A method and system for discharging configuration of intelligent energy storage cabinet with natural wind redundant output

Abstract

The present invention belongs to the technical field of wind power generation control, and in particular, relates to a method and system for discharging configuration of an intelligent energy storage cabinet with natural wind redundant output. The method firstly obtains historical wind speed, wind direction and output power data, and constructs a wind-wind direction output change function and a predicted change function; secondly, constructs a power demand prediction function and combines it with the wind-wind direction output change function to obtain a wind output surplus-difference function; thirdly, constructs an energy storage cabinet storage control model based on a particle swarm algorithm, a fuzzy PID control algorithm and a wind output surplus-difference function, and trains it using wind data, load demand data and energy storage cabinet status data; configures the trained model into an energy storage cabinet control subsystem, and dynamically adjusts the charge and discharge power of the energy storage cabinet according to real-time predicted wind and load demand data and energy storage cabinet status data; the method improves the stability of wind power generation and reduces the phenomenon of load demand imbalance caused by wind fluctuations.

Description

Technical Field

[0001] The present invention belongs to the technical field of wind power generation control, and in particular relates to a discharge configuration method and system for an intelligent energy storage cabinet with natural wind redundant output. Background Art

[0002] When a key component (such as a wind vane) fails, the existing wind turbine generator set usually adopts a single fault protection mechanism, that is, immediate shutdown, which leads to unnecessary downtime and power generation loss. In addition, redundant operation methods such as using historical data or adjacent wind turbine data have problems such as data lag and wake effect, making it difficult to accurately predict wind direction changes; at the same time, the application of existing energy storage systems in wind power generation is not intelligent enough, and most of them adopt fixed charging and discharging strategies, which cannot be dynamically adjusted to adapt to the volatility of wind power generation and changes in grid load. In addition, although there are many grid dispatching and balancing methods in the existing technology, most of them adopt multi-energy balancing, but these methods lack control accuracy when applied to the control of a single power generation type, making it difficult to achieve the balance of local microgrids.

[0003] For example, a Chinese patent application with publication number CN118412935A discloses a method for optimizing the stable output power of offshore wind-hydrogen coupled power generation. This method adjusts the power output power plan by considering the hydrogen inventory, the discrete degree and trend of wind power generation power changes when transmitting electricity to the power grid through wind power generation, the electricity used to produce hydrogen, and the power of hydrogen power generation.

[0004] The above prior art has the following problems: due to the volatility of wind force and wind direction, the existing wind power generation produces output redundancy or insufficient output, resulting in an imbalance in the power grid. It is still very difficult to accurately balance the fluctuations generated by wind power generation by controlling the storage and discharge time points and corresponding power output of the energy storage cabinet. To this end, the present invention provides a discharge configuration method and system for an intelligent energy storage cabinet with natural wind redundant output. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a method and system for discharging configuration of an intelligent energy storage cabinet with natural wind redundant output. The method first obtains historical wind speed, wind direction and output power data, and constructs a wind-wind direction output change function and a predicted change function; secondly, a power demand prediction function is constructed and combined with the wind-wind direction output change function to obtain a wind output surplus-difference function; thirdly, an energy storage cabinet storage control model is constructed based on a particle swarm algorithm, a fuzzy PID control algorithm and a wind output surplus-difference function, and is trained using wind data, load demand data and energy storage cabinet status data; the trained model is configured in the energy storage cabinet control subsystem, and the charging and discharging power of the energy storage cabinet is dynamically adjusted according to the real-time predicted wind and load demand data and the energy storage cabinet status data; the present invention improves the stability of wind power generation and reduces the phenomenon of load demand imbalance caused by wind fluctuations.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for discharging and configuring an intelligent energy storage cabinet with natural wind redundant output, comprising:

[0008] S1. Obtain historical wind power generation wind speed, wind direction and corresponding output power time series data, and use the acquired data to construct a wind force-wind direction output change function and a wind force-wind direction output change curve and a wind force-wind direction prediction change function;

[0009] S2. Obtain the load demand data corresponding to the same historical time period in S1, construct a power demand prediction function and a corresponding power demand prediction change curve through a support vector machine, and align the wind force-wind direction output change function and the power demand prediction function in the time dimension to obtain a wind output surplus-difference function and a wind output surplus-difference curve;

[0010] S3. Constructing an energy storage cabinet storage control model based on a particle swarm algorithm and an enhanced control algorithm, and embedding a wind output surplus-difference function and a wind output surplus-difference curve into the energy storage cabinet storage control model. Using wind speed, wind direction, wind power corresponding to output power, load demand data, and energy storage cabinet status data, the energy storage cabinet storage control model is trained to obtain a trained energy storage cabinet storage control model.

