Intermittent energy output modeling prediction method and system based on Gaussian process

Abstract

The invention relates to an intermittent energy output modeling prediction method and system based on a Gaussian process, and the method comprises the steps: carrying out the modeling of intermittent energy output based on a Bernstein polynomial, calculating a mean value and a covariance function of the intermittent energy output, and constructing a Gaussian process prior form; obtaining a historical observation set of intermittent energy output, optimizing a hyper-parameter in a Gaussian process prior form through maximum likelihood estimation according to a noise observation point of each track sample in the historical observation set on discrete time, and solving a covariance matrix of a Bernstein coefficient; and calculating a mean value and a covariance matrix of the posterior distribution of the Bernstein coefficient according to the observed intermittent energy output at multiple moments, thereby obtaining the joint distribution of the intermittent energy output in the future test time. Compared with the prior art, the method has the advantages that the latest observed data can be effectively utilized to predict the generating capacity of the next time period, and energy scheduling and optimization with quick response are realized.

Description

Technical Field

[0001] This invention relates to the field of intermittent energy output prediction, and in particular to a method and system for intermittent energy output modeling and prediction based on Gaussian processes. Background Technology

[0002] During power grid emergency recovery, the randomness of factors such as intermittent power output, equipment repair time, and mobile emergency resource allocation time has a significant impact on the formulation and implementation of the system recovery plan. Traditional static modeling methods are difficult to address this issue. To address this, the invention disclosed in CN110795841A presents a mathematical modeling method for the uncertainty of intermittent energy output, comprising: Step 1: Expectation-Maximization Algorithm Initialization: Solving for the initial values ​​of model parameters using a hard clustering algorithm; Step 2: Expectation Calculation: Performing probability calculations for the latent variables of each data point; Step 3: Maximization: Deriving the iterative formula for the Gaussian mixture model parameters with weighted data; Step 4: Repeating Steps 2-3 until convergence.

[0003] This scheme models the uncertainty of intermittent energy output using a Gaussian mixture model based on weighted data and the expectation-maximization algorithm. However, it calculates model parameters based on the probability of prior distribution parameters, resulting in large data fluctuations. Furthermore, the parameter update process requires repeated iterations of the entire scheme, leading to a large amount of data computation. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the existing technology by providing a method and system for modeling and predicting intermittent energy output based on Gaussian processes, which can effectively use the latest observed data to predict the power generation in the next time period, and achieve rapid response energy dispatch and optimization.

[0005] The objective of this invention can be achieved through the following technical solutions: A method for modeling and predicting intermittent energy output based on Gaussian processes includes the following steps: Based on Bernstein polynomials, intermittent energy output is modeled, and the Gaussian process prior form of intermittent energy output is constructed based on the mean and covariance function of the intermittent energy output obtained from the modeling. Obtain a historical observation set of intermittent energy output. Based on the noise observation points of each trajectory sample in the historical observation set at discrete time, optimize the hyperparameters in the prior form of the Gaussian process through maximum likelihood estimation, and solve the covariance matrix of the Bernstein coefficients. Based on the intermittent energy output observed at multiple times, the mean and covariance matrix of the posterior distribution of the Bernstein coefficients are calculated by solving the hyperparameters and the covariance matrix of the Bernstein coefficients, thereby obtaining the joint distribution of intermittent energy output during future test times.

[0006] Furthermore, the expression for the intermittent energy output obtained by modeling the intermittent energy output based on Bernstein polynomials is as follows: In the formula, Modeling results for intermittent energy output, It is 4 is a random variable, representing the Bernstein coefficient; For a reduced Bernstein function space, For linear transformation, for Bernstein function space of order, , To divide the time interval Divided One interval, For index parameters, for Bernstein basis functions of order.

[0007] Furthermore, the a priori form of the Gaussian process with intermittent energy output is: In the formula, for The mean, for The covariance function, and For different points in time; The The specific expression for calculating the covariance function is as follows: In the formula, for The specific function value of the covariance function. and These are the variances of the periodic and smooth components, respectively. and For the feature length scale, Indicates the period.

[0008] Furthermore, the expression for the noisy observation points of each trajectory sample in the historical observation set in discrete time is: In the formula, For all trajectory samples of the power output process dimensional vector, A historical observation set for intermittent energy output. This represents the number of noisy observation points in the trajectory sample. In the last trajectory sample of the historical observation set for intermittent energy output The output value at each noise observation point For the intermittent energy output model in the last trajectory sample of the historical observation set The values ​​at each noise observation point This represents the output vector of the intermittent energy output model. For noise vectors, In the last trajectory sample of the historical observation set for intermittent energy output Noise vector at each noise observation point Each element in the matrix is ​​independently and identically distributed with zero mean and variance. Gaussian noise.

