A method for correcting predicted power data of a photovoltaic power generation system

Abstract

The application discloses a correction method for predicted power data of a photovoltaic power generation system, and the method comprises the following steps: constructing a power deviation sequence based on real-time power sequence data and original predicted power sequence data; performing time discretization processing on the power deviation sequence and constructing a one-dimensional heat conduction equation; iteratively solving by using a finite difference method to output a deviation heat field value sequence; fusing and correcting the original predicted power sequence data based on the deviation heat field value sequence, and determining the effective iterative result corresponding to each time node based on residual error; updating the data corresponding to the first time node in the original predicted power sequence data based on the deviation heat field value sequence; for the second time node, re-executing the heat conduction equation iterative solving process until a preset termination condition is met, and updating the data; and obtaining corrected power sequence data. The scheme can correct the predicted power data of the photovoltaic power generation system.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power generation technology, and in particular to a method for correcting predicted power data of a photovoltaic power generation system. Background Technology

[0002] Predicted power data for photovoltaic (PV) power generation systems is crucial for grid dispatch and the integration of new energy sources into the grid. Current technologies for predicting PV power data often rely on weather classification and statistical learning modeling, with energy storage devices compensating for errors. However, when meteorological conditions change rapidly or cloud movement causes continuous fluctuations in PV power, the prediction results obtained using this method are prone to over- or under-correction at local time points, thus affecting the accuracy of the predicted power data for PV power generation systems. Summary of the Invention

[0003] This invention provides a method for correcting the predicted power data of a photovoltaic power generation system, which can correct the predicted power data of the photovoltaic power generation system and improve the matching degree between the corrected prediction results and the actual operating data.

[0004] In a first aspect, the present invention provides a method for correcting predicted power data of a photovoltaic power generation system, comprising: A power deviation sequence is constructed based on real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system; wherein, the power deviation sequence is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points; The power deviation sequence is discretized over time, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient. The one-dimensional heat conduction equation is solved iteratively using the finite difference method, and a sequence of deviation heat field values ​​is output. The sequence of deviation heat field values ​​indicates the deviation heat field value at each time point. The iteration step size used in the iterative solution process is adaptively adjusted based on the actual power fluctuation amplitude. The original predicted power sequence data is fused and corrected based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data. Based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results corresponding to each time node are determined. For the first time node, the power data corresponding to the first time node in the original predicted power sequence data is updated based on the deviation thermal field value sequence; for the second time node, the iterative solution process of the heat conduction equation is re-executed until the preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition; After updating the power data for all time points in the original predicted power sequence data, the corrected power sequence data is obtained.

[0005] Secondly, the present invention also provides a device for correcting predicted power data of a photovoltaic power generation system, comprising: A power deviation sequence construction module is used to construct a power deviation sequence based on real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system; wherein, the power deviation sequence is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points; A one-dimensional heat conduction equation construction module is used to perform time discretization processing on the power deviation sequence and construct a one-dimensional heat conduction equation based on the power deviation sequence and the deviation change gradient. The deviation thermal field value sequence determination module is used to iteratively solve the one-dimensional heat conduction equation using the finite difference method and output the deviation thermal field value sequence; wherein, the deviation thermal field value sequence is used to indicate the deviation thermal field value at each time node; the iteration step size used in the iterative solution process is adaptively adjusted based on the actual power fluctuation amplitude; The effective iteration result determination module is used to fuse and correct the original predicted power sequence data based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data, and to determine the effective iteration result corresponding to each time node based on the residual between the intermediate corrected power sequence data and the real-time power sequence data. The data update module is used to update the power data corresponding to the first time node in the original predicted power sequence data based on the deviation thermal field value sequence for the first time node; and to re-execute the iterative solution process of the heat conduction equation for the second time node until a preset termination condition is met, and update the power data corresponding to the second time node in the original predicted power sequence data based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition; The corrected power sequence data determination module is used to obtain corrected power sequence data after updating the power data corresponding to all time points in the original predicted power sequence data.

[0006] Thirdly, this invention also provides an electronic device, comprising: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method for correcting the predicted power data of a photovoltaic power generation system as provided in any embodiment of the present invention.

[0007] Fourthly, embodiments of the present invention also provide a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a method for correcting predicted power data of a photovoltaic power generation system as provided in any embodiment of the present invention.

[0008] The technical solution of this invention constructs a power deviation sequence using real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system. This power deviation sequence indicates the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points. The power deviation sequence is time-discreteized, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient. This ensures that the power deviation value has smooth diffusion characteristics in the time dimension, avoiding abrupt changes in the overall prediction curve caused by abnormal deviation values ​​at a single time point, thus providing continuous constraints for subsequent numerical iterations. The one-dimensional heat conduction equation is iteratively solved using the finite difference method, outputting a deviation heat field value sequence. This deviation heat field value sequence indicates the deviation heat field value at each time point, and the iteration step size used in the iterative solution is adaptively adjusted based on the actual power fluctuation amplitude, thus adapting to the power fluctuation characteristics at different time periods during the solution process. The original predicted power sequence data is then fused based on the deviation heat field value sequence. The process involves several steps: First, intermediate corrected power sequence data is obtained. Based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results for each time node are determined. This allows for error assessment of the heat conduction model output, ensuring that subsequent photovoltaic power correction processes are based on stable solutions. For the first time node, the power data corresponding to that time node in the original predicted power sequence data is updated based on the deviation thermal field value sequence. For the second time node, the iterative solution process of the heat conduction equation is re-executed until a preset termination condition is met. The power data corresponding to that time node in the original predicted power sequence data is then updated based on the latest obtained deviation thermal field value sequence. The first time node is defined as the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition, thus achieving data updates for each node. After updating the power data for all time nodes in the original predicted power sequence data, the corrected power sequence data is obtained. This approach enables the modeling of the continuous propagation characteristics of prediction deviations in photovoltaic power generation systems over time. Furthermore, it establishes an adaptive solution mechanism for different power fluctuation stages during the iterative model solving process. This allows the model to simultaneously consider stability during periods of high fluctuation and computational efficiency during periods of low fluctuation, thereby correcting the predicted power data of the photovoltaic power generation system and improving the matching degree between the corrected prediction results and the actual operating data.

[0009] The above description of the invention is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0010] The above and other features, advantages, and aspects of the various embodiments of the present invention will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0011] Figure 1 This is a flowchart illustrating a method for correcting predicted power data of a photovoltaic power generation system according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating another method for correcting predicted power data of a photovoltaic power generation system provided in an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of a device for correcting predicted power data of a photovoltaic power generation system provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of an electronic device for implementing a method for correcting predicted power data of a photovoltaic power generation system, as provided in an embodiment of the present invention. Detailed Implementation

[0012] Embodiments of the present invention will now be described in more detail with reference to the accompanying drawings. While some embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of the invention. It should be understood that the accompanying drawings and embodiments are for illustrative purposes only and are not intended to limit the scope of protection of the invention.

[0013] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.

[0014] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0015] It should be noted that the concepts of "first" and "second" mentioned in this invention are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0016] It should be noted that the terms "a" and "a plurality of" used in this invention are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0017] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0018] Figure 1 This is a flowchart illustrating a method for correcting predicted power data of a photovoltaic power generation system according to an embodiment of the present invention. This embodiment is applicable to situations where predicted power data of a photovoltaic power generation system needs correction. The method can be executed by a device for correcting predicted power data of the photovoltaic power generation system. This device can be implemented in software and / or hardware and is generally integrated into any electronic device with network communication capabilities, such as a mobile terminal, PC, or server. Figure 1 As shown, the method for correcting the predicted power data of a photovoltaic power generation system according to an embodiment of the present invention may include the following process: S110. Based on the real-time power sequence data and the original predicted power sequence data of the photovoltaic power generation system, construct a power deviation sequence; wherein, the power deviation sequence is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points.

