Power smooth control method and system for water-light-storage complementary system

Abstract

The invention discloses a water-light-storage complementary system power smooth control method and system. The method comprises the following steps: extracting a characteristic index representing an operation state from operation data of a hydroelectric generating set; calculating objective weights of the indexes; dividing operation state grades; a hydroelectric generating set operation state online evaluation model based on CNN-GSVM is established, and the hydroelectric generating set operation state is output; photovoltaic power is decomposed into a low-frequency component and a high-frequency detail component by combining an adaptive local weighted regression smoothing algorithm with wavelet decomposition, the low-frequency component serves as photovoltaic grid-connected power, and the negative part power of the high-frequency detail component is borne by hydroelectric generating sets with different performance levels. And the power of the positive value part of the high-frequency detail component and the negative value part which is not absorbed by the hydroelectric generating set is borne by the energy storage battery, so that smooth control of the photovoltaic power is realized. According to the method, the real-time operation state of the hydroelectric generating set is considered, and the electric energy quality of photovoltaic power generation is further improved while the photovoltaic power fluctuation is stabilized according to the water-light-storage complementary characteristic.

Description

Technical Field

[0001] The present invention relates to the technical field of clean energy power generation, and more specifically to a power smoothing control method and system for a water-photovoltaic-storage complementary system. Background Art

[0002] The hydro-photovoltaic complementary power generation system is an intelligent microgrid system based on hydropower and photovoltaic power generation technologies. Since hydropower and photovoltaic power generation have different working principles, they are complementary in energy output. By configuring appropriate energy storage devices, the hydro-photovoltaic complementary power generation system can balance the energy output of photovoltaic power generation and hydropower generation to achieve reliable and stable power supply.

[0003] As a key equipment in the field of hydropower generation, the operating status assessment of hydropower generating units is crucial to ensure the safe, reliable and stable operation of power stations. With the continuous complexity, integration and intelligence level of the unit structure, traditional online monitoring and regular maintenance can no longer meet the needs of timely grasp of the operating status and performance change trends of the unit. How to use the unit operation data to extract feature information and realize accurate and rapid assessment of the unit operation status has become a major challenge in the field of intelligent operation and maintenance of large-scale equipment. Therefore, the operating status assessment of hydropower units is of great significance to help power station managers better understand the operating status of the unit, timely discover potential failure risks, formulate effective maintenance strategies, and maximize the utilization and operating efficiency of the unit.

[0004] Since traditional hydropower unit status assessment methods have the problem of poor data applicability, there are too many subjective factors in the process of assigning indicator weights, resulting in insufficient objectivity of the evaluation results. Moreover, most of the evaluation results are relative values ​​rather than absolute values, and the evaluation results show great differences as the data changes. Deep learning methods have been applied to hydropower unit performance evaluation to a certain extent, but their algorithm models have high requirements for data sets, and the evaluation results of training sets often lack certain theoretical basis. If traditional assessment methods are combined with deep learning, the shortcomings of the above-mentioned assessment methods can be effectively solved and the advantages of the two can be combined, so that the evaluation results have good data applicability and objectivity.

[0005] Therefore, based on the evaluation of the operating status of the hydropower unit, relying on the rapid adjustment characteristics of the hydropower generator unit and the adjustment capacity of the hydropower station reservoir, the photovoltaic power is decomposed into low-frequency components and high-frequency detail components. The low-frequency components are used as the grid-connected power, and the high-frequency detail components are borne by the hydropower units with different performances. The energy storage equipment bears the high-frequency detail components that are not absorbed, so as to achieve the purpose of smoothing the photovoltaic power generation curve and improving the quality of photovoltaic power generation. Summary of the invention

[0006] In view of this, the present invention provides a power smoothing control method and system for a water-photovoltaic-storage complementary system that solves at least some of the above-mentioned technical problems, which can realize power smoothing control of the water-photovoltaic-storage complementary system and achieve the purpose of further improving the quality of photovoltaic power generation.

[0007] To achieve the above object, the technical solution adopted by the present invention is:

[0008] In a first aspect, the present invention provides a power smoothing control method for a water-photovoltaic-storage complementary system, the method comprising the following steps:

[0009] S1. Obtain the operating data of the hydropower unit and the photovoltaic power station, and extract characteristic indicators representing the operating status of the hydropower unit;

[0010] S2. Using the coefficient of variation method to improve the CRITIC method, calculate the objective weight of the characteristic index;

[0011] S3. The TOPSIS method is improved by introducing Manhattan distance and Z score standardization, and the improved TOPSIS method is used to classify the operating status of the hydropower units in combination with the objective weights;

[0012] S4. The result of the classification of the operating status of the hydropower unit is used as the output of the CNN-GSVM model, and an online evaluation model of the operating status of the hydropower unit based on CNN-GSVM is established to realize real-time monitoring and evaluation of the operating status of the hydropower unit;

[0013] S5. Based on the adaptive local weighted regression smoothing algorithm combined with wavelet decomposition, the photovoltaic power is decomposed into low-frequency components and high-frequency detail components. The low-frequency components are used as photovoltaic grid-connected power. The negative part of the high-frequency detail components is borne by hydropower units with different operating status levels, and the positive part of the high-frequency detail components and the negative part of the power not absorbed by the hydropower units are borne by energy storage. According to the power generation characteristics of photovoltaic power stations and hydropower units, a coordinated control strategy for the water-photovoltaic-storage complementary system is formulated to achieve smooth control of photovoltaic power.

[0014] Furthermore, in step S1, the indicator categories of characteristic indicators characterizing the operating status of the hydropower unit include: swing, vibration, pressure pulsation and displacement speed.