[0009] Furthermore, the process of optimizing the hyperparameters in the prior form of the Gaussian process through maximum likelihood estimation specifically involves: make Represents the mean vector, The covariance function representing the power output process calculated at the noise observation point is expressed as follows: In the formula, For covariance function The relevant hyperparameter vectors, In the last trajectory sample of the historical observation set for intermittent energy output The model prediction mean for each noise observation point; Assumption 1, of which The mean is 1. If the hyperparameters in the prior form of the Gaussian process are a unit vector, then the expression for the optimization objective, which optimizes the hyperparameters through maximum likelihood estimation, is: In the formula, the optimization parameters are: , and .

[0010] Furthermore, the process of solving the covariance matrix of the Bernstein coefficients includes: Within the time frame Select an evaluation time set In this evaluation time set Calculate the covariance function at each time point in the time series. , construct Linear equations with unknown components: In the formula, for The optimal estimate, , ; By selecting the evaluation time point, the above... The linear equation with unknown components is of full rank, thus the covariance matrix of the Bernstein coefficients can be obtained. .

[0011] Furthermore, the specific process for solving the joint distribution of intermittent energy output during the future test period is as follows: On the day the intermittent energy output is predicted, based on the observed... The distribution of intermittent energy output at a given moment is used to predict the distribution of intermittent energy output at future moments, specifically: Record the observed At that moment, the power output was Calculate the mean of the posterior distribution of the Bernstein coefficient. Covariance Matrix Based on the modeling model of intermittent energy output, the future test time can be calculated. Joint distribution of intermittent energy output .

[0012] Furthermore, the mean of the posterior distribution of the Bernstein coefficients Covariance Matrix The calculation expression is: In the formula, .

[0013] Furthermore, the intermittent energy source is wind power and / or photovoltaic energy.

[0014] This invention also provides an intermittent energy output modeling and prediction system for implementing the above-described intermittent energy output modeling and prediction method based on Gaussian processes, comprising: The intermittent energy output modeling module is used to model intermittent energy output based on Bernstein polynomials, and to construct the Gaussian process prior form of intermittent energy output based on the mean and covariance function of the intermittent energy output obtained from the modeling. The parameter estimation module is used to obtain the historical observation set of intermittent energy output. Based on the noise observation points of each trajectory sample in the historical observation set in discrete time, it optimizes the hyperparameters in the prior form of the Gaussian process through maximum likelihood estimation and solves the covariance matrix of the Bernstein coefficients. The measurement module is used to calculate the mean and covariance matrix of the posterior distribution of the Bernstein coefficients based on the intermittent energy output at multiple observed times, by solving the hyperparameters and the covariance matrix of the Bernstein coefficients, thereby obtaining the joint distribution of intermittent energy output during future test times.

[0015] Compared with the prior art, the present invention has the following advantages: (1) This invention models intermittent energy output based on Bernstein polynomials and constructs a prior form of intermittent energy output Gaussian process. Based on this, the hyperparameters and covariance matrix of the Bernstein coefficients in the prior form of Gaussian process are optimized by maximum likelihood estimation according to the historical observation set of intermittent energy output. On the prediction day, the mean and covariance matrix of the posterior distribution of the Bernstein coefficients can be solved based on the observed intermittent energy output to obtain the joint distribution of intermittent energy output in the future test time. This scheme effectively captures the stochastic characteristics and temporal correlations in intermittent power output data through Gaussian processes. It not only provides point predictions but also generates associated uncertainty intervals, giving it a significant advantage in handling highly volatile energy sources such as solar and wind power. On the other hand, based on Bernstein polynomial modeling, it implements a parameter self-learning mechanism, enabling various stochastic information to be continuously updated according to new data during operation. This allows for real-time optimization of the characterization of random variables, improving the accuracy and robustness of recovery scheduling decisions. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for modeling and predicting intermittent energy output based on a Gaussian process, provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the structure of an intermittent energy output modeling and prediction system based on Gaussian processes provided in an embodiment of the present invention; Figure 3This is a schematic block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0018] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0019] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0020] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship in which the product of this invention is usually placed during use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0021] It should be noted that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0022] Furthermore, terms such as "horizontal" and "vertical" do not imply that components must be absolutely horizontal or suspended, but rather that they can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal than "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0023] Example 1 During power grid emergency restoration, the randomness of factors such as intermittent power output, equipment repair time, and the deployment time of mobile emergency resources has a significant impact on the formulation and implementation of the system's restoration plan, which traditional static modeling methods struggle to address. To enhance the model's adaptability, this embodiment introduces uncertainty modeling and parameter self-learning mechanisms, enabling various random information to be continuously updated based on new data during operation. This allows for real-time optimization of the characterization of random variables, improving the accuracy and robustness of restoration scheduling decisions.

