Wind, solar, and hydropower storage active power coordinated real-time control method and system

Abstract

The present invention discloses a method and system for real-time coordinated control of wind, solar, and hydropower storage active power, which solves the problem that existing methods cannot consider the wind and solar power output correlation output curve, and the distribution network operation status model only relies on the wind power and photovoltaic output curve and the load curve and the status of the energy storage equipment, resulting in the distribution network operation status model output volatility and frequency regulation success rate cannot be guaranteed. The method includes: combining the wind and solar power probability distribution function and the improved dynamic time bending algorithm to quantitatively describe the wind and solar power output correlation; constructing a coordinated scheduling model based on the wind and solar power output correlation and cascading failure simulation; the present invention defines the wind and solar power probability distribution function of the active output of wind power generation and photovoltaic power generation through the probability distribution method, and quantitatively describes the wind and solar power output correlation based on the improved dynamic time bending algorithm, so as to ensure that the coordinated scheduling model effectively combines and considers the quantitatively expressed wind and solar power output correlation output curve when adaptively searching for the optimal solution parameters, and can more accurately predict and simulate the changes in wind and solar power output.

Description

Technical Field

[0001] The present invention belongs to the field of power grid control technology, and specifically relates to a method and system for real-time coordinated control of wind, solar, and hydropower storage active power. Background Art

[0002] As my country's energy demand continues to grow, shifting the energy mix and promoting the development of clean energy in the power industry have become increasingly important. Combined wind, solar, hydro, and energy storage systems typically include multiple renewable energy sources, such as wind power, photovoltaic power, and hydropower, along with supporting energy storage systems. These systems work together to provide stable power output. However, the randomness and volatility of wind, photovoltaic, and hydropower can impact the safe and stable operation of the power grid.

[0003] Chinese patent CN116054152A discloses a method for the coordinated optimization control of wind, solar and storage participating in the distribution network source, grid, load and storage in consideration of economic benefits, which predicts the output of distributed wind and photovoltaic power generation; predicts the power load of the distribution network; obtains the status of distributed wind and solar power generation equipment, energy storage equipment and adjustable load in the distribution network; establishes a distribution network operation status model of source, grid, load and storage based on the wind power and photovoltaic output curves and load curves obtained from the predicted wind and photovoltaic power generation outputs and the status of the energy storage equipment; and calculates the power purchase cost of the upper power grid, the power generation cost of the distributed power source, the energy storage cost and the power sales cost. Price, under the premise of considering the stability of the power grid, the economic objective function of the distribution network is constructed; according to the distribution network operation status model and the distribution network economic objective function, the dispatching plan of the distribution network with the maximum economic benefit is obtained. However, when the existing method establishes the distribution network operation status model of the source, grid, load and storage, it is unable to effectively consider the output curve of the wind and solar power correlation. The distribution network operation status model only relies on the wind power and photovoltaic output curves and the load curves and the status of the energy storage equipment, resulting in the output volatility of the distribution network operation status model and the frequency regulation success rate cannot be guaranteed. To address the above problems, we proposed a real-time control method and system for the coordinated active power of wind, solar, hydro and storage. Summary of the Invention

[0004] The purpose of the present invention is to address the shortcomings of the existing technology and provide a real-time control method and system for the coordinated active power of wind, solar, hydro and storage. It solves the problem that when the existing method establishes the distribution network operation status model of source, grid, load and storage, it cannot effectively consider the wind and solar output correlation output curve. The distribution network operation status model only relies on the wind power and photovoltaic output curve and the load curve and the status of the energy storage equipment, resulting in the output volatility of the distribution network operation status model and the problem that the frequency regulation success rate cannot be guaranteed.

[0005] The present invention is implemented as follows: a method for real-time coordinated control of wind, solar, hydro and storage active power, comprising:

[0006] Obtain historical operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems. These parameters include output parameters of distributed wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems, environmental data, active power demand data, and grid fluctuation frequency.

[0007] Load historical operating parameters and use the probability distribution method to fit wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical operating parameters to obtain the wind and photovoltaic power generation active output probability distribution function;

[0008] Combined with the wind-solar probability distribution function and the improved dynamic time bending algorithm, the correlation between wind-solar output and power generation is quantitatively described.

[0009] A collaborative scheduling model is constructed based on the correlation of wind and solar power output and cascading failure simulation. With output volatility and frequency regulation success rate as constraints, the collaborative scheduling model is iteratively trained based on historical working parameters to output a converged collaborative scheduling model.

[0010] The real-time operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems within the collection period are obtained. The collaborative scheduling model is executed with the real-time operating parameters as input. The collaborative scheduling model adaptively searches for the optimal solution parameters for the real-time operating parameters and outputs the output parameters of wind power generation, photovoltaic power generation, and hydropower generation.

[0011] Preferably, the method of fitting the historical operating parameters of wind power generation, photovoltaic power generation, environmental data, and power grid fluctuation frequency in combination with the probability distribution method specifically includes:

[0012] Traverse the historical working parameters, extract wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency from the historical working parameters, and integrate wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency into a wind-solar signal set;

[0013] The wind and solar signal set is digitally low-pass filtered based on the Butterworth digital filter to obtain the interference signal filter set that eliminates noise interference and frequency deviation;

[0014] Among them, the interference signal filtering set is expressed as:

[0015]

[0016] Among them, x(n) is the output representation of the interference signal filtering set, A h represents the hth harmonic amplitude, H represents the harmonic order of the wind and solar signal set, L represents the digital low-pass filtering order of the digital filter, S(n) represents the input wind and solar signal set, B(ω h-Pos ), B(ω h-Neg ) represent the positive and negative frequency sampling angle differences of the wind and solar signal sets, Δf is the sampling frequency difference, f ris the rated frequency of wind and solar signal centralized parameters, f h is the frequency of the harmonic, f s is the signal sampling frequency, B eq (n) represents the Butterworth digital filter, arg(·) is the signal amplitude angle, and B(enbw) represents the filter bandwidth of the Butterworth digital filter;

[0017] The interference signal filtering set is fitted using discrete Fourier transform method to obtain a discrete fitting set;

[0018] The discrete fit set is expressed as:

[0019]

[0020] in, represents the discrete fitting set, x(n) represents the input value of the interference signal filtering set, F(n) is the Fourier transform function, α is the complex exponential coefficient, represents the variance of the interference signal filtered set;

[0021] Obtain a discrete fitting set, define a wind-solar probability distribution function based on the discrete fitting set, and obtain the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation;

[0022] The wind and solar probability distribution function expression is as follows:

[0023]

[0024] Where f(x) represents the probability distribution function of wind and solar power, n represents the discrete fitting set parameter, is the probability distribution function, k(x) is the probability kernel function, h represents the bandwidth, and β is the kernel parameter.

[0025] Preferably, the method for quantitatively describing the correlation between wind and solar power outputs by combining the wind and solar power probability distribution function and the improved dynamic time warping algorithm specifically includes:

[0026] Taking the wind-solar probability distribution function as a constraint, a wind-solar output joint constraint model is established. The wind-solar output joint constraint model is defined by the Copula connection function.