[0015] Furthermore, in step S2, the process of calculating the objective weight of the characteristic index based on the improved CRITIC method includes:

[0016] 1) Construct the original evaluation matrix:

[0017]

[0018] In the formula, x ijrepresents the jth index value of the i-th unit, m is the number of test units, and n is the number of indicators;

[0019] 2) Dimensionless processing of various indicators:

[0020] The positive indicator processing formula is:

[0021]

[0022] The reverse indicator processing formula is:

[0023]

[0024] 3) Calculate the mean and standard deviation of each indicator:

[0025]

[0026] In the formula, x j represents the jth index value; represents the mean value of the jth index; δ j represents the standard deviation of the jth index, x′ ij Represents x ij Normalized value;

[0027] 4) Calculate the coefficient of variation of the indicator:

[0028]

[0029] Where: v j is the coefficient of variation of the jth indicator;

[0030] 5) Calculate the conflict coefficient between indicators:

[0031]

[0032] In the formula, ρ pj represents the Pearson correlation coefficient between the pth indicator and the jth indicator; the indicator 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 of the pth indicator; represents the mean value of the jth indicator;

[0033] 6) Calculate the amount of information contained in each indicator:

[0034] G j =v j ·f j

[0035] In the formula, G jrepresents the amount of information contained in the jth indicator; f j represents the conflict coefficient of the jth indicator;

[0036] 7) Calculate the objective weight of the indicator based on the amount of information contained in the indicator:

[0037]

[0038] Where W j Represents the objective weight of the jth indicator.

[0039] Furthermore, in step S3, the process of using the improved TOPSIS method to classify the operating status of the hydropower unit includes:

[0040] 1) Calculate the weighted standardized evaluation matrix:

[0041] Z ij =W j ×x′ ij

[0042] In the formula, Z ij represents the weighted normalized matrix;

[0043] 2) Determine the positive ideal solution and the negative ideal solution:

[0044]

[0045] In the formula, represents the positive ideal solution of the jth index; represents the negative ideal solution of the jth index;

[0046] 3) Calculate the Manhattan distance between each unit and the positive ideal solution and the negative ideal solution:

[0047]

[0048] In the formula, represents the Manhattan distance between the ith unit and the positive ideal solution, represents the Manhattan distance between the ith unit and the negative ideal solution;

[0049] 4) The final evaluation result is obtained by comprehensive scoring and Z-score standardization and division into standardized intervals. The calculation formula is:

[0050]

[0051] In the formula, C i represents the comprehensive score of the i-th unit, Z i represents the Z score of the ith unit evaluation, μ represents the average of the comprehensive scores of each unit, and σ represents the standard deviation of the comprehensive scores of each unit.

[0052] Furthermore, the Z-score standardized interval is:

[0053] The performance of hydropower units is divided into four levels: 1-excellent, 2-good, 3-medium, and 4-poor;

[0054] The corresponding Z score interval is: Z ≥ 1, 0 <Z<1,-1<Z<0,Z≤-1。

[0055] Furthermore, in step S4, the online evaluation model of the operating status of the hydropower unit based on CNN-GSVM is composed of a CNN model and a GSVM model, and CNN is used to extract the features of the image, and these features are used as the input of the GSVM; wherein the CNN structure is input layer-convolution layer-pooling layer-convolution layer-pooling layer-fully connected layer, and the activation function is a RELU function; the result of the hydropower unit operating status classification is used as the output of the CNN-GSVM model, each convolution kernel of the CNN extracts features from the hydropower unit operating data, which are integrated into global features through the fully connected layer, and the output data of the fully connected layer is used as the input of the GSVM, and the GSVM network is trained to realize the recognition of the operating status of the hydropower unit.

[0056] Furthermore, the GSVM model is an SVM model optimized based on the Grey Wolf Algorithm GWO, and the optimization process includes:

[0057] 1) Assuming that the total number of wolves in the initial group is Q, set the Q×2-dimensional wolf group position vector X(0) = (c K ,g K )K=1,2,…,Q,c K and g K represents the parameters of the Kth wolf SVM model. The fitness of each gray wolf is evaluated and sorted according to the SVM recognition accuracy, and the initial optimal wolf position X is selected. α,β,δ ;

[0058] 2) The wolf pack follows the optimal wolf position for update. During the tth gray wolf search, the optimal wolf position X obtained last time is updated. α,β,δ Updated wolf pack locations as follows:

[0059] eX α,β,δ =e(c α,β,δ ,g α,β,δ )

[0060] a=2-2t / t max

[0061]

[0062] Where X(t+1) is the updated wolf pack position after the t-th search; X α，βδ is the 1×2 dimensional position vector of α, β, δ wolves in the t-th search; tmax is the maximum number of iterations; a is the convergence factor, which decreases linearly in [0,2] during the iteration process; e is the unit column vector; A and C are coefficient matrices; r1 and r2 are random number column vectors; I is the unit matrix; A 1,2,3 , C 1,2,3 is the coefficient matrix corresponding to α, β, δ wolves; X 1,2,3 Update the intermediate vectors for the positions of α, β, and δ wolves;

[0063] 3) Perform fitness evaluation and sorting on the updated wolf positions in X(t+1) to obtain the optimal values ​​of the penalty factor c and kernel function parameter g, and use them to optimize the SVM.

