Micro-grid stable operation control method and system

Abstract

The invention discloses a microgrid stable operation control method and system, and belongs to the technical field of microgrid operation control, and the method comprises the steps: obtaining the operation data of a microgrid system, and extracting the characteristic indexes related to the operation state of a photovoltaic power station from the operation data; calculating the weight of the characteristic index, dividing the operating state grade of the photovoltaic power station, and monitoring the operating state of the photovoltaic power station of the micro-grid system in real time through a CNN-SVM model; according to the operation state of a photovoltaic power station, photovoltaic power is decomposed into a low-frequency power component and a high-frequency power component, the low-frequency power component serves as photovoltaic grid-connected power, the negative part power of the high-frequency power component is borne by a hydroelectric generating set, the positive part of the high-frequency power component is absorbed by an energy storage device, and photovoltaic power station output fluctuation suppression is completed. And stable operation control of the micro-grid is realized. According to the invention, the operation efficiency and reliability of the micro-grid can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrid operation control, and more specifically, to a method and system for controlling the stable operation of a microgrid. Background Art

[0002] Currently, with the continuous growth of energy demand and the improvement of environmental protection requirements, microgrids, as a distributed energy system, have been widely applied. A water-light complementary power generation system is an intelligent microgrid system based on hydropower generation and photovoltaic power generation technologies. Due to the different working principles of hydropower generation and photovoltaic power generation, there is complementarity in their energy output, and it has the advantages of strong flexibility, high reliability, environmental protection, etc.; however, during actual operation, the photovoltaic power generation output in the microgrid system is not stable, and photovoltaic power generation has intermittency and uncertainty, and its output fluctuations may lead to instability of the grid frequency and voltage; moreover, currently, the operation control of the microgrid relies too much on manual intervention, with low efficiency, especially in areas with a high photovoltaic penetration rate, the stable operation of the microgrid system is severely affected.

[0003] Therefore, how to ensure the stable operation of the microgrid, improve the operation efficiency and reliability of the system, and study the method for controlling the stable operation of the microgrid is of great significance. Summary of the Invention

[0004] In view of this, the present invention provides a method and system for controlling the stable operation of a microgrid that at least solve the above partial technical problems, which is convenient for realizing the control of the stable operation of the microgrid and is beneficial to improving the operation efficiency and reliability of the microgrid.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] In the first aspect, the present invention provides a method for controlling the stable operation of a microgrid, and the method includes the following steps:

[0007] S1. Obtain the operation data of the microgrid system, and extract the characteristic indexes related to the operation state of the photovoltaic power station from the operation data;

[0008] S2. Calculate the weights of the characteristic indexes and divide the operation state levels of the photovoltaic power station;

[0009] S3. Take the division result of the operation state level of the photovoltaic power station as the output of the deep learning model, establish an evaluation model, and use the obtained data to input into the evaluation model to monitor the operation state of the photovoltaic power station in the microgrid system in real time;

[0010] S4. Decompose the photovoltaic power into a low-frequency power component and a high-frequency power component according to the operating state of the photovoltaic power station. The low-frequency power component is used as the photovoltaic grid-connected power. The negative part of the high-frequency power component is borne by the hydropower unit, and the positive part of the high-frequency power component is absorbed by the energy storage device, thereby suppressing the output fluctuation of the photovoltaic power station and realizing the stable operation control of the microgrid.

[0011] Preferably, in S1, the operation data of the microgrid system includes: hydropower output data, photovoltaic output data, energy storage power data, grid voltage and current data, grid load data, equipment status data, and environmental data.

[0012] Preferably, in S2, the process of calculating the weights of the characteristic indicators includes:

[0013] 1) Perform dimensionless processing on the characteristic indicators:

[0014] 2) Calculate the mean and standard deviation between the characteristic indicators. The formula is:

[0015]

[0016] In the formula, x j represents the value of the jth characteristic indicator; δ j represents the standard deviation of the jth characteristic indicator; represents the mean of the jth characteristic indicator; x ij represents the value of the jth characteristic indicator of the ith photovoltaic power generation unit; x i ′ j represents the normalization value of x ij ; m is the total number of photovoltaic power generation units; represents the normalization value of x j ;

[0017] 3) Calculate the coefficient of variation of the characteristic indicators:

[0018]

[0019] In the formula: v j is the coefficient of variation of the jth characteristic indicator;

[0020] 4) Calculate the conflict coefficient between the characteristic indicators:

[0021]

[0022] In the formula, ρ pj represents the Pearson correlation coefficient between the pth characteristic indicator and the jth characteristic indicator; the characteristic indicator number represented by p is not equal to j; X ip represents the value of the ith sample on the pth characteristic indicator; X ij represents the value of the ith sample on the jth characteristic indicator; represents the mean value of the p-th characteristic index; represents the mean value of the j-th characteristic index; f j represents the conflict coefficient of the j-th characteristic index; n is the total number of characteristic indexes;

[0023] 5) Calculate the information amount contained in each characteristic index:

[0024] G j = v j ·f j

[0025] In the formula, G j represents the information amount contained in the j-th characteristic index;

[0026] 6) Calculate the weight of the characteristic index according to the information amount contained in the characteristic index:

[0027]

[0028] In the formula, W j represents the weight of the j-th characteristic index.