[0027] The wind and solar power combined constraint model is defined as:

[0028]

[0029] Where C(W,V) represents the wind-solar output joint constraint function, f(Wx) and f(Vx) are the wind output distribution function and photovoltaic output distribution function respectively, and λ is the joint constraint coefficient of the wind-solar output joint constraint model;

[0030] Load the discrete fitting set, initialize the adjacency matrix of the discrete fitting set, and find the original distances of wind power generation and photovoltaic power generation parameters in the wind power output distribution function and photovoltaic output distribution function in the discrete fitting set based on the improved dynamic time warping algorithm;

[0031] The original distances of wind power generation and photovoltaic power generation parameters are filled into the adjacency matrix of the discrete fitting set, and the adjacency matrix is ​​updated based on the improved dynamic time warping algorithm;

[0032] Analyze the wind-solar output correlation interaction behavior based on the wind-solar output joint constraint model and generate a wind-solar output correlation relationship map;

[0033] Load the wind and solar power output correlation relationship map, and identify at least one set of wind and solar power output interaction cluster centers based on the Dijkstra path optimization algorithm;

[0034] Based on the fuzzy clustering algorithm, the Euclidean distance between the wind-solar output related scenes and the wind-solar output interaction cluster center is calculated, and the wind-solar output correlation relationship is mapped and assigned to the wind-solar output interaction cluster center with the nearest distance;

[0035] The clustering criterion function is used to determine the clustering criterion value of the wind-solar output interaction cluster center, the minimum clustering criterion value is selected as the wind-solar output interaction cluster center, and the wind-solar output correlation output curve is generated based on the wind-solar output interaction cluster center.

[0036] Preferably, the method for iteratively training the collaborative scheduling model based on historical working parameters specifically includes:

[0037] Obtain historical operating parameters, combine them with the wind-solar probability distribution function and the improved dynamic time warping algorithm to quantitatively describe the correlation between wind-solar output and the wind-solar output. Then divide the historical operating parameters into training and test sets in a 4:1 ratio.

[0038] Based on historical working parameters, a wind-solar-water-storage grid topology is constructed, and the wind-solar-water-storage grid topology curve is introduced into the wind-solar-water-storage grid topology, so that the wind-solar-water-storage grid topology can achieve adaptive dynamic changes;

[0039] Generate at least one set of decision trees based on random forests, use the cascading failure simulation model as the initial model, and learn the local features of the wind, solar, water and storage grid topology map based on the random forest decision trees;

[0040] Combined with global linear regression training, at least one set of collaborative scheduling models integrating wind, solar, water and storage grid topology maps and cascading failure simulation models is obtained;

[0041] Obtaining a training set, iteratively training the collaborative scheduling model using the training set, and determining whether the collaborative scheduling model has converged;

[0042] If the collaborative scheduling model converges, obtain the training set and generate the output volatility and frequency regulation success rate constraints of the collaborative scheduling model based on multiple conditional constraints;

[0043] Determine the final converged collaborative scheduling model based on output volatility, frequency regulation success rate combined constraint coefficient, and collaborative scheduling model objective function, and output the collaborative scheduling model;

[0044] Among them, the objective function of the collaborative scheduling model is:

[0045]

[0046] P z,t =S W P W,t +S V P V,t +S H P H,t +S Q P Q,t (11)

[0047] S=S W +S V +S H +S Q (12)

[0048] Among them, T is the number of cooperative scheduling cycles, V P , S are the combined constraint coefficients of output fluctuation coefficient and frequency regulation success rate, S W , S V , S H , S Q They represent wind constraint coefficient, photovoltaic constraint coefficient, hydraulic constraint coefficient, and energy storage constraint coefficient respectively, and P z,t represents the total output of the collaborative scheduling model within the collaborative scheduling period, P avg,t represents the average output of the collaborative scheduling model during the collaborative scheduling period, P W,t , P V,t , P H,t , P Q,t They are wind power output power, photovoltaic power output power, hydropower output power, and energy storage output power.

[0049] Preferably, the method for the collaborative scheduling model to adaptively search for optimal solution parameters for real-time working parameters specifically includes:

[0050] Obtain real-time working parameters, iteratively construct the convex hull of the real-time working parameters based on the augmented Lagrangian algorithm, and generate a parameter convex hull set;

[0051] Combined with the wind-solar probability distribution function and the improved dynamic time bending algorithm, the wind-solar output correlation is quantitatively described to generate a wind-solar output correlation output curve;

[0052] Based on the dual solution algorithm combined with cascading failure simulation, the parameter convex hull set is described as the output volatility of the main problem and the frequency regulation success rate of the sub-problem. Based on the objective function of the coordinated scheduling model, the optimal solutions for wind power output power, photovoltaic power output power, hydropower output power, and energy storage output power are calculated.

[0053] Obtain the optimal solutions for wind output power, photovoltaic output power, hydropower output power, and energy storage output power, and adaptively search for optimal solution parameters based on the optimal solutions for wind output power, photovoltaic output power, hydropower output power, and energy storage output power.

[0054] On the other hand, the present invention also provides a wind, solar, hydropower and storage active power coordinated real-time control system, which specifically includes:

[0055] The parameter acquisition module is used to obtain historical operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems. The historical operating parameters include output parameters of distributed wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems, environmental data, active power demand data, and grid fluctuation frequency.

[0056] The output correlation analysis module loads historical operating parameters and uses the probability distribution method to fit the wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical operating parameters to obtain the wind and photovoltaic power generation active output probability distribution function. The wind and photovoltaic power generation probability distribution function is combined with the improved dynamic time warping algorithm to quantitatively describe the wind and photovoltaic output correlation;

[0057] The model building module builds a collaborative scheduling model based on wind and solar power output correlation and cascading failure simulation. It uses output volatility and frequency regulation success rate as constraints, iteratively trains the collaborative scheduling model based on historical operating parameters, and outputs a converged collaborative scheduling model.

[0058] The parameter solution module is used to obtain the real-time operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems within the acquisition period. With the real-time operating parameters as input, the collaborative scheduling model is executed. The collaborative scheduling model adaptively searches for the optimal solution parameters for the real-time operating parameters and outputs the output parameters of wind power generation, photovoltaic power generation, and hydropower generation.

[0059] Preferably, the output correlation analysis module specifically includes:

[0060] A parameter traversal unit is used to traverse historical working parameters, extract wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency from the historical working parameters, and integrate wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency into a wind-solar signal set;

[0061] The noise interference unit performs digital low-pass filtering on the wind and solar signal set based on the Butterworth digital filter to obtain the interference signal filtering set that eliminates noise interference and frequency deviation;

[0062] The signal fitting unit adopts the discrete Fourier transform method to fit the interference signal filtering set to obtain a discrete fitting set;

[0063] The probability distribution definition unit is used to obtain a discrete fitting set, define a wind-solar probability distribution function based on the discrete fitting set, and obtain the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation.