[0019] Real-time power sequence data refers to the actual power output of the photovoltaic power generation system during power generation. Raw predicted power sequence data refers to the power output data obtained after predicting the power generation of the photovoltaic power generation system using relevant parameter data. For example, raw predicted power sequence data can be data obtained after prediction based on relevant prediction models. In this embodiment, both real-time power sequence data and raw predicted power sequence data are time series data, depicting power data at different time points; that is, each power data point is associated with corresponding time information.

[0020] Specifically, after obtaining the real-time power sequence data and the original predicted power sequence data corresponding to the photovoltaic power generation system, since both of them depict the power generation data at different time points, the specific power generation data at the same time point in the two sequence data can be subtracted under a unified time dimension, such as for the same time point, to determine the power deviation value at the same time point. This allows the construction of a power deviation sequence, which is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points. This power deviation value can be understood as the specific difference between the two power generation data.

[0021] As an optional but non-limiting implementation, a power deviation sequence is constructed based on real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system. This includes: acquiring real-time power sequence data corresponding to actual power generation by the photovoltaic power generation system, and original predicted power sequence data obtained based on a power prediction model; constructing an original power deviation sequence based on the power deviation values ​​between the real-time power sequence data and the original predicted power sequence data at each time point; and normalizing the original power deviation sequence to obtain the final power deviation sequence. This optional approach allows for processing of real-time power sequence data and original predicted power sequence data within a unified time dimension to construct the power deviation sequence, thereby ensuring that the deviation input in the subsequent thermal equation solution process possesses the characteristics of unified dimensions and continuous variation.

[0022] The power prediction model can be used to predict the photovoltaic power generation capacity of a photovoltaic power generation system. This power prediction model can be a pre-trained deep learning model / probabilistic statistical model, etc. For example, in this embodiment, the power prediction model can be a multi-weather model (capable of weather classification and statistical learning).

[0023] Specifically, we can first obtain the real-time power sequence data corresponding to the actual power generation of the photovoltaic power generation system, and the original predicted power sequence data obtained by predicting the photovoltaic power generation system based on the power prediction model. Of course, before constructing the specific power deviation sequence, we can perform time synchronization processing on the real-time power sequence data and the original predicted power sequence data to align the time information of the two sequence data. Then, we can determine the power deviation value of the two sequence data at the same time node, and construct the original power deviation sequence based on the power deviation value between the real-time power sequence data and the original predicted power sequence data at each time node. This original power deviation sequence can intuitively and originally characterize the degree of offset between the real-time power sequence data and the original predicted power sequence data. Understandably, since the power output of a photovoltaic power generation system varies significantly across different time periods, directly using the original power deviation sequence would lead to excessive influence of high-power periods on the overall deviation distribution. Therefore, in this embodiment, to eliminate the interference caused by power magnitude differences in subsequent deviation measurement, the original power deviation sequence is normalized. For example, normalization can be performed based on the maximum and minimum power deviation values ​​in the original power deviation sequence to unify the interval scale, resulting in the power deviation sequence after normalization. Normalization ensures that the deviation values ​​at all time points are constrained within a uniform interval scale, allowing the power deviation sequence to retain only relative change information, thereby preventing high-power periods from dominating the deviation evaluation.

[0024] For example, in this embodiment, the original power deviation sequence can be determined using the following formula: ; in, Indicates at time node The original power deviation value at the location (this original power deviation value is the original value obtained after subtraction); Indicates at time node The real-time power generation value at the location (the specific value can be determined based on real-time power sequence data); Indicates at time node The original predicted power value at the location (which can be determined based on the original predicted power sequence data); This represents the i-th time node.

[0025] After obtaining the original power deviation sequence, the maximum and minimum power deviation values ​​in the entire original power deviation sequence can be counted and denoted as follows: and Based on this, a linear scaling relationship is constructed to map each original power deviation value in the original power deviation sequence to a unified interval of [0,1]. Specifically, the normalization process can be performed using the following formula: ; in, Indicates at a time node The power deviation value obtained after normalization; Indicates at a time node The original power deviation value at the location; This represents the maximum power deviation value in the entire original power deviation sequence (i.e., the maximum value among all the original power deviation values ​​in the entire sequence). This represents the minimum power deviation in the entire original power deviation sequence (i.e., the minimum of all original power deviation values ​​in the entire sequence).

[0026] S120. The power deviation sequence is discretized over time, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient.

[0027] The deviation change gradient indicates the rate of deviation change between adjacent time points. That is, the deviation change gradient can be used to identify periods of rapid deviation change and periods of gradual deviation change throughout the entire time period, giving the power deviation sequence a structured temporal variation characteristic. In this embodiment, the deviation change gradient can be calculated based on the deviation power values ​​of the power deviation sequence between adjacent time points. For example, the deviation change gradient can be determined using the following formula: ; in, Indicates at a time node The gradient of deviation change at the location; Indicates at a time node The power deviation value at the location (already normalized); Indicates at a time node The power deviation value at the location (already normalized); Indicates time node and time nodes The time interval between them.

[0028] Specifically, after obtaining a stable power deviation sequence with time structure characteristics, the power deviation sequence can be time discretized to serve as the boundary input for the subsequent one-dimensional heat conduction equation. Based on the power deviation sequence and the deviation change gradient, the one-dimensional heat conduction equation is constructed.

[0029] As an optional but non-limiting implementation, the power deviation sequence is discretized over time, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient. This includes: discretizing the power deviation sequence over time at preset time intervals to obtain multiple time nodes and their corresponding power deviation values; constructing the driving source term of the one-dimensional heat conduction equation based on the deviation change gradient and gradient driving coefficients; assigning the power deviation values ​​corresponding to each time node to the corresponding time nodes of the one-dimensional heat diffusion equation as anchor points for the thermal field boundary; and constructing the one-dimensional heat conduction equation based on the driving source term and the heat diffusion law. This optional scheme transforms the original power deviation sequence into a continuous time field, and constructs the one-dimensional heat conduction equation based on the deviation change gradient. This allows the power deviation values ​​to exhibit smooth diffusion characteristics in the time dimension, avoiding abrupt changes in the overall prediction curve caused by abnormal deviation values ​​at a single time node, thus providing continuous constraints for subsequent numerical iterations.

[0030] The gradient driving coefficient is used to adjust the intensity of the influence of the deviation gradient on the thermal diffusion process.

[0031] Specifically, the power deviation sequence can be discretized at preset time intervals to obtain multiple time nodes and their corresponding power deviation values. This preset time interval can be customized based on actual needs; for example, it could be set to 15 minutes. Understanding this, time discretization allows the power deviation sequence to be segmented into time periods, transforming the continuous heat conduction differential equation into a computer-calculate discrete formula, facilitating the simulation of power deviation diffusion between adjacent time nodes. Then, based on the deviation change gradient and gradient driving coefficient, a driving source term for the one-dimensional heat conduction equation can be constructed. This driving source term reflects the disturbance effect of changes in photovoltaic operating status on deviation propagation. Subsequently, the power deviation values ​​corresponding to each time node are assigned to the corresponding time nodes of the one-dimensional heat diffusion equation as anchor points for the thermal field boundary; and based on the driving source term and the heat diffusion law, a one-dimensional heat conduction equation is constructed.

