Intelligent micro-grid frequency modulation and peak regulation and electric energy distribution method based on photovoltaic power generation

Abstract

The invention relates to an intelligent micro-grid frequency modulation and peak regulation and electric energy distribution method based on photovoltaic power generation, and relates to the field of intelligent micro-grids, and the method comprises the steps: carrying out the analysis to obtain a predicted fluctuation degree and an energy supply deviation amplitude, combining the predicted fluctuation degree with a historical fluctuation degree, and carrying out the calculation to obtain a fluctuation degree; a frequency modulation coefficient and a peak regulation distribution coefficient are obtained according to the fluctuation degree and energy supply deviation amplitude configuration; and randomly generating an electric energy distribution scheme, carrying out peak regulation and frequency modulation analysis according to the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, carrying out calculation according to a frequency modulation coefficient and a peak regulation distribution coefficient to obtain distribution fitness, carrying out iterative optimization evaluation, and obtaining an optimal distribution scheme to carry out electric energy distribution. According to the invention, the technical problems that the traditional method cannot efficiently cope with the conditions of load fluctuation and insufficient photovoltaic power generation, it is difficult to ensure that the energy storage unit can respond in time and provide stable and accurate electric energy support, so that the operation fluctuation of the micro-grid in a high-load period is large, and the frequency modulation and peak regulation difficulty is high can be solved.

Description

Technical Field

[0001] The present invention relates to the field of smart microgrids, and in particular to a method for frequency and peak regulation and electric energy distribution of a smart microgrid based on photovoltaic power generation. Background Art

[0002] With the widespread adoption of renewable energy, photovoltaic power generation has become the primary energy source in microgrids. However, the volatility and intermittent nature of photovoltaic power generation prevents it from consistently meeting load demands at all times, especially during periods of insufficient sunlight or peak demand. To compensate for this shortfall, energy storage systems are being introduced into microgrids to store excess power and provide support when demand increases.

[0003] Currently, energy storage units in smart microgrids mainly rely on traditional power distribution methods to regulate the charging and discharging process of the energy storage system. These methods are usually based on preset rules or simple load forecasts and cannot fully cope with the complex situations of load fluctuations and insufficient photovoltaic power generation. During periods of low photovoltaic power generation or high demand load, traditional methods often find it difficult to achieve accurate power distribution, resulting in the energy storage units being unable to provide sufficient power support in a timely and effective manner, which in turn causes greater volatility in the operation of the microgrid.

[0004] In addition, existing power distribution methods do not adequately support the frequency and peak-shaving capabilities of energy storage units, making it difficult to quickly respond to load fluctuations and ensure grid stability. This situation is particularly prominent during high-load periods in microgrids, which may lead to increased grid frequency fluctuations, affect power supply quality, and even cause power failures. Summary of the Invention

[0005] The present invention addresses the technical problems that traditional methods are unable to efficiently cope with load fluctuations and insufficient photovoltaic power generation, making it difficult to ensure that energy storage units can respond in a timely manner and provide stable and accurate power support, resulting in large fluctuations in the operation of microgrids during high-load periods and high difficulty in frequency and peak regulation. The present invention provides a method for frequency and peak regulation and power distribution of an intelligent microgrid based on photovoltaic power generation to solve the problem.

[0006] The technical solution of the present invention to solve the above technical problems is as follows: the present invention provides a method for frequency regulation and peak regulation and electric energy distribution of an intelligent microgrid based on photovoltaic power generation, the method is applied to a microgrid, and the microgrid includes a photovoltaic power generation group and an energy storage group, including: obtaining a historical photovoltaic power generation power sequence and a historical demand load power sequence in the microgrid, analyzing and obtaining a historical fluctuation, and performing photovoltaic power generation and demand load forecasting in a future time period according to the historical fluctuation to obtain a predicted photovoltaic power generation power sequence and a predicted demand load power sequence; analyzing and obtaining a predicted fluctuation and an energy supply deviation amplitude according to the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence. The predicted volatility and the historical volatility are combined to calculate the volatility, and the frequency regulation coefficient and the peak regulation allocation coefficient are configured according to the volatility and the energy supply deviation amplitude; the electric energy distribution plan of the energy storage unit is randomly generated, the peak regulation and frequency regulation analysis is performed according to the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, and the distribution fitness is calculated according to the frequency regulation coefficient and the peak regulation allocation coefficient, wherein the electric energy distribution plan includes the distribution power sequence in the future time period, and the distribution fitness is calculated according to the deviation of adjacent distribution powers; according to the distribution fitness, the electric energy distribution plan is iteratively optimized and evaluated, the optimal electric energy distribution plan is optimized, and the energy storage unit is controlled to distribute electric energy within the microgrid.

[0007] Optionally, the method for frequency regulation, peak regulation and power distribution of a smart microgrid based on photovoltaic power generation further includes: obtaining the photovoltaic power generation power of the photovoltaic power generation group at the past K historical moments in the microgrid, and the demand load power on the power consumption side, to obtain a historical photovoltaic power generation power sequence and a historical demand load power sequence, where K is a positive integer; analyzing the volatility of the historical photovoltaic power generation power sequence and the historical demand load power sequence, to obtain a historical photovoltaic volatility and a historical demand volatility, and to calculate the historical volatility; using integrated machine learning to construct a photovoltaic power generation prediction branch array and a demand load prediction branch array, wherein the photovoltaic power generation prediction branch array and the demand load prediction branch array respectively include P photovoltaic power generation prediction branches and P demand load forecasting branches, where P is a positive integer; Q is calculated based on the historical volatility and P, where Q is a positive integer; Q photovoltaic power generation forecasting branches and Q demand load forecasting branches are randomly selected, and photovoltaic power generation and demand load forecasting at K future moments in a future period are performed on the historical photovoltaic power generation power sequence and the historical demand load power sequence to obtain Q branch-predicted photovoltaic power generation power sequences and Q branch-predicted demand load power sequences; the mean of the Q branch-predicted photovoltaic power generation power and the Q branch-predicted demand load power at each future moment in the Q branch-predicted photovoltaic power generation power sequence and the Q branch-predicted demand load power sequence is calculated to obtain a predicted photovoltaic power generation power sequence and a predicted demand load power sequence.