[0064] Compared with the prior art, the embodiments of the present application have the following beneficial effects:

[0065] In an embodiment of the present invention, the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation is defined by a probability distribution method, and the wind-solar output correlation is quantitatively described based on an improved dynamic time bending algorithm, so as to ensure that the coordinated scheduling model can effectively combine and consider the wind-solar output correlation output curve represented quantitatively when adaptively searching for the optimal solution parameters. The changes in wind and solar output can be predicted and simulated more accurately, thereby adaptively searching for the charging and discharging strategies of energy storage power stations in different time periods and under different meteorological conditions, realizing the optimal configuration of energy storage resources, helping to reduce the phenomenon of wind and solar power abandonment, and improving the utilization efficiency of new energy. It also overcomes the problem that when the existing method establishes the distribution network operation status model of source, grid, load and storage, the wind-solar output correlation output curve cannot be effectively considered. The distribution network operation status model only relies on the wind power and photovoltaic output curves and the load curve and the status of the energy storage equipment, resulting in the output volatility and frequency regulation success rate of the distribution network operation status model cannot be guaranteed.

[0066] In an embodiment of the present invention, by combining the probability distribution method to fit the wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical working parameters, it is possible to accurately achieve a quantitative description of the correlation between wind and solar outputs. During frequency fitting, the Butterworth digital filter is used to perform digital low-pass filtering on the wind and solar signal set, thereby effectively eliminating noise interference and frequency deviation in the parameters. At the same time, the discrete Fourier transform method is used to fit the interference signal filtering set, and the wind and solar probability distribution function is defined based on the discrete fitting set. The probability distribution of the correlation between wind and solar outputs in different time periods can be predicted, providing strong support for the scheduling decision of the power system.

[0067] In the embodiment of the present invention, the wind-solar output joint constraint model is defined by the Copula connection function, so that the wind-solar output correlation constraint can be introduced while considering the uncertainty of the wind-solar output, thereby ensuring that the wind-solar output joint constraint model accurately analyzes and judges the wind-solar output correlation interaction behavior, and through the improved dynamic time bending algorithm, the original distance of the parameters is found, the adjacency matrix is ​​updated, and the wind-solar output correlation relationship mapping is efficiently generated, and finally the improved dynamic time bending algorithm is improved, so that the clustering criterion value of the wind-solar output interaction clustering center determined by the clustering criterion function is more accurate, and the time dependence and nonlinear relationship of the wind-solar output can be more accurately captured, thereby improving the prediction accuracy of the wind-solar output and ensuring the robustness of the collaborative scheduling model.

[0068] In an embodiment of the present invention, a collaborative scheduling model and a training construction method thereof are provided. The collaborative scheduling model uses a cascading failure simulation model as the initial model, and introduces a wind-solar output correlation output curve and a wind-solar-hydropower storage grid topology diagram into the initial model. Based on multi-condition constraints, the output volatility and frequency regulation success rate constraints of the collaborative scheduling model are generated, which helps to identify weak links in the power grid, predict possible fault propagation paths, reduce power grid volatility, and improve the primary frequency regulation success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a schematic diagram of the implementation flow of the wind, solar, and hydropower storage active collaborative real-time control method provided by the present invention.

[0070] Figure 2 The figure shows a flow chart of the implementation of the method for fitting historical working parameters such as wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency by combining the probability distribution method.

[0071] Figure 3 The figure shows a schematic diagram of the implementation process of the method for quantitatively describing the correlation of wind and solar power output by combining the wind and solar power probability distribution function and the improved dynamic time bending algorithm.

[0072] Figure 4 A schematic diagram of the implementation process of the collaborative scheduling model method based on iterative training of historical working parameters is shown.

[0073] Figure 5 It is a structural schematic diagram of the wind, solar, and water storage power grid topology provided by the present invention.

[0074] Figure 6 The figure shows the implementation flow of the collaborative scheduling model's method for adaptively searching for optimal solution parameters for real-time working parameters.

[0075] Figure 7 It is a structural diagram of the wind, solar, and hydropower storage active collaborative real-time control system provided by the present invention. DETAILED DESCRIPTION

[0076] Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the art to which this application belongs. The terms used in the specification of the application are only for the purpose of describing specific embodiments and are not intended to limit this application. The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. The terms "first", "second", etc. in the specification and claims of this application or the above-mentioned drawings are used to distinguish different objects, not to describe a specific order.

[0077] References herein to "embodiments" mean that a particular feature, structure, or characteristic described in connection with the embodiments may be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it constitute an independent or alternative embodiment that is mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0078] When the existing methods establish the distribution network operation status model of source, grid, load and storage, they cannot effectively consider the output curve of wind and solar power output correlation. The distribution network operation status model only relies on the wind power and photovoltaic output curve and the load curve and the status of the energy storage equipment, resulting in the output volatility and frequency regulation success rate of the distribution network operation status model cannot be guaranteed. To address the above problems, we propose a wind, photovoltaic, hydropower and storage active collaborative real-time control method. When executing the wind, photovoltaic, hydropower and storage active collaborative real-time control method, the historical working parameters of wind power generation, photovoltaic power generation, hydropower generation and energy storage system are first obtained. The wind power generation, photovoltaic power generation, environmental data and grid fluctuation frequency in the historical working parameters are fitted by the probability distribution method. Then, the wind and solar power probability distribution function and the improved dynamic time bending algorithm are combined to quantitatively describe the wind and solar power output correlation. Based on the wind and solar power output correlation and cascading failure simulation, a collaborative scheduling model is constructed. Finally, the real-time working parameters of wind power generation, photovoltaic power generation, hydropower generation and energy storage system within the acquisition period are obtained. The collaborative scheduling model adaptively searches for the optimal solution parameters for the real-time working parameters. In an embodiment of the present invention, the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation is defined by a probability distribution method, and the wind-solar output correlation is quantitatively described based on an improved dynamic time bending algorithm, so as to ensure that the coordinated scheduling model can effectively combine and consider the wind-solar output correlation output curve represented quantitatively when adaptively searching for the optimal solution parameters. The changes in wind and solar output can be predicted and simulated more accurately, thereby adaptively searching for the charging and discharging strategies of energy storage power stations in different time periods and under different meteorological conditions, realizing the optimal configuration of energy storage resources, helping to reduce the phenomenon of wind and solar power abandonment, and improving the utilization efficiency of new energy. It also overcomes the problem that when the existing method establishes the distribution network operation status model of source, grid, load and storage, the wind-solar output correlation output curve cannot be effectively considered. The distribution network operation status model only relies on the wind power and photovoltaic output curves and the load curve and the status of the energy storage equipment, resulting in the output volatility and frequency regulation success rate of the distribution network operation status model cannot be guaranteed.