[0032] For example, the process of constructing the one-dimensional heat conduction equation can be illustrated below: (1) In order to keep the one-dimensional heat conduction equation synchronized with the actual operating data, the power deviation sequence can be discretized in the time dimension according to a preset time interval. The preset time interval is set as follows: and stipulate By discretizing the data over several minutes, multiple time points and their corresponding power deviation values ​​can be obtained. These time points can form a thermal field time point sequence. At each discrete time node, the specific power deviation value in the power deviation sequence can be assigned to the corresponding time node of the one-dimensional heat diffusion equation as a boundary anchor point of the thermal field. This ensures that the subsequent one-dimensional heat conduction equation is always constrained by real-time data during the solution process. The specific boundary mapping relationship is as follows: ; in, This indicates that the one-dimensional heat conduction equation is at time node 1. Boundary values ​​at; Represents the time nodes in the power deviation sequence The power deviation value at each time point. There is a one-to-one mapping relationship between these two. Through this boundary mapping relationship, the power deviation value corresponding to each time point can directly participate in the solution process of the thermal equation, so that the heat diffusion process always revolves around the real deviation data, thereby forming a deviation propagation field with real-time constraint characteristics.

[0033] (2) To further enhance the thermal field's response to structural deviation changes, this embodiment constructs the driving source term of the one-dimensional heat conduction equation using the deviation change gradient and gradient driving coefficient, thereby coupling the thermal diffusion process of the deviation with the deviation change rate. Specifically, the driving source term can be constructed using the following formula: ; in, Indicates at a time node The driving source item at the location; Indicates the gradient driving coefficients; Indicates at a time node The deviation change gradient at a given point. By constructing this driving source term, time nodes with larger deviation change rates can obtain a stronger driving effect during the thermal diffusion process, thereby enhancing the response capability to periods of rapid fluctuation in photovoltaic power generation, while maintaining stable diffusion characteristics during periods of gradual deviation change.

[0034] (3) Assign the power deviation values ​​corresponding to each time node to the time nodes corresponding to the one-dimensional heat diffusion equation one by one, so as to serve as the thermal field boundary anchor points; and construct a one-dimensional heat conduction equation based on the driving source term and the heat diffusion law. Specifically, a heat diffusion function can be constructed to describe the propagation characteristics of the power deviation value in the time dimension, and the power deviation sequence can be represented as a time field function to simulate the diffusion characteristics of the power deviation value over time. Furthermore, assign the power deviation values ​​corresponding to each time node one by one to the time nodes corresponding to the one-dimensional heat diffusion equation one by one, so as to serve as the thermal field boundary anchor points, so that the power deviation propagates smoothly in the time dimension according to the heat diffusion law. The constructed one-dimensional heat conduction equation can be in the following form: ; in, This indicates that the one-dimensional heat conduction equation is at time node 1. The deviation thermal field value at the location; k represents the deviation diffusion coefficient, used to control the diffusion intensity of the deviation in the time dimension; This represents the rate of change of the deviation thermal field over time. This represents the curvature change of the deviation thermal field over the time dimension; Indicates at a time node The driving source term at the location can be used to reflect the disturbance effect of changes in photovoltaic operating status on deviation propagation.

[0035] It should be noted that in the one-dimensional heat conduction equation, It depicts a point in time. The deviation thermal field value at that point can be understood as the specific numerical value of the function at a single time node, representing the quantized result of the deviation at a single time node after thermal diffusion smoothing. Previously, when assigning values ​​to the thermal field boundary anchor points, It is itself the deviation thermal field value corresponding to the time field function. In the initial stage of solving the thermal equation, it is assigned the normalized deviation as a boundary constraint. After iterative calculation, it becomes the deviation thermal field value after diffusion smoothing.

[0036] As can be seen, after time discretization, assignment of thermal field boundary anchor points, and construction of driving source terms based on deviation change gradient, the final one-dimensional heat conduction equation can form a complete deviation thermal field structure over the entire time series. This deviation thermal field structure will serve as the basic model for subsequent iterative solutions using the variable step size finite difference method.

[0037] S130. The one-dimensional heat conduction equation is solved iteratively using the finite difference method, and the deviation heat field value sequence is output. The deviation heat field value sequence is used to indicate the deviation heat field value at each time node. The iteration step size used in the iterative solution process is adaptively adjusted based on the actual power fluctuation amplitude.

[0038] The deviation thermal field value sequence can be understood as the result obtained by iteratively solving the one-dimensional heat conduction equation. This deviation thermal field value sequence is used to indicate the deviation thermal field value at each time node. The deviation thermal field value is the specific power deviation value corresponding to a single time node after iterative solution (i.e., the value after thermal diffusion smoothing calculation).

[0039] Actual power fluctuation amplitude refers to the degree of fluctuation in real-time power generation in real-time power sequence data. This actual power fluctuation amplitude can be characterized based on the power fluctuation rate between adjacent time nodes in the real-time power sequence data.

[0040] Specifically, the finite difference method can be used to iteratively solve the one-dimensional heat conduction equation. The iteration step size can be adaptively adjusted based on the actual power fluctuation amplitude. For example, when the actual power fluctuation amplitude increases, the iteration step size can be decreased to enhance the stability of the numerical solution; conversely, when the actual power fluctuation amplitude is moderate, the iteration step size can be increased to reduce the number of iterations and improve computational efficiency. After iterative solving, a sequence of deviation thermal field values ​​can be output, indicating the deviation thermal field values ​​at each time point.

[0041] S140. Based on the deviation thermal field value sequence, the original predicted power sequence data is fused and corrected to obtain intermediate corrected power sequence data. Based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results corresponding to each time node are determined.

[0042] The intermediate corrected power sequence data can be understood as an intermediate data index used for residual determination. This data is used to assist in error evaluation during the iterative solution process.

[0043] The valid iteration results are used to indicate whether the solution to the next-dimensional heat conduction equation has reached stable convergence at the time node.

[0044] Specifically, the original predicted power sequence data can be fused and corrected based on the deviation thermal field value sequence. For example, since both the deviation thermal field value sequence and the original predicted power sequence data are associated with time node information, for any given time node, the specific deviation thermal field value in the deviation thermal field value sequence can be determined. This deviation thermal field value is then used to update the corresponding power value (i.e., the original predicted power value) in the original predicted power sequence data for that time node, thus achieving fusion correction. The corrected value can then be used as the corresponding value in the intermediate corrected power sequence data for that time node. Similarly, this principle can be applied to the values ​​of each time node, resulting in the intermediate corrected power sequence data. Furthermore, based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results corresponding to each time node can be determined.

[0045] S150. For the first time node, update the power data corresponding to the first time node in the original predicted power sequence data based on the deviation thermal field value sequence; for the second time node, re-execute the iterative solution process of the heat conduction equation until the preset termination condition is met, and update the power data corresponding to the second time node in the original predicted power sequence data based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition.

[0046] Specifically, after determining the valid iteration results for each time node, the first and second time nodes can be determined. The first time node is the time node where the valid iteration result meets the convergence condition, and the second time node is the time node where the valid iteration result does not meet the convergence condition. For the first time node, the power data (i.e., the corresponding original predicted power value) in the original predicted power sequence data corresponding to the first time node can be updated based on the deviation thermal field value sequence. For the second time node, the iterative solution process of the heat conduction equation can be re-executed, continuing multiple rounds of iterative updates until the preset termination condition is met. Then, the power data (i.e., the corresponding original predicted power value) in the original predicted power sequence data corresponding to the second time node can be updated based on the latest deviation thermal field value sequence obtained at the end of the iteration.

[0047] S160. After updating the power data corresponding to all time points in the original predicted power sequence data, the corrected power sequence data is obtained.

[0048] Among them, the corrected power sequence data is the result data obtained after correcting each power value in the original predicted power sequence data.

[0049] Specifically, after multiple rounds of iterative updates to the second time node and completing the power data update, it is equivalent to updating the power data corresponding to all time nodes in the original predicted power sequence data, thus obtaining the corrected power sequence data. This corrected power sequence data can then be used as input data for subsequent photovoltaic power prediction and grid dispatch calculations.