[0008] Optionally, the photovoltaic-based smart microgrid frequency regulation, peak regulation and power distribution method also includes: selecting the first historical photovoltaic power generation power at the first historical moment in the historical photovoltaic power generation power sequence, and randomly selecting multiple random historical photovoltaic power generation powers; calculating the average of the multiple random historical photovoltaic power generation powers and the fluctuation amplitude of the first historical photovoltaic power generation power to obtain a first photovoltaic fluctuation; continuing to calculate to obtain K photovoltaic fluctuations, and calculating the average to obtain a historical photovoltaic fluctuation; calculating the historical demand fluctuation according to the historical demand load power sequence; calculating the average of the historical photovoltaic fluctuation and the historical demand fluctuation to obtain a historical fluctuation.

[0009] Optionally, the method for frequency regulation, peak regulation and power distribution of a smart microgrid based on photovoltaic power generation further includes: collecting a set of sample historical photovoltaic power generation sequences based on the microgrid operation record log within the historical time, and using the photovoltaic power generation power sequence after each sample historical photovoltaic power generation sequence as a sample predicted photovoltaic power generation power sequence to obtain a set of sample predicted photovoltaic power generation sequences, and integrating to obtain a photovoltaic prediction sample data set; randomly selecting P photovoltaic prediction sample data from the photovoltaic prediction sample data set with replacement, and using integrated machine learning to train P photovoltaic power generation prediction branches; collecting a set of sample historical demand load power sequences based on the microgrid operation record log within the historical time, and using the demand load power sequence after each sample historical demand load power sequence as a sample predicted demand load power sequence to obtain a set of sample predicted demand load power sequences, and integrating to obtain a load prediction sample data set; randomly selecting P load prediction sample data from the load prediction sample data set with replacement, and using integrated machine learning to train P load power generation prediction branches; and combining the P photovoltaic power generation prediction branches and the P load power generation prediction branches respectively to obtain a photovoltaic power generation prediction branch array and a demand load prediction branch array.

[0010] Optionally, the method for frequency regulation, peak regulation and power distribution of a smart microgrid based on photovoltaic power generation also includes: analyzing and calculating to obtain a predicted volatility based on the predicted photovoltaic power generation power sequence and the predicted demand load power sequence; calculating the average of the predicted volatility and the historical volatility to obtain the volatility; calculating the power deviation amplitude of each demand load power and the corresponding photovoltaic power generation power based on the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, and calculating the average to obtain the energy supply deviation amplitude; configuring the volatility and the energy supply deviation amplitude as the frequency regulation coefficient and the peak regulation distribution coefficient.

[0011] Optionally, the method for frequency regulation, peak regulation and electric energy distribution of a smart microgrid based on photovoltaic power generation also includes: obtaining an energy supply distribution space for the energy storage unit to distribute energy to the power consumption side; randomly generating K first energy supply distribution powers in the energy supply distribution space as the first distribution power sequence for K future moments in a future time period to obtain a first electric energy distribution plan; calculating a first actual energy supply power sequence based on the first distribution power sequence and the predicted photovoltaic power generation power sequence; obtaining a first redundant energy supply power sequence by subtracting the corresponding predicted demand load power in the predicted demand load power sequence from each first actual energy supply power in the first actual energy supply power sequence; calculating a first distribution fitness based on the first redundant energy supply power sequence and the first distribution power sequence according to the frequency regulation coefficient and the peak regulation distribution coefficient, wherein the distribution fitness is calculated based on the deviation of adjacent distribution powers in the first distribution power sequence.

[0012] Optionally, the photovoltaic-based smart microgrid frequency and peak regulation and power distribution method further includes: calculating a first distribution fitness according to the first redundant energy supply power sequence and the first distribution power sequence, the frequency regulation coefficient and the peak regulation distribution coefficient, as shown in the following formula: Among them, F f is the allocation fitness, w1, w2 and w3 are weights, the sum of the three is 1, N is a positive integer, TP is the frequency modulation coefficient, TF is the peak modulation allocation coefficient, K is the number of redundant energy supply powers in the redundant energy supply power sequence, σ is the variance of K redundant energy supply powers in the redundant energy supply power sequence, G i is the redundant power supply at the i-th future moment, M i is the allocated power at the i-th future moment, M i+1 is the allocated power at the i+1th future moment.

[0013] Optionally, the method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation also includes: continuing to randomly generate power distribution plans, and calculating distribution fitness and iterative optimization; until the iterative optimization converges, retaining the power distribution plan with the largest distribution fitness as the optimal power distribution plan, and controlling the energy storage unit to perform power supply and distribution in the microgrid at K future moments in the future time period.