[0029] Preferably, in S2, the process of dividing the operation state level of the photovoltaic power station includes:

[0030] 1) Calculate the weighted standardized evaluation matrix:

[0031] Z ij = W j ×x i ′ j

[0032] In the formula, Z ij represents the matrix after weighted standardization; W j represents the weight of the j-th characteristic index;

[0033] 2) Determine the positive ideal solution and the negative ideal solution:

[0034]

[0035] In the formula, represents the positive ideal solution of the j-th characteristic index; represents the negative ideal solution of the j-th characteristic index;

[0036] 3) Calculate the distances between each photovoltaic power generation unit and the positive ideal solution and the negative ideal solution:

[0037]

[0038] In the formula, represents the Manhattan distance between the i-th photovoltaic power generation unit and the positive ideal solution, represents the Manhattan distance between the i-th photovoltaic power generation unit and the negative ideal solution;

[0039] 4) The standardized interval is obtained by using comprehensive scoring and Z-score standardization to divide the operating state level of the photovoltaic power station. The calculation formula is as follows:

[0040]

[0041] In the formula, C i represents the comprehensive score of the i-th photovoltaic power generation unit, which is obtained according to expert evaluation. Z i represents the evaluation Z-score of the i-th photovoltaic power generation unit. μ represents the average value of the comprehensive scores of each photovoltaic power generation unit, which is obtained by taking the average of the comprehensive scores of each photovoltaic power generation unit. σ represents the standard deviation of the comprehensive scores of each photovoltaic power generation unit, which is determined according to the empirical rule.

[0042] Preferably, the Z-score standardization interval is:

[0043] The operating state level of the photovoltaic power station is divided into three levels: excellent, general, and poor. The corresponding Z-score intervals are: Z≥1, 0<Z<1, -1<Z<0.

[0044] Preferably, in S3, the established evaluation model is composed of a CNN model and an SVM model. CNN is used to extract the features of the data, and these features are used as the input of the SVM. Among them, the CNN structure is the input layer - convolutional layer - pooling layer - convolutional layer - pooling layer - fully connected layer, and the activation function is the RELU function. The division result of the operating state level of the photovoltaic power station is used as the output of the CNN-SVM model. Each convolutional kernel of the CNN extracts features from the operating data of the photovoltaic power station and integrates them into global features through the fully connected layer. The output data of the fully connected layer is used as the input of the SVM, and the SVM network is trained to realize the identification of the operating state of the photovoltaic power station.

[0045] Preferably, the improved moth-flame optimization algorithm is used to solve and optimize the SVM model.

[0046] Preferably, in S4, the photovoltaic power is decomposed into a low-frequency power component and a high-frequency power component. The process includes:

[0047] 1) The regression smoothing algorithm is used to extract the low-frequency signal and the high-frequency signal, where:

[0048] The low-frequency signal is:

[0049]

[0050] In the formula, P low(t) represents the low-frequency power signal at time t; t is the current time point; t' is the time point within the local window Window; P(t') represents the original power signal at time point t'; τ controls the window size;

[0051] The high-frequency signal is:

[0052] P high (t) = P(t) - P low (t)

[0053] In the formula, P(t) is the original signal at time t;

[0054] 2) Use wavelet decomposition to decompose the signal into different frequency levels:

[0055]

[0056] In the formula, P g details (t) represents the power change amount of the g-th layer component; o is the number of signal data; A o is the approximation component coefficient; B o is the detail component coefficient; φ(·) is the scaling function; ψ(·) is the wavelet function; L is the total number of decomposition layers.

[0057] In a second aspect, the present invention further provides a microgrid stable operation control system, which is applied to the above-mentioned microgrid stable operation control method to achieve microgrid stable operation control. The system includes:

[0058] A data acquisition module, configured to acquire the operation data of the microgrid system and extract the characteristic indexes related to the operation state of the photovoltaic power station from the operation data;

[0059] A state division module, configured to calculate the weights of the characteristic indexes and divide the operation state levels of the photovoltaic power station;

[0060] A state monitoring module, configured to use the division result of the operation state level of the photovoltaic power station as the output of the deep learning model, establish an evaluation model, and use the acquired data to input the evaluation model to monitor the operation state of the photovoltaic power station in the microgrid system in real time;

[0061] An operation control module, configured to decompose the photovoltaic power in the microgrid into a low-frequency power component and a high-frequency power component according to the operation state of the photovoltaic power station. The low-frequency power component is used as the photovoltaic grid-connected power, the negative part of the high-frequency power component is borne by the hydropower unit, and the positive part of the high-frequency power component is absorbed by the energy storage device, so as to complete the suppression of the output fluctuation of the photovoltaic power station and achieve the stable operation control of the microgrid.