[0011] S4, configuring the trained energy storage cabinet storage control model into the energy storage cabinet control subsystem, using the wind force-wind direction prediction change function and the power demand prediction function to predict and obtain the corresponding wind force data and load demand data in real time, and monitor the energy storage cabinet status data in real time;

[0012] S5. Feedback the predicted wind power data, load demand data and energy storage cabinet status data to the energy storage cabinet storage control model, pre-configure the charge and discharge status of the energy storage cabinet, so that the power value stored or configured in real time by the energy storage cabinet is equal to the surplus or missing power value of wind power generation, and feed back the stored or configured power data to the wind power output surplus-difference curve for adjustment, so that the adjusted wind power output surplus-difference curve coincides with the x-axis in the coordinate system corresponding to the wind power output surplus-difference curve in real time.

[0013] Specifically, the steps of constructing the wind power output surplus-difference function and the wind power output surplus-difference curve include:

[0014] S201. Obtain wind speed sequence , the angle sequence between the wind direction and the normal of the blade plane , Wind power output series , electricity price and wind turbine operating cost per unit time, and pre-process the acquired data, including: , T is the window length of the input sequence;

[0015] S202: When the electricity price at the current moment is set to be greater than the operating cost of the wind turbine generator at the corresponding time point, the corresponding wind speed sequence and angle sequence are valid wind power generation sequences, otherwise they are invalid wind power generation sequences. The valid wind power generation input sequence is constructed based on the valid wind power generation sequence and the invalid wind power generation sequence. and invalid wind power input sequence , represent the wind speed, angle and wind power output value after dimensioning in the effective wind power input sequence, They represent the wind speed, angle and wind output power values ​​after dimensioning in the invalid wind power generation input sequence in turn;

[0016] S203, and The effective output prediction sub-model and the invalid output prediction sub-model inputted into the output prediction model are trained to obtain the trained effective output prediction sub-model and the invalid output prediction sub-model, and the effective wind force-wind direction output change function and the invalid wind force-wind direction output change function fitted by the radial basis function are built into the effective output prediction sub-model and the invalid output prediction sub-model respectively, which are specifically:

[0017] ;

[0018] in, represents the wind force-wind direction output variation function, represents the electricity price at time t, represents the wind turbine operating cost at time t, and They represent the minimum wind speed threshold and the maximum safe wind speed threshold for wind power generation, respectively. and They represent the effective wind force-wind direction output variation function and the invalid wind force-wind direction output variation function fitted by radial basis function respectively. Represents the modulo operation.

[0019] Specifically, the steps of constructing the wind power output surplus-difference function and the wind power output surplus-difference curve further include:

[0020] S204. Obtaining the load demand power sequence of the same time dimension as S201 , And the power demand prediction function is obtained by support vector machine training and fitting ;

[0021] S205, according to 、The remaining power of the energy storage cabinet at the current time t and The wind power output surplus-difference function is constructed as follows:

[0022] ;

[0023] in, Indicates the corresponding charging-discharging power value of the energy storage cabinet under effective wind power generation state, Indicates the corresponding power generation value of the wind turbine in the invalid wind power generation state;

[0024] S206. Construct a two-dimensional coordinate system with time as the horizontal coordinate and power as the vertical coordinate, align the wind force-wind direction output variation function and the wind force output surplus-difference function in time, and construct the corresponding wind force-wind direction output variation curve and wind force output surplus-difference curve in the two-dimensional coordinate system.

[0025] Specifically, the energy storage cabinet storage control model includes a parameter optimization sub-model and a parameter control sub-model; the steps of constructing the parameter optimization sub-model include:

[0026] S301. Based on the wind power output surplus-difference function, the input of the parameter optimization sub-model is constructed as follows: , the output is ,in, and In order to indicate the effective wind power generation state is the power value released by the timing energy storage cabinet and When it is negative, it is the stored power value corresponding to the energy storage cabinet; Indicates the power value of the wind power output configuration when the remaining power of the energy storage cabinet at time t is less than the load demand in the invalid wind power generation state; and are equal in value;

[0027] S302, constructing a parameter optimization sub-model fitness function based on the wind power output surplus-difference function and the input and output of the parameter optimization sub-model;

[0028] S303, setting the maximum number of iterations , the constructed parameter optimization sub-model input and output and fitness function are input into the particle swarm algorithm for iterative training. After N rounds of iterative training, it is equal to 0 or the number of iterations is When the search output is The power value stored or configured in the energy storage cabinet corresponding to the effective or invalid wind power generation state at the current time t is, Configure the energy storage cabinet with the adjusted wind output surplus-difference.

[0029] The fitness function of the parameter optimization sub-model is specifically:

[0030] .