[0050] The technical solution of this invention constructs a power deviation sequence using real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system. This power deviation sequence indicates the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points. The power deviation sequence is time-discreteized, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient. This ensures that the power deviation value has smooth diffusion characteristics in the time dimension, avoiding abrupt changes in the overall prediction curve caused by abnormal deviation values ​​at a single time point, thus providing continuous constraints for subsequent numerical iterations. The one-dimensional heat conduction equation is iteratively solved using the finite difference method, outputting a deviation heat field value sequence. This deviation heat field value sequence indicates the deviation heat field value at each time point, and the iteration step size used in the iterative solution is adaptively adjusted based on the actual power fluctuation amplitude, thus adapting to the power fluctuation characteristics at different time periods during the solution process. The original predicted power sequence data is then fused based on the deviation heat field value sequence. The process involves several steps: First, intermediate corrected power sequence data is obtained. Based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results for each time node are determined. This allows for error assessment of the heat conduction model output, ensuring that subsequent photovoltaic power correction processes are based on stable solutions. For the first time node, the power data corresponding to that time node in the original predicted power sequence data is updated based on the deviation thermal field value sequence. For the second time node, the iterative solution process of the heat conduction equation is re-executed until a preset termination condition is met. The power data corresponding to that time node in the original predicted power sequence data is then updated based on the latest obtained deviation thermal field value sequence. The first time node is defined as the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition, thus achieving data updates for each node. After updating the power data for all time nodes in the original predicted power sequence data, the corrected power sequence data is obtained. This approach enables the modeling of the continuous propagation characteristics of prediction deviations in photovoltaic power generation systems over time. Furthermore, it establishes an adaptive solution mechanism for different power fluctuation stages during the iterative model solving process. This allows the model to simultaneously consider stability during periods of high fluctuation and computational efficiency during periods of low fluctuation, thereby correcting the predicted power data of the photovoltaic power generation system and improving the matching degree between the corrected prediction results and the actual operating data.

[0051] Figure 2This is a flowchart illustrating another method for correcting predicted power data of a photovoltaic power generation system provided in an embodiment of the present invention. The technical solution of this embodiment further optimizes the process of iteratively solving the one-dimensional heat conduction equation using the finite difference method and outputting a sequence of deviation heat field values, based on the technical solutions of the above embodiments. This embodiment can be combined with various optional solutions in one or more of the above embodiments. Figure 2 As shown, the method for correcting the predicted power data of a photovoltaic power generation system according to an embodiment of the present invention may include the following process: S210. Based on the real-time power sequence data and the original predicted power sequence data of the photovoltaic power generation system, construct a power deviation sequence; wherein, the power deviation sequence is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points.

[0052] S220. The power deviation sequence is discretized over time, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient.

[0053] S230. Construct a power volatility function based on real-time power sequence data.

[0054] The power volatility function characterizes the intensity or volatility of power changes between adjacent time points. Using this function, the power volatility between adjacent time points throughout the entire time series can be determined, thereby identifying the rapidly changing and stable periods of actual photovoltaic power generation data within the entire time series.

[0055] Specifically, analysis can be performed based on real-time power sequence data to construct a power volatility function, thereby characterizing the intensity of power changes between adjacent time points. Understandably, the power volatility function can identify rapidly changing and stable regions in the actual photovoltaic power generation data throughout the time series. For rapidly changing regions, it means that power fluctuations are significant, so the heat diffusion solution process requires a finer time iteration granularity, i.e., a smaller iteration step size to avoid excessive deviation diffusion. For stable regions, it means that power fluctuations are relatively gentle, so the heat diffusion solution process can increase the iteration step size to improve computational efficiency.

[0056] For example, in this embodiment, the constructed power volatility function can be in the following form: ; in, Indicates at a time node Power fluctuation rate at the location; Indicates at a time node Real-time power generation value at the location; Show at the time node The real-time power generation value at the location.

[0057] As can be seen, this volatility index can identify rapidly changing and stable periods in photovoltaic output throughout the time series. For example, when determining... When the value exceeds a preset fluctuation threshold (which can be set to 0.05), it indicates significant power fluctuation. In this case, the iteration step size needs to be reduced to avoid excessive deviation propagation. If the fluctuation is not greater than the preset fluctuation threshold, it means that the power fluctuation in this interval is relatively smooth, so it is necessary to increase the iteration step size to improve the computational efficiency.

[0058] S240. Determine the iteration step size function based on the power fluctuation rate function; wherein, the iteration step size function is used to determine the iteration step size used in the iterative solution process.

[0059] Specifically, after determining the power volatility function, the iteration step size function can be determined based on the power volatility function. This iteration step size function is used to determine the iteration step size during the iterative solution process. In this embodiment, this method of dynamically determining the iteration step size (that is, adaptively and dynamically adjusting the iteration step size based on the power volatility between adjacent time nodes) can reduce the iteration step size when the power volatility changes significantly, thereby enhancing the stability of the numerical solution, and automatically increase the step size when the power change is gradual, thereby reducing the number of iterations and improving computational efficiency.

[0060] For example, in this embodiment, the constructed iteration step size function can be in the following form: ; in, Indicates at a time node The corresponding iteration step size at that location; This indicates the preset time interval (i.e., the time interval used in the time discretization process mentioned above). This represents the fluctuation adjustment coefficient, which is used to control the intensity of the influence of power fluctuation rate on step size changes; Indicates at a time node Power fluctuation rate at that location.

[0061] S250. Based on the iterative step size function, the one-dimensional heat conduction equation is solved iteratively using the finite difference method to obtain the sequence of deviation heat field values.

[0062] Specifically, after determining the iteration step size function, the corresponding iteration step size can be dynamically determined based on the iteration step size function. Then, the finite difference method can be used to iteratively solve the one-dimensional heat conduction equation, calculating the propagation process of the deviation heat field between each time node. During the specific iterative solution process, the original deviation heat field values ​​corresponding to each time node can be continuously updated, and a sequence of deviation heat field values ​​is obtained after the iteration is completed. The obtained sequence of deviation heat field values ​​is a continuous and smooth deviation distribution structure, which can maintain a consistent match with the fluctuation characteristics of power generation. Through this finite difference iterative calculation, the deviation heat field values ​​can be continuously updated in the time dimension, gradually forming a continuous and smooth deviation distribution structure under the action of heat diffusion law, and maintaining a dynamic matching relationship with the real-time photovoltaic power fluctuation characteristics.

[0063] For example, in this embodiment, the finite difference method is used to iteratively solve the one-dimensional heat conduction equation, and the specific expression of the difference iteration is as follows: ; in, This indicates the time node at the j-th iteration. The deviation thermal field value at that location; This indicates the time node at the (j+1)th iteration. The updated deviation thermal field value; Indicates the current time point The corresponding iteration step size at the given location; k represents the deviation diffusion coefficient; This indicates the time node at the j-th iteration. The deviation thermal field value at that location; This indicates the time node at the j-th iteration. The deviation thermal field value at that location; and Essentially, it refers to the same iteration round. The deviation thermal field values ​​corresponding to adjacent time nodes; Indicates at a time node The driver source item at that location.

[0064] S260. Based on the deviation thermal field value sequence, the original predicted power sequence data is fused and corrected to obtain intermediate corrected power sequence data. Based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results corresponding to each time node are determined.

