Power data interpolation compensation method, system, equipment and medium for grid-connected ride-through electric energy of new energy power plant

Abstract

The invention discloses a power data interpolation compensation method, system and device for grid-connected ride-through electric energy of a new energy power plant and a medium, and relates to the technical field of electrical measurement. The method comprises the steps of obtaining a power discrete data sequence, calculating to obtain a power fluctuation function, extracting fluctuation characteristic representation parameters, generating a dynamic fluctuation factor, constructing a dynamic interpolation primary function, and calculating a power sequence interpolation function to realize power compensation. The system comprises an acquisition module, a feature extraction module, a factor generation module, a primary function construction module and an interpolation calculation module. The method aims at fast nonlinear fluctuation of power flow in a new energy grid connection process, characteristic parameters capable of representing fluctuation essence are extracted in real time, core parameters such as scale factors and derivative weights of an interpolation function are dynamically adjusted through the characteristic parameters, a self-adaptive mixed interpolation model is formed, global nonlinear trends and local dynamic details are captured at the same time, and the method has the advantages of being simple in structure and convenient to operate. The problem of key dynamic feature loss caused by traditional fixed interval sampling is solved.

Description

Technical Field

[0001] This invention relates to the field of electrical measurement technology, and in particular to a method, system, equipment, and medium for power data interpolation compensation of grid-connected power energy of new energy power plants. Background Technology

[0002] With the rapid increase in the proportion of renewable energy generation in the power system, renewable energy grid connection technology has become one of the core areas of power system research. Renewable energy generation has significant intermittent, random, and fluctuating characteristics, resulting in highly complex and dynamically changing power flow distribution during grid connection. This makes it difficult for power flow analysis to provide high-frequency, high-precision power data by tracking power flow changes in real time, thus failing to guarantee the accuracy, fairness, and efficiency of power settlement.

[0003] In existing data acquisition systems, to ensure the stability and reliability of power grid operation, data acquisition typically employs fixed acquisition intervals to meet the needs of routine electricity metering and monitoring. However, with the large-scale integration of renewable energy generation into the grid, fixed-interval data acquisition methods cannot capture key dynamic characteristics, cannot adapt to the complexity of power flow changes, and their data quality is detrimental to subsequent electricity settlement.

[0004] Furthermore, the power data acquisition and monitoring control system relies on specific sensors, data acquisition terminals, and communication equipment. Its initial design goal is not to focus on the real-time processing of high-frequency dynamic data, but rather on the long-term stable operation and maintenance of the system. If the acquisition interval is reduced, all hardware, communication networks, and software systems in the current power data acquisition and monitoring control system for the scenario of new energy grid connection will need to be comprehensively upgraded, which will significantly increase the system construction and maintenance costs.

[0005] Meanwhile, traditional power data acquisition and monitoring control systems have low clock synchronization accuracy, making it difficult to support data synchronization under high-frequency acquisition and failing to meet the requirement of multi-node synchronous acquisition to ensure data consistency for new energy grid connection.

[0006] Existing power analysis methods mainly rely on static data acquisition techniques and fixed interpolation algorithms. These methods are effective in handling the stable operating environment of traditional power grids, but their adaptability is significantly insufficient in the context of renewable energy grid integration due to the rapid changes and nonlinear characteristics of power flow.

[0007] Traditional cubic Hermite interpolation constructs a cubic polynomial using known points and their derivatives to estimate data values ​​at unknown points. However, in renewable energy grid-connected scenarios, power flow data exhibits highly nonlinear fluctuations due to its intermittent and random nature. Cubic Hermite interpolation relies on the derivatives of local data points, making it difficult to accurately capture complex nonlinear trends. The interpolation results easily deviate from the actual power flow distribution, leading to distortions in power analysis. Traditional fractal interpolation generates self-similar interpolation curves through iterative function systems, suitable for simulating data sequences with fractal characteristics. However, fractal interpolation requires precise parameter settings, and the diverse data characteristics in renewable energy grid-connected scenarios make it difficult to determine uniform interpolation parameters. Traditional exponentially weighted moving average algorithms smooth data fluctuations by assigning exponentially decaying weights to historical data, but their strong dependence on historical data and the resulting weight allocation lead to a slow response to rapidly changing power flow data.

