Power output adjusting method of distributed power supply

Abstract

The invention discloses a power output adjusting method for a distributed power supply, and relates to the technical field of distributed power supply control, and the method comprises the steps: collecting the historical feature information of a power grid, predicting the features of the future power grid, and analyzing the power demand interval of the distributed power supply and the fluctuation parameters of the power grid; and based on the demand interval and the fluctuation parameter, optimizing power output to reduce impact on the power grid, obtaining optimal power, and adjusting output of the distributed power supply according to the optimal power. The technical problem of insufficient power adaptability caused by slow adjustment response time of the existing distributed power supply under the condition of complex power grid demand fluctuation is solved, and the technical effect of improving the adaptability of the power output of the distributed power supply and the power grid demand change is achieved.

Description

Technical Field

[0001] The present application relates to the technical field of distributed power supply control, and in particular to a power output regulation method of a distributed power supply. Background Art

[0002] With the increasing popularity of distributed power sources in power grids, their power output regulation capabilities have put forward higher requirements for the stability of power grid operation. Distributed power sources, including hydrogen fuel cells, photovoltaic power generation and wind power generation, due to their dynamic response characteristics, may cause operation deviations or even system fluctuations when connected to the grid due to insufficient matching between power output and grid demand. Especially in the case of frequent grid load fluctuations, the power regulation of distributed power sources needs to consider both load changes and adapt to future grid demand characteristics. However, existing regulation methods are usually based on real-time or short-term forecast data, lack systematic analysis, and have limited adaptability to the definition of power output ranges and fluctuation characteristics, making it difficult to meet the operation requirements of complex power grids. This technical limitation poses challenges to the stability of the power grid and the efficient use of distributed power sources. It is urgent to improve the matching degree and operation reliability of distributed power sources and power grids by improving regulation methods.

[0003] In the current related technologies, there is a technical problem that distributed power sources have insufficient power adaptability due to slow adjustment response time when the grid demand fluctuates complexly. Summary of the invention

[0004] The present application solves the technical problem of insufficient power adaptability of existing distributed power sources due to slow regulation response time under complex grid demand fluctuations by providing a power output regulation method for distributed power sources.

[0005] The present application provides a power output regulation method for a distributed power source, including:

[0006] Collect the historical grid characteristic information sequence of the grid side within the preset time range in the past, predict the grid characteristic information within the preset time range in the future, and obtain the predicted grid characteristic information sequence; according to the predicted grid characteristic information sequence, perform power output demand analysis and grid fluctuation analysis of the distributed power source to obtain the power output demand range and grid fluctuation parameters; within the power output demand range, according to the grid fluctuation parameters, with the purpose of reducing the impact of the distributed power source on the grid and the power output adaptation degree, optimize the power output regulation of the distributed power source to obtain the optimal output power; use the optimal output power to adjust the output power of the distributed power source.

[0007] The present application also provides a power output regulation system for a distributed power source, including:

[0008] An information sequence acquisition module, the information sequence acquisition module is used to collect the historical power grid characteristic information sequence of the power grid side within the past preset time range, predict the power grid characteristic information within the future preset time range, and obtain the predicted power grid characteristic information sequence; a demand interval acquisition module, the demand interval acquisition module is used to perform power output demand analysis and power grid fluctuation analysis of the distributed power source according to the predicted power grid characteristic information sequence, and obtain the power output demand interval and power grid fluctuation parameters; an adjustment optimization module, the adjustment optimization module is used to optimize the power output adjustment of the distributed power source within the power output demand interval and according to the power grid fluctuation parameters, with the purpose of reducing the impact of the distributed power source on the power grid and the power output adaptation degree, and obtain the optimal output power; an output power adjustment module, the output power adjustment module is used to use the optimal output power to adjust the output power of the distributed power source.

[0009] The power output regulation method of distributed power sources proposed in this application first collects historical characteristic information of the power grid, predicts future power grid characteristics, and analyzes the power demand range and power grid fluctuation parameters of the distributed power sources; based on the demand range and fluctuation parameters, the power output is optimized to reduce the impact on the power grid, obtain the optimal power, and adjust the output of the distributed power sources accordingly, achieving the technical effect of improving the adaptability of the power output of the distributed power sources to changes in power grid demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0011] Figure 1 A schematic diagram of a flow chart of a method for adjusting power output of a distributed power source provided in an embodiment of the present application;

[0012] Figure 2 A schematic diagram of a process flow for obtaining a predicted power grid characteristic information sequence of a power output regulation method for a distributed power source provided in an embodiment of the present application;

[0013] Figure 3 A schematic diagram of the structure of a power output regulation system for a distributed power source provided in an embodiment of the present application.