[0065] As an optional but non-limiting implementation, the original predicted power sequence data is fused and corrected based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data. This includes: for the target time node, determining the deviation thermal field value corresponding to the target time node based on the deviation thermal field value sequence, and determining the original predicted power value corresponding to the target time node based on the original predicted power sequence data; determining the maximum and minimum power deviation values ​​based on the original power deviation sequence; fusing and correcting the original predicted power value corresponding to the target time node based on the deviation thermal field value, maximum power deviation value, and minimum power deviation value to obtain intermediate corrected power value; and obtaining intermediate corrected power sequence data based on the intermediate corrected power values ​​corresponding to each target time node. Using this optional scheme, the original predicted power sequence data can be fused and corrected based on the obtained deviation thermal field value sequence, thereby restoring the deviation thermal field value, which has been processed by the heat conduction model and includes the propagation law of time-series deviation, to the original predicted power at the true power scale. This facilitates the subsequent analysis and determination of effective iterative results, verifying whether the heat conduction equation solution process has reached a stable and reliable convergence state.

[0066] The target time point can refer to any time point.

[0067] The maximum power deviation refers to the largest of all original power deviation values ​​in the entire original power deviation sequence. Conversely, the minimum power deviation refers to the smallest of all original power deviation values ​​in the entire original power deviation sequence. This original power deviation sequence is the one constructed before the normalization process described above.

[0068] Specifically, for a target time node, the deviation thermal field value corresponding to that target time node can be determined based on the deviation thermal field value sequence, and the original predicted power value corresponding to that target time node can be determined based on the original predicted power sequence data. Then, based on the original power deviation sequence, the maximum and minimum power deviation values ​​are determined. Next, based on the deviation thermal field value, maximum power deviation value, and minimum power deviation value corresponding to the target time node, the original predicted power value corresponding to that target time node can be fused and corrected to obtain an intermediate corrected power value. The intermediate corrected power value obtained at this point is equivalent to: the deviation thermal field value, which has been processed by the heat conduction model and includes the temporal deviation propagation law, being restored and superimposed onto the original predicted power value at the true power scale. Following the same principle, fusion and correction processing can be performed on all other target time nodes to determine the intermediate corrected power value corresponding to each time node. Finally, based on the intermediate corrected power values ​​corresponding to each target time node, intermediate corrected power sequence data can be obtained.

[0069] For example, in this embodiment, the obtained intermediate corrected power sequence data can be in the following form: ; in, Indicates time node The intermediate power value at the location; Represents the time node in the original predicted power sequence data. The corresponding original predicted power value at that location; This represents the sequence of deviation thermal field values ​​obtained through iterative solution; This represents the maximum power deviation (i.e., the maximum value among all the original power deviation values ​​in the entire sequence). This represents the minimum power deviation (i.e., the minimum of all original power deviation values ​​in the entire sequence).

[0070] It is evident that through such inverse normalization transformation, the deviation thermal field value obtained by iterative solution can be restored to the actual power scale and superimposed on the original predicted power in the real power scale, thereby obtaining an intermediate corrected power value that is closer to the real-time operating state. This facilitates the analysis and determination of the effective iterative results in the subsequent process, in order to verify whether the heat conduction equation solution process has reached a stable and reliable convergence state.

[0071] As an optional but non-limiting implementation, this method determines the valid iteration results for each time node based on the residuals between the intermediate corrected power sequence data and the real-time power sequence data. This includes: constructing a residual evaluation function based on the power residuals of the intermediate corrected power sequence data and the real-time power sequence data at each time node; and determining the valid iteration results for each time node based on the residual evaluation function and a preset residual threshold. Using this optional approach, a residual evaluation function can be constructed based on the residuals between the intermediate corrected power sequence data and the real-time power sequence data, thereby determining the valid iteration results for each time node.

[0072] The residual evaluation function reflects the degree of deviation between the "data after fusing and correcting the original predicted power sequence data with the biased thermal field value sequence (i.e., intermediate corrected power sequence data)" and the "real-time power sequence data". Essentially, this residual evaluation function can be used to measure and determine whether the solution process of the one-dimensional heat conduction equation has reached stable convergence.

[0073] The preset residual threshold can be set differently based on actual needs, and will not be specified in detail here.

[0074] Specifically, a residual evaluation function can be constructed based on the power residuals of intermediate corrected power sequence data and real-time power sequence data at various time points. The specific construction method or parameter operation relationship of the residual evaluation function can be differentiated according to actual needs. For example, the residual evaluation function can be defined as the ratio of intermediate corrected power sequence data to real-time power sequence data, and further scaling can be applied based on actual needs to facilitate direct comparison of convergence criteria. Then, based on the residual evaluation function and a preset residual threshold, the effective iteration results corresponding to each time point can be determined. For example, the specific residual evaluation value at a given time point can be determined based on the residual evaluation function, and the effective iteration result can be determined based on the relationship between this residual evaluation value and the preset residual threshold. If the residual evaluation value is not greater than the preset residual threshold, the effective iteration result at that time point satisfies the convergence condition; if the residual evaluation value is greater than the preset residual threshold, the effective iteration result at that time point does not satisfy the convergence condition.

[0075] For example, based on intermediate corrected power sequence data and real-time power sequence data, the specific power residual value at each time point can be determined. For instance, it can be determined using the following formula: ; in, Indicates at a time node The power residual value at the location; Indicates at a time node The real-time power generation value at the location (the specific value can be determined based on real-time power sequence data); Indicates time node The intermediate power correction value at that point.

[0076] It is understandable that the absolute residual values ​​between high-power periods and low-power periods cannot be directly compared using convergence criteria; they need to be unified to the same evaluation scale. Therefore, in this embodiment, the specific power residual values ​​obtained above can be further normalized. For example, the actual power generation can be used as the denominator to scale to the same evaluation scale, so that the same threshold can be used to determine whether the iterative solution at each time node meets the convergence condition. For example, in this embodiment, the constructed residual evaluation function can be in the following form: ; in, Indicates at a time node The residual evaluation value at the location; Indicates at a time node The power residual value at the location; Indicates at a time node The real-time power generation value at the location.

[0077] It is evident that this normalization process unifies the error evaluation at different power levels to the same scale, ensuring that the convergence judgment criteria remain consistent throughout the day. If the residual is not greater than the preset residual threshold, then the effective iteration result at that time point satisfies the convergence condition; if If the residual value is greater than the preset residual threshold, then the effective iteration result at that time point does not meet the convergence condition.

[0078] S270. For the first time node, update the power data corresponding to the first time node in the original predicted power sequence data based on the deviation thermal field value sequence; for the second time node, re-execute the iterative solution process of the heat conduction equation until the preset termination condition is met, and update the power data corresponding to the second time node in the original predicted power sequence data based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition.

[0079] As an optional but non-limiting implementation, for the first time node, the power data corresponding to the first time node in the original predicted power sequence data is updated based on the deviation thermal field value sequence; for the second time node, the iterative solution process of the heat conduction equation is re-executed until a preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence. This includes: determining the set of converged nodes and the set of non-converged nodes based on the valid iteration results corresponding to each time node; determining the deviation thermal field value corresponding to each first time node included in the set of converged nodes based on the deviation thermal field value sequence, and updating the power data corresponding to each first time node in the original predicted power sequence data based on each deviation thermal field value; for each second time node included in the set of non-converged nodes, the iterative solution process of the heat conduction equation is re-executed until a preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence. By adopting this optional scheme, the time-series filtering of the thermal equation iteration results can be performed based on the effective iteration results corresponding to each time node, and the original predicted power sequence data can be updated time-by-time according to the convergence state.

[0080] The convergent node set refers to the set of time points where the valid iteration results satisfy the convergence condition. Each time point included in the convergent node set is a first time point, and there must be at least one first time point in the convergent node set. Conversely, the non-converged node set refers to the set of time points where the valid iteration results do not satisfy the convergence condition. Each time point included in the non-converged node set is a second time point, and there must also be at least one second time point in the non-converged node set.