[0008] Therefore, there is currently a lack of effective interpolation compensation solutions for how to accurately collect changes in power data in real time under random power flow scenarios when new energy is connected to the grid. There is an urgent need to study a dynamic data interpolation compensation method for power analysis under random power flow scenarios when new energy is connected to the grid, so as to fully ensure the accuracy and reliability of power data collected by smart meters during long-term use. Summary of the Invention

[0009] Therefore, the technical problem to be solved by the present invention is that the traditional power data acquisition and monitoring control system, due to its fixed acquisition interval, is difficult to capture the rapid dynamic changes in power flow.

[0010] The above-mentioned technical problems are solved by the following technical solutions: A power data interpolation compensation method for grid-connected power generation in new energy power plants includes: acquiring a discrete power data sequence under grid-connected scenarios; calculating the cumulative deviation after removing the mean from the discrete power data sequence and removing the local linear trend of the cumulative deviation to obtain a power fluctuation function; analyzing the multi-scale characteristics of the power fluctuation function, extracting fluctuation feature characterization parameters, and quantifying the fractal characteristics of the sequence under different fluctuation amplitudes; calculating the power fractal energy distribution based on the fluctuation feature characterization parameters, and generating a dynamic fluctuation factor by combining time-series weighting processing; based on the dynamic fluctuation factor, and combining the local and global statistical characteristics of the discrete power data sequence, introducing multi-parameter dynamic correction to the standard interpolation basis function to construct a dynamic interpolation basis function that adapts to the local dynamic details of power fluctuations; and calculating the power sequence interpolation function based on the dynamic interpolation basis function and conventional coefficient weights to achieve power data compensation.

[0011] In a preferred embodiment of the power data interpolation compensation method for grid-connected power generation of new energy power plants according to the present invention: the step of analyzing the multi-scale characteristics of the power fluctuation function and extracting fluctuation feature characterization parameters includes: constructing q-order fluctuation functions with different moment orders based on the power fluctuation function to describe the influence of fluctuations of different amplitudes; establishing the proportional relationship between the q-order fluctuation function and the time scale; and solving for the fluctuation feature characterization parameters that can quantify the fractal characteristics of the sequence by fitting the proportional relationship.

[0012] In a preferred embodiment of the power data interpolation compensation method for grid-connected power transmission of new energy power plants according to the present invention: the step of calculating the power fractal energy distribution based on the fluctuation characteristic characterization parameters and generating a dynamic fluctuation factor by combining time-series weighted processing includes: calculating the power fractal energy distribution at different time scales by integrally coupling the fluctuation characteristic characterization parameters at different moment orders; introducing a time-series weighted algorithm to the power fractal energy distribution to extract multi-scale fluctuation characteristics and describe the overall change trend of the power fractal energy distribution; and using the overall change trend as the dynamic fluctuation factor.

[0013] In a preferred embodiment of the power data interpolation compensation method for grid-connected power generation of new energy power plants according to the present invention: the step of introducing multi-parameter dynamic correction to the standard interpolation basis function based on the dynamic fluctuation factor and combined with the local and global statistical characteristics of the power discrete data sequence to construct the dynamic interpolation basis function includes: obtaining the standard interpolation basis function; based on the dynamic fluctuation factor and combined with the local and global mean and variance parameters of the power discrete data sequence; introducing the multi-parameter dynamic correction to the standard interpolation basis function to generate the dynamic interpolation basis function that can adapt to power fluctuations.

[0014] In a preferred embodiment of the power data interpolation compensation method for grid-connected power generation of new energy power plants according to the present invention: the fluctuation characteristic characterization parameter is the generalized Hurst exponent; the construction of q-order fluctuation functions under different moment orders includes: taking the power fluctuation function to the power of q; summing and averaging the result of the q-th power, and then taking 1 / q to obtain the q-order fluctuation function; the step of obtaining the fluctuation characteristic characterization parameter that can quantify the fractal characteristics of the sequence by fitting the proportional relationship includes: establishing the power law relationship between the average value of the q-order fluctuation function on the time scale and the time scale; fitting the power law relationship by the least squares method to obtain the generalized Hurst exponent.

