Data-driven photovoltaic power generation estimation method and system based on Betti curve

Abstract

The invention discloses a Betti curve-based data-driven photovoltaic power generation estimation method and system, and the method comprises the steps: collecting first object data, carrying out the first reconstruction calculation, and obtaining a first feature of a first object; and performing statistical calculation according to the first object data to obtain a second feature of the first object. And performing a first conversion task for the specific information of the first object data to obtain a first conversion task result. And splicing the first feature, the second feature and the first conversion task result to form a feature space, inputting the feature space to a first prediction model, and outputting a prediction result of the first object. According to the method, the complex dynamic mode of the time sequence is captured, the diversification of the feature space is realized, and the generalization ability of the prediction model is improved. According to the method, multi-dimensional features are combined, an efficient machine learning algorithm is adopted, high-precision estimation of the photovoltaic power generation power is achieved, and meanwhile the adaptability of the model to nonlinear and multi-scale features is improved.

Description

Technical Field

[0001] The present invention relates to the field of new energy technology, and in particular to a data-driven photovoltaic power generation estimation method and system based on a Betty curve. Background Art

[0002] In recent years, with the continuous growth of global demand for clean energy, photovoltaic power generation has been widely used as a green and efficient form of energy. However, since photovoltaic power generation depends on weather conditions and environmental factors, its output power has significant randomness and volatility. Traditional photovoltaic power generation estimation methods mainly rely on physical models and empirical formulas. Although these methods can provide preliminary estimation results under certain conditions, they often have large estimation errors and insufficient real-time performance when faced with complex and changeable weather conditions and long-term data fluctuations. With the development of big data and machine learning technology, data-driven photovoltaic power generation estimation methods have gradually become a research hotspot. Among them, how to use deep-level feature extraction and modeling of complex time series has become the key to improving estimation accuracy.

[0003] Existing data-driven photovoltaic power generation estimation methods usually adopt the idea of ​​combining time series analysis with machine learning models. In terms of feature extraction, these methods mainly rely on simple statistical features (such as mean, standard deviation, etc.), but often fail to fully explore the topological structure information in the time series. In addition, most methods ignore the high-order relationships and complex dynamic characteristics of time series, and only use shallow models for regression prediction, which makes it difficult to deal with the nonlinear and multi-scale characteristics of photovoltaic power generation. Especially in the face of extreme weather and environmental changes, the prediction accuracy and robustness of traditional methods are further reduced. Therefore, there is an urgent need for an innovative estimation method that can effectively capture the deep topological features of time series and combine them with multidimensional statistical features. Summary of the invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is that the traditional method only uses basic statistical features and ignores the topological structure information in the time series. The present invention can effectively extract the deep topological features of the time series by introducing the Betti curve. Existing methods perform poorly when faced with multi-scale time series and nonlinear relationships. The present invention combines multi-dimensional feature input and the XGBoost model to achieve efficient modeling of complex dynamic characteristics. Traditional methods usually do not consider the periodicity and category properties of the time dimension. The present invention enhances the ability to capture time dimension features through unique hot encoding of specific information.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: a data-driven photovoltaic power generation estimation method based on the Betti curve, comprising: collecting first object data, performing a first reconstruction calculation, and obtaining a first feature of the first object.

[0007] Statistical calculation is performed based on the first object data to obtain a second feature of the first object.

[0008] A first conversion task is performed on the specific information of the first object data to obtain a first conversion task result.

[0009] The first feature, the second feature and the result of the first conversion task are concatenated to form a feature space, which is input into the first prediction model to output the prediction result of the first object.

[0010] As a preferred solution of the data-driven photovoltaic power generation estimation method based on the Betty curve described in the present invention, wherein: the collecting of the first object data and the first analysis include collecting the time-related data of the first object, converting the time series into point cloud data, and filtering and calculating to obtain the first feature.

[0011] As a preferred solution of the data-driven photovoltaic power generation estimation method based on the Betti curve described in the present invention, wherein: performing statistical calculations based on the first object data to obtain the second feature of the first object includes extracting multiple statistical features based on the first object data.

[0012] As a preferred scheme of the data-driven photovoltaic power generation estimation method based on the Betti curve described in the present invention, wherein: the first conversion task for the specific information of the first object data includes encoding the specific information of the first object data and converting the specific information into the first conversion task result.