[0081] Specifically, based on the valid iteration results corresponding to each time node, all time nodes can be divided to determine the set of converged nodes and the set of non-converged nodes. Then, based on the deviation thermal field value sequence, the deviation thermal field values ​​corresponding to each first time node in the converged node set can be determined, and the power data (i.e., the corresponding original predicted power value) of each first time node in the original predicted power sequence data can be updated based on the obtained deviation thermal field values. Furthermore, for each second time node in the non-converged node set, the iterative solution process of the heat conduction equation can be re-executed until a preset termination condition is met, and the power data (i.e., the corresponding original predicted power value) of the second time node in the original predicted power sequence data can be updated based on the latest obtained deviation thermal field value sequence. The preset termination condition can be that the valid iteration result corresponding to the second time node obtained after re-executing the iterative solution process satisfies the convergence condition.

[0082] For example, the principle of updating the corresponding power data in the original predicted power sequence data can be explained below: (1) First, a convergence flag function is established, which can logically judge the residual evaluation value at each time node so that time nodes that meet the convergence condition enter the update sequence, and time nodes that do not meet the convergence condition enter the re-iteration sequence. Its expression is as follows: ; in, Indicates at a time node The convergence status indicator value at the location; Indicates at a time node The residual evaluation value at the location; This indicates the preset residual threshold.

[0083] It can be seen that when When , it indicates that the valid iteration result corresponding to this time node satisfies the convergence condition. The convergence flag indicates that the valid iteration result corresponding to that time node does not meet the convergence condition. This convergence flag function allows for the division of all time nodes in the entire time series into a set of converged nodes and a set of non-converged nodes, thus providing a clear and structured processing path for subsequent power update processes.

[0084] (2) After obtaining the set of convergence nodes, the deviation thermal field values ​​corresponding to each time node (i.e., the first time node) in the set of convergence nodes can be determined based on the deviation thermal field value sequence. Furthermore, the power data corresponding to each first time node in the original predicted power sequence data can be updated one by one based on the obtained deviation thermal field values. The specific update expression can be in the following form: ; in, This represents the time nodes in the original predicted power sequence data. The updated power data is obtained here; Indicates at a time node The convergence status indicator value at the location; Indicates time node The intermediate power value at the location; Represents the time node in the original predicted power sequence data. The corresponding power data (i.e., the original predicted power value).

[0085] As can be seen, through this update relationship, when the time node meets the convergence condition, the intermediate corrected power value obtained above can be directly used for update and replacement. When the time node does not meet the convergence condition, the power data at the corresponding position of the original predicted power sequence data remains unchanged, thereby ensuring the continuity and stability of the predicted power sequence data in the time dimension.

[0086] (3) For time nodes that do not meet the convergence condition, that is, for each second time node included in the set of non-converged nodes, the iterative solution process of the heat conduction equation can be re-executed to further correct the deviation propagation result. In order to control the stability of the re-iteration process, an iterative update function can be constructed at the non-converged time nodes so that the deviation thermal field value continues to evolve on the basis of the previous round of solution. Its expression is as follows: ; in, Indicates the time point of non-convergence The deviation thermal field value in the next iteration; Indicates the time point of non-convergence The deviation thermal field value in the current iteration; This indicates the time node in the real-time power sequence data. Real-time power generation value at the location; Indicates time node The intermediate power value at the location; This represents the iterative adjustment coefficient. This update mechanism utilizes the error between real-time power and intermediate correction power to continuously correct the deviation thermal field value until a preset termination condition is met, gradually bringing the deviation diffusion result at the unconverged time point closer to the actual operating state.

[0087] S280. After updating the power data corresponding to all time points in the original predicted power sequence data, the corrected power sequence data is obtained.

[0088] The technical solution of this invention constructs a power deviation sequence using real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system. This power deviation sequence indicates the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points. The power deviation sequence is time-discreteized, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient. This ensures that the power deviation value has smooth diffusion characteristics in the time dimension, avoiding abrupt changes in the overall prediction curve caused by abnormal deviation values ​​at a single time point, thus providing continuous constraints for subsequent numerical iterations. A power volatility function is constructed based on the real-time power sequence data. An iteration step size function is determined based on the power volatility function, used to determine the iteration step size during the iterative solution process. Based on the iteration step size function, the one-dimensional heat conduction equation is iteratively solved using the finite difference method to obtain a deviation heat field value sequence. This deviation heat field value sequence indicates the deviation heat field value at each time point, and the iteration step size used during the iterative solution process is adaptively adjusted based on the actual power fluctuation amplitude, thereby enabling continuous numerical iterations during the solution process. The process adapts to the power fluctuation characteristics of different time periods; it fuses and corrects the original predicted power sequence data based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data, and determines the effective iteration results corresponding to each time node based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, thereby evaluating the error of the heat conduction model output results so that the subsequent photovoltaic power correction process is based on a stable solution; for the first time node, the power data corresponding to the first time node in the original predicted power sequence data is updated based on the deviation thermal field value sequence; for the second time node, the iterative solution process of the heat conduction equation is re-executed until the preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition, realizing data updates for each node; after the power data corresponding to all time nodes in the original predicted power sequence data are updated, the corrected power sequence data is obtained. This approach enables the modeling of the continuous propagation characteristics of prediction deviations in photovoltaic power generation systems over time. Furthermore, it establishes an adaptive solution mechanism for different power fluctuation stages during the iterative model solving process. This allows the model to simultaneously consider stability during periods of high fluctuation and computational efficiency during periods of low fluctuation, thereby correcting the predicted power data of the photovoltaic power generation system and improving the matching degree between the corrected prediction results and the actual operating data.

[0089] Figure 3This is a schematic diagram of a device for correcting predicted power data of a photovoltaic power generation system according to an embodiment of the present invention. This embodiment is applicable to situations where predicted power data of a photovoltaic power generation system needs to be corrected. This device can be implemented in software and / or hardware and is generally integrated into any electronic device with network communication capabilities, such as a mobile terminal, PC, or server. Figure 3 As shown, the device for correcting predicted power data of a photovoltaic power generation system according to an embodiment of the present invention may include a power deviation sequence construction module 310, a one-dimensional heat conduction equation construction module 320, a deviation thermal field value sequence determination module 330, an effective iteration result determination module 340, a data update module 350, and a corrected power sequence data determination module 360. Wherein: The power deviation sequence construction module 310 is used to construct a power deviation sequence based on the real-time power sequence data and the original predicted power sequence data of the photovoltaic power generation system; wherein, the power deviation sequence is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points; The one-dimensional heat conduction equation construction module 320 is used to perform time discretization processing on the power deviation sequence and construct a one-dimensional heat conduction equation based on the power deviation sequence and the deviation change gradient. The deviation thermal field value sequence determination module 330 is used to iteratively solve the one-dimensional heat conduction equation using the finite difference method and output the deviation thermal field value sequence; wherein, the deviation thermal field value sequence is used to indicate the deviation thermal field value at each time node; the iteration step size used in the iterative solution process is adaptively adjusted based on the actual power fluctuation amplitude; The effective iteration result determination module 340 is used to fuse and correct the original predicted power sequence data based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data, and to determine the effective iteration result corresponding to each time node based on the residual between the intermediate corrected power sequence data and the real-time power sequence data. The data update module 350 is used to update the power data corresponding to the first time node in the original predicted power sequence data based on the deviation thermal field value sequence for the first time node; and to re-execute the iterative solution process of the heat conduction equation for the second time node until a preset termination condition is met, and update the power data corresponding to the second time node in the original predicted power sequence data based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition; The corrected power sequence data determination module 360 ​​is used to obtain corrected power sequence data after updating the power data corresponding to all time nodes in the original predicted power sequence data.

