prediction model based on multivariate linear regression, with wind power potential index, photovoltaic potential index, daily average temperature, and daily average humidity as input, and wind and solar power generation for one day as output; perform model training based on the historical power generation data and historical meteorological data, use the least squares method or other optimization algorithms to perform parameter estimation, and save the trained power generation prediction model;
[0025] Obtain future meteorological data, including light intensity data, daylight duration data, wind speed data, temperature data, and humidity data, calculate power generation potential indicators and input them into the power generation prediction model, obtain power generation prediction values and organize them into power generation prediction sequences.
[0026] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, the load prediction method is as follows:
[0027] Arrange the preprocessed historical load data into a load sequence in chronological order;
[0028] Constructing a load forecasting model, wherein the load forecasting model is any one of a SARIMA model and a Holt-Winters model;
[0029] Training and testing the load forecasting model based on the load sequence;
[0030] The trained load forecasting model is saved and load forecasting is performed. Each time a load forecast is performed, an average power load forecast value for the day is obtained, which is added to the load sequence for the next load forecast. The load forecast is performed iteratively, and a load forecast sequence consisting of load forecast values is intercepted from the load sequence.
[0031] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, the method of electricity price prediction is as follows:
[0032] Arrange the historical electricity price data into a historical electricity price sequence in chronological order, add a timestamp tag to each historical electricity price data in the historical electricity price sequence, encode the timestamp tag to form a timestamp tag sequence; the timestamp tag is a tag of the date corresponding to the historical electricity price data, including working days, weekends, and statutory holidays;
[0033] Time-aligning the historical electricity price sequence, the historical power generation sequence, the load sequence, and the timestamp tag sequence;
[0034] Constructing an electricity price prediction model; the electricity price prediction model is any one of a recurrent neural network, a long short-term memory network, and a gated recurrent unit; the input is a label sequence consisting of a historical power generation sequence, a load sequence, and a timestamp label, and the output is an electricity price prediction sequence;
[0035] Performing training, verification, testing and tuning of the electricity price prediction model, and saving the trained electricity price prediction model;
[0036] A future timestamp label sequence is constructed and input into the electricity price prediction model together with the power generation prediction sequence and the load prediction sequence to obtain an electricity price prediction sequence.
[0037] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, the storage capacity is the storage capacity of the multi-source energy storage system for the next day; the method for determining the storage capacity is as follows:
[0038] The historical power storage data is counted and averaged, recorded as ;
[0039] Based on the power generation forecast sequence, the mean value of the power generation data for the next n days is calculated and recorded as , n is a positive integer; the mean of the load data for the next n days is calculated based on the load forecast sequence, and is recorded as ; Based on the electricity price forecast sequence, calculate the mean of the electricity price data for the next n days, denoted as ;
[0040] The formula for calculating the power storage capacity is as follows:
[0041] ;
[0042] Among them, S represents the storage capacity; Indicates the power generation data of the previous day; Indicates the load data of the previous day; Indicates the electricity price data of the previous day; , , are all weight coefficients.
[0043] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, the method of generating the discharge sequence is as follows:
[0044] Count the historical discharge data, divide a day into m time slices, where m is a positive integer; calculate the mean of the historical discharge data of each time slice in any day, and record the mean of the historical discharge data of the jth time slice on the qth day as ; Calculate the expected discharge amount for the jth time slice in the future day , the formula is as follows:
[0045] ;
[0046] Where Q is the total number of days involved in the historical discharge data;
[0047] Will The discharge sequence is organized in chronological order, and the value range of j is 1, 2, ..., m.
[0048] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, wherein: the energy storage strategy is the charging power of each energy storage device in each time slice;
[0049] The energy storage device includes an electrochemical energy storage device, a mechanical energy storage device, a thermal energy storage device, a chemical energy storage device, and a supercapacitor;
[0050] The method for energy storage strategy planning is as follows:
[0051] Collect the current storage capacity of each energy storage device, and set the minimum energy storage threshold, maximum energy storage threshold, and maximum charging power for each energy storage device; obtain the electricity purchase price for each time slice in the future day;
[0052] Construct an energy storage strategy model, including objective function, constraints, and state update equations;
[0053] An optimization algorithm is used to solve the energy storage strategy model to obtain an energy storage strategy that minimizes the objective function.