[0024] Intermittent power sources such as wind and solar power can be modeled using a Gaussian Process (GP). This model can effectively predict power generation for the next time period using the latest observed data, enabling rapid energy dispatch and optimization. The Gaussian Process effectively captures the stochastic characteristics and temporal correlations in intermittent power output data, providing not only point predictions but also generating associated uncertainty intervals, giving it a significant advantage in handling highly volatile energy sources such as solar and wind power.

[0025] like Figure 1 As shown, this embodiment provides a method for modeling and predicting intermittent energy output based on Gaussian processes, including the following steps: S1: Based on Bernstein polynomials, intermittent energy output is modeled, and the Gaussian process prior form of intermittent energy output is constructed based on the mean and covariance function of the intermittent energy output obtained from the modeling. S2: Obtain the historical observation set of intermittent energy output, optimize the hyperparameters in the prior form of the Gaussian process by maximum likelihood estimation based on the noise observation points of each trajectory sample in the historical observation set in discrete time, and solve the covariance matrix of the Bernstein coefficients. S3: Based on the intermittent energy output observed at multiple times, the mean and covariance matrix of the posterior distribution of the Bernstein coefficients are calculated by solving the hyperparameters and the covariance matrix of the Bernstein coefficients, thereby obtaining the joint distribution of intermittent energy output during future test times.

[0026] The following is a detailed description of each step: 1. Modeling process of intermittent energy output stochastic process in step S1 This model models intermittent energy output data as a continuous-time stochastic process and projects it onto a low-order function space composed of Bernstein polynomials, thereby ensuring continuity in the prediction time domain. The Bernstein basis functions are defined as follows: Assuming time interval Divided into interval Then the piecewise Bernstein basis functions can cover the entire interval. The Bernstein function space of order can be represented as ,in The specific form is as follows: In order to maintain throughout the entire time period To maintain continuity, use a linear transformation. Construct a reduced Bernstein function space (Z×P dimensional). Then the power output Modeled as: in It is 4D random variables, representing Bernstein coefficients, whose mean and covariance functions are expressed using . and express. The mean and covariance functions are respectively used as and express.

[0027] Bernstein polynomials are a set of elegant and unique polynomials, initially proposed by the Soviet mathematician Ernst Bernstein in 1912. They were primarily used to provide a constructive proof of the Weierstrass approximation theorem, which states that any continuous function on a closed interval can be approximated by polynomials with arbitrary precision. Their core value lies in their excellent properties such as nonnegativity, unit factorization (i.e., the sum of the function values ​​of all polynomials at any point in the domain [0,1] is always 1), and recursive computation. These properties make them highly stable in numerical computation and have a clear geometric meaning. For this reason, Bernstein polynomials form the mathematical foundation of Bézier curves, generating smooth curves through linear combinations of control points.

[0028] 2. Step S2: Maximum likelihood fitting process based on prior distribution of historical data Suppose that a Gaussian process prior is constructed based on a reduced Bernstein function space. To save computation time, we first assume its covariance function. For a specific form containing periodic and smooth components, it is represented as follows: in, and These are the variances of the periodic and smooth components, respectively. and For the feature length scale, The period is indicated and is usually set to 24 hours to capture daily periodicity.

[0029] Subsequently, the prior distribution of the Gaussian process is estimated based on past observations of the power output process. Let... Time range Historical power output observation sets during comparable periods (e.g., observation sets for specific working days). (Note: The original text contains some formatting errors and inconsistencies. A more accurate translation would require the full context.) A trajectory sample representing the power output process. ,and This indicates that the trajectory is in discrete time. Noise observation points on the surface Represents all trajectory samples dimensional vector, that is: in , Is the model in The value of each sampling point, It is a noise vector, where each element is independently and identically distributed with zero mean and variance. Gaussian noise.

[0030] make Represents the mean vector. The covariance function representing the power output process calculated at the sampling points is: in It is a hyperparameter vector associated with the covariance function k(t, t'). Here, we assume... 1, of which The mean is 1. A unit vector of dimension. Then, given past observation data... In the case of optimizing the hyperparameters of a Gaussian process using maximum likelihood estimation, the optimization objective can be expressed as: The optimization parameters are: .