[0014] Explanation of the reference numerals: information sequence acquisition module 10 , demand interval acquisition module 20 , adjustment optimization module 30 , output power adjustment module 40 . DETAILED DESCRIPTION

[0015] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0016] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0017] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0018] The present application provides a method for adjusting the power output of a distributed power source. Figure 1 As shown, the method includes:

[0019] Step S100, collect the historical grid characteristic information sequence of the grid side in the past preset time range, predict the grid characteristic information in the future preset time range, and obtain the predicted grid characteristic information sequence. Specifically, for the distributed power source connected to the grid side (taking hydrogen fuel cells as an example, its output power adjustment has time response characteristics), first start the data acquisition system, use a high-precision power sensor to collect grid demand power data every Δt time in the past time period of T1, and form a historical grid characteristic information sequence as the basic data for prediction. Then select a suitable prediction model based on the data characteristics and accuracy requirements. For example, when selecting the ARIMA model, first check the stability of the historical data. If it is not stable, differentiate it to make it stable, and then use the autocorrelation and partial autocorrelation function diagrams to determine the ARIMA model parameters, and then input the historical data into the training optimization model to grasp the law of power demand changes in the grid. The trained model is then used to predict the grid characteristic information sequence of the future duration T2. ​​After obtaining the prediction results, the root mean square error, mean absolute error and other evaluation indicators are used for comparison and analysis with some historical verification data. If the error is large, the model parameter settings, data processing process or model replacement are reviewed. Through continuous adjustment and optimization, the prediction results are ensured to be accurate and reliable, providing a basis for the power output adjustment of distributed power sources.

[0020] In one possible implementation, Figure 2 As shown, the historical grid characteristic information sequence of the grid side in the past preset time range is collected, and the grid characteristic information in the future preset time range is predicted to obtain the predicted grid characteristic information sequence. Step S100 further includes step S110, obtaining the adjustment response time of the distributed power supply for power regulation. Specifically, first, the adjustment response time of the distributed power supply for power regulation is obtained. For different types of distributed power supplies, such as hydrogen fuel cells, wind turbines or photovoltaic cells, the adjustment response time will vary due to factors such as their own physical properties, control mechanisms, and connected power conversion equipment. For example, hydrogen fuel cells may have a power regulation response time between several seconds and tens of seconds due to their chemical reaction process, the response speed of the gas supply system, and the dynamic characteristics of the power electronic converter. Through special testing equipment and experimental methods, the response of the distributed power supply under different power regulation instructions is monitored and analyzed, and the time from issuing the power regulation instruction to the actual power output reaching a stable state is recorded. The average value is taken for multiple measurements to obtain a more accurate adjustment response time.

[0021] Step S120, calculate the sum of the adjustment response time and the preset time length as the preset time range. Specifically, after obtaining the adjustment response time of the distributed power source, calculate the sum of the time length and the preset time length to determine the preset time range. The setting of the preset time length is usually based on the empirical understanding of the law of change of the grid characteristic information and the comprehensive consideration of the prediction accuracy requirements. For example, if it is found based on the analysis of previous grid data that the grid demand power will show a more obvious change trend within a few minutes, and in order to ensure that the prediction results can cover a sufficiently long time span to adapt to the power adjustment plan of the distributed power source, the preset time length can be set to a few minutes. Assuming that the adjustment response time of the distributed power source is measured to be 10 seconds and the preset time length is set to 3 minutes (i.e. 180 seconds), the sum of the two, 190 seconds, is the preset time range. The preset time range will be used as the time window basis for collecting historical grid characteristic information sequences and predicting future grid characteristic information.

[0022] Step S130, collect the historical grid characteristic information sequence of the grid side in the past preset time range, predict the grid characteristic information in the future preset time range, and obtain the predicted grid characteristic information sequence. Specifically, according to the determined preset time range, comprehensively collect the historical grid characteristic information sequence of the grid side in the past preset time range, use high-precision power monitoring equipment, such as smart meters, power sensors, etc., to monitor and record the voltage, current, power and other parameters of the grid in real time, collect data at a certain time interval (such as every second or every few milliseconds), so as to construct a historical grid characteristic information sequence, and after accumulating enough historical data, use a suitable prediction method and model to predict the grid characteristic information in the future preset time range to obtain a predicted grid characteristic information sequence, for example, an ARIMA model based on time series analysis or a neural network model based on machine learning can be used. Taking the ARIMA model as an example, the stationarity of the historical power grid characteristic information sequence is first tested and processed to determine the model's parameters such as the autoregressive order, differential order, and moving average order. Then, the model is trained and optimized using historical data. Finally, the trained model is applied to predict the power grid characteristic information within a preset time range in the future to obtain the predicted power grid characteristic information sequence, which provides a key reference for the subsequent power output adjustment of distributed power sources, enabling them to adapt to the changing trend of the power grid in advance and achieve more efficient and stable grid-connected operation.

[0023] In a possible implementation, the historical power grid characteristic information sequence of the power grid side within the past preset time range is collected, and the power grid characteristic information within the future preset time range is predicted to obtain the predicted power grid characteristic information sequence. Step S130 further includes step S131, according to the preset time range, the power demand of the power grid side in the historical time is collected to obtain a set of sample historical power grid characteristic information sequences, and the power demand of the preset time range after each sample historical power grid characteristic information sequence is collected to obtain a set of sample predicted power grid characteristic information sequences. Specifically, according to the determined preset time range, the power demand of the power grid side in the historical time is collected, and for different time intervals, multiple sections of power grid demand data are sequentially extracted with the preset time range as the span, and each section of such data sequence constitutes a sample historical power grid characteristic information sequence. For example, if the preset time range is 10 minutes, a 10-minute power demand data is selected from the historical power grid data of the past hours or even days at a certain time interval (such as 30 minutes), so as to obtain a series of sample historical power grid characteristic information sequences, and the set of these sequences is the sample historical power grid characteristic information sequence set. After collecting each sample historical power grid characteristic information sequence, the demand power data of the same preset time range after the corresponding sequence is collected. These subsequent data sequences constitute the sample prediction power grid characteristic information sequence set. This step is like intercepting segments of "past" and "future" demand power data fragments in the long river of historical data with the preset time range as the scale, providing rich and logically related sample data for subsequent model training.