[0031] Specifically, the steps of constructing the parameter control sub-model include:

[0032] S304: construct the current input state of the parameter control submodel based on the acquired wind power data, load demand data and parameter optimization submodel search output data ;

[0033] S305, setting the current moment action execution trigger information according to the current moment input state, including:

[0034] When the parameter optimization submodel search output is When the first execution trigger information is triggered, the energy storage cabinet is controlled to perform a discharge operation, and the specific discharge value is , so that Real time is equal to 0;

[0035] When the parameter optimization submodel search output is When the second execution trigger information is triggered, the energy storage cabinet is controlled to perform charging configuration operation. The specific value of the charging configuration is , so that Real time is equal to 0;

[0036] When the parameter optimization submodel search output is When the third execution trigger information is triggered, the wind power generation device is controlled to perform the discharge configuration operation. The specific value of the discharge configuration is , so that = and Real time is equal to 0;

[0037] When the wind power generation is ineffective and Greater than When , the fourth execution trigger information is triggered, that is, the energy storage cabinet is directly controlled to discharge to meet the load demand, so that the energy storage cabinet discharge power value is equal to ;

[0038] S306: Constructing the current moment execution action based on the current moment action execution trigger information ,in, Indicates triggering the execution action corresponding to the first execution trigger information. Indicates triggering the execution action corresponding to the second execution trigger information. Indicates triggering the execution action corresponding to the third execution trigger information. Indicates triggering the execution action corresponding to the fourth execution trigger information.

[0039] An intelligent energy storage cabinet discharge configuration system for natural wind redundant output, comprising: a data processing module, a wind power output module, a demand prediction module and an output configuration module;

[0040] The data processing module is used for real-time collection and preprocessing of wind speed, wind direction and corresponding output power time series data of wind turbines, load demand data and working status data of energy storage cabinets;

[0041] The wind power output module includes a wind power output unit and a wind power prediction unit;

[0042] The wind power output unit constructs a wind power-wind direction output variation function and a wind power-wind direction output variation curve using the pre-processed wind power generation wind speed, wind direction and corresponding output power time series data;

[0043] The wind force prediction unit constructs a wind force-wind direction prediction change function using pre-processed historical wind power generation wind speed and wind direction data;

[0044] The demand forecasting module uses the pre-processed load demand data to construct a power demand forecasting function and a corresponding power demand forecasting change curve through a support vector machine.

[0045] Specifically, the output configuration module includes a surplus function unit, a model training unit, a prediction unit and an output configuration unit;

[0046] The surplus-difference function unit obtains a wind output surplus-difference function and a wind output surplus-difference curve by using the constructed wind force-wind direction output variation function and the power demand prediction function;

[0047] The model training unit uses a particle swarm algorithm and an enhanced control algorithm to construct an energy storage cabinet storage control model, and embeds a wind power output surplus-difference function and a wind power output surplus-difference curve into the energy storage cabinet storage control model, uses wind power data and load demand data to obtain energy storage cabinet state data, trains the energy storage cabinet storage control model, and obtains a trained energy storage cabinet storage control model;

[0048] The prediction unit predicts and obtains corresponding wind power data and load demand data in real time through the wind power-wind direction prediction change function and the power demand prediction function, and monitors the energy storage cabinet status data in real time;

[0049] The output configuration unit uses the predicted wind data, load demand data and monitored energy storage cabinet status data to pre-configure the charging and discharging status of the energy storage cabinet through the energy storage cabinet storage control model, so that the power value stored or configured in real time by the energy storage cabinet is equal to the surplus or missing power value of wind power generation, and feeds back the stored or configured power data to the wind power output surplus-gap curve and adjusts it.

[0050] A computer-readable storage medium stores computer instructions, which, when executed, execute the steps of a method for discharging a smart energy storage cabinet with redundant output of natural wind power.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] In view of the deficiencies in the prior art, the present invention realizes accurate prediction of wind power output through high-precision wind force-wind direction output variation function and power demand prediction function, combined with wind force-wind direction prediction variation function; secondly, by constructing wind power output surplus-difference function and curve, the difference between wind power generation and grid load can be evaluated in real time; then, based on particle swarm algorithm, enhanced control algorithm and energy storage cabinet storage control model of wind power output surplus-difference function and curve, the wind power surplus-difference configuration strategy of the energy storage cabinet can be dynamically adjusted to ensure that the energy storage cabinet stores electric energy when wind power generation is surplus and releases electric energy when wind power generation is insufficient; finally, by feeding back the predicted wind force and load demand and energy storage cabinet status data to the energy storage cabinet storage control model, real-time and accurate configuration of wind power surplus-difference is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of a method for discharging a smart energy storage cabinet with natural wind redundant output according to Embodiment 1 of the present invention;

[0054] Figure 2 This is a principle flow chart of a method for discharging a smart energy storage cabinet with natural wind redundant output according to Embodiment 1 of the present invention;

[0055] Figure 3This is a module diagram of a discharge configuration system for an intelligent energy storage cabinet with natural wind redundant output according to Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0056] Example 1

[0057] See also Figure 1-Figure 2 , an embodiment provided by the present invention: a method for discharging configuration of an intelligent energy storage cabinet with natural wind redundant output, the steps comprising:

[0058] S1. Obtain historical wind power generation wind speed, wind direction and corresponding output power time series data, and use the acquired data to construct a wind force-wind direction output change function and a wind force-wind direction output change curve and a wind force-wind direction prediction change function;

[0059] Furthermore, the wind force-wind direction prediction change function in this embodiment is constructed by using a logistic regression algorithm and data obtained from weather forecasts;

[0060] Furthermore, in this embodiment, the wind direction prediction is obtained by simulation prediction through a three-dimensional wind simulation prediction algorithm; the three-dimensional simulation prediction algorithm in this embodiment includes a WRF algorithm, an ECMWF algorithm, a GFS algorithm or a numerical weather forecast algorithm;

[0061] S2. Obtain the load demand data corresponding to the same historical time period in S1, construct a power demand prediction function and a corresponding power demand prediction change curve through a support vector machine, and align the wind force-wind direction output change function and the power demand prediction function in the time dimension to obtain a wind output surplus-difference function and a wind output surplus-difference curve;

[0062] Furthermore, in this embodiment, the steps of constructing the wind power output surplus-difference function and the wind power output surplus-difference curve include:

[0063] S201. Obtain wind speed sequence , the angle sequence between the wind direction and the normal of the blade plane , Wind power output series , electricity price and wind turbine operating cost per unit time, and pre-process the acquired data, including: , T is the window length of the input sequence;

[0064] S202: When the electricity price at the current moment is set to be greater than the operating cost of the wind turbine generator at the corresponding time point, the corresponding wind speed sequence and angle sequence are valid wind power generation sequences, otherwise they are invalid wind power generation sequences. The valid wind power generation input sequence is constructed based on the valid wind power generation sequence and the invalid wind power generation sequence. and invalid wind power input sequence , represent the wind speed, angle and wind power output value after dimensioning in the effective wind power input sequence, They represent the wind speed, angle and wind output power values ​​after dimensioning in the invalid wind power generation input sequence in turn;

[0065] S203, and The effective output prediction sub-model and the invalid output prediction sub-model inputted into the output prediction model are trained to obtain the trained effective output prediction sub-model and the invalid output prediction sub-model, and the effective wind force-wind direction output change function and the invalid wind force-wind direction output change function fitted by the radial basis function are built into the effective output prediction sub-model and the invalid output prediction sub-model respectively, which are specifically:

[0066] ;

[0067] in, represents the wind force-wind direction output variation function, represents the electricity price at time t, represents the wind turbine operating cost at time t, and They represent the minimum wind speed threshold and the maximum safe wind speed threshold for wind power generation, respectively. and They represent the effective wind force-wind direction output variation function and the invalid wind force-wind direction output variation function fitted by radial basis function respectively. Represents the modulo operation.

[0068] Furthermore, in this embodiment, the angle between the wind direction and the normal line of the blade plane is used to control the predicted wind output. According to the nature of wind power generation, when the wind direction is perpendicular to the direction of the normal line of the wind turbine blade plane, the corresponding wind power generation efficiency is the minimum. Therefore, in this embodiment, the cosine value of the angle between the wind direction and the normal line of the wind turbine blade plane and the corresponding wind speed are used to construct a wind power generation adjustment factor. and Adjustment is made so that when the angle between the wind direction and the normal line of the wind turbine blade plane is larger, the corresponding wind force adjustment factor is larger, and the predicted wind force output value is closer to the actual value; among them, after taking the modulus There is no dimension, just take the corresponding wind speed value;

[0069] Furthermore, in this embodiment, the prediction of the effective wind power generation sequence and the invalid wind power generation sequence is to minimize the output cost of wind power generation while using wind power generation for demand configuration. When the time period corresponding to the invalid wind power generation sequence is reached, the output of the energy storage cabinet is configured by wind power generation, which can not only transmit the electric energy stored in the energy storage cabinet to reduce the cost of electric energy storage, but also reduce the amount and time of wind power output, further reducing the cost output corresponding to wind power generation. When entering the time period corresponding to the effective wind power generation sequence and there is redundancy in the load demand of wind power generation, the redundant electric energy is stored in the energy storage cabinet for standby electric energy storage, thereby improving the utilization efficiency of electric energy.

[0070] S204. Obtaining the load demand power sequence of the same time dimension as S201 , And the power demand prediction function is obtained by support vector machine training and fitting ;

[0071] S205, according to 、The remaining power of the energy storage cabinet at the current time t and The wind power output surplus-difference function is constructed as follows:

[0072] ;

[0073] in, Indicates the corresponding charging-discharging power value of the energy storage cabinet under effective wind power generation state, Indicates the corresponding power generation value of the wind turbine in the invalid wind power generation state;

[0074] S206. Construct a two-dimensional coordinate system with time as the horizontal coordinate and power as the vertical coordinate, align the wind force-wind direction output variation function and the wind force output surplus-difference function in time, and construct the corresponding wind force-wind direction output variation curve and wind force output surplus-difference curve in the two-dimensional coordinate system.