[0064] Furthermore, in step S5, the photovoltaic power is decomposed into low-frequency components and high-frequency detail components based on an adaptive local weighted regression smoothing algorithm combined with wavelet decomposition, and the decomposition process includes:

[0065] 1) Local weighted regression smoothing algorithm, used to extract low-frequency signals and high-frequency signals, including:

[0066] Linear Weighted Regression:

[0067] P low (t) = β0(t) + β1(t)(tt′)

[0068] Where P low (t) represents the smoothed low-frequency component, which represents the low-frequency power signal at time t; β0(t) and β1(t) are the parameters of local fitting; t is the current time point; t' is the time point in the local window Window;

[0069] Weighted least squares error:

[0070]

[0071] Where P(t') represents the original power signal at time point t'; w(t',t) is the weighting function;

[0072] The weighting function selects the Gaussian kernel, and the formula is:

[0073]

[0074] In the formula, τ controls the window size and determines the range of local weighting;

[0075] The smoothing formula after using the weighted function is:

[0076]

[0077] Low frequency signal:

[0078]

[0079] High frequency signal:

[0080] P high (t) = P(t) - P low (t)

[0081] Smoothness calculation:

[0082]

[0083] ΔP(t)=P low (t+1)-P low (t)

[0084] S max =max(|S(t)|)<S thresholds

[0085] Where S(t) represents smoothness; ΔP(t) is the power change per minute; P m Represents the total installed capacity of the system; S max is the maximum smoothness; S thresholds is the smoothness threshold set;

[0086] 2) Wavelet decomposition, used to decompose the signal into different frequency levels:

[0087]

[0088] Where 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.

[0089] Furthermore, in step S5, the coordinated control strategy of the water-photovoltaic-storage complementary system includes:

[0090] Coordination control condition formula:

[0091]

[0092] Where P g hydro (t) represents the output power of the g-th hydropower unit at time t; P g storage-discharge (t) represents the discharge power of the energy storage system for the g-th layer component at time t; P g storage-charge (t) represents the charging power of the energy storage system for the g-th layer component at time t; P g details(t) represents the power change of the g-th layer component; θ hydro is the output weight coefficient of the hydropower system; θ storage is the discharge power density coefficient of the energy storage system; P storage-max is the maximum charging and discharging power of the energy storage system:

[0093] The photovoltaic power formula after power smoothing is:

[0094]

[0095] Where P final (t) is the PV power after power smoothing.

[0096] In a second aspect, the present invention further provides a power smoothing control system for a water-photovoltaic-storage complementary system, which is applied to the above-mentioned power smoothing control method for a water-photovoltaic-storage complementary system to realize power smoothing control of the water-photovoltaic-storage complementary system, and the system comprises:

[0097] A data acquisition module is used to acquire the operating data of the hydropower unit and the photovoltaic power station, and to extract characteristic indicators representing the operating status of the hydropower unit;

[0098] An indicator weight calculation module is used to improve the CRITIC method by using the coefficient of variation method to calculate the objective weight of the characteristic indicator;

[0099] A status grade classification module is used to improve the TOPSIS method by introducing Manhattan distance and Z score standardization, and to classify the operating status of the hydropower unit using the improved TOPSIS method in combination with the objective weight;

[0100] The evaluation model building module is used to use the hydropower unit operation status classification results as the output of the CNN-GSVM model, establish an online evaluation model for the hydropower unit operation status based on CNN-GSVM, and realize real-time monitoring and evaluation of the hydropower unit operation status;

[0101] The power smoothing control module uses an adaptive local weighted regression smoothing algorithm combined with wavelet decomposition to decompose the photovoltaic power into a low-frequency component and a high-frequency detail component. The low-frequency component is used as the photovoltaic grid-connected power. The negative part of the high-frequency detail component is borne by hydropower units with different operating status levels, and the positive part of the high-frequency detail component and the negative part of the power not absorbed by the hydropower unit are borne by the energy storage. According to the power generation characteristics of the photovoltaic power station and the hydropower unit, a coordinated control strategy for the water-photovoltaic-storage complementary system is formulated to achieve smooth control of the photovoltaic power.

[0102] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising 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 the above-mentioned power smoothing control method of a water-photovoltaic-storage complementary system.

[0103] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0104] The present invention provides a power smoothing control method and system for a water-photovoltaic-storage complementary system. The improved CRITIC-TOPSIS method is used to preliminarily classify the operating status levels of hydropower units. The CNN-GSVM model is used to perform online evaluation of the operating status of the hydropower units, so that the operating status of the hydropower units can be grasped in real time, and the accuracy and robustness of the evaluation of the operating status of the hydropower units can be effectively improved. Furthermore, on the basis of the evaluation of the operating status of the hydropower units, relying on the rapid regulation characteristics of the hydro-generator units and the regulation capacity of the hydropower station reservoir, an adaptive local weighted regression smoothing algorithm is used in combination with wavelet decomposition to decompose the photovoltaic power into a low-frequency component and a high-frequency detail component. The low-frequency component is used as the grid-connected power, and the high-frequency detail component is borne by hydropower units with different performances respectively. The energy storage device bears the high-frequency detail component that is not absorbed, so as to achieve a smooth photovoltaic power generation curve, realize power smoothing control of the water-photovoltaic-storage complementary system, and improve the quality of photovoltaic power generation.

[0105] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0106] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0107] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0108] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0109] Figure 1 A schematic flow chart of a power smoothing control method for a water-photovoltaic-storage complementary system provided in an embodiment of the present invention.

[0110] Figure 2 A schematic diagram of a hydropower unit operating status assessment process based on CNN-GSVM provided in an embodiment of the present invention.

[0111] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0112] In order to make the purpose, 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 in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0113] In the description of the present invention, it should be noted that: in some processes described in the specification and drawings of this application, multiple operations appearing in a specific order are included, but it should be clearly understood that these operations may not be performed in the order in which they appear in this document or may be performed in parallel. In addition, various serial numbers are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0114] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0115] See also Figure 1 As shown, the embodiment of the present invention provides a power smoothing control method for a water-photovoltaic-storage complementary system, which mainly includes the following steps:

[0116] S1. Read the actual operation data of the photovoltaic power station and the hydropower unit from the Supervisory Control And Data Acquisition (SCADA) system, and extract the main characteristic indicators representing the operation status of the hydropower unit according to the annual operation data curve of the hydropower unit;

[0117] S2. Using the coefficient of variation method to improve the CRITIC method, calculate the objective weight of the characteristic index;