[0062] In a third aspect, the present invention further provides an electronic device, including a processor and a memory. The memory stores machine-executable instructions that can be executed by the processor, and the processor executes the machine-executable instructions to implement a method for controlling the stable operation of a microgrid as described above.

[0063] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0064] The present invention provides a method and a system for controlling the stable operation of a microgrid. First, the present invention preliminarily divides the operation state level of a photovoltaic power station, and monitors the operation state of the photovoltaic power station through a CNN-SVM model, improving the accuracy and robustness of the evaluation of the operation state of the photovoltaic power station, and enabling real-time grasp of the operation state of the photovoltaic power station. Further, the photovoltaic power is decomposed into a low-frequency power component and a high-frequency power component. The low-frequency power component is used as the photovoltaic grid-connected power, the negative part of the high-frequency power component is borne by a hydropower unit, and the positive part of the high-frequency power component is absorbed by an energy storage device, completing the suppression of the output power fluctuation of the photovoltaic power station, achieving a smooth photovoltaic power generation curve, realizing the stable operation control of the microgrid, and improving the operation efficiency and reliability of the microgrid.

[0065] Other features and advantages of the present invention will be described in the following specification, and some of them will become obvious from the specification or be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained by the structures specifically pointed out in the written specification and the drawings.

[0066] The technical solutions of the present invention will be further described in detail below with reference to the drawings and embodiments. Description of the Drawings

[0067] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0068] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, but do not constitute a limitation to the present invention.

[0069] Figure 1 It is a schematic flowchart of a method for controlling the stable operation of a microgrid provided by an embodiment of the present invention.

[0070] Figure 2 It is a schematic flowchart of the evaluation of the operation state of a photovoltaic power station based on CNN-SVM provided by an embodiment of the present invention.

[0071] Figure 3 This is a schematic structural diagram of a microgrid stable operation control system provided by an embodiment of the present invention.

[0072] Figure 4 This is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Specific embodiments

[0073] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention.

[0074] In the description of the present invention, it should be noted that: in some processes described in the specification and drawings of this application, there are multiple operations that appear in a specific order. However, it should be clearly understood that these operations may not be executed in the order in which they appear herein or may be executed in parallel. In addition, various serial numbers, etc. are only for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0075] Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed present invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0076] See Figure 1 As shown, an embodiment of the present invention provides a method for controlling the stable operation of a microgrid, mainly including the following steps:

[0077] S1. Read the operation data of the microgrid system (photovoltaic-hydro complementary power generation system) from the Supervisory Control And Data Acquisition (SCADA) system, and extract the main characteristic indicators representing the operation status of the photovoltaic power station according to the operation data curve of the photovoltaic power station therein;

[0078] S2. Calculate the weights of the characteristic indicators and divide the operation status levels of the photovoltaic power station;

[0079] S3. Use the division result of the operation status level of the photovoltaic power station as the output of the CNN-SVM model, establish an online evaluation model for the operation status of the photovoltaic power station based on CNN-SVM, and realize the real-time monitoring and evaluation of the operation status of the photovoltaic power station;

[0080] S4. According to the operating state of the photovoltaic power station, the photovoltaic power is decomposed into a low-frequency power component and a high-frequency power component. The low-frequency power component is used as the photovoltaic grid-connected power, the negative part of the high-frequency power component is borne by the hydropower unit, and the positive part of the high-frequency power component is absorbed by the energy storage device, thereby suppressing the output fluctuation of the photovoltaic power station and realizing the stable operation control of the microgrid.

[0081] The following details the specific implementation manner and working principle of the method of the present invention:

[0082] In the embodiment of the present invention, the characteristic indexes of the operating state of the photovoltaic power station can be mainly divided into characteristic indexes such as light intensity, actual power generation, peak power output, and conversion efficiency of photovoltaic modules.

[0083] In the embodiment of the present invention, the calculation process of the weight in the above S2 is as follows:

[0084] First, construct the original evaluation matrix:

[0085]

[0086] In the formula, m is the total number of photovoltaic units, n is the number of indexes, and x ij represents the j-th index value of the i-th photovoltaic unit.