[0013] As a preferred solution of the data-driven photovoltaic power generation estimation method based on the Betty curve described in the present invention, wherein: the first object includes but is not limited to photovoltaic power generation equipment, and the first object data includes a historical photovoltaic power generation time series and a historical load time series.

[0014] The first reconstruction calculation includes but is not limited to phase space reconstruction using Takens embedding theory.

[0015] The first feature is a topological feature, and the second feature is a statistical feature.

[0016] The first conversion task is to convert the date feature into a category feature, and the result of the first conversion task is the category feature.

[0017] The first predictive model includes, but is not limited to, a machine learning regressor.

[0018] As a preferred solution of the data-driven photovoltaic power generation estimation method based on the Betti curve of the present invention, wherein: the first reconstruction calculation to obtain the first feature of the first object includes using Takens embedding theory to perform phase space reconstruction and convert the time series into point cloud data. n-1} is a one-dimensional time series. Given the delay parameter τ and the dimension parameter d, the Takens phase space reconstruction point corresponding to time t is defined as:

[0019] z t :=(x t ,x t+τ ,…,x t+(d-1)τ )∈R d

[0020] Vietoris-Rips filtering is applied to the point cloud data to generate the filter complex and calculate the persistence graph, from which the 0-dimensional and 1-dimensional Betty curves are extracted as the topological features of the time series as the first feature.

[0021] As a preferred scheme of the data-driven photovoltaic power generation estimation method based on the Betty curve described in the present invention, wherein: the statistical calculation based on the first object data to obtain the second feature of the first object includes calculating the absolute energy feature, the average value feature, the standard deviation feature, the skewness feature, the kurtosis feature, the absolute average change feature, the average second-order difference center feature and the median feature.

[0022] The first conversion task is performed, and the first conversion task result obtained includes: for date information, it is converted into a category feature through one-hot encoding. A date has 12 possible month attributes, 31 possible day attributes, and 7 possible week attributes. The one-hot encoding is vectorized to obtain a 50-dimensional category feature vector.

[0023] A data-driven photovoltaic power generation estimation system based on Betty curve, characterized by: comprising:

[0024] The feature calculation module collects the first object data, performs a first reconstruction calculation, and obtains a first feature of the first object. The module performs a statistical calculation based on the first object data and obtains a second feature of the first object.

[0025] The conversion module performs a first conversion task on the specific information of the first object data to obtain a first conversion task result.

[0026] The prediction module concatenates the first feature, the second feature and the result of the first conversion task to form a feature space, inputs the feature space into the first prediction model, and outputs the prediction result of the first object.

[0027] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0028] A computer-readable storage medium stores a computer program, which implements the steps of the method described above when executed by a processor.

[0029] The beneficial effects of the present invention are as follows: the time series is converted into point cloud data using Takens embedding theory, and a persistence graph is obtained through Vietoris-Rips filtering calculation, from which the topological features of the Betty curve are extracted. The complex dynamic patterns of the time series are effectively captured, providing a unique high-dimensional feature description for subsequent predictions. Compared with traditional simple statistical features, this method can describe the nonlinear dynamic structure of the time series and improve the ability to characterize photovoltaic power changes under complex weather conditions. A variety of statistical features including mean, standard deviation, skewness, etc. are extracted to reflect the distribution characteristics of the time series. The low-dimensional statistical information that is not fully covered in the topological features is supplemented, the feature space is diversified, and the generalization ability of the prediction model is improved.

[0030] The date features are converted into category features through unique hot encoding, which further supplements the time dimension attribute of the time series. The feature space's ability to parse time information is enhanced, enabling the model to better capture periodic patterns. Topological features, statistical features, and category features are spliced ​​into feature space and input into the XGBoost regression model for prediction. Combining multidimensional features and adopting efficient machine learning algorithms, high-precision estimation of photovoltaic power generation is achieved, while improving the model's adaptability to nonlinear and multi-scale characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0032] Figure 1 The present invention provides an overall flow chart of a data-driven photovoltaic power generation estimation method and system based on the Betty curve according to the first embodiment of the present invention.

[0033] Figure 2 A periodic time series diagram of a data-driven photovoltaic power generation estimation method and system based on a Betti curve provided in the first embodiment of the present invention.

[0034] Figure 3A chaotic time series diagram of a data-driven photovoltaic power generation estimation method and system based on a Betty curve provided in the first embodiment of the present invention.