[0118] S3. The TOPSIS method is improved by introducing Manhattan distance and Z score standardization, and the improved TOPSIS method is used to classify the operating status of the hydropower units in combination with the objective weights;

[0119] S4. The result of the classification of the operating status of the hydropower unit is used as the output of the CNN-GSVM model, and an online evaluation model of the operating status of the hydropower unit based on CNN-GSVM is established to realize real-time monitoring and evaluation of the operating status of the hydropower unit;

[0120] S5. Based on the adaptive local weighted regression smoothing algorithm combined with wavelet decomposition, the photovoltaic power is decomposed into low-frequency components and high-frequency detail components. The low-frequency components are used as photovoltaic grid-connected power. The negative part of the high-frequency detail components is borne by hydropower units with different operating status levels, and the positive part of the high-frequency detail components and the negative part of the power not absorbed by the hydropower units are borne by energy storage. According to the power generation characteristics of photovoltaic power stations and hydropower units, a coordinated control strategy for the water-photovoltaic-storage complementary system is formulated to achieve smooth control of photovoltaic power.

[0121] The specific implementation manner and working principle of the method of the present invention are described in detail below:

[0122] In the above step S1, the main characteristic indicators of the operating status of the hydropower unit are as follows:

[0123] In the embodiment of the present invention, the comprehensive evaluation index of the operation status of the hydropower unit can be divided into four types of characteristics, namely swing (A1-A4), vibration (B1-B6), pressure pulsation (C1-C4) and displacement speed (D1-D2), with a total of 30 indicators. The details are as follows:

[0124] The swing (A1-A4) includes: A1-upper guide X / Y swing; A2-lower guide X / Y swing; A3-water guide X / Y swing; A4-fixed machine X / Y / Z swing;

[0125] Vibration (B1-B6) includes: B1-X / Y / Z vibration of upper frame; B2-X / Y / Z vibration of lower frame; B3-X / Y / Z vibration of top cover; B4-horizontal / vertical vibration of core 1; B5-horizontal / vertical vibration of core 2; B6-horizontal / vertical vibration of core 3;

[0126] Pressure pulsation (C1-C4) includes: C1-tail inlet pressure pulsation; C2-volute pressure pulsation; C3-guide vane pressure pulsation; C4-top cover pressure pulsation;

[0127] Displacement speed (D1-D2) includes: D1-driving bearing displacement; D2-speed.

[0128] In the embodiment of the present invention, the objective weight calculation process based on the improved CRITIC method in the above step S2 is as follows:

[0129] 1) Construct the original evaluation matrix:

[0130]

[0131] In the formula, x ij It represents the jth index value of the ith unit, m is the number of test units, and n is the number of indicators.

[0132] 2) Dimensionless processing of various indicators:

[0133] The processing formula for positive indicators (the larger the indicator value, the better) is as follows:

[0134]

[0135] The processing formula for the reverse indicator (the smaller the indicator value, the better) is as follows:

[0136]

[0137] 3) Calculate the standard deviation between various indicators:

[0138]

[0139] In the formula, represents the mean value of the jth index; δ j represents the standard deviation of the jth index, x′ ij Represents the data of the i-th row and j-th column after the original elementary matrix is ​​normalized (i.e. x ij Normalized value).

[0140] 4) Calculate the coefficient of variation of all indices:

[0141]

[0142] Where: v j is the coefficient of variation of the jth indicator.

[0143] 5) Calculate the conflict coefficient between indicators:

[0144]

[0145] In the formula, ρ pj represents the Pearson correlation coefficient between the pth indicator and the jth indicator; the indicator 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 of the pth indicator; represents the mean of the j-th indicator.

[0146] 6) Calculate the amount of information contained in each indicator:

[0147] G j =v j ·f j, j=1,2,…,n

[0148] In the formula, G j represents the amount of information contained in the jth indicator, f j Represents the conflict coefficient of the jth indicator.

[0149] 7) Calculate the objective weight of the indicator based on the amount of information contained in the indicator:

[0150]

[0151] Where W j Represents the objective weight of the j-th indicator.

[0152] In the embodiment of the present invention, the operation status evaluation process of the hydropower unit based on the improved TOPSIS method in the above step S3 is as follows:

[0153] 1) Calculate the standardized evaluation matrix:

[0154] Z ij =W j ×x' ij

[0155] In the formula, Z ij represents the weighted normalized matrix;

[0156] 2) Determine the positive ideal solution and negative ideal solution

[0157] Among them, the positive ideal solution For all evaluation indicators to achieve the optimal solution, negative ideal solution The solution that achieves the worst result for each evaluation index is:

[0158]

[0159] 3) Calculate the Manhattan distance between each index value and the positive ideal solution and the negative ideal solution:

[0160]

[0161] In the formula, represents the Manhattan distance between the ith unit and the positive ideal solution, represents the Manhattan distance between the ith unit and the negative ideal solution;

[0162] 4) The comprehensive score and the standardized interval divided by Z score are used to obtain the final evaluation result as follows:

[0163]

[0164] In the formula, C iIt represents the comprehensive score of each unit data, Z i represents the Z score of the ith unit evaluation, μ represents the average of the comprehensive scores of each unit, and σ represents the standard deviation of the comprehensive scores of each unit.

[0165] In this embodiment, the Z-score normalization interval is as follows:

[0166] The performance of hydropower units is divided into four levels: 1-excellent, 2-good, 3-medium, and 4-poor;

[0167] The corresponding Z score interval is: Z ≥ 1, 0 <Z<1,-1<Z<0,Z≤-1。

[0168] Further, combined with Figure 2 As shown, in the above step S4, the modeling process of the online evaluation model of the hydropower unit operation status based on CNN-GSVM fusion is as follows:

[0169] 1) Normalize the original index data of the hydropower units and the preliminary operating status assessment results (the operating status levels of the hydropower units divided);

[0170] 2) Randomly divide the normalized data samples into training sets and test sets to ensure data diversity and universality during model training;

[0171] 3) Input the training set into the CNN model and continuously adjust the model parameters to make the loss function converge to obtain a stable and accurate feature representation;

[0172] 4) Use the fully connected layer of the trained CNN model to extract feature vectors and use them as input for the SVM classifier to further train the SVM;

[0173] 5) The test set is input into the CNN-GSVM model to identify the operating status of the hydropower unit and evaluate its classification accuracy and robustness.