[0014] The beneficial effects of the present invention are as follows: by acquiring the historical photovoltaic power generation power sequence and the historical demand load power sequence in the microgrid, analyzing and obtaining the historical fluctuation, and predicting the photovoltaic power generation and demand load in the future time period according to the historical fluctuation, obtaining the predicted photovoltaic power generation power sequence and the predicted demand load power sequence; then, according to the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, analyzing and obtaining the predicted fluctuation and the energy supply deviation amplitude, combining the predicted fluctuation and the historical fluctuation, calculating and obtaining the fluctuation, and configuring and obtaining the frequency modulation coefficient and the peak regulation distribution coefficient according to the fluctuation and the energy supply deviation amplitude; then randomly generating the electric energy distribution plan for the energy storage unit, and performing the forecast according to the predicted photovoltaic power generation power sequence and the predicted demand load power sequence. Peak shaving and frequency regulation analysis, according to the frequency regulation coefficient and the peak shaving distribution coefficient, calculate the distribution fitness, wherein the power distribution plan includes the distribution power sequence in the future time period, and the distribution fitness is calculated according to the deviation of the adjacent distribution power; further according to the distribution fitness, iterative optimization evaluation of the power distribution plan is performed to optimize the optimal power distribution plan; finally, according to the optimal power distribution plan, the energy storage unit is controlled to distribute power within the microgrid; that is, by optimizing the power distribution of the energy storage unit, the demand load can be accurately matched during peak power consumption periods, the scientificity and accuracy of power distribution can be improved, and the negative impact of photovoltaic power generation fluctuations can be effectively alleviated. At the same time, it is ensured that the energy storage unit can quickly respond to load fluctuations and achieve efficient frequency and peak regulation, thereby greatly improving the overall operation stability and reliability of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a process for frequency and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation provided by the present invention;

[0016] Figure 2 This is a flow chart of obtaining predicted photovoltaic power generation power sequence and predicted demand load power sequence in a photovoltaic power generation-based smart microgrid frequency and peak regulation and power distribution method provided by the present invention. DETAILED DESCRIPTION

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0018] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0019] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0020] Examples, such as Figure 1 As shown, an embodiment of the present invention provides a method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation. The method is applied to a microgrid including a photovoltaic generator set and an energy storage unit, and specifically includes the following steps:

[0021] S1: Obtain the historical photovoltaic power generation power sequence and the historical demand load power sequence in the microgrid, analyze and obtain the historical fluctuation, and predict the photovoltaic power generation and demand load in the future time period according to the historical fluctuation to obtain the predicted photovoltaic power generation power sequence and the predicted demand load power sequence.

[0022] Further, if Figure 2 As shown, step S1 of the present invention further includes:

[0023] The photovoltaic power generation power of the photovoltaic power generation group in the past K historical moments in the microgrid and the required load power on the power consumption side are obtained to obtain a historical photovoltaic power generation power sequence and a historical required load power sequence, where K is a positive integer.

[0024] Specifically, query the microgrid operation log to obtain the photovoltaic power generation power of the photovoltaic power generation group in the past K historical moments in the microgrid. The photovoltaic power generation power usually fluctuates with weather changes (such as light intensity) and time (such as the difference between day and night). Therefore, this data is crucial for analyzing the stability of the photovoltaic power generation system and effectively predicting future power generation conditions. K is a positive integer. The specific value of K can be set according to the actual scenario, such as 20, that is, the photovoltaic power generation power at the past K (20) historical moments (monitoring time points, such as data monitoring every 2 minutes) is selected, and the K photovoltaic power generation powers are arranged in chronological order to obtain a historical photovoltaic power generation power sequence.

[0025] On the other hand, the demand load power on the electricity consumption side at the past K historical moments is obtained. The demand load refers to the power required by the electricity consumption side of the microgrid at the same moment, which is usually affected by weather, time and electricity consumption behavior. Among them, the collection time point of the demand load power and the photovoltaic power generation power is consistent. Then, the K demand load powers are arranged in chronological order to obtain the historical demand load power sequence.

[0026] The fluctuations of the historical photovoltaic power generation sequence and the historical demand load power sequence are analyzed to obtain the historical photovoltaic fluctuation and the historical demand fluctuation, and the historical fluctuation is calculated.

[0027] Furthermore, the present invention further comprises the steps of:

[0028] The method comprises the following steps: selecting a first historical photovoltaic power generation power at a first historical moment in the historical photovoltaic power generation power sequence, and randomly selecting multiple random historical photovoltaic power generation powers; calculating an average of the multiple random historical photovoltaic power generation powers and a fluctuation amplitude of the first historical photovoltaic power generation power to obtain a first photovoltaic fluctuation; continuing to calculate to obtain K photovoltaic fluctuations, and calculating an average to obtain a historical photovoltaic fluctuation; calculating a historical demand fluctuation according to the historical demand load power sequence; and calculating an average of the historical photovoltaic fluctuation and the historical demand fluctuation to obtain a historical fluctuation.

[0029] Specifically, first, the first historical photovoltaic power generation power of the first historical moment (any one of the K historical moments) is selected in the historical photovoltaic power generation power sequence, and a predetermined number of multiple random historical photovoltaic power generation powers are randomly selected in the historical photovoltaic power generation power sequence, where the predetermined number can be set according to the amount of data. For example, the predetermined number is set to 20%, that is, 20% of the historical photovoltaic power generation powers are randomly selected from the historical photovoltaic power generation power sequence and set as random historical photovoltaic power generation powers. For example, assuming that the number of historical photovoltaic power generation powers in the historical photovoltaic power generation power sequence is 20, the number of random historical photovoltaic power generation powers is 4.