[0087] Then, perform dimensionless processing on each index; further, calculate the average value and standard deviation between the dimensionless processed indexes. The formula is:

[0088]

[0089] In the formula, x j represents the j-th index value; δ j represents the standard deviation of the j-th index; represents the mean value of the j-th index; x ij represents the j-th index value of the i-th photovoltaic power generation unit; x i ′ j represents the normalization value of x ij ; m is the total number of photovoltaic power generation units;

[0090] Further, calculate the coefficient of variation of the indexes:

[0091]

[0092] In the formula: v j is the coefficient of variation of the j-th index;

[0093] Further, calculate the conflict coefficient between the indexes:

[0094]

[0095] Where ρ pj represents the Pearson correlation coefficient between the p-th index and the j-th index; the index number represented by p is not equal to j; X ip represents the value of the i-th sample on the p-th index; X ij represents the value of the i-th sample on the j-th index; represents the mean value of the p-th index; represents the mean value of the j-th index; f j represents the conflict coefficient of the j-th index; n is the total number of indexes;

[0096] Furthermore, calculate the information amount contained in each index:

[0097] G j = v j ·f j

[0098] Where G j represents the information amount contained in the j-th index;

[0099] Furthermore, calculate the weight of the index according to the information amount contained in the index:

[0100]

[0101] Where W j represents the weight of the j-th index.

[0102] Furthermore, the process of dividing the operation state level of the photovoltaic power station in the embodiment of the present invention includes:

[0103] 1) Calculate the standardized evaluation matrix:

[0104] Z ij = W j × x i ′ j

[0105] Where Z ij represents the matrix after weighted standardization;

[0106] 2) Determine the positive ideal solution and the negative ideal solution

[0107] Among them, the positive ideal solution is the solution where each evaluation index reaches the optimal, and the negative ideal solution is the solution where each evaluation index reaches the worst. Specifically:

[0108]

[0109] 3) Calculate the Manhattan distances between each index value and the positive ideal solution and the negative ideal solution:

[0110]

[0111] In the formula, represents the Manhattan distance between the i-th photovoltaic power generation unit and the positive ideal solution, represents the Manhattan distance between the i-th photovoltaic power generation unit and the negative ideal solution;

[0112] 4) Obtain the evaluation result by using comprehensive scoring and dividing the standardized interval through Z-score standardization, which is calculated as follows:

[0113]

[0114] In the formula, C i represents the comprehensive score of the data of each photovoltaic power generation unit, Z i represents the Z-score of the i-th photovoltaic power generation unit evaluation, μ represents the average value of the comprehensive scores of each photovoltaic power generation unit, and σ represents the standard deviation of the comprehensive scores of each photovoltaic power generation unit.

[0115] In this embodiment, the Z-score standardization interval is: The operating state level of the photovoltaic power station is divided into three levels: excellent, general, and poor, and the corresponding Z-score intervals are: Z≥1, 0<Z<1, -1<Z<0.

[0116] Furthermore, as shown in Figure 2 the modeling process of the online evaluation model for the operating state of the photovoltaic power station based on CNN-SVM fusion in the above S3 is as follows:

[0117] 1) Perform normalization processing on the original index data of the photovoltaic power station and the preliminary operating state evaluation results (the divided operating state levels of the photovoltaic power station); 2) Randomly divide the normalized data samples into a training set and a test set to ensure data diversity and universality during the model training process; 3) Input the training set into the CNN model, and by continuously adjusting the model parameters, make the loss function converge to obtain a stable and accurate feature representation; 4) Use the fully connected layer of the trained CNN model to extract feature vectors, which are used as the input of the SVM classifier to further train the SVM; 5) Input the test set into the CNN-SVM model to realize the identification of the operating state of the photovoltaic power station and evaluate its classification accuracy and robustness.

[0118] In the embodiment of the present invention, the structure of the performance evaluation model for hydropower units based on CNN-SVM is as follows:

[0119] The CNN-SVM model consists of a CNN model and an SVM model optimized based on the improved moth-to-flame algorithm. CNN is used to extract image features and these features are used as the input of the SVM. The CNN structure is input layer-convolution layer-pooling layer-convolution layer-pooling layer-fully connected layer, and the activation function is the RELU function. Each convolution kernel of CNN extracts features from the operating data of the hydropower unit, which are integrated into global features through the fully connected layer. The output data of the fully connected layer is used as the input of the SVM, and the SVM network is trained to realize the recognition of the operating status of the photovoltaic power station.