[0015] In a preferred embodiment of the power data interpolation compensation method for grid-connected power energy of new energy power plants according to the present invention: the time-series weighted algorithm is an exponentially weighted moving average algorithm; the introduction of the time-series weighted algorithm into the power fractal energy distribution includes: introducing a fractal periodic term into the exponentially weighted moving average algorithm; substituting the power fractal energy distribution, calculating the moving weighted average of the power fractal energy distribution, and obtaining the dynamic fluctuation factor.

[0016] In a preferred embodiment of the power data interpolation compensation method for grid-connected power generation of new energy power plants described in this invention: the standard interpolation basis function is a standard cubic Hermitian basis function; the dynamic interpolation basis function is a time-varying Hermitian piecewise interpolation basis function; the multi-parameter dynamic correction includes: introducing a maximum volatility factor and an exponential decay factor to suppress interpolation overshoot; introducing a minimum volatility factor and a gradient gain factor to quickly respond to weak fluctuations; introducing local and global mean and variance parameters to dynamically scale the basis function to adapt to load baseline drift and volatility.

[0017] A power data interpolation compensation system for grid-connected power generation in new energy power plants includes: an acquisition module for acquiring discrete power data sequences under grid-connected scenarios of new energy power plants; a feature extraction module for calculating the cumulative deviation after removing the mean from the discrete power data sequences, removing the local linear trend of the cumulative deviation to obtain a power fluctuation function, analyzing the multi-scale characteristics of the power fluctuation function, and extracting fluctuation feature characterization parameters; a factor generation module for calculating the power fractal energy distribution based on the fluctuation feature characterization parameters, and generating a dynamic fluctuation factor by combining time-series weighting processing; a basis function construction module for introducing multi-parameter dynamic correction to the standard interpolation basis function based on the dynamic fluctuation factor and the local and global statistical characteristics of the discrete power data sequences to construct a dynamic interpolation basis function; and an interpolation calculation module for calculating the power sequence interpolation function based on the dynamic interpolation basis function and conventional coefficient weights to achieve power data compensation.

[0018] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the power data interpolation compensation method for grid-connected power energy of a new energy power plant as described above.

[0019] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the power data interpolation compensation method for grid-connected power energy of a new energy power plant as described above.

[0020] The beneficial effects of this invention are as follows: In response to the rapid nonlinear fluctuations in power flow during the grid connection of new energy sources, feature parameters that can characterize the nature of the fluctuations are extracted in real time. By dynamically adjusting the scaling factor, derivative weight, and other core parameters of the interpolation function through the feature parameters, an adaptive hybrid interpolation model is formed to simultaneously capture global nonlinear trends and local dynamic details, thus solving the problem of loss of key dynamic features caused by traditional fixed-interval sampling. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Obviously, the drawings described below only relate to some embodiments of the present invention and are not intended to limit the present invention.

[0022] Figure 1 A flowchart is shown for the power data interpolation compensation method for grid-connected power energy of new energy power plants. Figure 2 The acquisition interval is shown. The photovoltaic power generation data is 15 minutes from the next day. Figure 3 The cumulative bias of the power series after mean removal is shown. ; Figure 4 The generalized Hurst exponent is shown. How it varies with the order of the moment q; Figure 5 The time scale is shown. Power fractal energy distribution under ; Figure 6 The results of the FEWMA interpolation function are shown. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0024] The terminology used in this invention is that which is currently widely used in the art in consideration of the function of the invention; however, these terms may vary according to the intent of those skilled in the art, precedent, or new technology in the art. Furthermore, specific terms may be chosen by the applicant, and in such cases, their detailed meanings will be described in the detailed description of the invention. Therefore, the terms used in this specification should not be construed as simple names, but rather based on their meanings and the overall description of the invention.

[0025] Example 1, referring to Figure 1 This embodiment provides a power data interpolation compensation method for grid-connected power generation of new energy power plants, including: S1: Obtain the discrete power data sequence under the new energy grid connection scenario, establish the original data foundation to be analyzed, and ensure that subsequent processing can be based on the real power flow state of the power grid, so as to provide an accurate input source for subsequent fluctuation feature extraction and interpolation calculation.