[0030] Next, the mean of the multiple random historical photovoltaic power generation powers is calculated to obtain the mean of the random historical photovoltaic power generation powers; the fluctuation amplitude of the mean of the random historical photovoltaic power generation powers and the first historical photovoltaic power generation powers is further calculated, where the fluctuation amplitude is the ratio of the absolute value of the power difference between the mean of the random historical photovoltaic power generation powers and the first historical photovoltaic power generation powers to the mean of the random historical photovoltaic power generation powers, and the fluctuation amplitude is set as the first photovoltaic fluctuation degree, which represents the fluctuation amplitude of the first historical photovoltaic power generation power and other historical photovoltaic power generation powers. Then, using the same method, the fluctuation amplitude of other historical photovoltaic power generation powers in the historical photovoltaic power generation sequence is continued to be calculated to obtain K photovoltaic fluctuation degrees, and the mean of the K photovoltaic fluctuation degrees is calculated, and the mean calculation result is set as the historical photovoltaic fluctuation degree. The larger the historical photovoltaic fluctuation degree, the stronger the uncertainty of photovoltaic power generation at the K historical moments.

[0031] On the other hand, using the same algorithm used to calculate the historical photovoltaic fluctuation, K demand fluctuations are calculated based on the historical demand load power sequence, and the K demand fluctuations are averaged to obtain the historical demand fluctuation. The historical demand fluctuation reflects the volatility of the microgrid's demand-side power at K historical moments. Finally, the historical photovoltaic fluctuation and the historical demand fluctuation are averaged, and the average calculation result is set as the historical fluctuation. The historical fluctuation combines the fluctuation characteristics of photovoltaic power generation and demand load, reflecting the overall volatility level of the microgrid. By calculating the historical fluctuation, the overall fluctuation state of the microgrid can be obtained, providing an important basis for optimizing power distribution and frequency and peak regulation.

[0032] Integrated machine learning is used to construct a photovoltaic power generation prediction branch array and a demand load prediction branch array, wherein the photovoltaic power generation prediction branch array and the demand load prediction branch array respectively include P photovoltaic power generation prediction branches and P demand load prediction branches, where P is a positive integer.

[0033] Furthermore, the present invention further comprises the steps of:

[0034] According to the microgrid operation record logs in the historical time, a set of sample historical photovoltaic power generation sequences is collected, and the photovoltaic power generation sequence after each sample historical photovoltaic power generation sequence is used as the sample predicted photovoltaic power generation sequence to obtain a set of sample predicted photovoltaic power generation sequences, which are integrated to obtain a photovoltaic prediction sample data set; P photovoltaic prediction sample data are randomly selected from the photovoltaic prediction sample data set with replacement, and integrated machine learning is used to train P photovoltaic power generation prediction branches; according to the microgrid operation record logs in the historical time, a set of sample historical demand load power sequences is collected, and the demand load power sequence after each sample historical demand load power sequence is used as the sample predicted demand load power sequence to obtain a set of sample predicted demand load power sequences, which are integrated to obtain a load prediction sample data set; P load prediction sample data are randomly selected from the load prediction sample data set with replacement, and integrated machine learning is used to train P load power generation prediction branches; the P photovoltaic power generation prediction branches and the P load power generation prediction branches are respectively combined to obtain a photovoltaic power generation prediction branch array and a demand load prediction branch array.

[0035] Specifically, first, based on the microgrid operation record log within the historical time (such as the last month), sample historical photovoltaic power generation power sequences in multiple different time periods are collected to obtain a set of sample historical photovoltaic power generation power sequences; then, the photovoltaic power generation power sequence after each sample historical photovoltaic power generation power sequence is used as a sample predicted photovoltaic power generation power sequence to obtain a set of sample predicted photovoltaic power generation power sequences; then, the set of sample historical photovoltaic power generation power sequences and the set of sample predicted photovoltaic power generation power sequences are integrated to obtain a photovoltaic prediction sample data set. The photovoltaic prediction sample data set is then divided into P equal parts, where P is a positive integer. The specific value of P can be set according to the actual prediction requirements, such as setting P to 20 to obtain P sample data sets; further, P times are selected with replacement from the P sample data sets to obtain the first set of photovoltaic prediction sample data, and the same method is used to iteratively select P times to obtain P sets of photovoltaic prediction sample data.

[0036] Then, based on machine learning, P photovoltaic power generation prediction branches are constructed. For example, a BP neural network is used to construct a photovoltaic power generation prediction branch. The photovoltaic power generation prediction branch is used to learn based on historical photovoltaic power generation data and then predict future photovoltaic power generation power. The branch includes an input layer, multiple hidden layers, and an output layer. The input data of the input layer is a historical photovoltaic power generation power sequence, and the output data of the output layer is a predicted photovoltaic power generation power sequence. Then, the sample historical photovoltaic power generation sequence is used as input and the sample predicted photovoltaic power generation sequence is used as output. The P photovoltaic prediction sample data are used to perform supervised training on the P photovoltaic power generation prediction branches, wherein the training process is as follows: first, the historical photovoltaic power generation sequence is passed to the hidden layer of the neural network through the input layer, a series of weighted calculations and activation function processing are performed, and the output layer outputs a predicted value; then the mean square error loss function is used to calculate the error between the output of the neural network and the actual predicted photovoltaic power generation power (target output), and the gradient of the error relative to the weight of each neuron is calculated through the back propagation algorithm. Back propagation updates the weights and biases in the neural network through the gradient descent method, so that the output is closer to the target output; then the above steps (forward propagation, error calculation, back propagation and weight update) are repeated, and the network weights are adjusted at each iteration to make the prediction results more and more accurate until the error converges, and the trained P photovoltaic power generation prediction branches are obtained.