[0174] In the embodiment of the present invention, the structure of the hydropower unit performance evaluation model based on CNN-GSVM is as follows:

[0175] The CNN-GSVM model consists of a CNN model and a SVM model (GSVM) optimized based on the Grey Wolf Algorithm (GWO). CNN is used to extract image features and these features are used as the input of GSVM. The CNN structure is input layer-convolution layer-pooling layer-convolution layer-pooling layer-fully connected layer, and the activation function is RELU function. The evaluation results obtained by the improved CRITIC-TOPSIS algorithm are used as the output of the CNN-GSVM input layer. 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 GSVM, and the GSVM network is trained to realize the recognition of the operating status of the hydropower unit.

[0176] In a specific implementation, the gray wolf algorithm optimizes the SVM process as follows:

[0177] The Gray Wolf Optimization Algorithm (GWO) is a swarm intelligence optimization algorithm developed by simulating the behavior of gray wolves hunting prey in nature. The specific form of the Gray Wolf Optimization Algorithm is as follows:

[0178] 1) Assuming that the total number of wolves in the initial group is Q, set the Q×2-dimensional wolf group position vector X(0) = (c K , g K ), K=1,2,…,Q,c K and g K represents the parameters of the Kth wolf SVM model. The fitness of each gray wolf is evaluated and sorted according to the SVM recognition accuracy, and the initial optimal wolf position X is selected. α,β,δ ;

[0179] 2) The wolf pack follows the optimal wolf position for update. During the tth gray wolf search, the optimal wolf position X obtained last time is updated. α,β,δ Update the wolf pack position, such as X(t0)→X(t+1), as follows:

[0180] eX α,β,δ =e(c α,β,δ ,g α,β,δ )

[0181] a=2-2t / t max

[0182]

[0183] Where X(t+1) is the updated wolf pack position after the t-th search; X α，β，δ is the 1×2 dimensional position vector of α, β, δ wolves in the t-th search; t max is the maximum number of iterations; a is the convergence factor, which decreases linearly in [0,2] during the iteration process; e is the K-dimensional unit column vector; A and C are coefficient matrices; r1 and r2 are random number column vectors; I is the K-order unit matrix; A1,2,3 , C 1,2,3 is the coefficient matrix corresponding to α, β, δ wolves; X 1,2,3 Update the intermediate vectors for the positions of α, β, and δ wolves;

[0184] 3) Evaluate and sort the updated wolf positions in X(t+1) and finally output X α.best =(c best ,g best ), that is, the optimal values ​​of the penalty factor c and the kernel function parameter g are obtained and used to optimize the SVM.

[0185] In a specific embodiment, the CNN feature extraction operation process is as follows:

[0186] The calculation formula of CNN convolution process is:

[0187]

[0188] In the formula, Y is the feature subgraph; f(*) is a nonlinear activation function; w is the weight; B is the bias; l is the lth layer of the network; k is the kth convolution; * is the convolution operation.

[0189] The pooling layer expression is:

[0190]

[0191] Where β is the weight; down is the subsampling function; w is the weight vector.

[0192] The output layer is the Softmax layer, and its expression is:

[0193]

[0194] In the formula, S h is the value input to each unit of Softmax.

[0195] The SVM classification process is as follows:

[0196] The mathematical model of a nonlinear classification hyperplane optimization problem is:

[0197]

[0198] The constraints are:

[0199]

[0200] In the formula, x e is the input feature of SVM; eis the label corresponding to the input feature; M is the number of input features; ω′ is the normal vector of the hyperplane; ξ is the slack variable; b is the offset of the hyperplane; C * It is a penalty factor, which is generally greater than 0. The larger the value, the greater the penalty for misclassified samples, the higher the accuracy of the training samples, but the lower the generalization ability of the model.

[0201] The optimal classification decision function is obtained by using the Lagrange multiplier method:

[0202]

[0203] In the formula, and b * is the Lagrange multiplier; K(x e ,y e ) is the kernel function.

[0204] Furthermore, the evaluation indicators of the model test results are as follows:

[0205] In the embodiment of the present invention, the accuracy (Acc), recall (Rec) and macro average (F1) are used to evaluate the model test results, and the calculation formula is as follows:

[0206]

[0207] In the formula, TP is the number of positive samples whose model evaluation results are consistent with the actual situation; TN is the number of negative samples whose model evaluation results are consistent with the actual situation; FP is the number of positive samples whose model evaluation results are positive and the actual situation is negative; FN is the number of negative samples whose model evaluation results are negative and the actual situation is positive.

[0208] In this embodiment, under the unimproved and improved CRITIC-TOPSIS algorithms, the operating status identification results of the hydropower units obtained by the following four models are compared, as shown in the following Table 1.

[0209] Table 1 Comparison of operating status identification results of hydropower units

[0210]

[0211] It can be seen from the comparison results that the improved CRITIC-TOPSIS algorithm (the method in the present invention) can significantly improve the accuracy and robustness of the identification of the operating status of the hydropower unit.