[0079] The embodiment of the present invention provides a method for real-time control of wind, solar, and hydropower storage active power coordination. Figure 1 The figure shows a flow chart of the implementation of the wind, solar, hydro and storage active power coordinated real-time control method, which specifically includes:

[0080] Step S10, obtaining historical operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems;

[0081] It should be noted that historical working parameters include but are not limited to distributed wind power generation, photovoltaic power generation, hydropower generation, energy storage system output parameters, environmental data, active power demand data, and grid fluctuation frequency. After obtaining the historical working parameters, the data can also be processed for outliers and missing values, and the sine value processing of the angular features in the historical parameters of wind power generation can be performed, and the characteristic numerical transformation processing of the environmental data can be performed, thereby constructing a multi-source heterogeneous historical working parameter feature set.

[0082] Step S20, loading historical operating parameters, fitting the wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical operating parameters using a probability distribution method, and obtaining a wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation;

[0083] Step S30, combining the wind-solar probability distribution function and the improved dynamic time warping algorithm to quantitatively describe the wind-solar output correlation;

[0084] Step S40: construct a collaborative scheduling model based on wind and solar power output correlation and cascading failure simulation, use output volatility and frequency regulation success rate as constraints, iteratively train the collaborative scheduling model based on historical working parameters, and output a converged collaborative scheduling model.

[0085] Step S50, obtain the real-time operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems within the collection period, use the real-time operating parameters as input, execute the collaborative scheduling model, and the collaborative scheduling model adaptively searches for the optimal solution parameters for the real-time operating parameters and outputs the wind power generation, photovoltaic power generation, and hydropower generation output parameters.

[0086] In an embodiment of the present invention, the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation is defined by a probability distribution method, and the wind-solar output correlation is quantitatively described based on an improved dynamic time bending algorithm, so as to ensure that the coordinated scheduling model can effectively combine and consider the wind-solar output correlation output curve represented quantitatively when adaptively searching for the optimal solution parameters. The changes in wind and solar output can be predicted and simulated more accurately, thereby adaptively searching for the charging and discharging strategies of energy storage power stations in different time periods and under different meteorological conditions, realizing the optimal configuration of energy storage resources, helping to reduce the phenomenon of wind and solar power abandonment, and improving the utilization efficiency of new energy. It also overcomes the problem that when the existing method establishes the distribution network operation status model of source, grid, load and storage, the wind-solar output correlation output curve cannot be effectively considered. The distribution network operation status model only relies on the wind power and photovoltaic output curves and the load curve and the status of the energy storage equipment, resulting in the output volatility and frequency regulation success rate of the distribution network operation status model cannot be guaranteed.

[0087] The embodiment of the present invention provides a method for fitting historical working parameters such as wind power generation, photovoltaic power generation, environmental data, and power grid fluctuation frequency by combining the probability distribution method. Figure 2 The following is a schematic diagram of a method for fitting wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in historical operating parameters using a probability distribution method. The method for fitting wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in historical operating parameters using a probability distribution method specifically includes:

[0088] Step S101, traversing historical operating parameters, extracting wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency from the historical operating parameters, and integrating wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency into a wind-solar signal set;

[0089] Step S102: performing digital low-pass filtering on the wind and solar signal set based on a Butterworth digital filter to obtain an interference signal filtered set that eliminates noise interference and frequency deviation;

[0090] In the embodiment of the present invention, considering that wind and solar signals (such as wind speed, solar irradiance, etc.) are often interfered with by various noises during the acquisition process, including sensor noise, electromagnetic interference, etc., the Butterworth low-pass filter can effectively suppress these high-frequency noises and retain the low-frequency components in the signal, thereby improving the signal-to-noise ratio of the signal, thereby facilitating the fitting of wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in historical working parameters.

[0091] Among them, the interference signal filtering set is expressed as:

[0092]

[0093] Among them, x(n) is the output representation of the interference signal filtering set, A h Indicates the hth harmonic amplitude. In this embodiment, H represents the harmonic order of the wind and solar signal set, L represents the digital low-pass filter order of the digital filter. In this embodiment, the harmonic order and the low-pass filter order can be 3-20 times, S(n) represents the input wind and solar signal set, B(ω h-Pos ), B(ω h-Neg ) represent the positive and negative frequency sampling angle differences of the wind and solar signal sets, Δf is the sampling frequency difference, f r is the rated frequency of wind and solar signal centralized parameters, f h is the frequency of the harmonic, f s is the signal sampling frequency, B eq (n) represents the Butterworth digital filter, arg(·) is the signal amplitude angle, and B(enbw) represents the filter bandwidth of the Butterworth digital filter;

[0094] In step S102, when the wind and solar signal set is digitally low-pass filtered based on the Butterworth digital filter, unnecessary high-frequency components can be removed by defining the positive and negative frequency sampling angle differences of the Butterworth digital filter and the filtering bandwidth of the digital filter, making the collaborative scheduling model based on these signals more accurate.

[0095] Step S103, using discrete Fourier transform method to fit the interference signal filtered set to obtain a discrete fitting set;

[0096] The discrete fit set is expressed as:

[0097]

[0098] in, represents the discrete fitting set, x(n) represents the input value of the interference signal filtering set, F(n)

[0099] is the Fourier transform function, α is the complex exponential coefficient, represents the variance of the interference signal filtered set;

[0100] Step S104, obtaining a discrete fitting set, defining a wind-solar probability distribution function based on the discrete fitting set, and obtaining the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation;

[0101] The wind and solar probability distribution function expression is as follows:

[0102]

[0103] Where f(x) represents the probability distribution function of wind and solar power, n represents the discrete fitting set parameter, is the probability distribution function, k(x) is the probability kernel function, h represents the bandwidth, and β is the kernel parameter. In this embodiment, the kernel parameter can be 0.1-0.5.

[0104] In an embodiment of the present invention, by combining the probability distribution method to fit the wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical working parameters, it is possible to accurately achieve a quantitative description of the correlation between wind and solar outputs. During frequency fitting, the Butterworth digital filter is used to perform digital low-pass filtering on the wind and solar signal set, thereby effectively eliminating noise interference and frequency deviation in the parameters. At the same time, the discrete Fourier transform method is used to fit the interference signal filtering set, and the wind and solar probability distribution function is defined based on the discrete fitting set. The probability distribution of the correlation between wind and solar outputs in different time periods can be predicted, providing strong support for the scheduling decision of the power system.

[0105] The embodiment of the present invention provides a method for quantitatively describing the correlation between wind and solar power output by combining wind and solar power probability distribution function and improved dynamic time bending algorithm. Figure 3 A schematic diagram of the implementation process of a method for quantitatively describing the correlation between wind and solar power outputs by combining a wind and solar power probability distribution function and an improved dynamic time warping algorithm is shown. The method for quantitatively describing the correlation between wind and solar power outputs by combining a wind and solar power probability distribution function and an improved dynamic time warping algorithm specifically includes:

[0106] Step S201: establishing a wind-solar output joint constraint model with the wind-solar probability distribution function as a constraint, wherein the wind-solar output joint constraint model is defined by a Copula connection function;

[0107] The wind and solar power combined constraint model is defined as:

[0108]

[0109] Wherein, C(W,V) represents the wind-solar output joint constraint function, f(Wx) and f(Vx) are the wind output distribution function and the photovoltaic output distribution function, respectively, and λ is the joint constraint coefficient of the wind-solar output joint constraint model. In this embodiment, the joint constraint coefficient of the model can be 1-8.