[0090] The technical solution of this invention constructs a power deviation sequence using real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system through a power deviation sequence construction module. This power deviation sequence indicates the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points. A one-dimensional heat conduction equation construction module discretizes the power deviation sequence over time and constructs a one-dimensional heat conduction equation based on the power deviation sequence and the deviation change gradient. This ensures that the power deviation value has smooth diffusion characteristics in the time dimension, avoiding abrupt changes in the overall prediction curve caused by abnormal deviation values ​​at a single time point, thus providing continuous constraints for subsequent numerical iterations. A deviation thermal field value sequence determination module iteratively solves the one-dimensional heat conduction equation using the finite difference method, outputting a deviation thermal field value sequence. This deviation thermal field value sequence indicates the deviation thermal field value at each time point, and the iteration step size used in the iterative solution is adaptively adjusted based on the actual power fluctuation amplitude, thus adapting to the power fluctuation characteristics at different time periods during the solution process. An effective iteration result determination module determines the results based on the deviation thermal field value sequence. The original predicted power sequence data is fused and corrected to obtain intermediate corrected power sequence data. Based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results corresponding to each time node are determined, thereby evaluating the error of the heat conduction model output results and ensuring that the subsequent photovoltaic power correction process is based on a stable solution. For the first time node, the data update module updates the power data corresponding to the first time node in the original predicted power sequence data based on the deviation thermal field value sequence. For the second time node, the iterative solution process of the heat conduction equation is re-executed until a preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence. The first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition, achieving data update for each node. After the corrected power sequence data determination module has updated the power data corresponding to all time nodes in the original predicted power sequence data, the corrected power sequence data is obtained. This approach enables the modeling of the continuous propagation characteristics of prediction deviations in photovoltaic power generation systems over time. Furthermore, it establishes an adaptive solution mechanism for different power fluctuation stages during the iterative model solving process. This allows the model to simultaneously consider stability during periods of high fluctuation and computational efficiency during periods of low fluctuation, thereby correcting the predicted power data of the photovoltaic power generation system and improving the matching degree between the corrected prediction results and the actual operating data.

[0091] As an optional but non-limiting implementation, the power deviation sequence construction module 310 includes a sequence data acquisition unit, an original power deviation sequence construction unit, and a normalization processing unit. Wherein: The sequence data acquisition unit is used to acquire the real-time power sequence data corresponding to the actual power generation of the photovoltaic power generation system, and the original predicted power sequence data obtained based on the power prediction model. The original power deviation sequence construction unit is used to construct the original power deviation sequence based on the power deviation value between the real-time power sequence data and the original predicted power sequence data at each time point; The normalization processing unit is used to normalize the original power deviation sequence to obtain the power deviation sequence.

[0092] As an optional but non-limiting implementation, the one-dimensional heat conduction equation construction module 320 includes a time discretization processing unit, a driving source term construction unit, and a one-dimensional heat conduction equation construction unit. Wherein: The time discretization processing unit is used to perform time discretization processing on the power deviation sequence according to a preset time interval to obtain multiple time nodes and the power deviation value corresponding to each time node. A driving source term construction unit is used to construct a driving source term for a one-dimensional heat conduction equation based on the deviation change gradient and the gradient driving coefficient. The one-dimensional heat conduction equation construction unit is used to assign the power deviation values ​​corresponding to each time node to the time node corresponding to the one-dimensional heat diffusion equation, so as to serve as the thermal field boundary anchor point; and to construct the one-dimensional heat conduction equation based on the driving source term and the heat diffusion law.

[0093] As an optional but non-limiting implementation, the deviation thermal field value sequence determination module 330 includes a power fluctuation rate function construction unit, an iteration step size function determination unit, and an iterative solution unit. Wherein: A power volatility function construction unit is used to construct a power volatility function based on the real-time power sequence data; An iterative step size function determination unit is used to determine an iterative step size function based on the power volatility function; wherein, the iterative step size function is used to determine the iterative step size used in the iterative solution process; The iterative solution unit is used to iteratively solve the one-dimensional heat conduction equation based on the iterative step size function using the finite difference method to obtain the sequence of deviation heat field values.

[0094] As an optional but non-limiting implementation, the effective iteration result determination module 340 includes a data determination unit, an extremum determination unit, a fusion correction unit, and an intermediate correction power sequence data determination unit. Wherein: The data determination unit is used to determine the deviation thermal field value corresponding to the target time node based on the deviation thermal field value sequence, and to determine the original predicted power value corresponding to the target time node based on the original predicted power sequence data. The extreme value determination unit is used to determine the maximum and minimum power deviation values ​​based on the original power deviation sequence. The fusion correction unit is used to fuse and correct the original predicted power value corresponding to the target time node based on the deviation thermal field value, the maximum power deviation value, and the minimum power deviation value corresponding to the target time node, to obtain an intermediate corrected power value. The intermediate correction power sequence data determination unit is used to obtain the intermediate correction power sequence data based on the intermediate correction power values ​​corresponding to each target time node.

[0095] As an optional but non-limiting implementation, the effective iteration result determination module 340 further includes a residual evaluation function construction unit and an effective iteration result determination unit. Wherein: The residual evaluation function construction unit is used to construct a residual evaluation function based on the power residuals of the intermediate corrected power sequence data and the real-time power sequence data at each time node. The effective iteration result determination unit is used to determine the effective iteration result corresponding to each time node based on the residual evaluation function and the preset residual threshold.

[0096] As an optional but non-limiting implementation, the data update module 350 includes a set determination unit, a first update unit, and a second update unit. Wherein: The set determination unit is used to determine the set of converged nodes and the set of non-converged nodes based on the effective iteration results corresponding to each time node. The first update unit is used to determine the deviation thermal field value corresponding to each first time node in the convergence node set based on the deviation thermal field value sequence, and update the power data corresponding to each first time node in the original predicted power sequence data based on each deviation thermal field value. The second update unit is used to re-execute the iterative solution process of the heat conduction equation for each second time node included in the set of unconverged nodes until the preset termination condition is met, and update the power data corresponding to the second time node in the original predicted power sequence data based on the latest obtained deviation thermal field value sequence.

[0097] The device for correcting the predicted power data of a photovoltaic power generation system provided in this embodiment of the invention can be used to execute the method for correcting the predicted power data of a photovoltaic power generation system, and has the corresponding functional modules and beneficial effects for executing the method for correcting the predicted power data of a photovoltaic power generation system.

[0098] It is worth noting that the various units and modules included in the above-mentioned device are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the protection scope of the embodiments of the present invention.

[0099] Figure 4 This is a schematic diagram of an electronic device for implementing a method for correcting predicted power data of a photovoltaic power generation system, as provided in an embodiment of the present invention. The following refers to... Figure 4 The diagram illustrates a structural schematic of an electronic device 410 suitable for implementing embodiments of the present invention. The terminal devices in these embodiments may include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 4 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0100] like Figure 4 As shown, the electronic device 410 includes at least one processor 411 and a memory, such as a read-only memory (ROM) 412 and a random access memory (RAM) 413, communicatively connected to the at least one processor 411. The memory stores computer programs executable by the at least one processor. The processor 411 can perform various appropriate actions and processes based on the computer program stored in the ROM 412 or loaded from storage unit 418 into the RAM 413. The RAM 413 may also store various programs and data required for the operation of the electronic device 410. The processor 411, ROM 412, and RAM 413 are interconnected via a bus 414. An input / output (I / O) interface 415 is also connected to the bus 414.

[0101] Multiple components in electronic device 410 are connected to input / output (I / O) interface 415, including: input unit 416, such as keyboard, mouse, etc.; output unit 417, such as various types of monitors, speakers, etc.; storage unit 418, such as disk, optical disk, etc.; and communication unit 419, such as network card, modem, wireless transceiver, etc. Communication unit 419 allows electronic device 410 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0102] Processor 411 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 411 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 411 executes the method for correcting the predicted power data of the photovoltaic power generation system provided in any embodiment of the present invention.