[0212] Further, in the embodiment of the present invention, in the above step S5, the photovoltaic power decomposition process using adaptive local weighted regression smoothing and wavelet transform is as follows:

[0213] 1) Local weighted regression smoothing is a low-pass filter, which is mainly used to extract low-frequency and high-frequency signals. The smoothed signal P is obtained by weighted fitting the original signal P(t) in the local time window Window. low (t). The window size can be adaptively adjusted according to the smoothness threshold to ensure that the smoothed low-frequency signal does not exceed the preset smoothness. The specific formula is as follows:

[0214] Linear Weighted Regression:

[0215] P low (t) = β0(t) + β1(t)(tt′)

[0216] Where P low (t) represents the smoothed low-frequency component, which represents the low-frequency power signal at time t; β0(t) and β1(t) are the parameters of local fitting; t is the current time point; t' is the time point in the local window Window;

[0217] Weighted least squares error:

[0218]

[0219] Where P(t') represents the original power signal at time point t'; w(t', t) is a weighting function, which represents the contribution weight at time t' to time t; Window is the time window of local weighted regression, which represents a set of time points t' selected near time point t;

[0220] The weighting function usually chooses the Gaussian kernel, and the formula is:

[0221]

[0222] In the formula, τ controls the window size and determines the range of local weighting.

[0223] Smoothing formula after using weighting function:

[0224]

[0225] Low frequency signal:

[0226]

[0227] High frequency signal:

[0228] P high (t) = P(t) - P low (t)

[0229] Smoothness calculation: Volatility is used to measure the degree of change of a signal at adjacent time points. It is an important criterion for judging the smoothing effect of a signal to ensure that the power change is not too drastic. The smoothness calculation formula is as follows:

[0230]

[0231] ΔP(t)=P low (t+1)-P low (t)

[0232] S max =max(|S(t)|)<S thresholds

[0233] Where ΔP(t) is the power change per minute; P m Indicates the total installed capacity of the system. max is the maximum smoothness; S thresholds is the smoothness threshold set;

[0234] 2) Wavelet decomposition: Wavelet decomposition is a signal processing technique that can decompose signals into different frequency levels. In this step, we perform multi-level wavelet decomposition on high-frequency signals to capture the detailed changes of signals at different frequencies, which is helpful for signal reconstruction and subsequent power scheduling.

[0235]

[0236] Where 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 layers;

[0237] In an embodiment of the present invention, the coordinated control strategy of the water-photovoltaic-storage complementary system in the above step S5 is as follows:

[0238] Coordination control condition formula:

[0239]

[0240] Where P g hydro (t) represents the output power of the g-th hydropower unit at time t; P g storage-discharge (t) represents the discharge power of the energy storage system for the g-th layer component at time t; P g storage-charge (t) represents the charging power of the energy storage system for the g-th layer component at time t; P g details (t) represents the power change of the g-th layer component; θhydro is the output weight coefficient of the hydropower system; θ storage is the discharge power density coefficient of the energy storage system; P storage-max is the maximum charging and discharging power of the energy storage system.

[0241] Energy storage overall charging and discharging:

[0242]

[0243] SOC battery state formula:

[0244]

[0245]

[0246] Where SOC(t) is the battery state of charge at time t; E s (t) is the actual remaining capacity of the battery at time t; E max is the maximum capacity of the battery.

[0247] The photovoltaic power formula after power smoothing is:

[0248]

[0249] 1 minute and 10 minute volatility calculation formula:

[0250]

[0251] Where P final (t) is the PV power after power smoothing.

[0252] From the description of the above embodiments, those skilled in the art can know that: the present invention provides a power smoothing control method for a water-photovoltaic-storage complementary system, which reads historical operation data and actual operation data of photovoltaic power stations and hydropower units from a data acquisition and monitoring control system, and extracts main characteristic indicators characterizing the operation status of the hydropower units according to the annual operation data curve of the hydropower units; improves the CRITIC method by using the coefficient of variation method to calculate the objective weight of each characteristic indicator; improves the TOPSIS method by introducing Manhattan distance and Z score standardization to obtain the initial division of the operation status level of the hydropower units; uses the obtained initial division result of the operation status level of the hydropower units as the convolutional neural network (Convolutional Neural Network-Grey Wolf Optimization Support Vector Machine) of the Grey Wolf Optimization Support Vector Machine. The output of the CNN-GSVM model is used to establish an online evaluation model for the operating status of hydropower units based on the fusion of CNN-GSVM; the photovoltaic power is decomposed into low-frequency components and high-frequency detail components using an adaptive local weighted regression smoothing algorithm combined with wavelet decomposition. The low-frequency component is used as the photovoltaic grid-connected power, and the negative part of the high-frequency detail component is borne by hydropower units with different performance levels. The positive part of the high-frequency detail component and the negative part of the power not absorbed by the hydropower unit are borne by the energy storage device (lithium iron phosphate battery). According to the power generation characteristics of photovoltaic power stations and hydropower units, a coordinated control strategy for the water-photovoltaic-storage complementary system is formulated to achieve smooth control of photovoltaic power.

[0253] The present invention takes into account the real-time operating status of the hydropower unit and, based on the complementary characteristics of water, photovoltaic and storage, further improves the power quality of photovoltaic power generation while smoothing out photovoltaic power fluctuations.

[0254] Furthermore, the present invention also provides a power smoothing control system for a water-photovoltaic-storage complementary system, which is applied to a power smoothing control method for a water-photovoltaic-storage complementary system described in the above embodiment to perform power smoothing control for the water-photovoltaic-storage complementary system, and the system comprises:

[0255] A data acquisition module is used to acquire the operating data of the hydropower unit and the photovoltaic power station, and to extract characteristic indicators representing the operating status of the hydropower unit;

[0256] An indicator weight calculation module is used to improve the CRITIC method by using the coefficient of variation method to calculate the objective weight of the characteristic indicator;

[0257] A status grade classification module is used to improve the TOPSIS method by introducing Manhattan distance and Z score standardization, and to classify the operating status of the hydropower unit using the improved TOPSIS method in combination with the objective weight;

[0258] The evaluation model building module is used to use the hydropower unit operation status classification results as the output of the CNN-GSVM model, establish an online evaluation model for the hydropower unit operation status based on CNN-GSVM, and realize real-time monitoring and evaluation of the hydropower unit operation status;

[0259] The power smoothing control module uses an adaptive local weighted regression smoothing algorithm combined with wavelet decomposition to decompose the photovoltaic power into a low-frequency component and a high-frequency detail component. The low-frequency component is used as the photovoltaic grid-connected power. The negative part of the high-frequency detail component is borne by hydropower units with different operating status levels, and the positive part of the high-frequency detail component and the negative part of the power not absorbed by the hydropower unit are borne by the energy storage. According to the power generation characteristics of the photovoltaic power station and the hydropower unit, a coordinated control strategy for the water-photovoltaic-storage complementary system is formulated to achieve smooth control of the photovoltaic power.