[0075] This process achieves accurate matching and cost savings of wind power generation and load demand by constructing wind power output surplus-difference function and curve. Specifically, firstly, wind speed, wind direction and power data are preprocessed, and effective and invalid output prediction sub-models are trained respectively. Then, the wind power output surplus-difference function is constructed by combining load demand data and energy storage cabinet status data to accurately reflect the dynamic difference between wind power generation and load demand. By aligning the wind force-wind direction output change function and the wind power output surplus-difference function in the time dimension and generating the corresponding curve, real-time monitoring and adjustment of wind power generation fluctuations are achieved. The problem of grid imbalance caused by wind power generation fluctuations is effectively solved, the stability and reliability of the grid are improved, and the energy utilization efficiency is optimized. In addition, by distinguishing effective and invalid wind power generation sequences to construct different output prediction functions, the wind power output can be adjusted better according to the power generation cost and electricity price, thereby reducing the cost of wind power generation.

[0076] S3. Constructing an energy storage cabinet storage control model based on a particle swarm algorithm and an enhanced control algorithm, and embedding a wind output surplus-difference function and a wind output surplus-difference curve into the energy storage cabinet storage control model. Using wind speed, wind direction, wind power corresponding to output power, load demand data, and energy storage cabinet status data, the energy storage cabinet storage control model is trained to obtain a trained energy storage cabinet storage control model.

[0077] Furthermore, in this embodiment, the energy storage cabinet storage control model includes a parameter optimization sub-model and a parameter control sub-model; the steps of constructing the parameter optimization sub-model include:

[0078] S301. Based on the wind power output surplus-difference function, the input of the parameter optimization sub-model is constructed as follows: , the output is ,in, and In order to indicate the effective wind power generation state is the power value released by the timing energy storage cabinet and When it is negative, it is the stored power value corresponding to the energy storage cabinet; Indicates the power value of the wind power output configuration when the remaining power of the energy storage cabinet at time t is less than the load demand in the invalid wind power generation state; and are equal in value;

[0079] S302: Based on the wind power surplus-difference function and the input and output of the parameter optimization sub-model, a parameter optimization sub-model fitness function is constructed. The parameter optimization sub-model fitness function is specifically:

[0080] ;

[0081] S303, setting the maximum number of iterations , the constructed parameter optimization sub-model input and output and fitness function are input into the particle swarm algorithm for iterative training. After N rounds of iterative training, it is equal to 0 or the number of iterations is When the search output is The power value stored or configured in the energy storage cabinet corresponding to the effective or invalid wind power generation state at the current time t is, The adjusted wind output surplus-difference value is configured for the energy storage cabinet;

[0082] Furthermore, the steps of constructing the parameter control sub-model in this embodiment include:

[0083] S304: construct the current input state of the parameter control submodel based on the acquired wind power data, load demand data and parameter optimization submodel search output data ;

[0084] S305, setting the current moment action execution trigger information according to the current moment input state, including:

[0085] When the parameter optimization submodel search output is When the first execution trigger information is triggered, the energy storage cabinet is controlled to perform a discharge operation, and the specific discharge value is , so that Real time is equal to 0;

[0086] When the parameter optimization submodel search output is When the second execution trigger information is triggered, the energy storage cabinet is controlled to perform charging configuration operation. The specific value of the charging configuration is , so that Real time is equal to 0;

[0087] When the parameter optimization submodel search output is When the third execution trigger information is triggered, the wind power generation device is controlled to perform the discharge configuration operation. The specific value of the discharge configuration is , so that = and Real time is equal to 0;

[0088] When the wind power generation is ineffective and Greater than When , the fourth execution trigger information is triggered, that is, the energy storage cabinet is directly controlled to discharge to meet the load demand, so that the energy storage cabinet discharge power value is equal to ;

[0089] S306: Constructing the current moment execution action based on the current moment action execution trigger information ,in, Indicates triggering the execution action corresponding to the first execution trigger information. Indicates triggering the execution action corresponding to the second execution trigger information. Indicates triggering the execution action corresponding to the third execution trigger information. Indicates triggering the execution action corresponding to the fourth execution trigger information;

[0090] S307, constructing an energy storage cabinet storage control model based on the SAC algorithm and the PID algorithm, inputting the constructed current moment input state, current moment action execution trigger information and current moment execution action into the energy storage cabinet storage control model for training, and obtaining the trained energy storage cabinet storage control model;

[0091] S4, configuring the trained energy storage cabinet storage control model into the energy storage cabinet control subsystem, using the wind force-wind direction prediction change function and the power demand prediction function to predict and obtain the corresponding wind force data and load demand data in real time, and monitor the energy storage cabinet status data in real time;