[0110] Step S202: loading the discrete fitting set, initializing the adjacency matrix of the discrete fitting set, and finding the original distances of wind power generation and photovoltaic power generation parameters in the wind power output distribution function and photovoltaic output distribution function in the discrete fitting set based on the improved dynamic time warping algorithm;

[0111] Step S203, filling the original distances of wind power generation and photovoltaic power generation parameters into the adjacency matrix of the discrete fitting set, and updating the adjacency matrix based on the improved dynamic time warping algorithm;

[0112] It should be noted that in the discrete fitting set, the adjacency matrix is ​​used to represent the connection relationship between each node (in this case, the wind power generation and photovoltaic power generation projects). First, an empty adjacency matrix is ​​initialized, and its size depends on the number of nodes. Then, according to the calculated original distance, the corresponding value is filled into the corresponding position of the adjacency matrix. If there is no direct connection between two nodes (that is, the distance is infinite or exceeds a certain threshold), it can be represented by a special value in the adjacency matrix.

[0113] Step S204 : analyzing the wind-solar output correlation interaction behavior based on the wind-solar output joint constraint model, and generating a wind-solar output correlation relationship map.

[0114] It's important to note that the wind-solar joint output constraint model uses Copula functions to analyze the interactive behavior of wind and solar output correlations. Copula functions are important tools for describing correlations between random variables. They connect the joint distribution of multiple random variables with their marginal distributions. Introducing Copula functions in the wind-solar joint output constraint model more accurately simulates the correlation between wind and photovoltaic power generation. Using Copula functions, a joint probability distribution model for wind and photovoltaic power generation is constructed. This requires considering their spatial correlation and time series characteristics.

[0115] Step S205: Loading the wind and solar power output correlation relationship map, and identifying at least one set of wind and solar power output interaction cluster centers based on Dijkstra path optimization algorithm analysis;

[0116] Step S206: Calculate the Euclidean distance between the wind / solar output related scenes and the wind / solar output interaction clustering center based on the fuzzy clustering algorithm, and assign the wind / solar output correlation relationship mapping to the wind / solar output interaction clustering center with the closest distance;

[0117] Step S207: using a clustering criterion function to determine the clustering criterion value of the wind-solar output interaction cluster center, selecting the minimum clustering criterion value as the wind-solar output interaction cluster center, and generating a wind-solar output correlation output curve based on the wind-solar output interaction cluster center.

[0118] In this embodiment, through steps S201-S207, the wind-solar output joint constraint model is defined with the Copula connection function, so that the wind-solar output correlation constraint can be introduced while considering the uncertainty of the wind-solar output, thereby ensuring that the wind-solar output joint constraint model accurately analyzes and judges the wind-solar output correlation interaction behavior, and through the improved dynamic time bending algorithm to find the original distance of the parameters and update the adjacency matrix, the wind-solar output correlation relationship mapping is efficiently generated, and finally the improved dynamic time bending algorithm is improved, so that the clustering criterion value of the wind-solar output interaction clustering center determined by the clustering criterion function is more accurate, and the time dependence and nonlinear relationship of the wind-solar output can be more accurately captured, thereby improving the prediction accuracy of the wind-solar output and ensuring the robustness of the collaborative scheduling model.

[0119] The embodiment of the present invention provides a method for iteratively training a collaborative scheduling model based on historical working parameters. Figure 4 A schematic diagram of the implementation process of a method for iteratively training a collaborative scheduling model based on historical working parameters is shown. The method for iteratively training a collaborative scheduling model based on historical working parameters specifically includes:

[0120] Step S301: Obtain historical operating parameters, combine the wind-solar probability distribution function and the improved dynamic time warping algorithm to quantitatively describe the wind-solar output correlation, and divide the historical operating parameters into a training set and a test set in a ratio of 4:1;

[0121] Step S302: constructing a wind-solar-water-storage grid topology map based on historical operating parameters, introducing wind-solar output correlation output curves into the wind-solar-water-storage grid topology map, and enabling the wind-solar-water-storage grid topology map to achieve adaptive dynamic changes;

[0122] It should be noted that Figure 5A schematic diagram of a wind, solar, hydropower, and energy storage grid topology is shown. In this diagram, the serial number represents an actual power device or system component, such as a wind farm, photovoltaic power station, hydropower station, or energy storage facility. As a unique identifier for a node, the serial number helps clearly distinguish and identify each node. The wind, solar, hydropower, and energy storage grid topology can effectively explore the correlation between wind power output, photovoltaic power output, hydropower output, and energy storage output power. When constructing the wind, solar, hydropower, and energy storage grid topology, simulation software (such as Matlab Simulink) can be used to simulate and verify the drawn topology. This includes simulating the operation of the grid under different operating conditions to verify the rationality and stability of the topology.

[0123] Step S303: generating at least one set of decision trees based on random forests, using the cascading failure simulation model as an initial model, and learning local features of the wind, solar, water storage, and power grid topology map based on the random forest decision trees;

[0124] Step S304 , combining global linear regression training to obtain at least one set of collaborative scheduling models integrating wind, solar, water storage, and power grid topology maps and cascading failure simulation models.

[0125] In this embodiment of the present invention, learning local features of a wind, solar, hydro, and storage grid topology using a random forest decision tree allows for calculating the reduction in impurity or impact on model accuracy for each feature when constructing the random forest, thereby assessing feature importance. This helps us understand which features have the greatest impact on the output characteristics of the wind, solar, hydro, and storage grid. Based on the results of the feature importance analysis, features with the greatest impact on the local characteristics of the grid topology are extracted. These features may include geographic location, weather conditions, seasonal variations, and so on.

[0126] Step S305: obtaining a training set, and iteratively training the collaborative scheduling model using the training set;

[0127] Step S306, determining whether the collaborative scheduling model has converged;

[0128] Step S307: If the collaborative scheduling model converges, obtain a training set, and generate output volatility and frequency regulation success rate constraints of the collaborative scheduling model based on multiple conditional constraints;

[0129] If the collaborative scheduling model has not converged, return to step S305.