[0024] Step S132, using the sample historical power grid characteristic information sequence set and the sample predicted power grid characteristic information sequence set as supervised training data, and training the power grid characteristic prediction channel until the training converges. Specifically, after successfully obtaining the sample historical power grid characteristic information sequence set and the sample predicted power grid characteristic information sequence set, they are used as supervised training data for the training of the power grid characteristic prediction channel. The power grid characteristic prediction channel can be a neural network model built based on a deep learning architecture, such as a long short-term memory network (LSTM) or a convolutional neural network (CNN), etc., or it can be a traditional time series prediction model such as an autoregressive moving average (ARIMA) model. After the model structure is modified, during the training process, the sample historical power grid characteristic information sequence is used as the input of the model, and the corresponding sample predicted power grid characteristic information sequence is used as the expected output. The internal parameters of the model (such as weights and biases in the neural network) are continuously adjusted to minimize the difference between the predicted output and the expected output. Optimization algorithms such as stochastic gradient descent (SGD) or its variants Adagrad and Adadelta are used. Training continues until the model reaches a training convergence state. The basis for judging convergence is to set a threshold of a loss function (such as mean square error MSE). When the value of the loss function is less than the threshold, the model is considered to have converged, that is, the model has fully learned the changing laws and characteristics of the power demand of the power grid from the sample data, and can more accurately predict future power grid characteristic information based on the input historical data.

[0025] Step S133, input the historical power grid characteristic information sequence into the power grid characteristic prediction channel, and predict and output to obtain the predicted power grid characteristic information sequence. Specifically, when the power grid characteristic prediction channel training converges, the previously collected historical power grid characteristic information sequence is input into the channel. The model will perform in-depth analysis and processing on the input historical power grid characteristic information sequence based on the rules and characteristics learned during the training process, so as to predict and output the power grid characteristic information sequence within the preset time range in the future. The predicted output sequence is the predicted power grid characteristic information sequence we ultimately need, which will serve as the key basis for the subsequent distributed power supply power output adjustment. For example, if it is predicted that the power demand of the power grid will show an upward trend in the future, the distributed power supply can adjust its own power output strategy in advance, increase the output power to better match the power grid demand, and achieve efficient and stable grid-connected operation, while reducing the impact on the power grid and improving the power output adaptation degree.

[0026] Step S200, according to the predicted grid characteristic information sequence, the power output demand analysis and grid fluctuation analysis of the distributed power source are performed to obtain the power output demand interval and the grid fluctuation parameter. Specifically, for the obtained predicted grid characteristic information sequence, the power output demand analysis of the distributed power source is first performed, that is, the maximum and minimum required power values ​​are extracted from the sequence to construct the power output demand interval. The interval determines the power output range of the distributed power source to adapt to the grid demand. If it is determined that the future grid demand power fluctuates in a certain interval, the distributed power source needs to dynamically adjust the power output in this interval. Analyze the grid fluctuation, first calculate the mean of the predicted grid characteristic information sequence, then randomly extract multiple values ​​in the sequence to calculate its mean, and finally use the deviation percentage of the two as the grid fluctuation parameter. This parameter intuitively reflects the degree of power fluctuation of the grid demand. If the deviation is large, the distributed power source needs to respond flexibly to reduce the impact. If the deviation is small, it can be operated according to the stable strategy to improve efficiency.

[0027] In one possible implementation, based on the predicted grid characteristic information sequence, power output demand analysis and grid fluctuation analysis of the distributed power source are performed to obtain a power output demand range and grid fluctuation parameters. Step S200 further includes step S210, extracting the maximum demand power and the minimum demand power within the predicted grid characteristic information sequence, and constructing a power output demand range for the distributed power source to output power to the grid side. Specifically, the grid characteristic information sequence is analyzed and predicted, and the maximum demand power and the minimum demand power are accurately extracted from it. The extraction process requires detailed comparison and screening of each data point in the sequence. For example, when faced with a prediction sequence containing demand power data corresponding to many time nodes, the maximum and minimum values ​​are recorded by checking each data point one by one. It is assumed that the predicted grid characteristic information sequence covers the grid demand power data recorded every 1 minute in the next hour. After traversing these 60 data points, it is determined that the maximum demand power is 80kW and the minimum demand power is 30kW. These two extreme values ​​are used to construct the power output demand interval of the distributed power source to the grid side, that is, [30kW, 80kW]. The interval clearly defines the upper and lower limits of the power output of the distributed power source in the future period, so that it can reasonably adjust its own power output according to this interval to better match the needs of the grid, ensure the stability and reliability of the power supply, and avoid adverse effects on the operation of the grid due to excessive or low power output.