[0092] Furthermore, in this embodiment, the wind force data includes wind speed data and angle data between the wind direction and the normal line of the blade plane;

[0093] S5. Feedback the predicted wind power data, load demand data and energy storage cabinet status data to the energy storage cabinet storage control model, pre-configure the energy storage cabinet charge and discharge status, so that the power value stored or configured in real time by the energy storage cabinet is equal to the surplus or missing power value of wind power generation, and feed back the stored or configured power data to the wind power output surplus-difference curve for adjustment, so that the adjusted wind power output surplus-difference curve coincides with the x-axis in the coordinate system corresponding to the wind power output surplus-difference curve in real time. Furthermore, the x-axis in the coordinate system corresponding to the wind power output surplus-difference curve is the time parameter axis.

[0094] This process achieves precise optimization and real-time adjustment of the storage control of the energy storage cabinet by constructing a wind power output surplus-difference function and curve, and combining the particle swarm algorithm and the enhanced control algorithm. Specifically, firstly, based on the wind power output surplus-difference function, a parameter optimization sub-model is constructed, and the particle swarm algorithm is used to optimize the charging and discharging power values ​​of the energy storage cabinet under effective and ineffective wind power generation states to ensure that the power output of the energy storage cabinet matches the surplus or missing power value of wind power generation. Among them, the fitness function of the parameter optimization sub-model takes into account the remaining power and load demand of the energy storage cabinet to ensure the rationality and feasibility of the optimization result. Secondly, based on the acquisition The wind data, load demand data and the output of the parameter optimization sub-model are used to build a parameter control sub-model. By setting different execution trigger information, real-time control of the charging and discharging status of the energy storage cabinet is achieved. Finally, the SAC algorithm and PID algorithm are used to build the energy storage cabinet storage control model. By training the model, it can automatically adjust the charging and discharging status of the energy storage cabinet according to the real-time predicted wind data and load demand data, ensuring that the wind output surplus-difference curve coincides with the x-axis in the coordinate system corresponding to the wind output surplus-difference curve in real time, thereby achieving the dynamic balance of the power grid and the real-time charging and discharging configuration of the energy storage cabinet corresponding to the power value at the corresponding time point.

[0095] Example 2

[0096] See also Figure 3 , another embodiment provided by the present invention: a smart energy storage cabinet discharge configuration system with natural wind redundant output, comprising: a data processing module, a wind power output module, a demand prediction module, an output configuration module and a visualization module;

[0097] Data processing module, used for real-time collection and preprocessing of wind turbine wind speed, wind direction and corresponding output power time series data, load demand data and energy storage cabinet working status data;

[0098] Wind output module, used for constructing wind output function and curve; wind output module includes wind output unit and wind prediction unit;

[0099] The wind power output unit uses the pre-processed wind power generation wind speed, wind direction and corresponding output power time series data to construct a wind force-wind direction output change function and a wind force-wind direction output change curve;

[0100] The wind prediction unit uses the pre-processed historical wind speed and wind direction data to construct a wind power-wind direction prediction change function;

[0101] The demand forecasting module uses the preprocessed load demand data and the support vector machine to construct a power demand forecasting function and a corresponding power demand forecasting change curve;

[0102] The output configuration module is used to construct an output surplus-difference function, and use the output surplus-difference function and the configured energy storage cabinet storage control model to adjust the wind power generation surplus-difference power value in real time; the output configuration module includes a surplus-difference function unit, a model training unit, a prediction unit and an output configuration unit;

[0103] The surplus-difference function unit obtains the wind output surplus-difference function and the wind output surplus-difference curve by using the constructed wind force-wind direction output variation function and the power demand prediction function;

[0104] The model training unit uses a particle swarm algorithm and an enhanced control algorithm to construct an energy storage cabinet storage control model, and embeds a wind power output surplus-difference function and a wind power output surplus-difference curve into the energy storage cabinet storage control model, uses wind power data and load demand data to obtain energy storage cabinet state data, trains the energy storage cabinet storage control model, and obtains a trained energy storage cabinet storage control model;

[0105] A prediction unit, used to predict and obtain corresponding wind power data and load demand data in real time through a wind power-wind direction prediction change function and a power demand prediction function, and to monitor energy storage cabinet status data in real time;

[0106] The output configuration unit uses the predicted wind power data, load demand data and monitored energy storage cabinet status data, and pre-configures the energy storage cabinet charging and discharging status through the energy storage cabinet storage control model, so that the power value stored or configured in real time by the energy storage cabinet is equal to the surplus or missing power value of wind power generation, and feeds back the stored or configured power data to the wind power output surplus-gap curve and adjusts it;

[0107] The visualization module is used to visualize the constructed wind force-wind direction output change curve, the power demand forecast change curve, and the wind output surplus-difference curve before and after adjustment, and to make the wind output surplus-difference curve coincide with the x-axis in real time.