[0130] Step S308: Determine the final converged collaborative scheduling model based on the output volatility, the frequency regulation success rate combined constraint coefficient, and the collaborative scheduling model objective function, and output the collaborative scheduling model;

[0131] Among them, the objective function of the collaborative scheduling model is:

[0132]

[0133] P z,t =S W P W,t +S V P V,t +S H P H,t +S Q P Q,t (11)

[0134] S=S W +S V +S H +S Q (12)

[0135] Among them, T is the number of cooperative scheduling cycles, V P , S are the combined constraint coefficients of output fluctuation coefficient and frequency regulation success rate, S W , S V , S H , S Q Respectively represent the wind constraint coefficient, photovoltaic constraint coefficient, hydraulic constraint coefficient, and energy storage constraint coefficient. In this embodiment, the wind constraint coefficient, photovoltaic constraint coefficient, hydraulic constraint coefficient, and energy storage constraint coefficient have a value range of 0.1-0.6. z,t represents the total output of the collaborative scheduling model within the collaborative scheduling period, P avg,t represents the average output of the collaborative scheduling model during the collaborative scheduling period, P W,t , P V,t , P H,t , P Q,t They are wind power output power, photovoltaic power output power, hydropower output power, and energy storage output power.

[0136] In an embodiment of the present invention, a collaborative scheduling model and a training construction method thereof are provided. The collaborative scheduling model uses a cascading failure simulation model as the initial model, and introduces a wind-solar output correlation output curve and a wind-solar-hydropower storage grid topology diagram into the initial model. Based on multi-condition constraints, the output volatility and frequency regulation success rate constraints of the collaborative scheduling model are generated, which helps to identify weak links in the power grid, predict possible fault propagation paths, reduce power grid volatility, and improve the primary frequency regulation success rate.

[0137] The embodiment of the present invention provides a method for adaptively searching for optimal solution parameters for real-time working parameters in a collaborative scheduling model. Figure 6 The following is a schematic diagram showing the implementation process of a method for adaptively searching for optimal solution parameters for real-time working parameters using a collaborative scheduling model. The method for adaptively searching for optimal solution parameters using a collaborative scheduling model specifically includes:

[0138] Step S401: obtaining real-time working parameters, iteratively constructing the convex hull of the real-time working parameters based on the augmented Lagrangian algorithm, and generating a parameter convex hull set;

[0139] It should be noted that the iterative construction of the convex hull of real-time operating parameters using the augmented Lagrangian algorithm enables the collaborative scheduling model to effectively solve distributed mixed-integer linear programming problems, ensuring convergence and transforming constrained optimization problems into unconstrained ones, thereby simplifying the problem-solving process. This transformation enables the algorithm to more efficiently handle complex constraints, particularly in practical application scenarios such as power systems, enabling rapid identification of optimal frequency regulation parameters that satisfy various operational constraints.

[0140] Step S402: combining the wind-solar probability distribution function and the improved dynamic time warping algorithm to quantitatively describe the wind-solar output correlation and generate a wind-solar output correlation output curve;

[0141] In this embodiment, the method of quantitatively describing the correlation between wind and solar power outputs is combined with the wind-solar probability distribution function and the improved dynamic time warping algorithm and steps S201 to S207 in the embodiment of the present invention.

[0142] Step S403: Based on the dual solution algorithm combined with cascading failure simulation, the parameter convex hull set is described as the output volatility of the main problem and the frequency regulation success rate of the sub-problem. The optimal solution of wind power output power, photovoltaic power output power, hydropower output power, and energy storage output power is calculated based on the objective function of the coordinated scheduling model.

[0143] Step S404: Obtain the optimal solutions of wind power output power, photovoltaic power output power, hydraulic power output power, and energy storage output power, and adaptively search for optimal solution parameters based on the optimal solutions of wind power output power, photovoltaic power output power, hydraulic power output power, and energy storage output power.

[0144] In this embodiment, when adaptively searching for optimal solution parameters based on the optimal solutions of wind output power, photovoltaic output power, hydropower output power, and energy storage output power, the adaptive search algorithm may be an ant colony algorithm or a simulated annealing algorithm.

[0145] The embodiment of the present invention provides a wind, solar, hydropower and storage active power coordinated real-time control system. Figure 7 The structure diagram of the wind, solar, hydropower and storage active power coordinated real-time control system is shown. The wind, solar, hydropower and storage active power coordinated real-time control system specifically includes:

[0146] Parameter acquisition module 100, for obtaining historical operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems, wherein the historical operating parameters include output parameters of distributed wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems, environmental data, active power demand data, and grid fluctuation frequency;

[0147] Output correlation analysis module 200 loads historical operating parameters and uses a probability distribution method to fit wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical operating parameters to obtain wind and photovoltaic power generation active output probability distribution functions. This module then uses the wind and photovoltaic power generation probability distribution functions and an improved dynamic time warping algorithm to quantitatively describe the wind and photovoltaic output correlation.

[0148] Model building module 300 builds a collaborative scheduling model based on wind and solar power output correlation and cascading failure simulation. It uses output volatility and frequency regulation success rate as constraints, iteratively trains the collaborative scheduling model based on historical operating parameters, and outputs a converged collaborative scheduling model.

[0149] The parameter solution module 400 is used to obtain the real-time operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems within the collection period. The collaborative scheduling model uses the real-time operating parameters as input. The collaborative scheduling model adaptively searches for the optimal solution parameters for the real-time operating parameters and outputs the wind power generation, photovoltaic power generation, and hydropower generation output parameters.

[0150] In this embodiment, the output correlation analysis module 200 specifically includes:

[0151] The parameter traversal unit 210 is used to traverse the historical working parameters, extract wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency from the historical working parameters, and integrate the wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency into a wind-solar signal set;

[0152] The noise interference unit 220 performs digital low-pass filtering on the wind and solar signal set based on a Butterworth digital filter to obtain an interference signal filtered set that eliminates noise interference and frequency deviation;

[0153] The signal fitting unit 230 uses a discrete Fourier transform method to fit the interference signal filtered set to obtain a discrete fitting set;

[0154] The probability distribution definition unit 240 is used to obtain a discrete fitting set, define a wind-solar probability distribution function based on the discrete fitting set, and obtain the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation.

[0155] It should be noted that the wind, solar, hydropower and storage active power coordinated real-time control system in the embodiment of the present invention corresponds to the above-mentioned wind, solar, hydropower and storage active power coordinated real-time control method. The explanations, examples, beneficial effects and other parts of its relevant contents can refer to the corresponding contents in the wind, solar, hydropower and storage active power coordinated real-time control method, and will not be repeated here.

[0156] On the other hand, an embodiment of the present invention further provides a computer-readable storage medium storing computer program instructions, which can be executed by a processor. When the computer program instructions are executed, the method of any of the above embodiments is implemented.

[0157] The computer-readable storage medium (e.g., memory) herein can be a volatile memory or a non-volatile memory, or can include both a volatile memory and a non-volatile memory. As an example and not limitation, the non-volatile memory can include a read-only memory (ROM), a programmable ROM (PROM), an electrically programmable ROM (EPROM), an electrically erasable programmable ROM (EEPROM), or a flash memory. The volatile memory can include a random access memory (RAM), which can act as an external cache memory. As an example and not limitation, RAM can be obtained in a variety of forms, such as synchronous RAM (DRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), and direct Rambus RAM (DRRAM). The storage device of the disclosed aspects is intended to include, but is not limited to, these and other suitable types of memories.