[0054] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, the objective function calculation formula is as follows:
[0055] ;
[0056] Where L represents the objective function; represents the charging power of the i-th energy storage device at the j-th time slice; represents the price of electricity purchased by the multi-source energy storage system from the grid in the jth time slice; represents the time corresponding to the midpoint of the j-th time slice; Indicates the peak load period; represents the charging power of the supercapacitor in the jth time period; represents the charging power of the electrochemical energy storage device in the jth time period; represents the storage capacity of the thermal energy storage device at the end of the j-1th time slice; Indicates the duration of any time slice; represents the bonus coefficient of the supercapacitor; represents the bonus coefficient of the electrochemical energy storage device; It is the heat loss coefficient of the thermal energy storage device, which indicates the amount of heat loss per unit time for each unit energy stored in the thermal energy storage device.
[0057] As a preferred solution of the centralized control method of the multi-source energy storage system of the present invention, the constraint conditions are as follows:
[0058] ;
[0059] in, represents the storage capacity of the i-th energy storage device in the j-th time slice; represents the minimum energy storage threshold of the i-th energy storage device; represents the maximum energy storage threshold of the i-th energy storage device; represents the maximum charging power of the i-th energy storage device;
[0060] The state update equation is as follows:
[0061] ;
[0062] in, It represents the storage capacity of the i-th energy storage device in the j+1-th time slice.
[0063] In a second aspect, the present invention provides a centralized control system for a multi-source energy storage system, including a data acquisition module, a power prediction module, a power storage planning module, a discharge planning module, a power storage solution module, and a monitoring and display module; wherein:
[0064] The data acquisition module is used to collect historical data, meteorological data, energy storage device status and perform data cleaning;
[0065] The power prediction module is used to perform power price prediction, power generation prediction, and load prediction based on historical data;
[0066] The power storage planning module is used to calculate the power storage capacity for the next day based on the power price forecast, power generation forecast and load forecast;
[0067] The discharge planning module is used to generate a discharge sequence based on historical data;
[0068] The power storage solution module is used to plan energy storage strategies based on the power storage capacity and discharge sequence;
[0069] The monitoring and display module is used to visually display the energy storage device status, energy storage strategy, and discharge sequence.
[0070] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:
[0071] Through electricity data forecasting based on historical data, we can accurately predict future electricity prices, power generation and load, and then determine the appropriate amount of storage capacity, avoiding over-configuration or under-configuration of energy storage capacity, thereby improving the overall efficiency of the energy storage system, and being able to better cope with fluctuations in the electricity market and the uncertainty of renewable energy generation, balance electricity supply and demand, reduce grid fluctuations, promote the consumption and utilization of renewable energy, reduce wind and solar power abandonment, and promote the sustainable development of renewable energy.
[0072] It realizes effective coordinated management of different energy storage devices and formulates more accurate charging strategies for each energy storage device, which helps to extend the service life of the energy storage device and reduce maintenance costs. It ensures the optimal configuration and efficient use of energy storage resources while meeting system requirements, ensures stable system operation, reduces operating costs, and improves the economy of the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS
[0073] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. 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 creative labor. Among them:
[0074] Figure 1 A flowchart of a centralized control method for a multi-source energy storage system provided by the present invention;
[0075] Figure 2 A flow chart of a method for predicting power generation provided by the present invention;
[0076] Figure 3 A flow chart of a method for predicting electricity prices provided by the present invention;
[0077] Figure 4 A schematic diagram of the structural principle of a centralized control system of a multi-source energy storage system provided by the present invention. DETAILED DESCRIPTION
[0078] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.