[0028] Step S220, perform grid fluctuation analysis based on the predicted grid characteristic information sequence to obtain grid fluctuation parameters. Specifically, after constructing the power output demand interval, perform grid fluctuation analysis based on the predicted grid characteristic information sequence, first calculate the mean of the predicted grid characteristic information sequence, sum all the demand power data in the sequence, and then divide it by the total number of data points. For example, for the above-mentioned prediction sequence containing 60 data points, add all the data and divide it by 60 to obtain the mean, and randomly extract multiple predicted grid characteristic information from the predicted grid characteristic information sequence. The number of extractions can be determined according to the actual situation and the analysis accuracy requirements, such as extracting 20 data points. The mean of 20 randomly selected data points is calculated, and the deviation percentage of the two means is calculated and used as the grid fluctuation parameter. Specifically, assuming that the sequence mean is μ and the mean of the randomly selected data points is μ', the grid fluctuation parameter δ=(μ'-μ) / μ*100%. The grid fluctuation parameter can intuitively reflect the degree of fluctuation of the grid demand power. If the deviation percentage is large, it indicates that the grid demand power fluctuates violently. Distributed power sources need to have stronger adaptability and flexibility in the power output adjustment process to cope with such fluctuations and reduce the impact on the grid. On the contrary, if the deviation percentage is small, it means that the grid demand power is relatively stable. Distributed power sources can follow a more stable strategy when outputting power, reducing unnecessary frequent adjustment operations, thereby improving operating efficiency and reducing equipment losses.

[0029] In a possible implementation, a power grid fluctuation analysis is performed based on the predicted power grid characteristic information sequence to obtain power grid fluctuation parameters, and step S220 further includes step S221, calculating the mean of the predicted power grid characteristic information sequence to obtain average predicted power grid characteristic information. Specifically, for the mean calculation of the predicted power grid characteristic information sequence, it is necessary to accumulate and sum all data points in the sequence, and then divide by the total number of data points. For example, if the predicted power grid characteristic information sequence contains the power grid demand data recorded every 1 minute in the next 30 minutes, a total of 30 data points, the 30 data points are added in sequence, assuming that the sum is S, then the average predicted power grid characteristic information μ=S / 30, through calculation, an average demand power value representing the overall level of the predicted sequence can be obtained, reflecting the approximate central trend of the power grid demand within the preset time range in the future, and providing a benchmark reference value for further analysis of power grid fluctuations.

[0030] Step S222, randomly extract multiple predicted power grid characteristic information in the predicted power grid characteristic information sequence, calculate the mean, and obtain random average predicted power grid characteristic information. Specifically, after obtaining the average predicted power grid characteristic information, a plurality of predicted power grid characteristic information are randomly extracted in the predicted power grid characteristic information sequence, and the number of extractions can be determined according to the actual situation and the requirements for analysis accuracy, such as extracting 10 data points. Using a random number generation algorithm, 10 different index positions are randomly selected within the index range of the sequence, and then the predicted power grid characteristic information data at the corresponding positions are obtained, and the 10 randomly extracted data points are accumulated and summed, and then divided by the number of extractions 10 to obtain the random average predicted power grid characteristic information μ'. For example, the 10 randomly extracted data points are P1, P2, ..., P10, respectively, then μ'=(P1+P2+...+P10) / 10. Random sampling and mean calculation can obtain the local average situation in the prediction sequence, which is compared with the overall average predicted power grid characteristic information, thereby providing a basis for evaluating the degree of fluctuation of power grid demand.

[0031] Step S223, calculate the deviation percentage between the random average predicted grid characteristic information and the average predicted grid characteristic information as a grid fluctuation parameter. Specifically, calculate the deviation percentage between the random average predicted grid characteristic information μ' and the average predicted grid characteristic information μ, and calculate according to the formula δ=(μ'-μ) / μ*100%. For example, if μ=50kW, μ'=55kW, then the deviation percentage δ=(55-50) / 50*100%=10%. The deviation percentage δ is the grid fluctuation parameter, which intuitively reflects the degree of fluctuation of the grid demand power within the prediction time range. If the deviation percentage value is large, it means that the fluctuation of the grid demand power is more severe. The distributed power source needs to respond more cautiously when adjusting the power output, and adopt a more flexible adjustment strategy to reduce the impact on the grid and ensure the stable operation of the grid; on the contrary, if the deviation percentage value is small, it means that the grid demand power is relatively stable, and the distributed power source can operate in a relatively stable power output mode, reducing unnecessary adjustment operations, improving operation efficiency and reducing equipment loss and energy consumption.

[0032] Step S300, within the power output demand interval, according to the power grid fluctuation parameters, with the purpose of reducing the impact of distributed power sources on the power grid and the degree of power output adaptation, the power output regulation optimization of the distributed power source is performed to obtain the optimal output power. Specifically, within the power output demand interval, the initial optimization parameters are first set, including the initialization iteration number, the search step size and the random number seed, etc., and the upper and lower limits of the interval are clearly defined. Then, the random number generation function is used to generate the initial output power in the interval, which is used as the starting guess value of the distributed power source. Subsequently, based on this initial output power, the average predicted power grid characteristic information and the power grid fluctuation parameters, the power regulation fitness is calculated according to a specific formula. This fitness comprehensively reflects the matching degree of the output power with the power grid demand and the impact on the power grid. Entering the iterative optimization phase, a new candidate value is generated within the interval according to the current output power and search step size, and its fitness is calculated and compared with the current optimal value. If it is higher, the optimal output power and fitness are updated, and the number of iterations is updated at the same time until the preset number of iterations is reached or the convergence conditions are met. The final output power obtained at the time of convergence is the optimal output power that can minimize the impact on the power grid and improve the adaptability of power output according to the power grid fluctuation parameters within the interval. The distributed power source will adjust the power output accordingly to achieve efficient and stable coordinated operation with the power grid.