[0026] S2: Calculate the cumulative deviation after removing the mean from the discrete power data sequence, and remove the local linear trend of the cumulative deviation to obtain the power fluctuation function.

[0027] This removes the interference of mean background and local linear trends in the power data, thereby highlighting the essential fluctuation characteristics of the current sequence and ensuring that subsequent fractal analysis can focus on the complex details of nonlinear fluctuations rather than simple trend changes.

[0028] S3: Analyze the multi-scale characteristics of the power fluctuation function, extract fluctuation feature characterization parameters, and quantify the fractal characteristics of the sequence under different fluctuation amplitudes.

[0029] It is used to effectively distinguish and quantify the fractal characteristics of large fluctuations such as sudden power output changes and background noise with similar small-amplitude disturbances, thereby providing a precise quantitative basis driven by both physics and data for dynamic power interpolation.

[0030] S4: Calculate the power fractal energy distribution based on the fluctuation characteristic characterization parameters, and generate a dynamic fluctuation factor by combining time-series weighting processing.

[0031] Introducing a fractal periodic term can compensate for the lag in the response of the traditional exponentially weighted moving average algorithm to rapidly changing power flows, enabling the generated volatility factor to simultaneously adapt to both the long-term memory characteristics and short-term sudden changes in the load.

[0032] S5: Based on the dynamic fluctuation factor and combined with the local and global statistical characteristics of the power discrete data sequence, a multi-parameter dynamic correction is introduced to the standard interpolation basis function to construct a dynamic interpolation basis function that adapts to the local dynamic details of power fluctuation.

[0033] The dynamic correction mechanism is used to suppress overshoot during the interpolation process and provide a fast response under weak fluctuations. It also adapts to load baseline drift and high variance ranges, solving the problem that traditional fixed interpolation basis functions are not good at capturing complex nonlinear fluctuations.

[0034] S6: Based on the dynamic interpolation basis function and the conventional coefficient weights, the power sequence interpolation function is calculated to achieve power data compensation.

[0035] This enables high-precision data reconstruction that simultaneously captures global nonlinear trends and local dynamic details, effectively solving the problem of key dynamic feature loss caused by traditional fixed-interval sampling without increasing hardware sampling costs.

[0036] It should be noted that with the increasing proportion of renewable energy generation, its significant intermittent, random, and volatile characteristics lead to a highly complex and dynamically changing power flow distribution. Traditional power data acquisition and monitoring control systems typically use fixed acquisition intervals, which cannot capture key dynamic characteristics when faced with rapidly changing power flow fluctuations. If the acquisition interval is reduced simply to obtain high-frequency data, a comprehensive upgrade of existing hardware, communication networks, and software systems is required, resulting in high construction and maintenance costs. In addition, existing interpolation methods have obvious limitations: traditional cubic Hermitian interpolation relies on local derivatives, making it difficult to capture nonlinear changes and easily leading to distorted results; fractal interpolation requires precise parameter settings and is difficult to adapt to changing data characteristics; while the exponentially weighted moving average algorithm can smooth fluctuations, it has a lag in response to rapidly changing power flow.

[0037] Therefore, to address the aforementioned problems, this invention integrates the fractal fluctuation characteristics of power sequences at different time scales and amplitudes with an exponentially weighted moving average algorithm that incorporates fluctuation trends. This enables real-time extraction of feature parameters that characterize the essence of fluctuations, providing a dual-driven basis for interpolation based on both physical and data aspects. This invention solves the compatibility problem between power data fractal interpolation and cubic Hermitian interpolation by dynamically adjusting core parameters such as the scaling factor and derivative weights of the interpolation function using feature parameters, forming an adaptive hybrid interpolation model. This method can simultaneously capture the global nonlinear trend and local dynamic details of grid-connected renewable energy and power, overcoming the lag and inadequacy of traditional algorithms. Thus, without increasing hardware costs, it effectively solves the problem of key dynamic feature loss caused by traditional fixed-interval sampling, significantly improving the accuracy and reliability of interpolated data.