[0120] The improved moth-to-fire algorithm adopted by the present invention is introduced below:

[0121] 1) Moth-to-flame optimization algorithm

[0122] The moth-to-fire optimization algorithm is an intelligent evolutionary algorithm inspired by the horizontal positioning mechanism of moths. When flying at night, moths complete positioning by fixing a certain angle with the moonlight. Since moonlight can be approximately regarded as parallel light, moths can fly in a straight line. When the moth flies around a point light source, since the moth is still flying at a certain angle with the light, the flight path will appear in a spiral shape, and eventually the moth will hit the point light source. This is the "moth-to-fire" phenomenon.

[0123] Inspired by the phenomenon of moths flying into flames, Mirjalili proposed the Moth-to-Fire Optimization (MFO) algorithm in 2015. The MFO algorithm has a moth population and a flame population. The moth population can be understood as a universal population in evolutionary algorithms, and the flame population is the population with the best historical composition of all moth individuals during the evolution process, and its size is the same as the moth population. Moth individuals are updated using a spiral function centered on the flame individual. When the newly generated moth individual is better than the flame individual, the flame individual is replaced by the new moth individual. The spiral update function is as follows:

[0124]

[0125] Among them, M ij and M i ' j is the value of the jth decision variable of the i-th moth individual before and after the update; F ij is the jth decision variable of the i-th flame individual; D ij is the distance between the flame individual and the moth individual; parameter t is a random number between -1 and 1; parameter b is usually set to 1.

[0126] 2) Improve the moth-to-flame optimization algorithm

[0127] The reasons for the poor convergence of MFO are: ① The center of the spiral update formula is the flame individual F ij , when the moth individual Mij When not updated to a better position than F ij The update formula center will remain unchanged all the time, which greatly limits the ability of the algorithm to overcome premature convergence. ② The flame individual is the historical optimal solution of its corresponding moth individual, and the update of the moth individual is only affected by its corresponding flame individual, that is, there is no communication between moth individuals, which will also make the optimization result fall into a local optimum.

[0128] Aiming at the problem of poor convergence of the MFO algorithm, an Improved Moth-flame Optimization Algorithm (IMFO) is proposed in the present invention. IMFO improves the MFO algorithm from three aspects: update formula, linear flight path inspiration, and flame population update strategy.

[0129] (a) Improvement of update formula

[0130] The IMFO spiral update formula is generally similar to the MFO algorithm, but there are also the following three improvements:

[0131] ① The center of the spiral changes from the flame individual F ij to the mean value 0.5(F ij +M ij ) of the flame individual and the moth individual. Since the moth individual M ij changes at each iteration, the center of the spiral also changes at each iteration, which is beneficial to jumping out of the local optimal solution.

[0132] ② The formula of the distance influence parameter c is improved through multiple experiments and trial and error.

[0133] ③ The absolute value symbol of the distance D ij between the moth individual and the flame individual is removed. If D ij is restricted to a positive number, the search process will reduce the search possibility by half.

[0134] The improved spiral update formula is as follows:

[0135]

[0136] Among them, the parameter t is a random number between 0 and 2; the parameter b is a constant parameter; c is the distance influence parameter.

[0137] (b) Linear flight path inspiration

[0138] The MFO algorithm only inspires the spiral flight path of moths and does not inspire the linear flight path. In the embodiments of the present invention, the creation of the lunar population Mo is further inspired. In the single-objective optimization algorithm, the lunar population Mo can be a population composed of the current global optimum in the entire search process, and the number of individuals in the population is 1; in the multi-objective optimization algorithm, the lunar population Mo is an external archive set maintained by a specific multi-objective mechanism, and the number of individuals in the population is the same as that of the moth population.

[0139] The linear update formula of moths is as follows:

[0140]

[0141] where Mo rj is the j-th decision variable of the r-th lunar individual; r means randomly selecting an individual from the lunar population to participate in the update. Since the lunar population represents the optimal solution or set of optimal solutions in the current entire search process, linear update will accelerate convergence.

[0142] (c) Flame population update strategy

[0143] In the single-objective optimization algorithm, the newly generated moth individual M i ′ can directly compare with the flame individual F i through the objective value to determine whether to use the newly generated moth individual M i ′ to replace the flame individual F i . However, in the multi-objective optimization algorithm, simply relying on the objective function value may not necessarily compare the newly generated moth individual M i ′ with the flame individual F i . If the newly generated moth individual M i ′ dominates the flame individual F i , then M i ′ is used to replace F i ; if F i dominates M i ′, then F i remains unchanged; if F i and M i ′ do not dominate each other, a random individual Mo r is selected from the lunar population to replace F i . This improvement not only retains the newly generated excellent individuals, but also enhances the communication between individuals, which is beneficial to improving the convergence speed and avoiding falling into local optima.

[0144] In a specific embodiment, the accuracy rate (Acc), recall rate (Rec), and macro-average value (F1) are used to evaluate the model test results, and the recognition results of the operating status of a photovoltaic power station obtained by the following 4 models are compared, as shown in Table 1 below.