[0108] Example 3

[0109] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, a method for discharging a smart energy storage cabinet with natural wind redundant output is implemented.

[0110] A computer-readable storage medium stores computer instructions, which, when executed, execute a method for discharging configuration of an intelligent energy storage cabinet with natural wind redundant output.

[0111] 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 modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are within the protection of the present invention.

[0112] If the disclosed technical solution involves personal information, the product using the disclosed technical solution has clearly informed the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the disclosed technical solution involves sensitive personal information, the product using the disclosed technical solution has obtained the individual's separate consent before processing the sensitive personal information, and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, clear and prominent signs are set to inform that the personal information collection scope has been entered and personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that he or she agrees to the collection of his or her personal information; or on the device that processes personal information, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload his or her personal information; among them, the personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. A method for discharging and configuring an intelligent energy storage cabinet with natural wind redundant output, characterized in that: include: S1. Obtain historical wind power generation wind speed, wind direction and corresponding output power time series data, and use the acquired data to construct a wind force-wind direction output change function and a wind force-wind direction output change curve and a wind force-wind direction prediction change function; S2. Obtain the load demand data corresponding to the same historical time period in S1, construct a power demand prediction function and a corresponding power demand prediction change curve through a support vector machine, and align the wind force-wind direction output change function and the power demand prediction function in the time dimension to obtain a wind output surplus-difference function and a wind output surplus-difference curve; S3. Constructing an energy storage cabinet storage control model based on a particle swarm algorithm and an enhanced control algorithm, and embedding a wind output surplus-difference function and a wind output surplus-difference curve into the energy storage cabinet storage control model. Using wind speed, wind direction, wind power corresponding to output power, load demand data, and energy storage cabinet status data, the energy storage cabinet storage control model is trained to obtain a trained energy storage cabinet storage control model. S4, configuring the trained energy storage cabinet storage control model into the energy storage cabinet control subsystem, using the wind force-wind direction prediction change function and the power demand prediction function to predict and obtain the corresponding wind force data and load demand data in real time, and monitor the energy storage cabinet status data in real time; S5, feeding back the predicted wind power data, load demand data and energy storage cabinet status data to the energy storage cabinet storage control model, pre-configuring the charge and discharge status of the energy storage cabinet, so that the power value stored or configured in real time by the energy storage cabinet is equal to the surplus or missing power value of wind power generation, and feeding back the stored or configured power data to the wind power output surplus-difference curve for adjustment, so that the adjusted wind power output surplus-difference curve coincides with the x-axis in the coordinate system corresponding to the wind power output surplus-difference curve in real time; The steps of constructing the wind power output surplus-difference function and the wind power output surplus-difference curve include: S201. Obtain wind speed sequence , the angle sequence between the wind direction and the normal of the blade plane , Wind power output series , electricity price and wind turbine operating cost per unit time, and pre-process the acquired data, including: , T is the window length of the input sequence; S202: When the electricity price at the current moment is set to be greater than the operating cost of the wind turbine generator at the corresponding time point, the corresponding wind speed sequence and angle sequence are valid wind power generation sequences, otherwise they are invalid wind power generation sequences. The valid wind power generation input sequence is constructed based on the valid wind power generation sequence and the invalid wind power generation sequence. and invalid wind power input sequence , represent the wind speed, angle and wind power output value after dimensioning in the effective wind power input sequence, They represent the wind speed, angle and wind output power values ​​after dimensioning in the invalid wind power generation input sequence in turn; S203, and The effective output prediction sub-model and the invalid output prediction sub-model inputted into the output prediction model are trained to obtain the trained effective output prediction sub-model and the invalid output prediction sub-model, and the effective wind force-wind direction output change function and the invalid wind force-wind direction output change function fitted by the radial basis function are built into the effective output prediction sub-model and the invalid output prediction sub-model respectively, which are specifically: ； in, represents the wind force-wind direction output variation function, represents the electricity price at time t, represents the wind turbine operating cost at time t, and They represent the minimum wind speed threshold and the maximum safe wind speed threshold for wind power generation, respectively. and They represent the effective wind force-wind direction output variation function and the invalid wind force-wind direction output variation function fitted by radial basis function respectively. Represents modulo operation; S204. Obtaining the load demand power sequence of the same time dimension as S201 , And the power demand prediction function is obtained by support vector machine training and fitting ; S205, according to 、The remaining power of the energy storage cabinet at the current time t and The wind power output surplus-difference function is constructed as follows: ； in, Indicates the corresponding charging-discharging power value of the energy storage cabinet under effective wind power generation state, Indicates the corresponding power generation value of the wind turbine in the invalid wind power generation state; S206, constructing a two-dimensional coordinate system with time as the horizontal coordinate and power as the vertical coordinate, aligning the wind force-wind direction output change function and the wind force output surplus-difference function in time, and constructing corresponding wind force-wind direction output change curve and wind force output surplus-difference curve in the two-dimensional coordinate system; The energy storage cabinet storage control model includes a parameter optimization sub-model and a parameter control sub-model; the steps of constructing the parameter optimization sub-model include: S301. Based on the wind power output surplus-difference function, the input of the parameter optimization sub-model is constructed as follows: , the output is ,in, and In order to indicate the effective wind power generation state is the power value released by the timing energy storage cabinet and When it is negative, it is the stored power value corresponding to the energy storage cabinet; Indicates the power value of the wind power output configuration when the remaining power of the energy storage cabinet at time t is less than the load demand in the invalid wind power generation state; and are equal in value; S302, constructing a parameter optimization sub-model fitness function based on the wind power output surplus-difference function and the input and output of the parameter optimization sub-model; S303, setting the maximum number of iterations , the constructed parameter optimization sub-model input and output and fitness function are input into the particle swarm algorithm for iterative training. After N rounds of iterative training, it is equal to 0 or the number of iterations is When the search output is The power value stored or configured in the energy storage cabinet corresponding to the effective or invalid wind power generation state at the current time t is, The adjusted wind output surplus-difference value is configured for the energy storage cabinet; The fitness function of the parameter optimization sub-model is specifically: ； The steps of constructing the parameter control sub-model include: S304: construct the current input state of the parameter control submodel based on the acquired wind power data, load demand data and parameter optimization submodel search output data ; S305, setting the current moment action execution trigger information according to the current moment input state, including: When the parameter optimization submodel search output is When the first execution trigger information is triggered, the energy storage cabinet is controlled to perform a discharge operation, and the specific discharge value is , so that Real time is equal to 0; When the parameter optimization submodel search output is When the second execution trigger information is triggered, the energy storage cabinet is controlled to perform charging configuration operation. The specific value of the charging configuration is , so that Real time is equal to 0; When the parameter optimization submodel search output is When the third execution trigger information is triggered, the wind power generation device is controlled to perform the discharge configuration operation. The specific value of the discharge configuration is , so that = and Real time is equal to 0; When the wind power generation is ineffective and Greater than When , the fourth execution trigger information is triggered, that is, the energy storage cabinet is directly controlled to discharge to meet the load demand, so that the energy storage cabinet discharge power value is equal to ; S306: Constructing the current moment execution action based on the current moment action execution trigger information ,in, Indicates triggering the execution action corresponding to the first execution trigger information. Indicates triggering the execution action corresponding to the second execution trigger information. Indicates triggering the execution action corresponding to the third execution trigger information. Indicates triggering the execution action corresponding to the fourth execution trigger information.