[0038] Example 2, refer to Figure 1 This is the second embodiment of the present invention. Based on embodiment 1, this embodiment further refines and explains each step, including: This process involves acquiring discrete power data sequences for new energy grid-connected scenarios; calculating the cumulative deviation after removing the mean from the discrete power data sequences and removing the local linear trend of the cumulative deviation to obtain the power fluctuation function; analyzing the multi-scale characteristics of the power fluctuation function, extracting fluctuation feature characterization parameters, and quantifying the fractal characteristics of the sequence under different fluctuation amplitudes; calculating the power fractal energy distribution based on the fluctuation feature characterization parameters, and generating a dynamic fluctuation factor by combining time-series weighting processing; based on the dynamic fluctuation factor and combining the local and global statistical characteristics of the discrete power data sequences, introducing multi-parameter dynamic correction to the standard interpolation basis function to construct a dynamic interpolation basis function that adapts to the local dynamic details of power fluctuations; and calculating the power sequence interpolation function based on the dynamic interpolation basis function and conventional coefficient weights to achieve power data compensation.

[0039] This invention first eliminates the DC component and background trend interference in the original power data by removing the mean and local trends, allowing subsequent analysis to focus on the nonlinear fluctuation characteristics of the power signal. Furthermore, by extracting fractal features and combining them with a time-weighted dynamic fluctuation factor as a parameter to adjust the shape of the interpolation function, the interpolation basis function can adaptively adjust according to the severity of power fluctuations, historical trends, and current statistical characteristics. This overcomes the limitation of traditional static interpolation methods that apply a uniform processing approach to all data segments, maintaining smoothness during stable periods and capturing details during periods of severe fluctuation, thereby improving the accuracy of the interpolation results in reproducing the actual power flow of the power grid.

[0040] Specifically, the analysis of the multi-scale characteristics of the power fluctuation function and the extraction of fluctuation feature characterization parameters include: constructing q-order fluctuation functions with different moment orders based on the power fluctuation function to describe the influence of fluctuations with different amplitudes; establishing the proportional relationship between the q-order fluctuation function and the time scale; and solving for the fluctuation feature characterization parameters that can quantify the fractal characteristics of the sequence by fitting the proportional relationship.

[0041] New energy power data exhibits multifractal characteristics, namely large-amplitude, dramatic fluctuations and small-amplitude background noise, each following different statistical laws. By introducing different moments q, this invention can separately calculate the statistical characteristics of large and small fluctuations in the data. Establishing and fitting the proportional relationship between the fluctuation function and the time scale aims to mathematically define the strength of this self-similarity. This ensures that the extracted feature parameters are not merely simple statistics, but reflect the physical meaning of the complex structure of the power signal at different scales, providing theoretical support for accurately distinguishing the abrupt changes in the signal and avoiding information loss caused by single-scale analysis.

[0042] Specifically, the calculation of power fractal energy distribution based on fluctuation characteristic characterization parameters and the generation of dynamic fluctuation factors by combining time-series weighting processing include: calculating power fractal energy distribution at different time scales by integrally coupling fluctuation characteristic characterization parameters at different moment orders; introducing a time-series weighting algorithm to the power fractal energy distribution to extract multi-scale fluctuation characteristics and describe the overall trend of power fractal energy distribution; and using the overall trend as the dynamic fluctuation factor.

[0043] By integrating features dispersed across different moment orders through integral coupling, an index reflecting the total energy state of power fluctuations at a specific time scale is formed. A time-weighted algorithm is introduced to dynamically track energy distribution. This approach considers both historical fluctuation states and gives greater weight to changes at the current moment, allowing the generated dynamic fluctuation factor to smoothly reflect the evolution of power fluctuation characteristics. This effectively solves the problem of parameter instability caused by relying solely on instantaneous values, while also overcoming the slow response of traditional averaging algorithms to rapid changes, ensuring the timeliness of the interpolation adjustment factor.

[0044] Specifically, based on the dynamic fluctuation factor and combined with the local and global statistical characteristics of the power discrete data sequence, a multi-parameter dynamic correction is introduced into the standard interpolation basis function to construct the dynamic interpolation basis function. This includes: obtaining the standard interpolation basis function; combining the local and global mean and variance parameters of the power discrete data sequence based on the dynamic fluctuation factor; and introducing a multi-parameter dynamic correction into the standard interpolation basis function to generate a dynamic interpolation basis function that can adapt to power fluctuations.