[0033] In a possible implementation, within the power output demand interval, according to the power grid fluctuation parameters, the power output regulation optimization of the distributed power source is performed to reduce the impact of the distributed power source on the power grid and the degree of power output adaptation, and the optimal output power is obtained. Step S300 further includes step S310, randomly generating a first output power within the power output demand interval. Specifically, within the determined power output demand interval, the first output power is randomly generated by a random number generation algorithm. For example, if the power output demand interval is [20kW, 60kW], the random number generator may generate a value such as 35kW as the first output power. The randomly generated power value is the starting point of the subsequent optimization process. It has randomness within the entire power output demand interval, and can provide a preliminary attempt direction for exploring the optimal output power to avoid falling into a local optimal solution.

[0034] Step S320, calculate the first power regulation fitness of the first output power according to the first output power, the average predicted grid characteristic information and the grid fluctuation parameter. Specifically, after obtaining the first output power, calculate the corresponding first power regulation fitness according to the first output power, the average predicted grid characteristic information and the grid fluctuation parameter according to a specific formula. Assuming that the average predicted grid characteristic information is 40kW, the grid fluctuation parameter is 0.15, and the first output power is 35kW, according to the formula (PRA is the power regulation adaptability, and is the weight, is the output power, is the average power demand in the average predicted grid characteristic information, D is the grid fluctuation parameter), let =0.6, =0.4, substituting the value into the first power regulation fitness value, the power regulation fitness is a comprehensive evaluation index, which reflects the current first output power in meeting grid demand and reducing grid impact. The larger the value, the closer the output power is to the ideal output state under the current conditions.

[0035] Step S330, continue to perform power output regulation optimization in the power output demand interval until convergence, and obtain the optimal output power according to the power regulation fitness. Specifically, continue to perform power output regulation optimization in the power output demand interval. In each iteration process, based on the current output power state, a new output power candidate value is generated in the power output demand interval through a certain optimization strategy (such as random search, gradient descent or other intelligent optimization algorithm), and the power regulation fitness of each candidate value is calculated according to the above formula. As the iteration proceeds, the power regulation fitness of the newly generated candidate value is continuously compared with the previous optimal value. If the power regulation fitness of a candidate value is higher, it is updated to the new optimal output power. This iterative process will continue until the convergence condition is met. The convergence condition can reach a preset number of iterations (such as 100 times), or the change in power regulation fitness after multiple consecutive iterations is less than a minimum value (such as 0.001). When the iteration converges, the output power with the highest power regulation fitness finally determined is the optimal output power. The optimal output power can fully consider the grid fluctuation parameters within a given power output demand range, minimize the impact of distributed power sources on the grid, and achieve the best power output adaptation level, thereby ensuring stable and efficient coordinated operation between distributed power sources and the grid.

[0036] In a possible implementation, a first power regulation fitness of the first output power is calculated based on the first output power, the average predicted grid characteristic information, and the grid fluctuation parameter. Step S320 further includes step S321, where a first power regulation fitness of the first output power is calculated based on the first output power, the average predicted grid characteristic information, and the grid fluctuation parameter, as follows: ; Among them, PRA is the power regulation adaptability, and is the weight, is the output power, is the average power demand in the average predicted grid characteristic information, and D is the grid fluctuation parameter. Specifically, when calculating the first power regulation fitness, the required data is first clarified, namely, the first output power randomly generated in the power output demand interval, the average power demand calculated by predicting the grid characteristic information sequence, the grid fluctuation parameter characterizing the degree of fluctuation of the grid power demand, and the weights pre-set according to the characteristics of distributed power sources, grid operation requirements, and the degree of emphasis on different aspects. and Then plug these data into the formula , as assumed 40kW, 50kW, is 0.2, 0.6 The PRA value is 0.4, and the two terms of the formula are calculated and added to get a PRA value of 2.98. The PRA value obtained by this calculation is of great significance. The larger the PRA value is, the better the first output power is in terms of comprehensive consideration of the adaptability to the average demand power and the adaptability to grid fluctuations. In the subsequent power output regulation optimization, this value is used as a key indicator to evaluate the pros and cons of different output powers. By comparing the fitness values, a better output power is selected, and the optimal output power is gradually approached, so as to achieve the goal of reducing the impact on the grid and improving the adaptability of power output according to the grid fluctuation parameters within the power output demand range of distributed power sources, and ensuring the stable and efficient coordinated operation with the grid.

[0037] In one possible implementation, the power output regulation optimization is continued within the power output demand interval until convergence, and the optimal output power is obtained according to the power regulation fitness screening. Step S330 further includes step S331, and the power output regulation optimization is continued within the power output demand interval until the convergence optimization times are reached and the optimization converges. Specifically, the iterative process of power output regulation optimization is started within the power output demand interval, an initial iteration count variable is set, the initial value is 0, and the number of convergence optimizations is determined, for example, set to 100 times. In each iteration, a new output power candidate value is generated within the power output demand interval according to the current output power state and a specific optimization strategy. The optimization strategy can be based on a random search method, that is, a small randomly generated increment or decrement is added on the basis of the current output power. The range of this random increment or decrement can be determined according to the size of the power output demand interval and the requirements for optimization accuracy. Alternatively, a gradient-based optimization algorithm can be used to determine the direction and step size of power adjustment by calculating the gradient of power regulation fitness to output power. For example, if the current output power is 40kW, a random search strategy is used to randomly generate an increment within the range of [-2kW, 2kW], such as 1.5kW, then the new candidate output power is 41.5kW. According to the newly generated candidate output power and the average predicted grid characteristic information and grid fluctuation parameters, the power regulation fitness is calculated according to the established formula. As the iteration continues, the iteration count variable gradually increases.