[0103] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing a method for correcting predicted power data of a photovoltaic power generation system as shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication unit 419, or installed from storage unit 418, or installed from read-only memory (ROM) 412. When the computer program is executed by processor 411, it performs the functions defined in the method for correcting predicted power data of a photovoltaic power generation system according to embodiments of the present invention.

[0104] The names of the messages or information exchanged between the multiple devices in the embodiments of the present invention are for illustrative purposes only and are not intended to limit the scope of these messages or information.

[0105] The electronic device provided in this embodiment of the invention and the method for correcting the predicted power data of the photovoltaic power generation system provided in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0106] This invention provides a computer storage medium storing a computer program that, when executed by a processor, implements the method for correcting predicted power data of a photovoltaic power generation system provided in the above embodiments.

[0107] It should be noted that the computer-readable medium described above in this invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0108] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0109] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0110] The units described in the embodiments of the present invention can be implemented in software or hardware. The names of the units are not, in some cases, intended to limit the specific unit.

[0111] 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-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0112] In the context of this invention, a machine-readable 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 machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. Machine-readable media can include, but are 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.

[0113] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-disclosed concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0114] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in sequential order. In certain circumstances, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of the invention. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0115] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for correcting predicted power data of a photovoltaic power generation system, characterized in that, The method includes: A power deviation sequence is constructed based on real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system; wherein, the power deviation sequence is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points; The power deviation sequence is discretized over time, and a one-dimensional heat conduction equation is constructed based on the power deviation sequence and the deviation change gradient. The one-dimensional heat conduction equation is solved iteratively using the finite difference method, and a sequence of deviation heat field values ​​is output. The sequence of deviation heat field values ​​indicates the deviation heat field value at each time point. The iteration step size used in the iterative solution process is adaptively adjusted based on the actual power fluctuation amplitude. The original predicted power sequence data is fused and corrected based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data. Based on the residual between the intermediate corrected power sequence data and the real-time power sequence data, the effective iteration results corresponding to each time node are determined. For the first time node, the power data corresponding to the first time node in the original predicted power sequence data is updated based on the deviation thermal field value sequence; for the second time node, the iterative solution process of the heat conduction equation is re-executed until the preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition; After updating the power data for all time points in the original predicted power sequence data, the corrected power sequence data is obtained.

2. The method according to claim 1, characterized in that, The power deviation sequence is constructed based on the real-time power sequence data and the original predicted power sequence data of the photovoltaic power generation system, including: Acquire the real-time power sequence data corresponding to the actual power generation of the photovoltaic power generation system, and the original predicted power sequence data obtained based on the power prediction model; Based on the power deviation between the real-time power sequence data and the original predicted power sequence data at each time point, an original power deviation sequence is constructed. The original power deviation sequence is normalized to obtain the power deviation sequence.

3. The method according to claim 1, characterized in that, The process of discretizing the power deviation sequence over time and constructing a one-dimensional heat conduction equation based on the power deviation sequence and the deviation change gradient includes: The power deviation sequence is discretized according to a preset time interval to obtain multiple time nodes and the power deviation value corresponding to each time node. Based on the aforementioned deviation change gradient and gradient driving coefficient, a driving source term for the one-dimensional heat conduction equation is constructed. The power deviation values ​​corresponding to each time node are assigned one by one to the time nodes corresponding to the one-dimensional heat diffusion equation to serve as thermal field boundary anchor points; and the one-dimensional heat conduction equation is constructed based on the driving source term and the heat diffusion law.

4. The method according to claim 1, characterized in that, The iterative solution of the one-dimensional heat conduction equation using the finite difference method outputs a sequence of deviation thermal field values, including: Based on the real-time power sequence data, a power volatility function is constructed; The iteration step size function is determined based on the power volatility function; wherein, the iteration step size function is used to determine the iteration step size used in the iterative solution process; Based on the iterative step size function, the one-dimensional heat conduction equation is solved iteratively using the finite difference method to obtain the sequence of deviation heat field values.

5. The method according to claim 1, characterized in that, The process of fusing and correcting the original predicted power sequence data based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data includes: For a target time node, the deviation thermal field value corresponding to the target time node is determined based on the deviation thermal field value sequence, and the original predicted power value corresponding to the target time node is determined based on the original predicted power sequence data; Based on the original power deviation sequence, determine the maximum and minimum power deviation values; Based on the deviation thermal field value, the maximum power deviation value, and the minimum power deviation value corresponding to the target time node, the original predicted power value corresponding to the target time node is fused and corrected to obtain the intermediate corrected power value. The intermediate correction power sequence data is obtained based on the intermediate correction power values ​​corresponding to each target time node.

6. The method according to claim 1, characterized in that, The determination of the effective iteration results corresponding to each time node based on the residual between the intermediate corrected power sequence data and the real-time power sequence data includes: Based on the power residuals of the intermediate corrected power sequence data and the real-time power sequence data at each time node, a residual evaluation function is constructed. Based on the residual evaluation function and the preset residual threshold, the effective iteration results corresponding to each time node are determined.

7. The method according to claim 1, characterized in that, For the first time node, the power data corresponding to the first time node in the original predicted power sequence data is updated based on the deviation thermal field value sequence; for the second time node, the iterative solution process of the heat conduction equation is re-executed until a preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence, including: Based on the valid iteration results corresponding to each time node, determine the set of converged nodes and the set of non-converged nodes; Based on the deviation thermal field value sequence, the deviation thermal field value corresponding to each first time node included in the convergence node set is determined, and the power data corresponding to each first time node in the original predicted power sequence data is updated based on each deviation thermal field value. For each second time node included in the set of unconverged nodes, the iterative solution process of the heat conduction equation is re-executed until the preset termination condition is met, and the power data corresponding to the second time node in the original predicted power sequence data is updated based on the latest obtained deviation thermal field value sequence.

8. A device for correcting predicted power data of a photovoltaic power generation system, characterized in that, The device includes: A power deviation sequence construction module is used to construct a power deviation sequence based on real-time power sequence data and original predicted power sequence data of a photovoltaic power generation system; wherein, the power deviation sequence is used to indicate the power deviation value between the real-time power sequence data and the original predicted power sequence data at different time points; A one-dimensional heat conduction equation construction module is used to perform time discretization processing on the power deviation sequence and construct a one-dimensional heat conduction equation based on the power deviation sequence and the deviation change gradient. The deviation thermal field value sequence determination module is used to iteratively solve the one-dimensional heat conduction equation using the finite difference method and output the deviation thermal field value sequence; wherein, the deviation thermal field value sequence is used to indicate the deviation thermal field value at each time node; the iteration step size used in the iterative solution process is adaptively adjusted based on the actual power fluctuation amplitude; The effective iteration result determination module is used to fuse and correct the original predicted power sequence data based on the deviation thermal field value sequence to obtain intermediate corrected power sequence data, and to determine the effective iteration result corresponding to each time node based on the residual between the intermediate corrected power sequence data and the real-time power sequence data. The data update module is used to update the power data corresponding to the first time node in the original predicted power sequence data based on the deviation thermal field value sequence for the first time node; and to re-execute the iterative solution process of the heat conduction equation for the second time node until a preset termination condition is met, and update the power data corresponding to the second time node in the original predicted power sequence data based on the latest obtained deviation thermal field value sequence; wherein, the first time node is the time node where the effective iteration result meets the convergence condition, and the second time node is the time node where the effective iteration result does not meet the convergence condition; The corrected power sequence data determination module is used to obtain corrected power sequence data after updating the power data corresponding to all time points in the original predicted power sequence data.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; Storage device for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method for correcting the predicted power data of the photovoltaic power generation system as described in any one of claims 1-7.

10. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform a method for correcting the predicted power data of a photovoltaic power generation system as described in any one of claims 1-7.

Images