[0158] As another aspect of an embodiment of the present invention, a computer device is provided. The computer device includes a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the method of any one of the above embodiments is implemented.

[0159] Memory, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer executable programs, and modules, such as the program instructions / modules corresponding to the wind, solar, hydropower, and hydropower coordinated real-time control method in the embodiments of the present application. The memory can include a program storage area and a data storage area. The program storage area can store an operating system and at least one application required for a function; the data storage area can store data created by using the wind, solar, hydropower, and hydropower coordinated real-time control method. Furthermore, the memory can include high-speed random access memory and non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, the memory can optionally include memory remotely located relative to the processor, and these remote memories can be connected to the local module via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0160] In summary, the present invention provides a method and system for real-time coordinated control of wind, solar, and hydropower storage. In an embodiment of the present invention, the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation is defined by a probability distribution method, and the wind-solar output correlation is quantitatively described based on an improved dynamic time bending algorithm, thereby ensuring that the coordinated scheduling model effectively combines and considers the quantified wind-solar output correlation output curve when adaptively searching for the optimal solution parameters. It can more accurately predict and simulate the changes in wind and solar output, thereby adaptively searching for the charging and discharging strategies of energy storage power stations in different time periods and different meteorological conditions, realizing the optimal configuration of energy storage resources, helping to reduce the phenomenon of wind and solar power abandonment, and improving the utilization efficiency of new energy. It also overcomes the problem that the existing method cannot effectively consider the wind-solar output correlation output curve when establishing the distribution network operation status model of source, grid, load and storage. The distribution network operation status model only relies on the wind power and photovoltaic output curves and the load curve and the status of the energy storage equipment, resulting in the output volatility and frequency regulation success rate of the distribution network operation status model cannot be guaranteed.

[0161] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the scope of protection of the invention. Obviously, the embodiments described are only some embodiments of the present invention, rather than all embodiments. Based on these embodiments, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in this field can still combine, add, delete or make other adjustments to the features in the various embodiments of the present invention according to the circumstances without conflict, without making creative work, so as to obtain different other technical solutions that do not deviate from the concept of the present invention in essence, and these technical solutions also fall within the scope of protection of the present invention.

Claims

1. A method for real-time control of wind, solar, and hydropower storage active power coordination, characterized in that: The wind, solar, and hydropower storage active power coordinated real-time control method includes: Obtain historical operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems. These parameters include output parameters of distributed wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems, environmental data, active power demand data, and grid fluctuation frequency. Load historical operating parameters and use the probability distribution method to fit wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical operating parameters to obtain the wind and photovoltaic power generation active output probability distribution function; Combined with the wind-solar probability distribution function and the improved dynamic time bending algorithm, the correlation between wind-solar output and power generation is quantitatively described. A collaborative scheduling model is constructed based on the correlation between wind and solar power output and the cascading failure simulation model. With output volatility and frequency regulation success rate as constraints, the collaborative scheduling model is iteratively trained based on historical operating parameters to output a converged collaborative scheduling model. Among them, using the wind-solar probability distribution function as a constraint, a wind-solar output joint constraint model is established. The original distance of the parameters is found and the adjacency matrix is ​​updated through an improved dynamic time warping algorithm. The adjacency matrix is ​​used to represent the connection relationship between wind power generation and photovoltaic power generation projects. Among them, the wind-solar-water-storage power grid topology is introduced into the wind-solar-water-storage power grid correlation output curve. The cascading failure simulation model is used as the initial model to learn the local characteristics of the wind-solar-water-storage power grid topology, and obtain a collaborative scheduling model that integrates the wind-solar-water-storage power grid topology and the cascading failure simulation model.

2. The method for real-time control of wind, solar, and hydropower storage active power coordination according to claim 1, characterized in that: The method further comprises: The real-time operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems within the collection period are obtained. The collaborative scheduling model is executed with the real-time operating parameters as input. The collaborative scheduling model adaptively searches for the optimal solution parameters for the real-time operating parameters and outputs the output parameters of wind power generation, photovoltaic power generation, and hydropower generation.

3. The method for real-time control of wind, solar, and hydropower storage active power coordination according to claim 1, characterized in that: The method of fitting the wind power generation, photovoltaic power generation, environmental data, and power grid fluctuation frequency in historical operating parameters by combining the probability distribution method specifically includes: Traverse the historical working parameters, extract wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency from the historical working parameters, and integrate wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency into a wind-solar signal set; The wind and solar signal set is digitally low-pass filtered based on the Butterworth digital filter to obtain the interference signal filter set that eliminates noise interference and frequency deviation; Among them, the interference signal filtering set is expressed as: Among them, x(n) is the output representation of the interference signal filtering set, A h represents the hth harmonic amplitude, H represents the harmonic order of the wind and solar signal set, L represents the digital low-pass filtering order of the digital filter, S(n) represents the input wind and solar signal set, B(ω h-Pos ), B(ω h-Neg ) represent the positive and negative frequency sampling angle differences of the wind and solar signal sets, Δf is the sampling frequency difference, f r is the rated frequency of wind and solar signal centralized parameters, f h is the frequency of the harmonic, f s is the signal sampling frequency, B eq (n) represents the Butterworth digital filter, arg(·) is the signal amplitude angle, and B(enbw) represents the filter bandwidth of the Butterworth digital filter; The interference signal filtering set is fitted using discrete Fourier transform method to obtain a discrete fitting set; The discrete fit set is expressed as: in, represents the discrete fitting set, x(n) represents the input value of the interference signal filtering set, F(n) is the Fourier transform function, α is the complex exponential coefficient, represents the variance of the interference signal filtered set; Obtain a discrete fitting set, define a wind-solar probability distribution function based on the discrete fitting set, and obtain the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation; The wind and solar probability distribution function expression is as follows: Where f(x) represents the probability distribution function of wind and solar power, n represents the discrete fitting set parameter, is the probability distribution function, k(x) is the probability kernel function, h represents the bandwidth, and β is the kernel parameter.

4. The method for real-time control of wind, solar, and hydropower storage active power coordination according to claim 1, characterized in that: The method of quantitatively describing the correlation between wind and solar power output by combining the wind and solar power probability distribution function and the improved dynamic time warping algorithm specifically includes: Taking the wind-solar probability distribution function as a constraint, a wind-solar output joint constraint model is established. The wind-solar output joint constraint model is defined by the Copula connection function. The wind and solar power combined constraint model is defined as: Where C(W,V) represents the wind-solar output joint constraint function, f(Wx) and f(Vx) are the wind output distribution function and photovoltaic output distribution function respectively, and λ is the joint constraint coefficient of the wind-solar output joint constraint model; Load the discrete fitting set, initialize the adjacency matrix of the discrete fitting set, and find the original distances of wind power generation and photovoltaic power generation parameters in the wind power output distribution function and photovoltaic output distribution function in the discrete fitting set based on the improved dynamic time warping algorithm; The original distances of wind power generation and photovoltaic power generation parameters are filled into the adjacency matrix of the discrete fitting set, and the adjacency matrix is ​​updated based on the improved dynamic time warping algorithm; Based on the wind-solar output joint constraint model, the wind-solar output correlation interaction behavior is analyzed, and a wind-solar output correlation relationship map is generated.