[0079] Example 1
[0080] This embodiment introduces a centralized control method for a multi-source energy storage system. Figure 1 , the method comprises the following steps:
[0081] S1: Collect historical data and perform data cleaning; the historical data includes historical electricity price data, historical power generation data, historical load data, historical power storage data, and historical discharge data;
[0082] The historical electricity price data is the historical data of the peak electricity price of daily residential electricity consumption;
[0083] The historical power generation data is the historical data of daily wind and solar power generation;
[0084] The historical load data is historical data of residents' daily average electricity load;
[0085] The historical power storage data is the historical data of the daily power storage amount of the multi-source energy storage system;
[0086] The historical discharge data is the historical data of the daily discharge amount of the multi-source energy storage system.
[0087] S2: Performing power data forecasting based on the historical data, including power price forecasting, power generation forecasting, and load forecasting;
[0088] Considering that electricity price data, power generation data, and load data each have different changing characteristics, this embodiment selects different methods to perform power generation forecasting, load forecasting, and electricity price forecasting respectively based on the characteristics of each type of data.
[0089] Reference Figure 2 , the method for predicting power generation is as follows:
[0090] Acquire historical meteorological data and perform data cleaning and normalization processing; the historical meteorological data includes historical light intensity data, historical daylight time data, historical wind speed data, historical temperature data, and historical humidity data;
[0091] The power generation potential index, including wind power potential index and photovoltaic potential index, is calculated as follows:
[0092] ;
[0093] in, It represents the wind power potential index; represents the wind power potential factor, which is determined by parameter fitting with the multivariate linear regression model; v represents the daily average wind speed; the daily average wind speed is obtained based on historical wind speed data;
[0094] ;
[0095] in, It represents the photoelectric potential index; represents the photoelectric potential factor, which is determined by parameter fitting with the multivariate linear regression model; G represents the daily average light intensity; T represents the daily average light time; the daily average light intensity is obtained according to the historical light intensity data, and the daily average light time is obtained according to the historical daylight time data;
[0096] Arranging the historical power generation data into a historical power generation sequence in chronological order;
[0097] Construct a power generation prediction model based on multivariate linear regression, with wind power potential index, photovoltaic potential index, daily average temperature, and daily average humidity as input, and wind and solar power generation for one day as output; perform model training based on the historical power generation data and historical meteorological data, use the least square method or other optimization algorithms to perform parameter estimation, and save the trained power generation prediction model; the daily average temperature is obtained according to the historical temperature data, and the daily average humidity is obtained according to the historical humidity data;
[0098] Obtain future meteorological data, including light intensity data, daylight duration data, wind speed data, temperature data, and humidity data, calculate power generation potential indicators and input them into the power generation prediction model, obtain power generation prediction values and organize them into a power generation prediction sequence; future meteorological data can be obtained through a variety of channels, mainly including meteorological department websites, commercial meteorological service providers, open source meteorological data platforms, etc.
[0099] Wind and solar power generation is directly affected by a variety of meteorological factors, such as wind speed, solar radiation intensity, temperature, etc. The multivariate linear regression model can effectively capture the linear relationship between these meteorological factors and power generation. By introducing multiple meteorological parameters as independent variables, the amount of electricity generated can be predicted more accurately. In addition, the linear regression model has strong explanatory power, is easy to understand and implement, and is suitable as an effective means of power generation prediction.
[0100] The load forecasting method is as follows:
[0101] Arrange the preprocessed historical load data into a load sequence in chronological order;
[0102] Constructing a load forecasting model, wherein the load forecasting model is any one of a SARIMA model and a Holt-Winters model;
[0103] Training and testing the load forecasting model based on the load sequence;
[0104] The trained load forecasting model is saved and load forecasting is performed. Each time a load forecast is performed, an average power load forecast value for one day is obtained, and the average power load forecast value is added to the load sequence for the next load forecast. The load forecast is performed iteratively, and a load forecast sequence consisting of load forecast values is intercepted from the load sequence.
[0105] Load data has obvious time series characteristics, such as seasonality, trend and periodicity, and is suitable for processing by time series forecasting models. Time series forecasting models can capture the law of load changes over time, including daily periodicity, the difference between weekdays and weekends, and seasonal changes. In addition, time series models perform well in processing continuous data forecasts, which helps to accurately predict future load demand.