[0038] Step S332, screen the output power with the largest output power regulation fitness to obtain the optimal output power. Specifically, when the iteration count variable reaches the preset number of convergence optimization times, i.e., 100 times, the optimization convergence is determined, and the optimal output power screening stage is entered. All output power regulation fitness calculated during the entire iteration process are compared to find the output power corresponding to the fitness with the largest value. For example, during the iteration process, a total of 100 different output power candidate values ​​and their corresponding power regulation fitness are generated. By comparing these fitness values ​​one by one, it is determined that the output power corresponding to the maximum fitness is 50kW, and this 50kW is the optimal output power finally obtained. The optimal output power can fully consider the grid fluctuation parameters within a given power output demand range, minimize the impact of distributed power sources on the grid, and achieve the best power output adaptation degree, thereby ensuring the stable and efficient coordinated operation between distributed power sources and the grid.

[0039] Step S400, using the optimal output power to adjust the output power of the distributed power source. Specifically, after determining the optimal output power, the control system of the distributed power source receives a command signal containing the value and starts a power adjustment mechanism. For photovoltaic power stations, the power output is changed by adjusting the working point of the photovoltaic array, and the optimal voltage and current values ​​are calculated based on environmental factors such as light intensity and temperature and the optimal output power. The output power is made to approach and stabilize at the optimal value by adjusting the inverter control parameters such as the PWM signal duty cycle; wind farms rely on adjusting the pitch angle and speed of the wind turbine, and calculate the appropriate set value according to the wind speed, wind direction and wind turbine operating status. The pitch angle is adjusted to control the amount of wind energy captured, and the speed is adjusted to optimize the power generation efficiency, so that the output power is stabilized at the optimal value; energy storage power stations such as lithium-ion battery energy storage systems achieve this by controlling the battery charge and discharge current, and determine the charge and discharge current value based on the optimal output power instruction and battery charge state, voltage, temperature and other parameters. During the adjustment, the battery status parameters are monitored in real time to ensure its safe and stable operation and prevent abnormal situations, so that the energy storage power station can continuously and reliably interact with the power grid according to the optimal output power, and achieve efficient coordinated operation of distributed power sources and the power grid.

[0040] The embodiment of the present application collects historical characteristic information of the power grid, predicts future power grid characteristics, and analyzes the power demand range of distributed power sources and power grid fluctuation parameters; based on the demand range and fluctuation parameters, the power output is optimized to reduce the impact on the power grid, obtain the optimal power, and adjust the output of the distributed power source accordingly, achieving the technical effect of improving the adaptability of the power output of the distributed power source to changes in power grid demand.

[0041] In the above, refer to Figure 1 The power output regulation method of the distributed power supply according to the embodiment of the present invention is described in detail. Figure 3 A power output regulation system of a distributed power source according to an embodiment of the present invention is described.

[0042] The power output regulation system of a distributed power source according to an embodiment of the present invention is used to solve the technical problem that the existing distributed power source has insufficient power adaptability due to slow regulation response time under complex power grid demand fluctuations, and achieves the technical effect of improving the adaptability of the power output of the distributed power source to the changes in power grid demand. The power output regulation system of the distributed power source includes: an information sequence acquisition module 10, a demand interval acquisition module 20, a regulation optimization module 30, and an output power regulation module 40.

[0043] The information sequence acquisition module 10 is used to acquire the historical power grid characteristic information sequence of the power grid side within the past preset time range, predict the power grid characteristic information within the future preset time range, and obtain the predicted power grid characteristic information sequence.

[0044] The demand interval acquisition module 20 is used to perform power output demand analysis and grid fluctuation analysis of the distributed power source according to the predicted grid characteristic information sequence, and obtain the power output demand interval and grid fluctuation parameters.

[0045] The regulation optimization module 30 is used to optimize the power output regulation of the distributed power source within the power output demand range according to the power grid fluctuation parameters, with the purpose of reducing the impact of the distributed power source on the power grid and the power output adaptation degree, so as to obtain the optimal output power.

[0046] The output power regulating module 40 is used to adopt the optimal output power to regulate the output power of the distributed power source.

[0047] The specific configuration of the information sequence acquisition module 10 will be described in detail below. As described above, the historical grid characteristic information sequence of the grid side within the past preset time range is collected, and the grid characteristic information within the future preset time range is predicted to obtain the predicted grid characteristic information sequence|, the information sequence acquisition module 10 further includes: a response time acquisition unit, the response time acquisition unit is used to obtain the adjustment response time of the distributed power source for power regulation; a time range calculation unit, the time range calculation unit is used to calculate the sum of the adjustment response time and the preset time length as the preset time range; a grid characteristic information prediction unit, the grid characteristic information prediction unit is used to collect the historical grid characteristic information sequence of the grid side within the past preset time range, predict the grid characteristic information within the future preset time range, and obtain the predicted grid characteristic information sequence.