5. The method for real-time control of wind, solar, and hydropower storage active power coordination according to claim 4, characterized in that: The method of quantitatively describing the correlation between wind and solar power output by combining the wind and solar power probability distribution function and the improved dynamic time warping algorithm specifically includes: Load the wind and solar power output correlation relationship map, and identify at least one set of wind and solar power output interaction cluster centers based on the Dijkstra path optimization algorithm; Based on the fuzzy clustering algorithm, the Euclidean distance between the wind-solar output related scenes and the wind-solar output interaction cluster center is calculated, and the wind-solar output correlation relationship is mapped and assigned to the wind-solar output interaction cluster center with the nearest distance; The clustering criterion function is used to determine the clustering criterion value of the wind-solar output interaction cluster center, the minimum clustering criterion value is selected as the wind-solar output interaction cluster center, and the wind-solar output correlation output curve is generated based on the wind-solar output interaction cluster center.

6. The method for real-time control of wind, solar, and hydropower storage active power coordination according to claim 5, characterized in that: The method for iteratively training the collaborative scheduling model based on historical working parameters specifically includes: Obtain historical operating parameters, combine them with the wind-solar probability distribution function and the improved dynamic time warping algorithm to quantitatively describe the correlation between wind-solar output and the wind-solar output. Then divide the historical operating parameters into training and test sets in a 4:1 ratio. Based on historical working parameters, a wind-solar-water-storage grid topology is constructed, and the wind-solar-water-storage grid topology curve is introduced into the wind-solar-water-storage grid topology, so that the wind-solar-water-storage grid topology can achieve adaptive dynamic changes; Generate at least one set of decision trees based on random forests, use the cascading failure simulation model as the initial model, and learn the local features of the wind, solar, water and storage grid topology map based on the random forest decision trees; Combined with global linear regression training, at least one set of collaborative scheduling models integrating wind, solar, water storage and power grid topology maps and cascading failure simulation models is obtained.

7. The method for real-time control of wind, solar, and hydropower storage active power coordination according to claim 6, characterized in that: The method for iteratively training the collaborative scheduling model based on historical working parameters specifically further includes: Obtaining a training set, iteratively training the collaborative scheduling model using the training set, and determining whether the collaborative scheduling model has converged; If the collaborative scheduling model converges, obtain the training set and generate the output volatility and frequency regulation success rate constraints of the collaborative scheduling model based on multiple conditional constraints; Determine the final converged collaborative scheduling model based on output volatility, frequency regulation success rate combined constraint coefficient, and collaborative scheduling model objective function, and output the collaborative scheduling model; Among them, the objective function of the collaborative scheduling model is: P z,t =S W P W,t +S V P V,t +S H P H,t +S Q P Q,t (11) S=S W +S V +S H +S Q (12) Among them, T is the number of cooperative scheduling cycles, V P , S are the combined constraint coefficients of output fluctuation coefficient and frequency regulation success rate, S W , S V , S H , S Q They represent wind constraint coefficient, photovoltaic constraint coefficient, hydraulic constraint coefficient, and energy storage constraint coefficient respectively, and P z,t represents the total output of the collaborative scheduling model within the collaborative scheduling period, P avg,t represents the average output of the collaborative scheduling model during the collaborative scheduling period, P W,t , P V,t , P H,t , P Q,t They are wind power output power, photovoltaic power output power, hydropower output power, and energy storage output power.

8. The method for real-time control of wind, solar, and hydropower storage active power coordination according to claim 7, characterized in that: The method of the collaborative scheduling model for adaptively searching for optimal solution parameters for real-time working parameters specifically includes: Obtain real-time working parameters, iteratively construct the convex hull of the real-time working parameters based on the augmented Lagrangian algorithm, and generate a parameter convex hull set; Combined with the wind-solar probability distribution function and the improved dynamic time bending algorithm, the wind-solar output correlation is quantitatively described to generate a wind-solar output correlation output curve; Based on the dual solution algorithm combined with cascading failure simulation, the parameter convex hull set is described as the output volatility of the main problem and the frequency regulation success rate of the sub-problem. Based on the objective function of the coordinated scheduling model, the optimal solutions for wind power output power, photovoltaic power output power, hydropower output power, and energy storage output power are calculated. Obtain the optimal solutions for wind output power, photovoltaic output power, hydropower output power, and energy storage output power, and adaptively search for optimal solution parameters based on the optimal solutions for wind output power, photovoltaic output power, hydropower output power, and energy storage output power.

9. A wind, solar, hydropower and storage active power coordinated real-time control system, configured to implement the wind, solar, hydropower and storage active power coordinated real-time control method according to any one of claims 1 to 8, characterized in that: The wind, solar, hydropower and storage active power coordinated real-time control system specifically includes: The parameter acquisition module is used to obtain historical operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems. The historical operating parameters include output parameters of distributed wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems, environmental data, active power demand data, and grid fluctuation frequency. The output correlation analysis module loads historical operating parameters and uses the probability distribution method to fit the wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency in the historical operating parameters to obtain the wind and photovoltaic power generation active output probability distribution function. The wind and photovoltaic power generation probability distribution function is combined with the improved dynamic time warping algorithm to quantitatively describe the wind and photovoltaic output correlation; The model building module builds a collaborative scheduling model based on wind and solar power output correlation and cascading failure simulation. It uses output volatility and frequency regulation success rate as constraints, iteratively trains the collaborative scheduling model based on historical operating parameters, and outputs a converged collaborative scheduling model. The parameter solution module is used to obtain the real-time operating parameters of wind power generation, photovoltaic power generation, hydropower generation, and energy storage systems within the acquisition period. With the real-time operating parameters as input, the collaborative scheduling model is executed. The collaborative scheduling model adaptively searches for the optimal solution parameters for the real-time operating parameters and outputs the output parameters of wind power generation, photovoltaic power generation, and hydropower generation.

10. The wind, solar, hydropower and storage active power coordinated real-time control system according to claim 9, characterized in that: The output correlation analysis module specifically includes: A parameter traversal unit is used to traverse historical working parameters, extract wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency from the historical working parameters, and integrate wind power generation, photovoltaic power generation, environmental data, and grid fluctuation frequency into a wind-solar signal set; The noise interference unit performs digital low-pass filtering on the wind and solar signal set based on the Butterworth digital filter to obtain the interference signal filtering set that eliminates noise interference and frequency deviation; The signal fitting unit adopts the discrete Fourier transform method to fit the interference signal filtering set to obtain a discrete fitting set; The probability distribution definition unit is used to obtain a discrete fitting set, define a wind-solar probability distribution function based on the discrete fitting set, and obtain the wind-solar probability distribution function of the active output of wind power generation and photovoltaic power generation.

Images