[0106] Reference Figure 3 , the method for electricity price prediction is as follows:
[0107] Arrange the historical electricity price data into a historical electricity price sequence in chronological order, add a timestamp tag to each historical electricity price data in the historical electricity price sequence, encode the timestamp tag, and form a timestamp tag sequence;
[0108] The timestamp label is a label of the date corresponding to the historical electricity price data, including working days, weekends, and statutory holidays;
[0109] Time-aligning the historical electricity price sequence, the historical power generation sequence, the load sequence, and the timestamp tag sequence;
[0110] Constructing an electricity price prediction model; the electricity price prediction model is any one of a recurrent neural network, a long short-term memory network, and a gated recurrent unit; the input is a label sequence consisting of a historical power generation sequence, a load sequence, and a timestamp label, and the output is an electricity price prediction sequence;
[0111] Performing training, verification, testing and tuning of the electricity price prediction model, and saving the trained electricity price prediction model;
[0112] A future timestamp label sequence is constructed and input into the electricity price prediction model together with the power generation prediction sequence and the load prediction sequence to obtain an electricity price prediction sequence.
[0113] S3: Determine the amount of storage capacity based on the historical storage data and the result of the power data prediction;
[0114] The storage capacity is the storage capacity of the multi-source energy storage system for the next day; the method for determining the storage capacity is as follows:
[0115] The historical power storage data is counted and averaged, recorded as ;
[0116] Based on the power generation forecast sequence, the mean value of the power generation data for the next n days is calculated and recorded as , n is a positive integer; the mean of the load data for the next n days is calculated based on the load forecast sequence, and is recorded as ; Based on the electricity price forecast sequence, calculate the mean of the electricity price data for the next n days, denoted as ;
[0117] The formula for calculating the power storage capacity is as follows:
[0118] ;
[0119] Among them, S represents the storage capacity; Indicates the power generation data of the previous day; Indicates the load data of the previous day; Indicates the electricity price data of the previous day; , , They are all weight coefficients and are set by technicians in this field according to actual needs.
[0120] If the price of electricity is going to rise in the next few days, the above formula tends to increase the current storage capacity and store more energy in advance to reduce the future storage cost; if the power generation of renewable energy will increase significantly in the next few days, the multi-source energy storage system will reduce the storage capacity before the predicted high power generation period, leaving enough space to absorb the upcoming excess renewable energy and avoid wasting renewable resources due to full load of energy storage devices. If the peak electricity consumption period will come in the next few days (such as increased cooling demand in summer), the storage capacity of the energy storage system will be increased to ensure that there is enough reserve energy to cope with the surge in load demand when the peak period arrives.
[0121] S4: Count the historical discharge data to generate a discharge sequence; the method is as follows:
[0122] Count the historical discharge data, divide a day into m time slices, where m is a positive integer; calculate the mean of the historical discharge data of each time slice in any day, and record the mean of the historical discharge data of the jth time slice on the qth day as ; Calculate the expected discharge amount for the jth time slice in the future day , the formula is as follows:
[0123] ;
[0124] Where Q is the total number of days involved in the historical discharge data;
[0125] Will The discharge sequence is organized in chronological order, and the value range of j is 1, 2, ..., m.
[0126] S5: Based on the storage amount and discharge sequence, perform energy storage strategy planning for the multi-source energy storage system.
[0127] The energy storage strategy is the charging power of each energy storage device in each time slice;
[0128] The energy storage devices include electrochemical energy storage devices, mechanical energy storage devices, thermal energy storage devices, chemical energy storage devices, and supercapacitors; among them, electrochemical energy storage devices such as lithium-ion batteries, sodium-sulfur batteries, lead-acid batteries, etc. are suitable for applications with fast response and high energy density; mechanical energy storage devices such as pumped storage, compressed air energy storage, and flywheel energy storage are suitable for large-scale, long-term energy storage; thermal energy storage devices store energy by heating or cooling media, and are suitable for heating, cooling or industrial processes; chemical energy storage devices such as hydrogen energy storage produce hydrogen through electrolysis of water, store energy, and convert it back into electrical energy through fuel cells when needed; supercapacitors provide extremely high power output and are suitable for occasions that require rapid charging and discharging.