[0037] On the other hand, based on the microgrid operation record log within a historical time period (such as the most recent month), a set of sample historical demand load power sequences is collected; then, the demand load power sequence following each sample historical demand load power sequence is used as a sample predicted demand load power sequence to obtain a set of sample predicted demand load power sequences; and the set of sample historical demand load power sequences and the set of sample predicted demand load power sequences are integrated to obtain a load forecast sample data set. The load forecast sample data set is further divided into P equal parts, and P random selections are made from the P data sets with replacement to obtain the first set of load forecast sample data. The data set is then iteratively selected P times to obtain P sets of load forecast sample data.

[0038] Then, based on the BP neural network, P load and power generation prediction branches are constructed. These branches consist of an input layer, multiple hidden layers, and an output layer. The input data of the input layer is a sample historical demand load power sequence, and the output data of the output layer is a sample predicted demand load power sequence. The P load and power generation prediction branches are then supervised trained using the P load forecast sample data until convergence, resulting in the trained P load and power generation prediction branches. The training method for these load and power generation prediction branches is the same as that for the photovoltaic power generation prediction branch, and is not further explained here.

[0039] Finally, the P photovoltaic power generation prediction branches are combined to form a photovoltaic power generation prediction branch array, and the P load power generation prediction branches are combined to form a demand load prediction branch array. By constructing the photovoltaic power generation prediction branch array and the demand load prediction branch array based on machine learning, the intelligence, accuracy, and efficiency of photovoltaic power generation and demand load prediction can be improved, thereby improving the accuracy and efficiency of subsequent power distribution optimization.

[0040] According to the historical volatility and P, Q is calculated, where Q is a positive integer; Q photovoltaic power generation prediction branches and Q demand load prediction branches are randomly selected, and photovoltaic power generation and demand load predictions are performed at K future moments in a future period on the historical photovoltaic power generation power sequence and the historical demand load power sequence to obtain Q branch predicted photovoltaic power generation power sequences and Q branch predicted demand load power sequences; the mean of the Q branch predicted photovoltaic power generation powers and the Q branch predicted demand load powers at each future moment in the Q branch predicted photovoltaic power generation power sequences and the Q branch predicted demand load power sequences are calculated to obtain a predicted photovoltaic power generation power sequence and a predicted demand load power sequence.

[0041] Specifically, the historical volatility is multiplied by P and rounded to obtain Q, where Q is a positive integer. For example, assuming the historical volatility is 18% and P is 20, Q is 4; further, Q photovoltaic power generation prediction branches are randomly selected from the P photovoltaic power generation prediction branches, and Q demand load prediction branches are randomly selected from the P load power generation prediction branches; then, the Q photovoltaic power generation prediction branches are used to predict the photovoltaic power generation at K future moments in a future period of the historical photovoltaic power generation power sequence, and Q branch predicted photovoltaic power generation power sequences are output; the Q demand load prediction branches are used to predict the demand load at K future moments in a future period of the historical demand load power sequence, and Q branch predicted demand load power sequences are obtained.

[0042] Finally, the mean of the Q branches predicted photovoltaic power generation power at each future moment in the Q branches predicted photovoltaic power generation power sequence is calculated to obtain the predicted photovoltaic power generation power sequence; the mean of the Q branches predicted demand load power at each future moment in the Q branches predicted demand load power sequence is calculated to obtain the predicted demand load power sequence.

[0043] By analyzing the volatility of photovoltaic power generation and demand load in the historical period, and selecting an appropriate number of prediction branches to predict photovoltaic power generation and demand load power based on the comprehensive volatility, the adaptability of the number of prediction branches to the actual fluctuation state can be improved. In the case of large fluctuations, more prediction branches can be added to improve the robustness and prediction accuracy of the model; in the case of small fluctuations, the number of prediction branches can be reduced to avoid overfitting and maintain the simplicity of the model. In this way, while ensuring the prediction accuracy, the consumption of computing resources can be saved, unnecessary computing burden can be reduced, and the prediction efficiency can be improved.

[0044] S2: According to the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, the predicted fluctuation and the energy supply deviation amplitude are analyzed and obtained, and the fluctuation is calculated by combining the predicted fluctuation and the historical fluctuation. According to the fluctuation and the energy supply deviation amplitude, the frequency regulation coefficient and the peak regulation allocation coefficient are configured.

[0045] Furthermore, step S2 of the present invention further includes:

[0046] According to the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, the predicted fluctuation is obtained by analysis and calculation; the average of the predicted fluctuation and the historical fluctuation is calculated to obtain the fluctuation; according to the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, the power deviation amplitude of each demand load power and the corresponding photovoltaic power generation power is calculated, and the average is calculated to obtain the energy supply deviation amplitude; the fluctuation and the energy supply deviation amplitude are configured as the frequency modulation coefficient and the peak regulation distribution coefficient.

[0047] Specifically, K predicted fluctuations are calculated based on the predicted photovoltaic power generation sequence, and the average is calculated to obtain the photovoltaic predicted fluctuation. The load predicted fluctuation is calculated based on the predicted demand load power sequence. Next, the photovoltaic predicted fluctuation and the load predicted fluctuation are averaged to obtain the predicted fluctuation. The average of the predicted fluctuation and the historical fluctuation is then calculated to obtain the fluctuation, which reflects the overall volatility of the microgrid during the current period.

[0048] Then, according to the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, the power deviation amplitude of each demand load power and the corresponding photovoltaic power generation power at the same time is calculated respectively, wherein the power deviation amplitude is the ratio of the absolute value of the power difference between the demand load power and the corresponding photovoltaic power generation power to the demand load power, and K historical power deviation amplitudes and K predicted power deviation amplitudes are obtained; then, the K historical power deviation amplitudes and the K predicted power deviation amplitudes are averaged, that is, the sum of the K historical power deviation amplitudes and the K predicted power deviation amplitudes is divided by 2K to obtain the energy supply deviation amplitude.