[0145] Table 1 Comparison of Identification Results of the Operating Status of Photovoltaic Power Stations

[0146] ACC REC F1 CNN 96.1％ 91.8％ 83.2％ SVM 95.2％ 81.9％ 86.1％ CNN-BP 97％ 91.8％ 87.1％ CNN-SVM 97％ 94.3％ 92.1％

[0147] It can be seen from the comparison results that the method of the present invention can significantly improve the accuracy and robustness of the identification of the operating status of photovoltaic power stations.

[0148] Furthermore, in the embodiment of the present invention, according to the operating status of the photovoltaic power station (for example, when the status level is poor, the Z - score interval is - 1 < Z < 0), the photovoltaic power is decomposed into a low - frequency power component and a high - frequency power component. The low - frequency power component represents the average output characteristics of the photovoltaic system on a longer time scale, while the high - frequency power component reflects the rapid fluctuations caused by factors such as weather changes in a short period of time. Among them:

[0149] Processing of the low - frequency power component:

[0150] Due to its relatively stable characteristics, the low - frequency power component can be directly used as the photovoltaic grid - connected power, that is, this part of the power is directly input into the grid to provide electrical energy for users. This method helps to maintain the basic load demand of the grid.

[0151] Processing of the high - frequency power component:

[0152] Negative part: For the negative part of the high - frequency power component, it means that the actual output of the photovoltaic system at a certain moment is lower than the expected value. At this time, the power generation of the hydro - generator set can be adjusted to compensate for this part of the missing power to ensure that the overall power balance of the grid is not affected.

[0153] Positive part: When there is a negative part in the high - frequency power component, it means that the electric power instantaneously generated by the photovoltaic system exceeds the current demand. This part of the excess electric energy can be stored in an energy storage device, such as a lithium iron phosphate battery energy storage system. This can not only avoid waste but also release the stored energy when needed to help regulate the grid. Through the above - mentioned method, the present invention can effectively suppress the rapid fluctuations of the output of the photovoltaic power station, reduce the impact on the grid, balance through the configuration of appropriate energy storage devices, facilitate reliable and stable power supply, and thus realize the dynamic operation control of the micro - grid and ensure the stable operation of the micro - grid.

[0154] In the embodiment of the present invention, preferably, adaptive locally weighted regression smoothing and wavelet transform can be used to complete the decomposition of photovoltaic power. The specific algorithm is as follows:

[0155] 1) Use the regression smoothing algorithm to extract the low - frequency signal and the high - frequency signal, where:

[0156] The low - frequency signal is:

[0157]

[0158] Wherein, P low (t) represents the low-frequency power signal at time t; t is the current time point; t' is the time point within the local window Window; P(t') represents the original power signal at time point t'; τ controls the window size;

[0159] The high-frequency signal is:

[0160] P high (t) = P(t) - P low (t)

[0161] Wherein, P(t) is the original signal at time t;

[0162] 2) Decompose the signal into different frequency levels by wavelet decomposition:

[0163]

[0164] Wherein, P g details (t) represents the power change amount of the g-th layer component; o is the number of signal data; A o is the approximation component coefficient; B o is the detail component coefficient; φ(·) is the scaling function; ψ(·) is the wavelet function; L is the total number of decomposition layers.

[0165] From the description of the above embodiments, those skilled in the art can know that: The present invention provides a method for controlling the stable operation of a microgrid. The present invention first preliminarily divides the operation state level of a photovoltaic power station, monitors the operation state of the photovoltaic power station through a CNN-SVM model, improves the accuracy and robustness of the evaluation of the operation state of the photovoltaic power station, and can grasp the operation state of the photovoltaic power station in real time; further, decomposes the photovoltaic power into a low-frequency power component and a high-frequency power component, the low-frequency power component is used as the photovoltaic grid-connected power, the negative part of the high-frequency power component is borne by the hydropower unit, and the positive part of the high-frequency power component is absorbed by the energy storage device, completing the suppression of the output fluctuation of the photovoltaic power station, achieving a smooth photovoltaic power generation curve, realizing the stable operation control of the microgrid, and improving the operation efficiency and reliability of the microgrid.