[0129] The method for energy storage strategy planning is as follows:
[0130] Collect the current storage capacity of each energy storage device, and set the minimum energy storage threshold, maximum energy storage threshold, and maximum charging power for each energy storage device; obtain the electricity purchase price for each time slice in the future day;
[0131] Construct an energy storage strategy model, including an objective function, constraints, and a state update equation; wherein the objective function calculation formula is as follows:
[0132] ;
[0133] Where L represents the objective function; represents the charging power of the i-th energy storage device at the j-th time slice; represents the price of electricity purchased by the multi-source energy storage system from the grid in the jth time slice; represents the time corresponding to the midpoint of the j-th time slice; Indicates the peak load period; represents the charging power of the supercapacitor in the jth time period; represents the charging power of the electrochemical energy storage device in the jth time period; represents the storage capacity of the thermal energy storage device at the end of the j-1th time slice; Indicates the duration of any time slice; represents the bonus coefficient of the supercapacitor; represents the bonus coefficient of the electrochemical energy storage device; The heat loss coefficient of the thermal energy storage device indicates the amount of heat loss per unit of energy stored in the thermal energy storage device per unit of time; , , They are all set by technicians in this field according to actual needs; the calculation of the objective function is a dimensionless operation.
[0134] The above objective function involves the total electricity purchase cost, supercapacitor fast response reward, electrochemical energy storage device cycle life cost, and thermal loss of thermal energy storage device. It fully considers the characteristics of each energy storage device and can not only meet the overall energy storage demand but also give full play to the advantages of each energy storage device.
[0135] The constraints are as follows:
[0136] ;
[0137] in, represents the storage capacity of the i-th energy storage device in the j-th time slice; represents the minimum energy storage threshold of the i-th energy storage device; represents the maximum energy storage threshold of the i-th energy storage device; Represents the maximum charging power of the i-th energy storage device.
[0138] The above constraints respectively constrain the total storage capacity, the discharge demand in each time slice, the storage capacity and charging power of each energy storage device, which can minimize the life loss of the energy storage device while meeting the energy storage and discharge requirements.
[0139] The state update equation is as follows:
[0140] ;
[0141] in, It represents the storage capacity of the i-th energy storage device in the j+1-th time slice.
[0142] The energy storage strategy model is solved by using an optimization algorithm or a computing tool (such as a MILP solver, etc.) to obtain an energy storage strategy that minimizes the objective function.
[0143] Example 2
[0144] This embodiment is the second embodiment of the present invention; it is based on the same inventive concept as embodiment 1, Figure 4 This embodiment introduces a centralized control system for a multi-source energy storage system, including a data acquisition module, a power prediction module, a power storage planning module, a discharge planning module, a power storage solution module, and a monitoring display module; wherein:
[0145] The data acquisition module is used to collect historical data, meteorological data, energy storage device status and perform data cleaning; the historical data includes historical electricity price data, historical power generation data, historical load data, historical power storage data, and historical discharge data; the energy storage device status includes the current storage capacity of each energy storage device;
[0146] The power prediction module is used to perform power price prediction, power generation prediction, and load prediction, and generate a power generation prediction sequence, a load prediction sequence, and a power price prediction sequence;
[0147] The power storage planning module is used to calculate the power storage capacity for the next day; the power storage capacity is calculated by combining historical power storage data with power forecast sequence, load forecast sequence, and electricity price forecast sequence;
[0148] The discharge planning module is used to generate a discharge sequence; the discharge sequence is calculated based on historical discharge data;
[0149] The power storage solution module is used to plan the energy storage strategy; considering the state of the energy storage device, the characteristics of each energy storage device and the total cost of purchasing electricity from the power grid, the energy storage strategy with the lowest cost and the smallest energy storage device loss is formulated;
[0150] The monitoring and display module is used to visualize the energy storage device status, energy storage strategy, and discharge sequence.
