CN115934691A - Method and device for determining short-term photovoltaic power - Google Patents

Abstract

The present invention proposes a method for determining short-term photovoltaic power, wherein the method includes: obtaining historical multi-source data of photovoltaic stations, and performing data cleaning on the historical multi-source data; decomposing and processing the cleaned historical multi-source data, obtaining The historical decomposition data corresponding to the historical multi-source data; the pre-built meta-extreme learning machine model is trained according to the historical decomposition data, and the trained meta-extreme learning machine model is obtained; the real-time operation data and real-time weather during the online operation of the photovoltaic array are obtained Data, input real-time operating data and real-time meteorological data into the trained Yuan extreme learning machine model to obtain short-term photovoltaic power data. Based on this, the method can effectively decompose the signal data and retain the original physical information for non-stationary signal data, and solve the problem of photovoltaic power prediction. Maintenance efficiency of photovoltaic unit equipment.

Description

Method and device for determining short-term photovoltaic power

technical field

The present invention relates to the technical field of artificial intelligence, in particular to a method and device for determining short-term photovoltaic power.

Background technique

As an important basis for human survival and development, the development and utilization of energy has been widely valued. The consumption rate of traditional fossil energy has increased significantly, causing irreversible environmental pollution, which in turn leads to extreme climate disasters. With social progress and development and economic construction entering a new stage, my country has put forward a new plan for the future energy development strategy, and determined the main theme of "green and low carbon". Due to its advantages of cleanliness, safety and high efficiency, solar energy has become the fastest growing renewable energy in recent years. With the increase of total photovoltaic installed capacity and large-scale grid-connected operation, it is difficult to effectively utilize solar energy, and there is an obvious problem of light abandonment.

In related technologies, the photovoltaic power prediction method is mainly based on data mining, and models in other fields are used to improve prediction accuracy.

However, in related technologies, if other domain models are applied to the field of photovoltaic power generation without selection and improvement, ideal results cannot be achieved. For short-term photovoltaic power forecasting, the difficulty lies in the fact that the original time series data is nonlinear and non-stationary. Therefore, there is an urgent need for a short-term photovoltaic power determination method to efficiently decompose the signal data and retain the original physical information for non-stationary signal data, so as to solve the validity problem of photovoltaic power prediction.

Contents of the invention

The present invention aims to solve one of the technical problems in the related art at least to a certain extent.

Therefore, the first purpose of the present invention is to propose a method for determining short-term photovoltaic power, which efficiently decomposes signal data and retains the original physical information for non-stationary signal data, so as to solve the problem of the validity of photovoltaic power prediction.

The second object of the present invention is to propose a device for determining short-term photovoltaic power.

A third object of the present invention is to propose a computer device.

A fourth object of the present invention is to provide a non-transitory computer-readable storage medium.

In order to achieve the above purpose, the embodiment of the first aspect of the present invention proposes a short-term photovoltaic power determination method, including:

Obtain the historical multi-source data of the photovoltaic field station, and perform data cleaning on the historical multi-source data;

Decomposing the cleaned historical multi-source data to obtain historical decomposition data corresponding to the historical multi-source data;

Training the pre-built meta-extreme learning machine model according to the historical decomposition data to obtain the trained meta-extreme learning machine model;

Obtain real-time operating data and real-time meteorological data during the online operation of the photovoltaic array, and input the real-time operating data and real-time meteorological data into the trained element extreme learning machine model to obtain short-term photovoltaic power data.

Optionally, in the embodiment of the present invention, the decomposing the cleaned historical multi-source data includes:

Decomposing the cleaned historical multi-source data into oscillating components by means of swarm hunting;

determining position information and irradiance information according to the oscillating component and the predefined driving force and cohesion;

Constructing the relationship between the location information and the decomposition data, and solving the group decomposition parameters;

According to preset conditions, an optimal solution of the group decomposition parameters and the number of groups is determined to obtain the decomposition data.

Optionally, in the embodiment of the present invention, the definition of the driving force is:

F _dr (n,i)=P _prey (n)-P _i (n-1)

Among them, F _dr (n,i) is the driving force, i is the component, and n is the number of steps. The hunting position information is defined as P _prey , and P _i (n-1) is the position information of component i at step n-1.

Optionally, in the embodiment of the present invention, the cohesion is defined as:

where F _coh (n,i) is the cohesion force, d and d _cr are the distance between components and the critical distance, respectively, M is the number of groups, P _i [n-1] is the position of component i in n-1 steps Information, P _j [n-1] is the position information of component j at n-1 steps.