[0166] Further, as shown in Figure 3 , the present invention also provides a control system for the stable operation of a microgrid, which is applied to a method for controlling the stable operation of a microgrid described in the above embodiments to realize the stable operation control of the microgrid. The system includes:

[0167] A data acquisition module, configured to acquire the operation data of the microgrid system and extract the characteristic indexes related to the operation state of the photovoltaic power station from the operation data;

[0168] A state division module, configured to calculate the weights of feature indicators and divide the operation state levels of a photovoltaic power station;

[0169] A state monitoring module, configured to use the division result of the operation state level of the photovoltaic power station as the output of a deep learning model, establish an evaluation model, and use the acquired data to input into the evaluation model to monitor the operation state of the photovoltaic power station in the microgrid system in real time;

[0170] An operation control module, configured to decompose the photovoltaic power in the microgrid into a low-frequency power component and a high-frequency power component according to the operation state of the photovoltaic power station. The low-frequency power component is used as the photovoltaic grid-connected power, the negative part of the high-frequency power component is borne by the hydropower unit, and the positive part of the high-frequency power component is absorbed by the energy storage device, so as to complete the suppression of the output power fluctuation of the photovoltaic power station and realize the stable operation control of the microgrid.

[0171] The microgrid stable operation control system provided by the embodiments of the present invention has the same implementation principle and the same technical effects as those of the foregoing method embodiments. For the sake of brief description, for the parts not mentioned in this embodiment, reference may be made to the corresponding content in the foregoing method embodiments, and details are not described herein again.

[0172] In addition, referring to Figure 4 As shown, the embodiments of the present invention further provide an electronic device, which may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13, and may further include a computer program stored in the memory 11 and executable on the processor 10. The processor executes the computer program to implement a microgrid stable operation control method in the foregoing method embodiments.

[0173] Among them, the processor 10 may be composed of integrated circuits in some embodiments. For example, it may be composed of a single packaged integrated circuit, or may be composed of multiple integrated circuits with the same or different functions, including the combination of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control core of the electronic device, connects various components of the entire electronic device through various interfaces and lines, and executes various functions of the electronic device and processes data by running or executing programs or modules stored in the memory 11 and calling data stored in the memory 11.

[0174] In one example, the microgrid system with optimized stable operation control in the present invention can switch to island mode operation and interact with systems such as charging piles and air conditioners within the regional scope. When the microgrid generates sufficient electricity, it preferentially charges electric vehicles or powers central air conditioners to ensure the normal operation of other important loads. And it uses energy storage devices to store electric energy for use during peak hours. This is beneficial to building a smart energy ecosystem, can effectively improve energy utilization efficiency, reduce carbon emissions, and at the same time provide more convenient and economical services for users.

[0175] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products, etc. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0176] It should be noted that the word "comprising" does not exclude the existence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the existence of a plurality of such components. The present invention can be implemented by means of hardware including several different components and by means of a suitably programmed computer.

[0177] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0178] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A microgrid stable operation control method, characterized in that: The method comprises the following steps: S1. Acquire the operation data of the microgrid system, and extract characteristic indicators related to the operation status of the photovoltaic power station from the operation data; S2. Calculate the weights of characteristic indicators and classify the operating status of the photovoltaic power station; S3, using the photovoltaic power station operation status classification result as the output of the deep learning model, establishing an evaluation model, and using the acquired operation data to input the evaluation model to monitor the photovoltaic power station operation status of the microgrid system in real time; S4. According to the operating status of the photovoltaic power station, the photovoltaic power is decomposed into a low-frequency power component and a high-frequency power component. The low-frequency power component is used as the photovoltaic grid-connected power, the negative part of the high-frequency power component is borne by the hydropower unit, and the positive part of the high-frequency power component is absorbed by the energy storage device, so as to suppress the output fluctuation of the photovoltaic power station and realize the stable operation control of the microgrid.

2. A microgrid stable operation control method according to claim 1, characterized in that: In S1, the operation data of the microgrid system includes: hydropower output data, photovoltaic output data, energy storage power data, grid voltage and current data, grid load data, equipment status data and environmental data.

3. A microgrid stable operation control method according to claim 1, characterized in that: In S2, the process of calculating the weight of the characteristic index includes: 1) Perform dimensionless processing on the characteristic index: 2) Calculate the mean and standard deviation of each characteristic index using the following formula: In the formula, x j represents the jth characteristic index value; δ j represents the standard deviation of the jth characteristic index; represents the average value of the jth characteristic index; x ij represents the jth characteristic index value of the i-th photovoltaic power generation unit; x i ' j Represents x ij Normalized value; m is the total number of photovoltaic power generation units; Represents x j The normalized value of 3) Calculate the coefficient of variation of the characteristic index: Where: v j is the coefficient of variation of the jth characteristic index; 4) Calculate the conflict coefficient between characteristic indicators: In the formula, ρ pj represents the Pearson correlation coefficient between the pth feature index and the jth feature index; the index number represented by p is not equal to j; X ip represents the value of the i-th sample on the p-th feature index; X ij Represents the value of the i-th sample on the j-th feature index; represents the mean of the pth characteristic index; represents the mean value of the jth characteristic index; f j represents the conflict coefficient of the jth characteristic index; n is the total number of indicators; 5) Calculate the amount of information contained in each characteristic indicator: G j =v j ·f j In the formula, G j Indicates the amount of information contained in the jth indicator; 6) Calculate the weight of the feature index according to the amount of information contained in the feature index: Where W j Represents the weight of the jth feature index.