[0048] Among them, the historical power grid characteristic information sequence of the power grid side within the preset time range in the past is collected, the power grid characteristic information within the preset time range in the future is predicted, and the predicted power grid characteristic information sequence is obtained. The power grid characteristic information prediction unit further includes: an information sequence set acquisition subunit, the information sequence set acquisition subunit is used to collect the demand power within the historical time of the power grid side according to the preset time range, obtain a sample historical power grid characteristic information sequence set, and collect the demand power within the preset time range after each sample historical power grid characteristic information sequence to obtain a sample predicted power grid characteristic information sequence set; a prediction channel training subunit, the prediction channel training subunit is used to use the sample historical power grid characteristic information sequence set and the sample predicted power grid characteristic information sequence set as supervised training data to train the power grid characteristic prediction channel until the training converges; a predicted power grid characteristic information sequence acquisition subunit, the predicted power grid characteristic information sequence acquisition subunit is used to input the historical power grid characteristic information sequence into the power grid characteristic prediction channel, and predict the output to obtain a predicted power grid characteristic information sequence.

[0049] The specific configuration of the demand interval acquisition module 20 will be described in detail below. As described above, according to the predicted grid characteristic information sequence, the power output demand analysis and grid fluctuation analysis of the distributed power source are performed to obtain the power output demand interval and grid fluctuation parameters. The demand interval acquisition module 20 further includes: a power output demand interval construction unit, the power output demand interval construction unit is used to extract the maximum demand power and the minimum demand power in the predicted grid characteristic information sequence, and construct the power output demand interval of the distributed power source outputting power to the grid side; a grid fluctuation analysis unit, the grid fluctuation analysis unit is used to perform grid fluctuation analysis according to the predicted grid characteristic information sequence to obtain grid fluctuation parameters.

[0050] Among them, a power grid fluctuation analysis is performed according to the predicted power grid characteristic information sequence to obtain a power grid fluctuation parameter, and the power grid fluctuation analysis unit further includes: a mean calculation subunit, the mean calculation subunit is used to calculate the mean of the predicted power grid characteristic information sequence to obtain average predicted power grid characteristic information; an average predicted power grid characteristic information acquisition subunit, the average predicted power grid characteristic information acquisition subunit is used to randomly extract multiple predicted power grid characteristic information in the predicted power grid characteristic information sequence, calculate the mean, and obtain random average predicted power grid characteristic information; a deviation percentage calculation subunit, the deviation percentage calculation subunit is used to calculate the deviation percentage between the random average predicted power grid characteristic information and the average predicted power grid characteristic information as a power grid fluctuation parameter.

[0051] The specific configuration of the regulation optimization module 30 will be described in detail below. As described above, within the power output demand interval, according to the grid fluctuation parameters, the power output regulation optimization of the distributed power source is performed to reduce the impact of the distributed power source on the grid and the power output adaptation degree, and the optimal output power is obtained. The regulation optimization module 30 further includes: a first output power generation unit, the first output power generation unit is used to randomly generate a first output power within the power output demand interval; a fitness calculation unit, the fitness calculation unit is used to calculate the first power regulation fitness of the first output power according to the first output power, the average predicted grid characteristic information and the grid fluctuation parameters; a power output regulation optimization unit, the power output regulation optimization unit is used to continue to perform power output regulation optimization within the power output demand interval until convergence, and obtain the optimal output power according to the power regulation fitness screening.

[0052] According to the first output power, the average predicted grid characteristic information and the grid fluctuation parameter, the first power regulation fitness of the first output power is calculated, and the fitness calculation unit further includes: a first power regulation fitness calculation subunit, and the first power regulation fitness calculation subunit is used to calculate the first power regulation fitness of the first output power according to the first output power, the average predicted grid characteristic information and the grid fluctuation parameter, as shown in the following formula: ; Among them, PRA is the power regulation adaptability, and is the weight, is the output power, is the average demand power in the average predicted grid characteristic information, and D is the grid fluctuation parameter.

[0053] Among them, the power output regulation optimization is continued in the power output demand interval until convergence, and the optimal output power is obtained according to the power regulation fitness screening. The power output regulation optimization unit further includes: an optimization convergence subunit, the optimization convergence subunit is used to continue the power output regulation optimization in the power output demand interval until the convergence optimization times are reached and the optimization converges; an optimal output power acquisition subunit, the optimal output power acquisition subunit is used to screen the output power with the largest output power regulation fitness to obtain the optimal output power.

[0054] The power output regulation system of a distributed power source provided in an embodiment of the present invention can execute the power output regulation method of a distributed power source provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0055] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0056] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A method for regulating power output of a distributed power source, characterized in that: The method is applied to a distributed power source, the distributed power source is connected to the power grid side, and the method includes: Collecting a historical grid characteristic information sequence of the grid side within a preset time range in the past, predicting the grid characteristic information within a preset time range in the future, and obtaining a predicted grid characteristic information sequence; According to the predicted grid characteristic information sequence, power output demand analysis and grid fluctuation analysis of the distributed power source are performed to obtain a power output demand interval and grid fluctuation parameters; Within the power output demand range, according to the power grid fluctuation parameters, the power output regulation and optimization of the distributed power source is performed to obtain the optimal output power in order to reduce the impact of the distributed power source on the power grid and the power output adaptation degree; The optimal output power is used to adjust the output power of the distributed power source.