Optionally, in the embodiment of the present invention, the training of the pre-built meta-extreme learning machine model according to the historical decomposition data to obtain the trained meta-extreme learning machine model includes:

Fusing the decomposed data and meteorological data, and dividing the fused decomposed data and meteorological data into multiple sub-data sequences;

The meta-extreme learning machine model is trained according to the plurality of sub-data series to obtain a trained meta-extreme learning machine model.

In order to achieve the above purpose, the embodiment of the second aspect of the present invention proposes a device for determining short-term photovoltaic power, including:

The first acquisition module is used to acquire the historical multi-source data of the photovoltaic field station, and perform data cleaning on the historical multi-source data;

A decomposition module, configured to decompose the cleaned historical multi-source data, and obtain historical decomposition data corresponding to the historical multi-source data;

A training module, configured to train a pre-built meta-extreme learning machine model according to the historical decomposition data, to obtain a trained meta-extreme learning machine model;

The second acquisition module is used to acquire real-time operating data and real-time meteorological data during the online operation of the photovoltaic array, and input the real-time operating data and real-time meteorological data into the trained element extreme learning machine model to obtain short-term photovoltaic power data.

Optionally, in the embodiment of the present invention, the decomposition module is also used for:

Decomposing the cleaned historical multi-source data into oscillating components by means of swarm hunting;

determining position information and irradiance information according to the oscillating component and the predefined driving force and cohesion;

Constructing the relationship between the location information and the decomposition data, and solving the group decomposition parameters;

According to preset conditions, an optimal solution of the group decomposition parameters and the number of groups is determined to obtain the decomposition data.

Optionally, in the embodiment of the present invention, the training module is also used for:

Fusing the decomposed data and meteorological data, and dividing the fused decomposed data and meteorological data into multiple sub-data sequences;

The meta-extreme learning machine model is trained according to the plurality of sub-data series to obtain a trained meta-extreme learning machine model.

To sum up, the method and device for determining the short-term photovoltaic power provided by the present invention first obtain the historical multi-source data of the photovoltaic field and clean the historical multi-source data, and then clean the historical multi-source data Perform decomposition processing to obtain historical decomposition data, train the pre-built meta-extreme learning machine model according to the historical decomposition data, obtain the trained meta-extreme learning machine model, and finally obtain real-time operating data and real-time weather during the online operation of the photovoltaic array Data, input real-time operating data and real-time meteorological data into the trained Yuan extreme learning machine model to obtain short-term photovoltaic power data. Based on this, the method can effectively decompose the signal data and retain the original physical information for non-stationary signal data, and solve the problem of photovoltaic power prediction based on the fusion model of the group decomposition algorithm and the meta-extreme learning machine. Power generation system operation strategy, photovoltaic site selection scheme, and improving the efficiency of photovoltaic unit equipment maintenance.

To achieve the above-mentioned purpose, the embodiment of the third aspect of the present invention proposes a computer device, including a memory, a processor, and a computer program stored in the memory and operable on the processor, and the processor executes the computer program. When the computer program is described, the method described in the embodiment of the first aspect of the present invention is realized.

In order to achieve the above purpose, the embodiment of the fourth aspect of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the computer program implemented in the embodiment of the first aspect of the present invention described method.

Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.

Description of drawings

The above and/or additional aspects and advantages of the present invention will become apparent and easy to understand from the following description of the embodiments in conjunction with the accompanying drawings, wherein:

FIG. 1 is a schematic flowchart of a method for determining short-term photovoltaic power provided by an embodiment of the present invention;

Fig. 2 is a schematic structural diagram of a device for determining short-term photovoltaic power provided by an embodiment of the present invention.

Detailed ways

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.

A method and device for determining short-term photovoltaic power according to an embodiment of the present invention will be described below with reference to the accompanying drawings.

Fig. 1 is a schematic flowchart of a method for determining short-term photovoltaic power provided by an embodiment of the present invention.

As shown in Figure 1, the determination method of short-term photovoltaic power includes the following steps:

Step S10: Obtain the historical multi-source data of the photovoltaic field station, and perform data cleaning on the historical multi-source data.

Among them, in the embodiment of the present invention, data cleaning is performed on historical multi-source data, including:

Unify the format of multi-source data of each photovoltaic field station;

Normalize multi-source data.

Step S20: Decompose the cleaned historical multi-source data, and obtain historical decomposition data corresponding to the historical multi-source data.