[0045] Standard interpolation basis functions typically assume uniform or smooth data variation, which does not reflect the actual situation of new energy grid connection. This invention introduces local and global mean and variance parameters, providing the interpolation function with a reference for the data distribution environment. By combining these statistical characteristics with a dynamic fluctuation factor, the basis function is modified. In regions with large data fluctuations and high variance, the curvature of the basis function is increased to accommodate abrupt changes, while in regions with stable data and low variance, the rate of change is reduced to maintain smoothness. This ensures that the interpolation curve does not exhibit excessive jitter due to local noise, nor does it lose key power inflection points due to excessive smoothing, thus achieving a balance between global trends and local details.

[0046] Specifically, the fluctuation characteristic characterization parameter is the generalized Hurst exponent; constructing the q-order fluctuation function under different moment orders includes: raising the power fluctuation function to the power power; summing and averaging the result raised to the power of q, and then raising it to the power of 1 / q to obtain the q-order fluctuation function; by fitting the proportional relationship, the fluctuation characteristic characterization parameter that can quantify the fractal characteristics of the sequence is obtained by: establishing the power law relationship between the average value of the q-order fluctuation function on the time scale and the time scale; fitting the power law relationship by the least squares method to obtain the generalized Hurst exponent.

[0047] Using the generalized Hurst exponent as the core feature parameter, and performing power-law calculations and averaging to extract the square root of the power fluctuation function, a multifractal mathematical structure is employed to eliminate trend fluctuation analysis. By fitting the power-law relationship using the least squares method, a robust linear slope can be calculated from noisy data; this slope is the Hurst exponent. The generalized Hurst exponent is an indicator that can describe the long-range correlation and roughness of time series, capturing the implicit nonlinear dynamics in photovoltaic or wind power data, and reflecting the essential characteristics of fluctuations more effectively than ordinary variance or standard deviation.

[0048] Specifically, the time-weighted algorithm is an exponentially weighted moving average algorithm; introducing a time-weighted algorithm into the power fractal energy distribution includes: introducing a fractal periodic term into the exponentially weighted moving average algorithm; substituting the power fractal energy distribution, calculating the moving weighted average of the power fractal energy distribution, and obtaining the dynamic fluctuation factor.

[0049] Traditional exponentially weighted moving averages, while considering historical weights, suffer from lag. This invention introduces a fractal periodic term, utilizing the inherent periodicity of new energy power generation to correct prediction biases. This improvement allows the algorithm to incorporate the periodic components of fractal characteristics, reducing the phase lag of traditional algorithms. This enables the generated dynamic fluctuation factor to more closely and timely match the actual trajectory of power fractal energy changes, thus providing accurate interpolation data during periods of rapid power flow changes.

[0050] Specifically, the standard interpolation basis function is the standard cubic Hermit basis function; the dynamic interpolation basis function is the time-varying Hermit piecewise interpolation basis function; the multi-parameter dynamic correction includes: introducing the maximum volatility factor and the exponential decay factor to suppress interpolation overshoot; introducing the minimum volatility factor and the gradient gain factor to quickly respond to weak volatility; introducing local and global mean and variance parameters to dynamically scale the basis function to adapt to load baseline drift and volatility.

[0051] While standard cubic Hermitian interpolation ensures derivative continuity, it is prone to overshoot when dealing with sharp fluctuations, resulting in interpolation points exceeding the range of adjacent data points. Introducing the maximum fluctuation factor and exponential decay factor aims to limit this overshoot and prevent spurious power spikes in the interpolation. Simultaneously, the minimum fluctuation factor and gradient gain factor provide gain when fluctuations are weak but the changing trend is clear, preventing sluggish interpolation response. Scaling using the mean and variance addresses the load baseline drift problem. The advantage of these comprehensive corrections is that they transform the static mathematical basis function into a dynamic model with constraints and adaptive capabilities, solving the problem that traditional interpolation methods cannot accurately follow data changes in high-frequency ride-through scenarios of new energy grid connection.