[0035] Figure 4 A 0-dimensional Betty curve diagram of a data-driven photovoltaic power generation estimation method and system based on a Betty curve provided in the first embodiment of the present invention.

[0036] Figure 5 A 1-dimensional Betty curve diagram of a data-driven photovoltaic power generation estimation method and system based on a Betty curve provided in the first embodiment of the present invention.

[0037] Figure 6 A 2-dimensional Betty curve diagram of a data-driven photovoltaic power generation estimation method and system based on a Betty curve provided in the first embodiment of the present invention.

[0038] Figure 7 Elia photovoltaic data diagram of a data-driven photovoltaic power generation estimation method and system based on a Betti curve provided in the second embodiment of the present invention.

[0039] Figure 8 The second embodiment of the present invention provides a data-driven photovoltaic power generation estimation method and system based on the Betty curve, and the prediction results of the method, NbeatsX model and continuous model for 10 consecutive days.

[0040] Fig. 9 A data-driven photovoltaic power generation estimation method and system based on a Betty curve provided in the second embodiment of the present invention is compared with the prediction results of the method and system not using topological features for 10 consecutive days. DETAILED DESCRIPTION

[0041] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0042] Example 1, reference Figure 1 to Figure 6 , which is an embodiment of the present invention, provides a data-driven photovoltaic power generation estimation method based on a Betty curve, comprising:

[0043] S1: Collect first object data, perform a first reconstruction calculation, and obtain a first feature of the first object.

[0044] In the present invention, the first object is a photovoltaic power generation device, the first object data includes a historical photovoltaic power generation time series and a historical load time series, the first reconstruction calculation is a phase space reconstruction using Takens embedding theory, and the first feature is a topological feature.

[0045] In order to analyze the topological properties of historical photovoltaic and historical load time series, it is necessary to first convert the time series into point cloud representation. The method used in the present invention is phase space reconstruction, and its theoretical basis comes from the following theorem:

[0046] Takens embedding theorem, if is an n-dimensional compact manifold, φ:R×M→M and f:M→R are almost everywhere smooth mappings. Given a delay parameter τ, for a point x in M, define x i =φ(i·τ,x), then, g:M→R 2n+1 ,x→(f(x),f(x1),…·,f(x 2n )) is the embedding mapping between manifolds.

[0047] Specifically, in the scenario to be studied in the present invention, if X = {x0, ..., x n-1} is a one-dimensional time series, then given the delay parameter τ and the dimension parameter d, the Takens phase space reconstruction point corresponding to time t is defined as:

[0048] z t :=(x t ,x t+τ ,…,x t+(d-1)τ )∈R d

[0049] Thus, given the parameters (d,τ), phase space reconstruction converts the time series into point cloud data.

[0050] Figure 2 is a schematic diagram of a periodic time series. Figure 3 It is a schematic diagram of chaotic time series.

[0051] On the basis of phase space reconstruction, Vietoris-Rips filtering is used to filter the point cloud data to obtain the filter complex and then the persistence graph is obtained based on it. Using the persistence graph, the Betti curve of the time series can be calculated.

[0052] Using the persistence graph, the Betti curve of the time series is calculated. For a specified dimension d, the d-dimensional persistence homology Betti curve is the curve formed by the number of free generators of the homology group in each layer of filtering. By definition, it is obvious that the 0-dimensional Betti curve is a curve that monotonically decreases along the filtering time axis, and the higher-dimensional Betti curve reflects the fine-grained topological properties of the time series phase space. Figure 3 Take the time series as an example, the 0-dimensional Betty curve is as follows Figure 4 , the 1-dimensional Betti curve is as follows Figure 5 , the 2D Betty curve is as follows Figure 6 .

[0053] It should be noted that the first object includes but is not limited to photovoltaic power generation equipment, and may also be wind power generation equipment and power load monitoring equipment. The first reconstruction calculation includes but is not limited to phase space reconstruction using Takens embedding theory, and may also be a sliding window method based on delay embedding and a method based on RecurrencePlot (RP).

[0054] In an optional embodiment of the present invention, the first object is a wind power generation device, and the first object data is data collected based on a historical wind speed time series and a wind power generation power time series.