Among them, in the embodiment of the present invention, decomposing and processing the cleaned historical multi-source data includes:

Decompose the cleaned historical multi-source data into oscillating components through group hunting method;

Determining position information and irradiance information according to the oscillation component and the pre-defined driving force and cohesion;

Construct the relationship between location information and decomposition data, and solve the group decomposition parameters;

According to the preset conditions, the optimal solution of the group decomposition parameters and the number of groups is determined, and the decomposition data are obtained.

And, in the embodiment of the present invention, the definition of driving force is:

F _dr (n,i)=P _prey (n)-P _i (n-1)

Among them, F _dr (n,i) is the driving force, i is the component, and n is the number of steps. The hunting position information is defined as P _prey , and P _i (n-1) is the position information of component i at step n-1.

And, in the embodiment of the present invention, the cohesion is defined as:

where F _coh (n,i) is the cohesion, d and d _cr are the distance between components and the critical distance, respectively, M is the number of groups, P _i [n-1] is the position of component i in n-1 steps Information, P _j [n-1] is the position information of component j at n-1 steps.

Further, in the embodiment of the present invention, the relationship between location information and decomposition data is constructed, and group decomposition parameters are solved, including:

P _i [n]=P _i [n-1]+δ(IR _i [n])

Among them, IR is the irradiance information, IR _i [n-1] is the irradiance information of component i at step n-1, and δ is one of the important parameters of group decomposition, which determines the adaptability of the group. Based on this, the relationship between the decomposed data y[n] and the position information can be constructed to solve the parameter δ:

where β is a parameter that affects the number of components. In order to determine the value of these parameters from the input data, the following conditions need to be met:

|Y _{δ, M} [k]| and |S[k]| represent the amplitudes of the original sequences of Y _{δ, M} [k] and S[k] after discrete Fourier transform, respectively. Y _{δ, M} [n] is a SWF represented by the two parameters δ, M, and S[n] is a single-component signal including non-stationary.

By finding the optimal solution of δ and M, the decomposed data Y _δ,M [n] is obtained.

Step S3: Train the pre-built meta-extreme learning machine model according to the historical decomposition data to obtain the trained meta-extreme learning machine model.

Among them, in the embodiment of the present invention, the pre-built meta-extreme learning machine model is trained according to the historical decomposition data, and the trained meta-extreme learning machine model is obtained, including:

Fuse the decomposed data and meteorological data, and divide the fused decomposed data and meteorological data into multiple sub-data sequences;

The meta-extreme learning machine model is trained according to a plurality of sub-data series to obtain a trained meta-extreme learning machine model.

Specifically, in the embodiment of the present invention, the meta-extreme learning machine model is trained according to multiple sub-data series to obtain a trained meta-extreme learning machine model, including:

The input subseries data trains a single hidden layer feed-forward network with an extreme learning machine:

是连接权重，ω_j是输入权重，b_j是偏差，N_h是隐层的数量。in,

is the connection weight, ω _j is the input weight, b _j is the bias, and N _h is the number of hidden layers.

In addition, except for the connection weight, other values are randomly selected, and the connection weight is calculated through the following steps.

Input N training data, the equation described above can be further expressed as a matrix vector H:

The output weights and the target for each output can be expressed as:

T=Hγ

The weight of the output connection is estimated by taking the inverse matrix of the Moore-Penrose H matrix. Each extreme learning machine in the meta-extreme learning machine network is obtained through sub-series data training, and the output connection weight is obtained by using all the data through extreme learning. Machine learning rules are obtained.

Step S4: Obtain real-time operation data and real-time weather data during the online operation of the photovoltaic array, and input the real-time operation data and real-time weather data into the trained Yuanji learning machine model to obtain short-term photovoltaic power data.

To sum up, the method for determining the short-term photovoltaic power provided by the present invention first obtains the historical multi-source data of the photovoltaic field and cleans the historical multi-source data, and then decomposes the cleaned historical multi-source data Process to obtain the historical decomposition data, train the pre-built meta-extreme learning machine model according to the historical decomposition data, obtain the trained meta-extreme learning machine model, and finally obtain the real-time operation data and real-time meteorological data during the online operation of the photovoltaic array, Input real-time operation data and real-time meteorological data into the trained Yuan extreme learning machine model to obtain short-term photovoltaic power data. Based on this, the method can effectively decompose the signal data and retain the original physical information for non-stationary signal data, and solve the problem of photovoltaic power prediction based on the fusion model of the group decomposition algorithm and the meta-extreme learning machine. Power generation system operation strategy, photovoltaic site selection scheme, and improving the efficiency of photovoltaic unit equipment maintenance.

Fig. 2 is a schematic structural diagram of a device for determining short-term photovoltaic power provided by an embodiment of the present invention.