[0151] The specific functions of the above modules are realized by referring to the relevant contents of the centralized control method of the multi-source energy storage system described in Example 1, which will not be described in detail.
[0152] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0153] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also make many forms without departing from the purpose and scope of protection of the present invention, which are all within the protection of the present invention.
Claims
1. A centralized control method for a multi-source energy storage system, characterized in that: The following steps are involved: S1: Collect historical data and perform data cleaning; the historical data includes historical electricity price data, historical power generation data, historical load data, historical power storage data, and historical discharge data; S2: Performing power data forecasting based on the historical data, wherein the power data forecasting includes power price forecasting, power generation forecasting, and load forecasting; S3: Determine the amount of storage capacity based on the historical storage data and the result of the power data prediction; The storage capacity is the storage capacity of the multi-source energy storage system for the next day; the method for determining the storage capacity is as follows: The historical power storage data is counted and averaged, recorded as ; Based on the power generation forecast sequence, the mean of the power generation data for the next n days is calculated and recorded as , n is a positive integer; the mean of the load data for the next n days is calculated based on the load forecast sequence, denoted as ; Calculate the mean of the electricity price data for the next n days based on the electricity price forecast sequence, denoted as ; The formula for calculating the power storage capacity is as follows: ; Among them, S represents the storage capacity; Indicates the power generation data of the previous day; Indicates the load data of the previous day; Indicates the electricity price data of the previous day; , , All are weight coefficients; S4: Counting the historical discharge data to generate a discharge sequence; S5: Based on the storage capacity and discharge sequence, perform energy storage strategy planning for the multi-source energy storage system; The energy storage strategy is the charging power of each energy storage device in each time slice; The energy storage device includes an electrochemical energy storage device, a mechanical energy storage device, a thermal energy storage device, a chemical energy storage device, and a supercapacitor; The method for energy storage strategy planning is as follows: Collect the current storage capacity of each energy storage device, and set the minimum energy storage threshold, maximum energy storage threshold, and maximum charging power for each energy storage device; obtain the electricity purchase price for each time slice in the future day; Construct an energy storage strategy model, including objective function, constraints, and state update equations; An optimization algorithm is used to solve the energy storage strategy model to obtain an energy storage strategy that minimizes the objective function.
2. A centralized control method for a multi-source energy storage system according to claim 1, characterized in that: The method for power generation prediction is as follows: Acquire historical meteorological data and perform data cleaning and normalization processing; the historical meteorological data includes historical light intensity data, historical daylight time data, historical wind speed data, historical temperature data, and historical humidity data; The power generation potential index, including wind power potential index and photovoltaic potential index, is calculated as follows: ; in, It represents the wind power potential index; represents the wind power potential factor; v represents the daily average wind speed; ; in, It represents the photoelectric potential index; represents the photoelectric potential factor; G represents the average daily light intensity; T represents the average daily light time; Arranging the historical power generation data into a historical power generation sequence in chronological order; Constructing a power generation prediction model based on multivariate linear regression, with wind power potential index, photovoltaic potential index, daily average temperature, and daily average humidity as input, and wind and solar power generation for one day as output; training the model based on the historical power generation data and historical meteorological data, using an optimization algorithm for parameter estimation, and saving the trained power generation prediction model; Obtain future meteorological data, including light intensity data, daylight duration data, wind speed data, temperature data, and humidity data, calculate power generation potential indicators and input them into the power generation prediction model, obtain power generation prediction values and organize them into power generation prediction sequences.
3. A centralized control method for a multi-source energy storage system according to claim 2, characterized in that: The load forecasting method is as follows: Arrange the preprocessed historical load data into a load sequence in chronological order; Constructing a load forecasting model, wherein the load forecasting model is any one of a SARIMA model and a Holt-Winters model; Training and testing the load forecasting model based on the load sequence; The trained load forecasting model is saved and load forecasting is performed. Each time a load forecast is performed, an average power load forecast value for the day is obtained, which is added to the load sequence for the next load forecast. The load forecast is performed iteratively, and a load forecast sequence consisting of load forecast values is intercepted from the load sequence.