2. The power output regulation method of a distributed power source according to claim 1, characterized in that: Collecting a historical grid characteristic information sequence of the grid side within a preset time range in the past, and predicting the grid characteristic information within a preset time range in the future, including: Obtaining a regulation response time for the distributed power source to perform power regulation; Calculating the sum of the adjustment response time and the preset time length as the preset time range; The historical power grid characteristic information sequence of the power grid side within a preset time range in the past is collected, and the power grid characteristic information within a preset time range in the future is predicted to obtain a predicted power grid characteristic information sequence.

3. The power output regulation method of a distributed power source according to claim 2, characterized in that: Collecting a historical grid characteristic information sequence of the grid side within a preset time range in the past, and predicting the grid characteristic information within a preset time range in the future, including: According to the preset time range, the power demand in the historical time of the power grid side is collected to obtain a set of sample historical power grid characteristic information sequences, and the power demand in the preset time range after each sample historical power grid characteristic information sequence is collected to obtain a set of sample predicted power grid characteristic information sequences; The sample historical power grid characteristic information sequence set and the sample predicted power grid characteristic information sequence set are used as supervised training data to train the power grid characteristic prediction channel until the training converges; The historical power grid characteristic information sequence is input into the power grid characteristic prediction channel, and the prediction output is used to obtain the predicted power grid characteristic information sequence.

4. The power output regulation method of a distributed power source according to claim 1, characterized in that: According to the predicted grid characteristic information sequence, the power output demand analysis and grid fluctuation analysis of the distributed power source are performed to obtain the power output demand interval and grid fluctuation parameters, including: Extracting the maximum power demand and the minimum power demand in the predicted power grid characteristic information sequence, and constructing a power output demand interval for the distributed power source to output power to the power grid side; A power grid fluctuation analysis is performed according to the predicted power grid characteristic information sequence to obtain power grid fluctuation parameters.

5. The power output regulation method of a distributed power source according to claim 4, characterized in that: Performing power grid fluctuation analysis according to the predicted power grid characteristic information sequence to obtain power grid fluctuation parameters includes: Calculating the mean of the predicted power grid characteristic information sequence to obtain average predicted power grid characteristic information; Randomly extracting a plurality of predicted power grid characteristic information in the predicted power grid characteristic information sequence, calculating the mean, and obtaining random average predicted power grid characteristic information; The deviation percentage between the random average predicted power grid characteristic information and the average predicted power grid characteristic information is calculated as a power grid fluctuation parameter.

6. The power output regulation method of a distributed power source according to claim 5, characterized in that: Within the power output demand range, according to the power grid fluctuation parameter, the power output regulation optimization of the distributed power source is performed for the purpose of reducing the impact of the distributed power source on the power grid and the power output adaptation degree, including: Randomly generating a first output power within the power output requirement interval; Calculating a first power regulation fitness of the first output power according to the first output power, average predicted grid characteristic information and grid fluctuation parameters; Continue to perform power output regulation optimization within the power output demand range until convergence, and obtain the optimal output power according to the power regulation fitness screening.

7. The power output regulation method of a distributed power source according to claim 6, characterized in that: According to the first output power, the average predicted grid characteristic information and the grid fluctuation parameter, a first power regulation fitness of the first output power is calculated and obtained as follows: ； Among them, PRA is the power regulation adaptability, and is the weight, is the output power, is the average demand power in the average predicted grid characteristic information, and D is the grid fluctuation parameter.

8. The power output regulation method of a distributed power source according to claim 6, characterized in that: Continuing to optimize the power output regulation within the power output demand range until convergence, and obtaining the optimal output power according to the power regulation fitness screening, including: Continue to perform power output regulation optimization within the power output demand range until the convergence optimization times are reached and the optimization converges; The output power with the largest output power adjustment adaptability is selected to obtain the optimal output power.

9. A power output regulation system for a distributed power source, characterized in that: The system is used to implement the power output regulation method of a distributed power source according to any one of claims 1 to 8, and the system comprises: An information sequence acquisition module, the information sequence acquisition module is used to collect the historical power grid characteristic information sequence of the power grid side within the past preset time range, predict the power grid characteristic information within the future preset time range, and obtain the predicted power grid characteristic information sequence; A demand interval acquisition module, the demand interval acquisition module is used to perform power output demand analysis and power grid fluctuation analysis of the distributed power source according to the predicted power grid characteristic information sequence, and obtain a power output demand interval and a power grid fluctuation parameter; A regulation optimization module, wherein the regulation optimization module is used to optimize the power output regulation of the distributed power source within the power output demand range according to the power grid fluctuation parameters, with the purpose of reducing the impact of the distributed power source on the power grid and the power output adaptation degree, so as to obtain the optimal output power; An output power regulation module is used to use the optimal output power to regulate the output power of the distributed power source.

10. The power output regulation system of a distributed power source according to claim 9, characterized in that: The information sequence acquisition module comprises: A response time acquisition unit, the response time acquisition unit is used to obtain the adjustment response time of the distributed power source for power regulation; A time range calculation unit, the time range calculation unit is used to calculate the sum of the adjustment response time and the preset time length as the preset time range; A power grid characteristic information prediction unit is used to collect the historical power grid characteristic information sequence of the power grid side within a preset time range in the past, predict the power grid characteristic information within a preset time range in the future, and obtain a predicted power grid characteristic information sequence.

Images