As shown in Figure 2, the device for determining short-term photovoltaic power includes the following modules:

The first acquisition module 100 is used to acquire the historical multi-source data of the photovoltaic field station, and perform data cleaning on the historical multi-source data;

Decomposition module 200, configured to decompose the cleaned historical multi-source data, and obtain historical decomposed data corresponding to the historical multi-source data;

The training module 300 is used to train the pre-built meta-extreme learning machine model according to the historical decomposition data to obtain the trained meta-extreme learning machine model;

The second acquisition module 400 is used to acquire real-time operating data and real-time meteorological data during the online operation of the photovoltaic array, input the real-time operating data and real-time meteorological data into the trained Yuan extreme learning machine model, and obtain short-term photovoltaic power data.

Wherein, in the embodiment of the present invention, the decomposition module 200 is also used for:

Decompose the cleaned historical multi-source data into oscillating components through group hunting method;

Determining position information and irradiance information based on the oscillating components and predefined driving force and cohesion;

Construct the relationship between location information and decomposition data, and solve the group decomposition parameters;

According to the preset conditions, the optimal solution of the group decomposition parameters and the number of groups is determined, and the decomposition data are obtained.

And, in the embodiment of the present invention, the training module 300 is also used for:

Fuse the decomposed data and meteorological data, and divide the fused decomposed data and meteorological data into multiple sub-data sequences;

The meta-extreme learning machine model is trained according to a plurality of sub-data series to obtain a trained meta-extreme learning machine model.

It should be noted that the foregoing explanations of the embodiment of the method for determining based on short-term photovoltaic power are also applicable to the device of this embodiment, and reference may be made to relevant descriptions of the above embodiments, which will not be repeated here.

To sum up, the device for determining short-term photovoltaic power provided by the present invention first acquires the historical multi-source data of the photovoltaic field and cleans the historical multi-source data, and then decomposes the cleaned historical multi-source data Process to obtain the historical decomposition data, train the pre-built meta-extreme learning machine model according to the historical decomposition data, obtain the trained meta-extreme learning machine model, and finally obtain the real-time operation data and real-time meteorological data during the online operation of the photovoltaic array, Input real-time operation data and real-time meteorological data into the trained Yuan extreme learning machine model to obtain short-term photovoltaic power data. Based on this, the method can effectively decompose the signal data and retain the original physical information for non-stationary signal data, and solve the problem of photovoltaic power prediction based on the fusion model of the group decomposition algorithm and the meta-extreme learning machine. Power generation system operation strategy, photovoltaic site selection scheme, and improving the efficiency of photovoltaic unit equipment maintenance.

In the description of this specification, descriptions referring to the terms "one embodiment", "some embodiments", "example", "specific examples", or "some examples" mean that specific features described in connection with the embodiment or example , structure, material or characteristic is included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the described specific features, structures, materials or characteristics may be combined in any suitable manner in any one or more embodiments or examples. In addition, those skilled in the art can combine and combine different embodiments or examples and features of different embodiments or examples described in this specification without conflicting with each other.

In addition, the terms "first" and "second" are used for descriptive purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly specifying the quantity of indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include at least one of these features. In the description of the present invention, "plurality" means at least two, such as two, three, etc., unless otherwise specifically defined.

Any process or method descriptions in flowcharts or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing custom logical functions or steps of a process , and the scope of preferred embodiments of the invention includes alternative implementations in which functions may be performed out of the order shown or discussed, including substantially concurrently or in reverse order depending on the functions involved, which shall It is understood by those skilled in the art to which the embodiments of the present invention pertain.

The logic and/or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced listing of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium, For use with instruction execution systems, devices, or devices (such as computer-based systems, systems including processors, or other systems that can fetch instructions from instruction execution systems, devices, or devices and execute instructions), or in conjunction with these instruction execution systems, devices or equipment for use. For the purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use in or in conjunction with an instruction execution system, device, or device. More specific examples (non-exhaustive list) of computer-readable media include the following: electrical connection with one or more wires (electronic device), portable computer disk case (magnetic device), random access memory (RAM), Read Only Memory (ROM), Erasable and Editable Read Only Memory (EPROM or Flash Memory), Fiber Optic Devices, and Portable Compact Disc Read Only Memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program can be printed, since the program can be read, for example, by optically scanning the paper or other medium, followed by editing, interpretation or other suitable processing if necessary. processing to obtain the program electronically and store it in computer memory.