4. A centralized control method for a multi-source energy storage system as claimed in claim 3, characterized in that: The method for electricity price prediction is as follows: Arrange the historical electricity price data into a historical electricity price sequence in chronological order, add a timestamp tag to each historical electricity price data in the historical electricity price sequence, encode the timestamp tag to form a timestamp tag sequence; the timestamp tag is a tag of the date corresponding to the historical electricity price data, including working days, weekends, and statutory holidays; Time-aligning the historical electricity price sequence, the historical power generation sequence, the load sequence, and the timestamp tag sequence; Constructing an electricity price prediction model; the electricity price prediction model is any one of a recurrent neural network, a long short-term memory network, and a gated recurrent unit; the input is a label sequence consisting of a historical power generation sequence, a load sequence, and a timestamp label, and the output is an electricity price prediction sequence; Performing training, verification, testing and tuning of the electricity price prediction model, and saving the trained electricity price prediction model; A future timestamp tag sequence is constructed and input into the electricity price prediction model together with the power generation prediction sequence and the load prediction sequence to obtain an electricity price prediction sequence.
5. A centralized control method for a multi-source energy storage system according to claim 4, characterized in that: The method for generating a discharge sequence is as follows: Count the historical discharge data, divide a day into m time slices, where m is a positive integer; calculate the mean of the historical discharge data of each time slice in any day, and record the mean of the historical discharge data of the jth time slice on the qth day as ; Calculate the expected discharge amount for the jth time slice in the future day , the formula is as follows: ; Wherein, Q is the total number of days involved in the historical discharge data; the value range of q is 1, 2, ..., Q; Will The discharge sequence is organized in chronological order, and the value range of j is 1, 2, ..., m.
6. A centralized control method for a multi-source energy storage system as claimed in claim 5, characterized in that: The objective function calculation formula is as follows: ; Wherein, L represents the objective function; i represents the number of the energy storage device; represents the charging power of the i-th energy storage device at the j-th time slice; represents the price of electricity purchased by the multi-source energy storage system from the grid in the jth time slice; represents the time corresponding to the midpoint of the j-th time slice; Indicates the peak load period; represents the charging power of the supercapacitor in the jth time period; represents the charging power of the electrochemical energy storage device in the jth time period; represents the storage capacity of the thermal energy storage device at the end of the j-1th time slice; represents the bonus coefficient of the supercapacitor; represents the bonus coefficient of the electrochemical energy storage device; is the heat loss coefficient of the thermal energy storage device, which indicates the amount of heat loss per unit of energy stored in the thermal energy storage device per unit of time; Indicates the duration of any time slice.
7. A centralized control method for a multi-source energy storage system according to claim 6, characterized in that: The constraints are as follows: ; in, represents the storage capacity of the i-th energy storage device in the j-th time slice; represents the minimum energy storage threshold of the i-th energy storage device; represents the maximum energy storage threshold of the i-th energy storage device; represents the maximum charging power of the i-th energy storage device; The state update equation is as follows: ; in, It represents the storage capacity of the i-th energy storage device in the j+1-th time slice.
8. A centralized control system for a multi-source energy storage system, which is used to implement a centralized control method for a multi-source energy storage system according to any one of claims 1 to 7, characterized in that: It includes data acquisition module, power forecasting module, power storage planning module, discharge planning module, power storage solution module and monitoring display module; The data acquisition module is used to collect historical data, meteorological data, energy storage device status and perform data cleaning; The power prediction module is used to perform power price prediction, power generation prediction, and load prediction based on historical data; The power storage planning module is used to calculate the power storage capacity for the next day based on the power price forecast, power generation forecast and load forecast; The discharge planning module is used to generate a discharge sequence based on historical data; The power storage solution module is used to plan energy storage strategies based on the power storage capacity and discharge sequence; The monitoring and display module is used to visually display the state of the energy storage device, the energy storage strategy, and the discharge sequence.