[0055] Takens embedding theory phase space reconstruction calculation is performed. For the wind speed time series, given the delay parameter τ = 10 minutes and the embedding dimension m = 3, the time series is mapped into a three-dimensional point cloud using the formula: xt = {xt, xt + τ, xt + 2τ}. Vietoris-Rips filtering is used to construct the complex, calculate the persistence graph and extract the 0-dimensional Betty curve (the number of connected components) and the 1-dimensional Betty curve (the number of cycles) as the topological features of the wind speed time series.

[0056] In an optional embodiment of the present invention, the first object is an electric load monitoring device, and the first object data is data collected based on historical electric load time series and voltage fluctuation time series. Takens embedding theory phase space reconstruction calculation is performed. For the electric load time series, given the delay parameter τ = 15 minutes and the embedding dimension m = 4, the formula: xt = {xt, xt + τ, xt + 2τ, xt + 3τ} is used to map it to a four-dimensional point cloud. The Alpha filtering algorithm is applied to the generated point cloud data to construct a topological complex and calculate the 0-dimensional and 1-dimensional Betty curves to describe the topological characteristics of the load time series.

[0057] In an optional embodiment of the present invention, the first reconstruction calculation method is a sliding window method based on delay embedding. For photovoltaic power generation equipment, based on the historical photovoltaic power generation power time series, the sliding window width is set to W = 60 minutes and the step length is S = 15 minutes. The time series segment in each sliding window is represented by delay embedding:

[0058] x t ={x t ,x t+S ,x t+2S ,…,x t+(W-1)S}

[0059] Among them, x t Represents a collection of time series.

[0060] Topological feature extraction: The Persistent Cohomology method is applied to the embedded point cloud of each sliding window to extract the persistence graph and calculate the Betti curve to describe the changes in the topological characteristics of the photovoltaic power generation.

[0061] In an optional embodiment of the present invention, the first reconstruction calculation method is a method based on Recurrence Plot (RP). For the time series of photovoltaic power generation equipment, the RP matrix is ​​constructed:

[0062] R(i,j)=Θ(∈-|x i -x j |)

[0063] Where Θ is the step function, ∈ is the distance threshold, and x i and x j is the embedded point.

[0064] The Cliques Topology algorithm is applied to the similarity network in the RP matrix to extract 0-dimensional and 1-dimensional Betti curves to describe the cyclic and connectivity characteristics of the time series.

[0065] S2: Perform statistical calculations based on the first object data to obtain a second feature of the first object.

[0066] For the historical weather, historical photovoltaic and historical load time series data, calculate their statistical characteristics. The specific characteristics and formulas involved are as follows:

[0067] Absolute energy, which is the sum of the squares of the components of the time series:

[0068] E=∑x i 2

[0069] The average value, i.e. the mean of the time series:

[0070]

[0071] Standard Deviation:

[0072]

[0073] Skewness, the third-order standardized moment of the time series:

[0074]

[0075] It is a measure that describes the direction and degree of skewness of the distribution of statistical data.

[0076] Kurtosis, the fourth-order standardized moment of the time series:

[0077]

[0078] It is a characteristic quantity that describes the height of the peak of the probability density distribution curve at the mean value.

[0079] The absolute average change is the mean of the absolute values ​​of the continuous change values ​​of the time series:

[0080]

[0081] Average second-order difference center: that is, the mean of the second-order change of the time series:

[0082]

[0083] The median is the median of each value in the time series arranged from small to large.

[0084] S3: Perform a first conversion task on the specific information of the first object data to obtain a first conversion task result.

[0085] In the present invention, the specific information of the first object data is date information, the first conversion task is to convert the date feature into a category feature, and the first conversion task result is the category feature.

[0086] Furthermore, one-hot encoding, that is, for a variable with n states, using e i Express its i-th state (here e i is an n-dimensional column vector with 1 only in the i-th component and 0 in the other components) Specifically, a date has 12 possible month attributes, 31 possible day attributes, and 7 possible week attributes. One-hot encoding uses a 50-dimensional feature vector to express the attribute information of the date through vectorization.

[0087] S4: Concatenate the first feature, the second feature, and the result of the first conversion task to form a feature space, input it into the first prediction model, and output the prediction result of the first object.

[0088] All features are connected to form a feature space. Finally, the feature space is input into a machine learning regressor for training to predict photovoltaics. The first prediction model of the present invention is the XGboos regressor. The XGBoost regressor is a powerful, flexible and efficient machine learning tool that provides excellent performance and a wide range of functions through a gradient boosting framework and is suitable for solving various complex regressor problems. Therefore, XGBoost is selected as the machine learning regressor.