[0049] The fluctuation degree is further set as the frequency modulation coefficient, wherein the greater the fluctuation degree, the greater the frequency modulation coefficient, which indicates that the frequency change fluctuation in the microgrid is greater, and the frequency modulation demand is greater; the energy supply deviation amplitude is set as the peak shaving allocation coefficient, wherein the greater the energy supply deviation amplitude, the greater the energy supply gap in the microgrid, which means that photovoltaic power generation cannot meet the demand during high load periods, and the energy storage unit needs a greater degree of support to ensure that the load is met, and the peak shaving demand to meet peak electricity consumption is greater.

[0050] S3: Randomly generate an electric energy distribution plan for the energy storage unit, perform peak-shaving and frequency regulation analysis based on the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, and calculate the distribution fitness according to the frequency regulation coefficient and the peak-shaving distribution coefficient. The electric energy distribution plan includes a distribution power sequence in a future time period, and the distribution fitness is calculated based on the deviation of adjacent distribution powers.

[0051] Furthermore, step S3 of the present invention further includes:

[0052] Obtain an energy supply distribution space for the energy storage unit to distribute energy to the power consumption side; randomly generate K first energy supply distribution powers in the energy supply distribution space as the first distribution power sequence for K future moments in a future time period, and obtain a first electric energy distribution plan; calculate and obtain a first actual energy supply power sequence based on the first distribution power sequence and the predicted photovoltaic power generation power sequence; use each first actual energy supply power in the first actual energy supply power sequence to subtract the corresponding predicted demand load power in the predicted demand load power sequence to obtain a first redundant energy supply power sequence.

[0053] Specifically, first, the energy distribution space for the energy storage unit to distribute energy to the power user is obtained, that is, the difference between the maximum power and the minimum power that the energy storage unit can discharge. Then, K first energy distribution powers are randomly generated within the energy distribution space. For example, K energy distribution powers are randomly selected in sequence within the energy distribution space as the first distribution power sequence for K future moments in the future time period to obtain a first power distribution plan. Then, based on the first distribution power sequence and the predicted photovoltaic power generation power sequence, the first distribution power and the predicted photovoltaic power generation power at the same moment are added together to obtain the first actual energy supply power, and the first actual energy supply power sequence is calculated in sequence.

[0054] Further, each first actual energy supply power in the first actual energy supply power sequence is subtracted from the corresponding predicted demand load power in the predicted demand load power sequence, that is, the first actual energy supply power at the same time is subtracted from the predicted demand load power, and set as the first redundant energy supply power, that is, the excess power of the energy storage unit, to obtain the first redundant energy supply power sequence.

[0055] According to the first redundant power supply power sequence and the first allocated power sequence, the first allocation fitness is calculated according to the frequency modulation coefficient and the peak modulation allocation coefficient, wherein the allocation fitness is calculated according to the deviation of adjacent allocated powers in the first allocated power sequence.

[0056] Furthermore, the present invention further comprises the steps of:

[0057] According to the first redundant energy supply power sequence and the first allocated power sequence, and in accordance with the frequency modulation coefficient and the peak modulation allocation coefficient, a first allocation fitness is calculated as follows:

[0058]

[0059] Among them, F f is the allocation fitness, w1, w2 and w3 are weights, the sum of the three is 1, N is a positive integer, TP is the frequency modulation coefficient, TF is the peak modulation allocation coefficient, K is the number of redundant energy supply powers in the redundant energy supply power sequence, σ is the variance of K redundant energy supply powers in the redundant energy supply power sequence, G i is the redundant power supply at the i-th future moment, M i is the allocated power at the i-th future moment, M i+1 is the allocated power at the i+1th future moment.

[0060] Specifically, a distribution fitness evaluation function is constructed, in which F fis the allocation fitness. The greater the allocation fitness, the better the overall effect of the power allocation scheme. w1 is the direct frequency regulation weight, w2 is the peak regulation weight, and w3 is the indirect frequency regulation weight. The sum of the three is 1. They can be set according to the degree of influence of the indicators on the allocation fitness. The greater the influence, the greater the corresponding weight. The three weights can also be dynamically adjusted according to the actual control needs of the microgrid. N is a positive integer, TP is the frequency regulation coefficient, TF is the peak regulation allocation coefficient, K is the number of redundant energy supply powers in the redundant energy supply power sequence, and σ is the variance of the K redundant energy supply powers in the redundant energy supply power sequence. The smaller the variance, the more stable the overall energy supply, and the better the frequency regulation effect. G i is the redundant energy supply power at the i-th future moment. The closer the redundant energy supply power is to 0, the smaller the deviation between the energy supply power and the power consumption side is. The energy supply just meets the power consumption side demand, that is, the peak power consumption is just met, and the peak load regulation effect is better. i is the allocated power at the i-th future moment, M i+1 =((i+1) / (i+1) ...

[0061] By constructing a distribution fitness evaluation function, we can comprehensively and quantitatively evaluate the performance of the power distribution scheme in frequency regulation and peak regulation, improve the precision and accuracy of the power distribution scheme evaluation, and provide a scientific basis for achieving efficient frequency and peak regulation, thereby ensuring the accuracy and reliability of the optimal power distribution scheme setting.

[0062] S4: performing iterative optimization evaluation of the electric energy distribution scheme according to the distribution adaptability, optimizing and obtaining the optimal electric energy distribution scheme, and controlling the energy storage unit to distribute electric energy within the microgrid.