[0031] To calculate the covariance matrix of the Bernstein coefficients The components, within the time range Select an evaluation time set And at these time points, the covariance function k(t, t') is calculated, thus forming a set of... Linear equations with unknown components: in, yes The optimal estimate, , By appropriately selecting the evaluation time point to ensure the linear equation system reaches full rank, the covariance matrix of the Bernstein coefficients can be obtained. .

[0032] 3. Step S3: Solving the posterior distribution based on the observed data After solving the maximum likelihood estimation problem and constructing a prior Gaussian process model, on the day the power output is predicted, it can be based on the observed... The power output at a given moment is used to predict the power output distribution at future moments. Record the observed power output at each moment. At that moment, the power output was The mean and covariance matrix of the posterior distribution of the Bernstein coefficients can be calculated by the following formula: in Based on this, we can then use the relational formula... Calculate the future test time Output value The joint distribution of .

[0033] This scheme models intermittent energy output based on Bernstein polynomials and constructs a Gaussian process prior form for intermittent energy output. Based on this, the hyperparameters and covariance matrix of the Bernstein coefficients in the Gaussian process prior form are optimized through maximum likelihood estimation using the historical observation set of intermittent energy output. On the prediction day, the mean and covariance matrix of the posterior distribution of the Bernstein coefficients can be solved based on the observed intermittent energy output, thus obtaining the joint distribution of intermittent energy output during the future test period. This scheme effectively captures the stochastic characteristics and temporal correlations in intermittent power output data through Gaussian processes. It not only provides point predictions but also generates associated uncertainty intervals, giving it a significant advantage in handling highly volatile energy sources such as solar and wind power. On the other hand, based on Bernstein polynomial modeling, it implements a parameter self-learning mechanism, enabling various stochastic information to be continuously updated according to new data during operation. This allows for real-time optimization of the characterization of random variables, improving the accuracy and robustness of recovery scheduling decisions.

[0034] Example 2 like Figure 2 As shown, this embodiment provides an intermittent energy output modeling and prediction system that implements the intermittent energy output modeling and prediction method based on Gaussian processes as described in Embodiment 1, comprising: The intermittent energy output modeling module is used to model intermittent energy output based on Bernstein polynomials, and to construct the Gaussian process prior form of intermittent energy output based on the mean and covariance function of the intermittent energy output obtained from the modeling. The parameter estimation module is used to obtain the historical observation set of intermittent energy output. Based on the noise observation points of each trajectory sample in the historical observation set in discrete time, it optimizes the hyperparameters in the prior form of the Gaussian process through maximum likelihood estimation and solves the covariance matrix of the Bernstein coefficients. The measurement module is used to calculate the mean and covariance matrix of the posterior distribution of the Bernstein coefficients based on the intermittent energy output at multiple observed times, by solving the hyperparameters and the covariance matrix of the Bernstein coefficients, thereby obtaining the joint distribution of intermittent energy output during future test times.

[0035] This embodiment also provides another intermittent energy output modeling and prediction system, including a memory and a processor. The memory stores a computer program, and the processor calls the computer program to execute the steps of the intermittent energy output modeling and prediction method based on Gaussian process as described in Embodiment 1.

[0036] This embodiment also provides a computer-readable storage medium storing a computer program, which is executed by a processor as described in Embodiment 1, a method for modeling and predicting intermittent energy output based on a Gaussian process.

[0037] Figure 3 A schematic block diagram of an electronic device that can be used to implement embodiments of the present disclosure is shown. Figure 3 As shown, the electronic device of the present invention includes a central processing unit (CPU), which can perform various appropriate actions and processes according to computer program instructions stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM can also store various programs and data required for device operation. The CPU, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0038] Multiple components in the device are connected to the I / O interface, including: input units such as keyboards and mice; output units such as various types of displays and speakers; storage units such as disks and optical discs; and communication units such as network interface cards (NICs), modems, and wireless transceivers. The communication unit allows the device to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0039] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0040] The computer program code used to implement the methods of the present invention can be written in any combination of one or more programming languages. This computer program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor or controller, the computer program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer program code can be executed entirely on the machine, partially on the machine, as a standalone software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0041] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium can be a machine-readable signal medium or a machine-readable storage medium. A computer-readable storage medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0042] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for modeling and predicting intermittent energy output based on Gaussian processes, characterized in that, Includes the following steps: Based on Bernstein polynomials, intermittent energy output is modeled, and the Gaussian process prior form of intermittent energy output is constructed based on the mean and covariance function of the intermittent energy output obtained from the modeling. Obtain a historical observation set of intermittent energy output. Based on the noise observation points of each trajectory sample in the historical observation set at discrete time, optimize the hyperparameters in the prior form of the Gaussian process through maximum likelihood estimation, and solve the covariance matrix of the Bernstein coefficients. Based on the intermittent energy output observed at multiple times, the mean and covariance matrix of the posterior distribution of the Bernstein coefficients are calculated by solving the hyperparameters and the covariance matrix of the Bernstein coefficients, thereby obtaining the joint distribution of intermittent energy output during future test times.

2. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 1, characterized in that, The expression for intermittent energy output obtained by modeling intermittent energy output based on Bernstein polynomials is as follows: In the formula, Modeling results for intermittent energy output, It is 4 is a random variable, representing the Bernstein coefficient; For a reduced Bernstein function space, For linear transformation, for Bernstein function space of order, , To divide the time interval Divided One interval, For index parameters, for Bernstein basis functions of order.

3. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 2, characterized in that, The a priori form of the intermittent energy output Gaussian process is: In the formula, for The mean, for The covariance function, and For different points in time; The The specific expression for calculating the covariance function is as follows: In the formula, for The specific function value of the covariance function. and These are the variances of the periodic and smooth components, respectively. and For the feature length scale, Indicates the period.

4. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 3, characterized in that, The expression for the noise observation points of each trajectory sample in the historical observation set in discrete time is: In the formula, For all trajectory samples of the power output process dimensional vector, A historical observation set for intermittent energy output. This represents the number of noisy observation points in the trajectory sample. In the last trajectory sample of the historical observation set for intermittent energy output The output value at each noise observation point For the intermittent energy output model in the last trajectory sample of the historical observation set The values ​​at each noise observation point This represents the output vector of the intermittent energy output model. For noise vectors, In the last trajectory sample of the historical observation set for intermittent energy output Noise vector at each noise observation point Each element in the matrix is ​​independently and identically distributed with zero mean and variance. Gaussian noise.

5. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 4, characterized in that, The process of optimizing the hyperparameters in the prior form of the Gaussian process through maximum likelihood estimation is specifically as follows: make Represents the mean vector, The covariance function representing the power output process calculated at the noise observation point is expressed as follows: In the formula, For covariance function The relevant hyperparameter vectors, In the last trajectory sample of the historical observation set for intermittent energy output The model prediction mean for each noise observation point; Assumption 1, of which The mean is 1. If the hyperparameters in the prior form of the Gaussian process are a unit vector, then the expression for the optimization objective, which optimizes the hyperparameters through maximum likelihood estimation, is: In the formula, the optimization parameters are: , and .

6. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 5, characterized in that, The process of solving the covariance matrix of the Bernstein coefficients includes: Within the time frame Select an evaluation time set In this evaluation time set Calculate the covariance function at each time point in the time series. , construct Linear equations with unknown components: In the formula, for The optimal estimate, , ; By selecting the evaluation time point, the above... The linear equation with unknown components is of full rank, thus the covariance matrix of the Bernstein coefficients can be obtained. .

7. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 6, characterized in that, The specific process for solving the joint distribution of intermittent energy output during the future test period is as follows: On the day the intermittent energy output is predicted, based on the observed... The distribution of intermittent energy output at a given moment is used to predict the distribution of intermittent energy output at future moments, specifically: Record the observed At that moment, the power output was Calculate the mean of the posterior distribution of the Bernstein coefficient. Covariance Matrix Based on the modeling model of intermittent energy output, the future test time can be calculated. Joint distribution of intermittent energy output .

8. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 7, characterized in that, The mean of the posterior distribution of the Bernstein coefficients Covariance Matrix The calculation expression is: In the formula, .

9. The method for modeling and predicting intermittent energy output based on Gaussian processes according to claim 1, characterized in that, The intermittent energy source is wind power and / or photovoltaic energy.

10. An intermittent energy output modeling and prediction system that implements the intermittent energy output modeling and prediction method based on Gaussian processes as described in any one of claims 1-9, characterized in that, include: The intermittent energy output modeling module is used to model intermittent energy output based on Bernstein polynomials, and to construct the Gaussian process prior form of intermittent energy output based on the mean and covariance function of the intermittent energy output obtained from the modeling. The parameter estimation module is used to obtain the historical observation set of intermittent energy output. Based on the noise observation points of each trajectory sample in the historical observation set in discrete time, it optimizes the hyperparameters in the prior form of the Gaussian process through maximum likelihood estimation and solves the covariance matrix of the Bernstein coefficients. The measurement module is used to calculate the mean and covariance matrix of the posterior distribution of the Bernstein coefficients based on the intermittent energy output at multiple observed times, by solving the hyperparameters and the covariance matrix of the Bernstein coefficients, thereby obtaining the joint distribution of intermittent energy output during future test times.

Images