4. A microgrid stable operation control method according to claim 3, characterized in that: In S2, the process of classifying the operating status of the photovoltaic power station includes: 1) Calculate the weighted standardized evaluation matrix: WITH ij =In j ×x′ ij In the formula, Z ij represents the weighted normalized matrix; 2) Determine the positive ideal solution and the negative ideal solution: In the formula, represents the positive ideal solution of the jth index; represents the negative ideal solution of the jth index; 3) Calculate the distance between each photovoltaic power generation unit and the positive ideal solution and the negative ideal solution: In the formula, represents the Manhattan distance between the ith photovoltaic power generation unit and the positive ideal solution, represents the Manhattan distance between the i-th photovoltaic power generation unit and the negative ideal solution; 4) The comprehensive score and Z score are used to standardize and divide the standardized interval to obtain the operating status level of the photovoltaic power station. The calculation formula is: In the formula, C i represents the comprehensive score of the ith photovoltaic power generation unit, Z i represents the Z score of the ith photovoltaic power generation unit evaluation, μ represents the average value of the comprehensive scores of each photovoltaic power generation unit, and σ represents the standard deviation of the comprehensive scores of each photovoltaic power generation unit.

5. A microgrid stable operation control method according to claim 4, characterized in that: The Z-score standardized interval is: The operating status of photovoltaic power stations is divided into three levels: excellent, general, and poor. The corresponding Z score range is: Z ≥ 1, 0 <Z<1,-1<Z<0。 6. A microgrid stable operation control method according to claim 1, characterized in that: In S3, the established evaluation model consists of a CNN model and an SVM model. The CNN model is used to extract the features of the operating data, and these features are used as the input of the SVM model. Among them, the CNN model structure is input layer-convolution layer-pooling layer-convolution layer-pooling layer-fully connected layer, and the activation function is the RELU function. The photovoltaic power station operation status classification result is used as the output of the CNN-SVM model. Each convolution kernel of the CNN model extracts features from the photovoltaic power station operation data, which are integrated into global features through the fully connected layer. The output data of the fully connected layer is used as the input of the SVM model, and the SVM model network is trained to realize the recognition of the photovoltaic power station operation status.

7. A microgrid stable operation control method according to claim 6, characterized in that: The improved moth-to-fire algorithm is used to solve and optimize the SVM model.

8. A microgrid stable operation control method according to claim 1, characterized in that: In S4, the photovoltaic power is decomposed into a low-frequency power component and a high-frequency power component. The process includes: 1) Use regression smoothing algorithm to extract low-frequency signals and high-frequency signals, where: The low frequency signal is: Where P low (t) represents the low-frequency power signal at time t; t is the current time point; t' is the time point in the local window Window; P(t') represents the original power signal at time point t'; τ controls the window size; The high frequency signal is: P high (t)=P(t)-P low (t) Where P(t) represents the original power signal at time t; 2) Use wavelet decomposition to decompose the signal into different frequency levels: Where P g details (t) represents the power change of the g-th layer component; o is the number of signal data; A o is the approximate component coefficient; B o is the detail component coefficient; φ(·) is the scaling function; ψ(·) is the wavelet function; L is the total number of decomposition levels.

9. A microgrid stable operation control system, characterized in that: When applied, a microgrid stable operation control method according to any one of claims 1 to 8 is executed to realize the stable operation control of the microgrid, and the system includes: A data acquisition module is used to acquire the operation data of the microgrid system and extract characteristic indicators related to the operation status of the photovoltaic power station from the operation data; The state classification module is used to calculate the weight of characteristic indicators and classify the operating state levels of photovoltaic power plants; A status monitoring module is used to use the photovoltaic power station operation status classification result as the output of the deep learning model, establish an evaluation model, and use the acquired data to input the evaluation model to monitor the photovoltaic power station operation status of the microgrid system in real time; The operation control module is used to decompose the photovoltaic power in the microgrid into low-frequency power components and high-frequency power components according to the operating status of the photovoltaic power station. The low-frequency power component is used as the photovoltaic grid-connected power, the negative part of the high-frequency power component is borne by the hydropower unit, and the positive part of the high-frequency power component is absorbed by the energy storage device, so as to suppress the output fluctuation of the photovoltaic power station and realize the stable operation control of the microgrid.

10. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores machine executable instructions that can be executed by the processor, and the processor executes the machine executable instructions to implement a microgrid stable operation control method as described in any one of claims 1-8.

Images