[0063] Furthermore, step S4 of the present invention further includes:

[0064] Continue to randomly generate power distribution plans, calculate distribution fitness and iterative optimization; until the iterative optimization converges, retain the power distribution plan with the largest distribution fitness as the optimal power distribution plan, and control the energy storage unit to distribute power in the microgrid at K future moments in the future time period.

[0065] Specifically, a second electric energy distribution scheme is continuously randomly generated in the energy supply distribution space, wherein the second electric energy distribution scheme is different from the first electric energy distribution scheme, and a second distribution fitness of the second electric energy distribution scheme is calculated; the iterative selection of the electric energy distribution scheme and the distribution fitness calculation are continued until a predetermined number of selections is reached (which can be set according to the optimization accuracy, such as 100 times), and multiple electric energy distribution schemes and multiple distribution fitnesses are output; then the electric energy distribution scheme with the largest distribution fitness is retained as the optimal electric energy distribution scheme, and according to the optimal electric energy distribution scheme, the energy storage unit is controlled to perform electric energy supply and distribution in the microgrid at K future moments in the future time period.

[0066] The embodiment of the present invention provides a method for frequency and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation, which has at least the following technical effects:

[0067] 1. By optimizing the power distribution of energy storage units, the demand load can be accurately matched during peak hours, improving the scientificity and accuracy of power distribution, effectively alleviating the negative impact of photovoltaic power generation fluctuations, and ensuring that the energy storage units can quickly respond to load fluctuations and achieve efficient frequency and peak regulation, thereby significantly improving the overall operational stability and reliability of the microgrid.

[0068] 2. By selecting an appropriate number of prediction branches based on historical fluctuations to predict photovoltaic power generation and demand load power, the adaptability of the number of prediction branches to the actual fluctuation state can be improved. In the case of large fluctuations, more prediction branches can be added to improve the robustness and prediction accuracy of the model; in the case of small fluctuations, the number of prediction branches can be reduced to avoid overfitting and maintain the simplicity of the model. In this way, while ensuring prediction accuracy, computing power resource consumption can be saved, unnecessary computational burden can be reduced, and prediction efficiency can be improved.

[0069] 3. By constructing a distribution fitness evaluation function, the performance of the power distribution scheme in frequency regulation and peak regulation can be comprehensively and quantitatively evaluated, the precision and accuracy of the power distribution scheme evaluation can be improved, and a scientific basis can be provided for achieving efficient frequency and peak regulation, thereby ensuring the accuracy and reliability of the optimal power distribution scheme setting.

[0070] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0071] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for frequency and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation, characterized in that: The method is applied to a microgrid including a photovoltaic generator set and an energy storage unit, and the method includes: Obtaining a historical photovoltaic power generation power sequence and a historical demand load power sequence within the microgrid, analyzing and obtaining a historical volatility, and forecasting photovoltaic power generation and demand load in a future period based on the historical volatility to obtain a predicted photovoltaic power generation power sequence and a predicted demand load power sequence; Analyze and obtain predicted fluctuation and energy supply deviation amplitude based on the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence, and the predicted demand load power sequence; calculate and obtain fluctuation based on the predicted fluctuation and the historical fluctuation; and configure and obtain a frequency modulation coefficient and a peak load allocation coefficient based on the fluctuation and the energy supply deviation amplitude; Randomly generate an electric energy distribution plan for the energy storage unit, perform peak shaving and frequency regulation analysis based on the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, and calculate the distribution fitness according to the frequency regulation coefficient and the peak shaving distribution coefficient, wherein the electric energy distribution plan includes a distribution power sequence in a future time period, and the distribution fitness is calculated based on the deviation of adjacent distribution powers; According to the distribution adaptability, an iterative optimization evaluation of the electric energy distribution scheme is performed to optimize and obtain the optimal electric energy distribution scheme, and the energy storage unit is controlled to distribute electric energy within the microgrid.

2. The method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation according to claim 1, characterized in that: Obtain the historical photovoltaic power generation power sequence and the historical demand load power sequence in the microgrid, analyze and obtain the historical fluctuation, and perform photovoltaic power generation and demand load forecasting in the future period according to the historical fluctuation to obtain the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, including: Obtain the photovoltaic power generation power of the photovoltaic generator set in the past K historical moments in the microgrid, as well as the required load power on the power consumption side, to obtain a historical photovoltaic power generation power sequence and a historical required load power sequence, where K is a positive integer; Analyze the fluctuations of the historical photovoltaic power generation sequence and the historical demand load power sequence to obtain the historical photovoltaic fluctuation and the historical demand fluctuation, and calculate the historical fluctuation; Using integrated machine learning, a photovoltaic power generation prediction branch array and a demand load prediction branch array are constructed, wherein the photovoltaic power generation prediction branch array and the demand load prediction branch array respectively include P photovoltaic power generation prediction branches and P demand load prediction branches, where P is a positive integer; Based on the historical volatility and P, Q is calculated, where Q is a positive integer; Randomly select Q photovoltaic power generation prediction branches and Q demand load prediction branches, perform photovoltaic power generation and demand load forecasting at K future moments in a future period on the historical photovoltaic power generation power sequence and the historical demand load power sequence, and obtain Q branch predicted photovoltaic power generation power sequences and Q branch predicted demand load power sequences; Calculate the mean of the Q branch predicted photovoltaic power generation powers and the Q branch predicted demand load powers at each future moment in the Q branch predicted photovoltaic power generation power sequences and the Q branch predicted demand load power sequences to obtain the predicted photovoltaic power generation power sequence and the predicted demand load power sequence.