[0052] It should be noted that with the increasing proportion of renewable energy generation, its significant intermittent, random, and volatile characteristics lead to a highly complex and dynamically changing power flow distribution. Traditional power data acquisition and monitoring control systems typically use fixed acquisition intervals, which cannot capture key dynamic characteristics when faced with rapidly changing power flow fluctuations. If the acquisition interval is reduced simply to obtain high-frequency data, a comprehensive upgrade of existing hardware, communication networks, and software systems is required, resulting in high construction and maintenance costs. In addition, existing interpolation methods have obvious limitations: traditional cubic Hermitian interpolation relies on local derivatives, making it difficult to capture nonlinear changes and easily leading to distorted results; fractal interpolation requires precise parameter settings and is difficult to adapt to changing data characteristics; while the exponentially weighted moving average algorithm can smooth fluctuations, it has a lag in response to rapidly changing power flow.

[0053] Therefore, to address the aforementioned problems, this invention integrates the fractal fluctuation characteristics of power sequences at different time scales and amplitudes with an exponentially weighted moving average algorithm that incorporates fluctuation trends. This enables real-time extraction of feature parameters that characterize the essence of fluctuations, providing a dual-driven basis for interpolation based on both physical and data aspects. This invention solves the compatibility problem between power data fractal interpolation and cubic Hermitian interpolation by dynamically adjusting core parameters such as the scaling factor and derivative weights of the interpolation function using feature parameters, forming an adaptive hybrid interpolation model. This method can simultaneously capture the global nonlinear trend and local dynamic details of grid-connected renewable energy and power, overcoming the lag and inadequacy of traditional algorithms. Thus, without increasing hardware costs, it effectively solves the problem of key dynamic feature loss caused by traditional fixed-interval sampling, significantly improving the accuracy and reliability of interpolated data.

[0054] Example 3, referring to Figure 1 This embodiment provides a power data interpolation compensation system for grid-connected power generation of new energy power plants, including a data acquisition module, a feature extraction module, a factor generation module, a basis function construction module, and an interpolation calculation module.

[0055] Specifically, the acquisition module is used to acquire discrete power data sequences in new energy grid-connected scenarios.

[0056] The data acquisition module, serving as the system's input interface, is responsible for reading raw power data at specific time intervals from smart meters, data acquisition terminals, or historical databases. Its primary purpose is to establish the foundation of the raw data to be analyzed, ensuring that subsequent processing is based on the actual power flow conditions of the power grid. The acquisition module can interface with hardware devices operating at different sampling frequencies, providing an accurate and standardized input source for subsequent fluctuation feature extraction and interpolation calculations.

[0057] Specifically, the feature extraction module is used to calculate the cumulative deviation after removing the mean from the discrete power data sequence, remove the local linear trend of the cumulative deviation to obtain the power fluctuation function, analyze the multi-scale characteristics of the power fluctuation function, and extract the fluctuation feature characterization parameters.

[0058] The core algorithm of the feature extraction module integrates detrended fluctuation analysis logic. First, it eliminates DC components and background trend interference in the original power data by removing the mean and local trends, thus highlighting the essential fluctuation characteristics of the current sequence. Then, it analyzes multi-scale characteristics and extracts parameters such as the generalized Hurst exponent to quantify the fractal features of the sequence under different fluctuation amplitudes. The feature extraction module can distinguish and quantify the statistical regularities of large fluctuations and background noise, providing a physical-level quantitative basis for subsequent interpolation and avoiding analytical biases caused by neglecting nonlinear characteristics in traditional methods.

[0059] Specifically, the factor generation module is used to calculate the power fractal energy distribution based on the fluctuation characteristic characterization parameters, and combine it with time-series weighting processing to generate dynamic fluctuation factors.

[0060] The factor generation module is responsible for transforming static fractal features into dynamic time-series parameters. It calculates the power fractal energy distribution through integral coupling and incorporates time-series weighting algorithms such as exponentially weighted moving averages for processing. Factor generation can extract multi-scale fluctuation characteristics and describe the overall trend of power fractal energy distribution, while using fractal periodic terms to correct prediction biases. The factor generation module can overcome the lag in response to rapidly changing power flows in traditional algorithms, enabling the generated dynamic fluctuation factors to simultaneously reflect both the long-term memory characteristics and short-term sudden changes in the load, ensuring the timeliness of the adjustment factors.