[0089] It should be noted that the first prediction model includes but is not limited to an XGboos regressor, and may also be a random forest regressor and a support vector machine regressor.

[0090] In an optional embodiment of the present invention, the first prediction model is a random forest regressor, which is an ensemble learning method based on decision trees. It improves prediction accuracy by weighted averaging the results of multiple decision trees and has strong robustness and generalization ability.

[0091] Set the number of decision trees N = 100, the maximum depth D = 10, and the minimum number of split samples to 5. Concatenate the first feature, the second feature, and the result of the first conversion task to form a feature space. Use the training set data of the historical photovoltaic power generation time series to build a random forest model. Input the feature space of the test data into the random forest regressor and output the predicted value of photovoltaic power generation.

[0092] Random forest can effectively handle the relationship between nonlinear features and is not prone to overfitting, making it suitable for complex multidimensional regression problems such as photovoltaic power generation prediction.

[0093] In an optional embodiment of the present invention, the first prediction model is a support vector machine regressor, which makes predictions by finding the best hyperplane between data points, has strong global optimization capabilities, and is particularly suitable for small sample and high-dimensional data problems.

[0094] Use the radial basis kernel function (RBF) with kernel parameter γ = 0.1 and regularization parameter C = 1. Input the concatenated feature space. Train the support vector machine regressor by minimizing the loss function within the error range (∈). Use the trained SVR model to predict the test data and output the estimated value of photovoltaic power generation.

[0095] SVR has strong processing capabilities for high-dimensional spatial data and can accurately fit nonlinear relationships in photovoltaic power generation prediction while avoiding over-reliance on the number of samples.

[0096] The computer device may be a server. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface is connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data cluster data of the power monitoring system. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a data-driven photovoltaic power generation estimation method based on a Betty curve is implemented.

[0097] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0098] Example 2, reference Figure 7 to Figure 9 , which is an embodiment of the present invention, provides a data-driven photovoltaic power generation estimation method and system based on the Betty curve. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0099] This paper selects the public dataset Elia as experimental data. Elia is the main power company in Belgium, responsible for the construction and maintenance of the country's power transmission and distribution infrastructure. Elia focuses on the construction and maintenance of electrical infrastructure, including high-voltage transmission lines and distribution networks. The Elia dataset is sampled at intervals of 15 minutes, that is, there are four data points per hour. The visualization of the data used is shown in the figure below. Figure 7 The descriptive statistics are shown in Table 1.

[0100] Table 1 Descriptive statistics

[0101] average value 339.5 Standard Deviation 535.1 Minimum 0 25％ 0 50％ 1.68 75％ 515.9 Maximum 2383.2

[0102] In the experiments of the present invention, the dataset is divided into a training set and a test set. The last 20% of the dataset (approximately three months, from September 2016 to December 2016) is used as a test set. The performance of the prediction is evaluated by the mean absolute error (MAE), the standardized mean absolute error (nMAE), and the standardized root mean square error (nRMSE). They are defined as follows (where y represents the true value of PV, Represents the predicted value of PV):

[0103]

[0104] The final experimental results are as follows Figure 8 , as shown in Table 2.

[0105] Table 2 Prediction results of this method, NbeatsX model and continuous model

[0106] Methods NbeatsX Continuous Model MAE 53.7 77.6 84.2 nMAE 0.028 0.038 0.041 nRMSE 0.051 0.088 0.100

[0107] from Figure 8 As can be seen from Table 2, the proposed method can effectively reduce the error of photovoltaic prediction: compared with other methods, the proposed method reduces MAE by at least 21.1, reduces nMAE by at least 1%, and reduces nRMSE by at least 3.7%.

[0108] Table 3 Prediction results of this method and those without topological features

[0109] Methods Remove topological features MAE 53.7 66.7 nMAE 0.028 0.032 nRMSE 0.051 0.068

[0110] from Fig. 9 As can be seen from Table 3, topological features can effectively reduce the error of photovoltaic prediction: compared with not using topological features, using topological features reduces MAE by 13%, nMAE by 0.4%, and nRMSE by 2.2%. In addition, from Fig. 9 It is not difficult to see that for some atypical clear sky days (such as the third, fourth, seventh and eighth days), using only the statistical characteristics of photovoltaics and loads and historical weather will produce deviations in the time scale (peak lag) or in the amplitude scale (peak overestimation). At this time, topological features can effectively improve the prediction accuracy.