[0260] An embodiment of the present invention provides a power smoothing control system for a water-photovoltaic-storage complementary system, and its implementation principle and technical effects are the same as those of the aforementioned method embodiment. For the sake of brief description, for parts not mentioned in this embodiment, reference can be made to the corresponding contents in the aforementioned method embodiment, which will not be repeated here.

[0261] In addition, refer to Figure 3 As shown, an embodiment of the present invention further provides an electronic device, which may include a processor 10, a memory 11, a communication bus 12 and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, and the processor executes the computer program to implement a power smoothing control method for a water-photovoltaic-storage complementary system in the above method embodiment.

[0262] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips. The processor 10 is the control core of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, 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.

[0263] It should be understood by those skilled in the art that the embodiments of the present invention may be provided as methods, systems or computer program products, etc. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention may take the form of a computer program product implemented on one or more storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0264] It should be noted that the word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several distinct components, and by means of a suitably programmed computer.

[0265] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0266] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may 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 the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A power smoothing control method for a water-photovoltaic-storage complementary system, characterized in that: The method comprises the following steps: S1. Obtain the operating data of the hydropower unit and the photovoltaic power station, and extract characteristic indicators representing the operating status of the hydropower unit; S2. Using the coefficient of variation method to improve the CRITIC method, calculate the objective weight of the characteristic index; S3. The TOPSIS method is improved by introducing Manhattan distance and Z score standardization, and the improved TOPSIS method is used to classify the operating status of the hydropower units in combination with the objective weights; S4. The result of the classification of the operating status of the hydropower unit is used as the output of the CNN-GSVM model, and an online evaluation model of the operating status of the hydropower unit based on CNN-GSVM is established to realize real-time monitoring and evaluation of the operating status of the hydropower unit; S5. Based on the adaptive local weighted regression smoothing algorithm combined with wavelet decomposition, the photovoltaic power is decomposed into low-frequency components and high-frequency detail components. The low-frequency components are used as photovoltaic grid-connected power. The negative part of the high-frequency detail components is borne by hydropower units with different operating status levels, and the positive part of the high-frequency detail components and the negative part of the power not absorbed by the hydropower units are borne by energy storage. According to the power generation characteristics of photovoltaic power stations and hydropower units, a coordinated control strategy for the water-photovoltaic-storage complementary system is formulated to achieve smooth control of photovoltaic power.

2. The power smoothing control method of a water-photovoltaic-storage complementary system according to claim 1 is characterized in that: In the step S1, the indicator categories of the characteristic indicators characterizing the operating status of the hydropower unit include: swing, vibration, pressure pulsation and displacement speed.

3. The power smoothing control method of a water-photovoltaic-storage complementary system according to claim 1 is characterized in that: In step S2, the process of calculating the objective weight of the characteristic index based on the improved CRITIC method includes: 1) Construct the original evaluation matrix: In the formula, x ij represents the jth index value of the i-th unit, m is the number of test units, and n is the number of indicators; 2) Dimensionless processing of various indicators: The positive indicator processing formula is: The reverse indicator processing formula is: 3) Calculate the mean and standard deviation of each indicator: In the formula, x j represents the jth index value; represents the mean value of the jth index; δ j represents the standard deviation of the jth index, x′ ij Represents x ij Normalized value; 4) Calculate the coefficient of variation of the indicator: Where: v j is the coefficient of variation of the jth indicator; 5) Calculate the conflict coefficient between indicators: In the formula, ρ pj represents the Pearson correlation coefficient between the pth indicator and the jth indicator; the indicator 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 of the pth indicator; represents the mean value of the jth indicator; 6) Calculate the amount of information contained in each indicator: G j =v j ·f j In the formula, G j represents the amount of information contained in the jth indicator; f j represents the conflict coefficient of the jth indicator; 7) Calculate the objective weight of the indicator based on the amount of information contained in the indicator: Where W j Represents the objective weight of the jth indicator.

4. The power smoothing control method of a water-photovoltaic-storage complementary system according to claim 3 is characterized in that: In step S3, the improved TOPSIS method is used to classify the operating status of the hydropower unit in combination with the objective weight, and the process 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 Manhattan distance between each unit and the positive ideal solution and the negative ideal solution: In the formula, represents the Manhattan distance between the ith unit and the positive ideal solution, represents the Manhattan distance between the ith unit and the negative ideal solution; 4) The final evaluation result is obtained by comprehensive scoring and Z-score standardization and division into standardized intervals. The calculation formula is: In the formula, C i represents the comprehensive score of the i-th unit, Z i represents the Z score of the ith unit evaluation, μ represents the average of the comprehensive scores of each unit, and σ represents the standard deviation of the comprehensive scores of each unit.

5. The power smoothing control method of a water-photovoltaic-storage complementary system according to claim 1 is characterized in that: In the step S4, the online evaluation model of the operating status of the hydropower unit based on CNN-GSVM is composed of a CNN model and a GSVM model. CNN is used to extract features, and these features are used as the input of the GSVM; wherein 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; the result of the classification of the operating status of the hydropower unit is used as the output of the CNN-GSVM model, each convolution kernel of the CNN extracts features from the operating data of the hydropower unit, which are integrated into global features through the fully connected layer, and the output data of the fully connected layer is used as the input of the GSVM, and the GSVM network is trained to realize the identification of the operating status of the hydropower unit.