3. The method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation according to claim 2, characterized in that: Analyze the fluctuations of the historical photovoltaic power generation sequence and the historical demand load power sequence to obtain the historical photovoltaic fluctuations and the historical demand fluctuations, and calculate the historical fluctuations, including: Selecting a first historical photovoltaic power generation power at a first historical moment in the historical photovoltaic power generation power sequence, and randomly selecting a plurality of random historical photovoltaic power generation powers; Calculating an average of the plurality of random historical photovoltaic power generation values and a fluctuation amplitude of the first historical photovoltaic power generation value to obtain a first photovoltaic fluctuation degree; Continue to calculate to obtain K photovoltaic fluctuations, and calculate the average to obtain the historical photovoltaic fluctuations; Calculating historical demand fluctuations based on the historical demand load power sequence; The average of the historical photovoltaic fluctuation and the historical demand fluctuation is calculated to obtain the historical fluctuation.

4. The method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation according to claim 2, characterized in that: Using integrated machine learning, we build a photovoltaic power generation prediction branch array and a demand load prediction branch array, including: According to the microgrid operation record logs in the historical time, a set of sample historical photovoltaic power generation sequence is collected, and the photovoltaic power generation sequence after each sample historical photovoltaic power generation sequence is used as a sample predicted photovoltaic power generation sequence to obtain a set of sample predicted photovoltaic power generation sequence, and integrate them to obtain a photovoltaic prediction sample data set; Randomly selecting P photovoltaic prediction sample data from the photovoltaic prediction sample data set with replacement, and using ensemble machine learning to train P photovoltaic power generation prediction branches; According to the microgrid operation record log in the historical time, a sample historical demand load power sequence set is collected, and the demand load power sequence after each sample historical demand load power sequence is used as the sample predicted demand load power sequence to obtain a sample predicted demand load power sequence set, which is integrated to obtain a load forecast sample data set; Randomly selecting P load forecast sample data from the load forecast sample data set with replacement, and using ensemble machine learning to train P load and power generation forecast branches; The P photovoltaic power generation prediction branches and the P load power generation prediction branches are respectively combined to obtain a photovoltaic power generation prediction branch array and a demand load prediction branch array.

5. The method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation according to claim 1, characterized in that: According to the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, the predicted fluctuation and the energy supply deviation amplitude are analyzed and obtained. The fluctuation is calculated by combining the predicted fluctuation and the historical fluctuation, and the frequency regulation coefficient and the peak regulation allocation coefficient are configured, including: Analyzing and calculating the predicted fluctuation according to the predicted photovoltaic power generation sequence and the predicted demand load power sequence; Calculating the average of the predicted volatility and the historical volatility to obtain volatility; According to the historical photovoltaic power generation power sequence, the historical demand load power sequence, the predicted photovoltaic power generation power sequence and the predicted demand load power sequence, the power deviation amplitude between each demand load power and the corresponding photovoltaic power generation power is calculated, and the average is calculated to obtain the energy supply deviation amplitude; The fluctuation degree and the energy supply deviation amplitude are configured as a frequency modulation coefficient and a peak regulation allocation coefficient.

6. The method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation according to claim 1, characterized in that: Randomly generate an electric energy distribution plan for the energy storage unit, perform peak regulation and frequency regulation analysis based on the predicted photovoltaic power generation sequence and the predicted demand load power sequence, and calculate the distribution fitness according to the frequency regulation coefficient and the peak regulation distribution coefficient, including: Obtaining an energy distribution space for the energy storage unit to distribute energy to the power consumption side; Randomly generate K first energy supply allocation powers in the energy supply allocation space as a first allocation power sequence at K future moments in a future time period, to obtain a first electric energy allocation plan; Calculating a first actual energy supply power sequence according to the first allocated power sequence and the predicted photovoltaic power generation power sequence; Obtain a first redundant energy supply power sequence by subtracting the corresponding predicted required load power in the predicted required load power sequence from each first actual energy supply power in the first actual energy supply power sequence; According to the first redundant power supply power sequence and the first allocated power sequence, the first allocation fitness is calculated according to the frequency modulation coefficient and the peak modulation allocation coefficient, wherein the allocation fitness is calculated according to the deviation of adjacent allocated powers in the first allocated power sequence.

7. The method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation according to claim 6, characterized in that: According to the first redundant energy supply power sequence and the first allocated power sequence, and in accordance with the frequency modulation coefficient and the peak modulation allocation coefficient, a first allocation fitness is calculated as follows: Among them, F f is the allocation fitness, w1, w2 and w3 are weights, the sum of the three is 1, N is a positive integer, TP is the frequency modulation coefficient, TF is the peak modulation allocation coefficient, K is the number of redundant energy supply powers in the redundant energy supply power sequence, σ is the variance of K redundant energy supply powers in the redundant energy supply power sequence, G i is the redundant power supply at the i-th future moment, M i is the allocated power at the i-th future moment, M i+1 is the allocated power at the i+1th future moment.

8. The method for frequency regulation and peak regulation and power distribution of a smart microgrid based on photovoltaic power generation according to claim 1, characterized in that: According to the distribution fitness, an iterative optimization evaluation of the electric energy distribution scheme is performed to optimize and obtain the optimal electric energy distribution scheme, including: Continue to randomly generate power distribution plans, calculate distribution fitness and iterative optimization; Until the iterative optimization converges, the power distribution scheme with the largest distribution fitness is retained as the optimal power distribution scheme, and the energy storage unit is controlled to distribute power in the microgrid at K future moments in the future period.

Images