It should be understood that various parts of the present invention can be realized by hardware, software, firmware or their combination. In the embodiments described above, various steps or methods may be implemented by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented by any one or a combination of the following techniques known in the art: a discrete Logic circuits, ASICs with suitable combinational logic gates, Programmable Gate Arrays (PGA), Field Programmable Gate Arrays (FPGA), etc.

Those of ordinary skill in the art can understand that all or part of the steps carried by the methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. During execution, one or a combination of the steps of the method embodiments is included.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing module, each unit may exist separately physically, or two or more units may be integrated into one module. The above-mentioned integrated modules can be implemented in the form of hardware or in the form of software function modules. If the integrated modules are realized in the form of software function modules and sold or used as independent products, they can also be stored in a computer-readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, and the like. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention, those skilled in the art can make the above-mentioned The embodiments are subject to changes, modifications, substitutions and variations.

Claims (10)

1. A method for determining short-term photovoltaic power, comprising: Obtain the historical multi-source data of the photovoltaic field station, and perform data cleaning on the historical multi-source data; Decomposing the cleaned historical multi-source data to obtain historical decomposition data corresponding to the historical multi-source data; Training the pre-built meta-extreme learning machine model according to the historical decomposition data to obtain the trained meta-extreme learning machine model; Obtain real-time operating data and real-time meteorological data during the online operation of the photovoltaic array, and input the real-time operating data and real-time meteorological data into the trained element extreme learning machine model to obtain short-term photovoltaic power data. 2. The determination method according to claim 1, wherein said decomposing the cleaned historical multi-source data includes: Decomposing the cleaned historical multi-source data into oscillating components by means of swarm hunting; determining position information and irradiance information according to the oscillating component and the predefined driving force and cohesion; Constructing the relationship between the location information and the decomposition data, and solving the group decomposition parameters; According to preset conditions, an optimal solution of the group decomposition parameters and the number of groups is determined to obtain the decomposition data. 3. determination method as claimed in claim 2 is characterized in that, the definition of described driving force is: F _dr (n,i)=P _prey (n)-P _i (n-1) Among them, F _dr (n,i) is the driving force, i is the component, and n is the number of steps. The hunting position information is defined as P _prey , and P _i (n-1) is the position information of component i at step n-1. 4. determination method as claimed in claim 2, is characterized in that, the definition of described cohesion is:

where F _coh (n,i) is the cohesion, d and d _cr are the distance between components and the critical distance, respectively, M is the number of groups, P _i [n-1] is the position of component i in n-1 steps Information, P _j [n-1] is the position information of component j at n-1 steps. 5. The determination method according to any one of claims 2-4, wherein the pre-built meta-extreme learning machine model is trained according to the historical decomposition data to obtain the trained meta-extreme learning machine model ,include: Fusing the decomposed data and meteorological data, and dividing the fused decomposed data and meteorological data into multiple sub-data sequences; The meta-extreme learning machine model is trained according to the plurality of sub-data series to obtain a trained meta-extreme learning machine model. 6. A device for determining short-term photovoltaic power, comprising: The first acquisition module is used to acquire the historical multi-source data of the photovoltaic field station, and perform data cleaning on the historical multi-source data; A decomposition module, configured to decompose the cleaned historical multi-source data, and obtain historical decomposition data corresponding to the historical multi-source data; A training module, configured to train a pre-built meta-extreme learning machine model according to the historical decomposition data, to obtain a trained meta-extreme learning machine model; The second acquisition module is used to acquire real-time operating data and real-time meteorological data during the online operation of the photovoltaic array, and input the real-time operating data and real-time meteorological data into the trained element extreme learning machine model to obtain short-term photovoltaic power data. 7. The determining device according to claim 6, wherein the decomposition module is also used for: Decomposing the cleaned historical multi-source data into oscillating components by means of swarm hunting; determining position information and irradiance information according to the oscillating component and the predefined driving force and cohesion; Constructing the relationship between the location information and the decomposition data, and solving the group decomposition parameters; According to preset conditions, an optimal solution of the group decomposition parameters and the number of groups is determined to obtain the decomposition data. 8. The determining device according to claim 6, wherein the training module is also used for: Fusing the decomposed data and meteorological data, and dividing the fused decomposed data and meteorological data into multiple sub-data sequences; The meta-extreme learning machine model is trained according to the plurality of sub-data series to obtain a trained meta-extreme learning machine model. 9. A computer device, characterized by comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the computer program, the The method described in any one of claims 1-5. 10. A non-transitory computer-readable storage medium on which a computer program is stored, wherein the computer program implements the method according to any one of claims 1-5 when executed by a processor.

Images