6. A power smoothing control method for a water-photovoltaic-storage complementary system according to claim 5, characterized in that: The GSVM model is an SVM model optimized based on the Grey Wolf Algorithm GWO. The optimization process includes: 1) Assuming that the total number of wolves in the initial group is Q, set the Q×2-dimensional wolf group position vector X(0) = (c K , g K )K=1,2,...,Q,c K and g K represents the parameters of the Kth wolf SVM model. The fitness of each gray wolf is evaluated and sorted according to the SVM recognition accuracy, and the initial optimal wolf position X is selected. α，β，δ ; 2) The wolf pack follows the optimal wolf position for update. During the tth gray wolf search, the optimal wolf position X obtained last time is updated. α，β，δ Updated wolf pack locations as follows: eX α，β，δ =e(c α，β，δ ,g α，β，δ ) α=2-2t / t max Where X(t+1) is the updated wolf pack position after the t-th search; X α，β，δ is the 1×2 dimensional position vector of α, β, δ wolves in the t-th search; t max is the maximum number of iterations; a is the convergence factor, which decreases linearly in [0, 2] during the iteration process; e is the unit column vector; A and C are coefficient matrices; r1 and r2 are random number column vectors; I is the unit matrix; A 1,2,3 , C 1，2，3 is the coefficient matrix corresponding to α, β, δ wolves; X 1,2，3 Update the intermediate vectors for the positions of α, β, and δ wolves; 3) Perform fitness evaluation and sorting on the updated wolf positions in X(t+1) to obtain the optimal values ​​of the penalty factor c and kernel function parameter g, and use them to optimize the SVM.

7. The power smoothing control method of a water-photovoltaic-storage complementary system according to claim 1 is characterized in that: In step S5, the photovoltaic power is decomposed into low-frequency components and high-frequency detail components based on an adaptive local weighted regression smoothing algorithm combined with wavelet decomposition. The decomposition process includes: 1) Local weighted regression smoothing algorithm, used to extract low-frequency signals and high-frequency signals, including: Linear Weighted Regression: P low (t)=β0(t)+β1(t)(t-t′) Where P low (t) represents the smoothed low-frequency component, which represents the low-frequency power signal at time t; β0(t) and β1(t) are the parameters of local fitting; t is the current time point; t' is the time point in the local window Window; Weighted least squares error: Where P(t') represents the original power signal at time point t'; w(t',t) is the weighting function; The weighting function selects the Gaussian kernel, and the formula is: In the formula, τ controls the window size and determines the range of local weighting; The smoothing formula after using the weighted function is: The low frequency signal is: The high frequency signal is: P high (t)=P(t)-P low (t) Smoothness calculation: ΔP(t)=P low (t+1)-P low (t) S max =max(|S(t)|)<S thresholds Where S(t) represents smoothness; P(t) is the original signal; ΔP(t) is the power change per minute; P m Represents the total installed capacity of the system; S max is the maximum smoothness; S thresholds is the smoothness threshold set; 2) Wavelet decomposition, used 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.

8. The power smoothing control method of a water-photovoltaic-storage complementary system according to claim 7 is characterized in that: In step S5, the coordinated control strategy of the water-solar-storage complementary system includes: Coordination control conditions: Where P g hydro (t) represents the output power of the g-th hydropower unit at time t; P g storage-discharge (t) represents the discharge power of the energy storage system for the g-th layer component at time t; P g storage-charge (t) represents the charging power of the energy storage system for the g-th layer component at time t; P g details (t) represents the power change of the g-th layer component; θ hydro is the output weight coefficient of the hydropower system; θ storage is the discharge power density coefficient of the energy storage system; P storage-max is the maximum charging and discharging power of the energy storage system: The photovoltaic power after power smoothing is: Where P final (t) is the PV power after power smoothing.

9. A power smoothing control system for a water-photovoltaic-storage complementary system, characterized in that: When applied, a power smoothing control method for a water-photovoltaic-storage complementary system according to any one of claims 1 to 8 is performed to realize power smoothing control of a water-photovoltaic-storage complementary system, the system comprising: A data acquisition module is used to acquire the operating data of the hydropower unit and the photovoltaic power station, and to extract characteristic indicators representing the operating status of the hydropower unit; An indicator weight calculation module is used to improve the CRITIC method by using the coefficient of variation method to calculate the objective weight of the characteristic indicator; A status grade classification module is used to improve the TOPSIS method by introducing Manhattan distance and Z score standardization, and to classify the operating status of the hydropower unit using the improved TOPSIS method in combination with the objective weight; The evaluation model building module is used to use the hydropower unit operation status classification results as the output of the CNN-GSVM model, establish an online evaluation model for the hydropower unit operation status based on CNN-GSVM, and realize real-time monitoring and evaluation of the hydropower unit operation status; The power smoothing control module uses an adaptive local weighted regression smoothing algorithm combined with wavelet decomposition to decompose the photovoltaic power into a low-frequency component and a high-frequency detail component. The low-frequency component is used as the photovoltaic grid-connected power. The negative part of the high-frequency detail component is borne by hydropower units with different operating status levels, and the positive part of the high-frequency detail component and the negative part of the power not absorbed by the hydropower unit are borne by the energy storage. According to the power generation characteristics of the photovoltaic power station and the hydropower unit, a coordinated control strategy for the water-photovoltaic-storage complementary system is formulated to achieve smooth control of the photovoltaic power.

10. An electronic device, characterized in that: It includes 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 power smoothing control method for a water-photovoltaic-storage complementary system as described in any one of claims 1-8.

Images