2. A discharge configuration system for an intelligent energy storage cabinet with natural wind redundant output, which is used to implement a discharge configuration method for an intelligent energy storage cabinet with natural wind redundant output as described in claim 1, characterized in that: include: Data processing module, wind power output module, demand forecasting module and output configuration module; The data processing module is used for real-time collection and preprocessing of wind speed, wind direction and corresponding output power time series data of wind turbines, load demand data and energy storage cabinet working status data; The wind power output module includes a wind power output unit and a wind power prediction unit; The wind power output unit constructs a wind power-wind direction output variation function and a wind power-wind direction output variation curve using the pre-processed wind power generation wind speed, wind direction and corresponding output power time series data; The wind force prediction unit constructs a wind force-wind direction prediction change function using pre-processed historical wind power generation wind speed and wind direction data; The demand forecasting module uses the pre-processed load demand data to construct a power demand forecasting function and a corresponding power demand forecasting change curve through a support vector machine.

3. The intelligent energy storage cabinet discharge configuration system for natural wind redundant output as claimed in claim 2, characterized in that: The output configuration module includes a surplus function unit, a model training unit, a prediction unit and an output configuration unit; The surplus-difference function unit obtains a wind output surplus-difference function and a wind output surplus-difference curve by using the constructed wind force-wind direction output variation function and the power demand prediction function; The model training unit uses a particle swarm algorithm and an enhanced control algorithm to construct an energy storage cabinet storage control model, and embeds a wind power output surplus-difference function and a wind power output surplus-difference curve into the energy storage cabinet storage control model, uses wind power data and load demand data to obtain energy storage cabinet state data, trains the energy storage cabinet storage control model, and obtains a trained energy storage cabinet storage control model; The prediction unit predicts and obtains corresponding wind power data and load demand data in real time through the wind power-wind direction prediction change function and the power demand prediction function, and monitors the energy storage cabinet status data in real time; The output configuration unit uses the predicted wind data, load demand data and monitored energy storage cabinet status data to pre-configure the charging and discharging status of the energy storage cabinet through the energy storage cabinet storage control model, so that the power value stored or configured in real time by the energy storage cabinet is equal to the surplus or missing power value of wind power generation, and feeds back the stored or configured power data to the wind power output surplus-gap curve and adjusts it.

Images