[0111] Embodiment 3 is an embodiment of the present invention, comprising a data-driven photovoltaic power generation estimation system based on the Betty curve, specifically:

[0112] The feature calculation module collects the first object data, performs a first reconstruction calculation, and obtains a first feature of the first object. The module performs a statistical calculation based on the first object data and obtains a second feature of the first object.

[0113] The conversion module performs a first conversion task on the specific information of the first object data to obtain a first conversion task result.

[0114] The prediction module concatenates the first feature, the second feature and the result of the first conversion task to form a feature space, inputs the feature space into the first prediction model, and outputs the prediction result of the first object.

[0115] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A data-driven photovoltaic power generation estimation method based on Betty curve, characterized in that: include: Collecting first object data, performing a first reconstruction calculation, and obtaining a first feature of the first object; Performing statistical calculations based on the first object data to obtain a second feature of the first object; Perform a first conversion task on the specific information of the first object data to obtain a first conversion task result; The first feature, the second feature and the result of the first conversion task are concatenated to form a feature space, which is input into the first prediction model to output the prediction result of the first object.

2. The data-driven photovoltaic power generation estimation method based on the Betty curve according to claim 1, characterized in that: The collecting of the first object data and performing the first analysis includes collecting time-related data of the first object, converting the time series into point cloud data, and filtering and calculating to obtain the first feature.

3. The data-driven photovoltaic power generation estimation method based on the Betty curve according to claim 2, characterized in that: The performing statistical calculation according to the first object data to obtain the second feature of the first object includes extracting a plurality of statistical features according to the first object data.

4. The data-driven photovoltaic power generation estimation method based on the Betty curve according to claim 3, characterized in that: The performing the first conversion task on the specific information of the first object data includes encoding the specific information of the first object data and converting the specific information into a first conversion task result.

5. The data-driven photovoltaic power generation estimation method based on the Betty curve according to claim 4, characterized in that: The first object includes but is not limited to photovoltaic power generation equipment, and the first object data includes a time series based on historical photovoltaic power generation power and a time series based on historical load; The first reconstruction calculation includes but is not limited to phase space reconstruction using Takens embedding theory; The first feature is a topological feature, and the second feature is a statistical feature; The first conversion task is to convert the date feature into a category feature, and the result of the first conversion task is a category feature; The first predictive model includes, but is not limited to, a machine learning regressor.

6. The data-driven photovoltaic power generation estimation method based on the Betty curve according to claim 5, characterized in that: The performing of the first reconstruction calculation to obtain the first feature of the first object includes using Takens embedding theory to perform phase space reconstruction to convert the time series into point cloud data; X = {x0, ···, x n-1 } is a one-dimensional time series. Given the delay parameter τ and the dimension parameter d, the Takens phase space reconstruction point corresponding to time t is defined as: z t :=(x t ,x t+τ ,···,x t+(d-1)τ )∈R d Vietoris-Rips filtering is applied to the point cloud data to generate the filter complex and calculate the persistence graph, from which the 0-dimensional and 1-dimensional Betty curves are extracted as the topological features of the time series as the first feature.

7. The data-driven photovoltaic power generation estimation method based on the Betty curve according to claim 6, characterized in that: The performing statistical calculation according to the first object data to obtain the second feature of the first object includes calculating an absolute energy feature, a mean value feature, a standard deviation feature, a skewness feature, a kurtosis feature, an absolute average change feature, an average second-order difference center feature, and a median feature; The first conversion task is performed, and the first conversion task result obtained includes: for date information, it is converted into a category feature through one-hot encoding. A date has 12 possible month attributes, 31 possible day attributes, and 7 possible week attributes. The one-hot encoding is vectorized to obtain a 50-dimensional category feature vector.

8. A dynamic load capacity management system for oil-immersed power transformers using the method according to any one of claims 1 to 7, characterized in that: A feature calculation module collects first object data, performs a first reconstruction calculation, and obtains a first feature of the first object; Performing statistical calculations based on the first object data to obtain a second feature of the first object; A conversion module, performing a first conversion task on specific information of the first object data to obtain a first conversion task result; The prediction module concatenates the first feature, the second feature and the result of the first conversion task to form a feature space, inputs the feature space into the first prediction model, and outputs the prediction result of the first object.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images