[0061] Specifically, the basis function construction module is used to introduce multi-parameter dynamic corrections to the standard interpolation basis function based on the dynamic fluctuation factor and the local and global statistical characteristics of the power discrete data sequence, thereby constructing a dynamic interpolation basis function.

[0062] The basis function construction module is the core building block for adaptive interpolation. It utilizes dynamic fluctuation factors and statistical properties of the data, such as mean and variance, to modify the standard cubic Hermitian basis function. Multi-parameter correction is introduced to adjust the shape of the basis function; for example, it suppresses overshoot using the maximum fluctuation factor, improves response speed under weak fluctuations using the gradient gain factor, and adapts to baseline drift using statistical parameters. The basis function construction module transforms fixed-shape mathematical basis functions into dynamic models capable of adapting to the local dynamic details of power fluctuations, thus solving the problem of insufficient adaptability of traditional interpolation basis functions in complex nonlinear fluctuation scenarios.

[0063] Specifically, the interpolation calculation module is used to calculate the power sequence interpolation function based on the dynamic interpolation basis function and the conventional coefficient weights, thereby realizing power data compensation.

[0064] The interpolation module is responsible for performing the final numerical calculations. It combines the constructed dynamic basis functions with conventional interpolation coefficients to generate interpolation functions for continuous or high-frequency power sequences. While ensuring the accuracy of data trends, the interpolation module supplements missing information between sampling points, achieving high-precision data reconstruction that simultaneously captures global nonlinear trends and local dynamic details. Without increasing hardware sampling costs, it effectively solves the problem of losing key dynamic features caused by traditional fixed-interval sampling, significantly improving the accuracy and reliability of electricity metering and analysis data.

[0065] Example 4, refer to Figure 1 This embodiment provides a power data interpolation compensation method for grid-connected power generation of new energy power plants, including: Step A01: Collect data on the current grid connection scenario of new energy power plants, with a data collection interval of [missing information]. A discrete power data sequence within a day, where The power data at time t can be denoted as .

[0066] like Figure 2 As shown, this case uses a data collection interval. This is the photovoltaic power generation data for the next day, 15 minutes from now.

[0067] Step A02: For a power data sequence of length N Calculate the cumulative deviation of the power series after removing the mean. Its calculation expression is as follows: Its main purpose is to remove the influence of the mean in the power data in order to highlight the fluctuation characteristics of the current sequence.

[0068] like Figure 3 The figure shows the cumulative deviation of the obtained power sequence. .

[0069] Step A03: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context.] Divided into Given an interval of length s, we can know... For each subinterval v of length s, the power fluctuation function is obtained. Its calculation expression is as follows: In the formula A first-order linear polynomial is fitted for each subinterval v to describe the local linear trend of each subinterval, and its calculation expression is as follows: In the formula Denotes the center of subinterval v. and For the coefficient solution: , Power fluctuation function That is, to further remove the local linear trend within each interval, the remaining trend reflects the fluctuation characteristic function of each sub-interval v of length s.

[0070] Step A04: Calculate the current discrete power sequence on a time scale of Fractal fluctuation characteristics under generalized Hurst exponent The generalized Hurst exponent is defined and calculated as follows: In the formula For time scale parameters, usually , representing the current time collection interval , Represents power at The average value over the time scale, where q is the order of moments, is used to describe the fluctuation amplitude of the current power sequence relative to the generalized Hurst exponent. The impact, when hour, The focus will be on characterizing the fractal features of sequences under large fluctuations, when hour, The focus will be on characterizing the fractal features of small fluctuations such as background noise, with L being a constant coefficient used to describe the power at different moment orders q. Average over time scale and generalized Hurst exponent A fixed proportional relationship. When the moment order is q, the q-th order oscillation function... With power fluctuation function The relational expression is as follows: Least squares fitting can usually be used. This allows us to directly obtain the generalized Hurst exponent. .

[0071] like Figure 4 As shown, this is the result of least squares fitting in this case. This allows us to directly obtain the generalized Hurst exponent. result.

[0072] Step A05: Calculate the time scale Power fractal energy distribution under The time scale is as follows: Fractal energy distribution with different moment orders q The calculation expression is as follows: Reflects different time scales The contribution to fractal energy is calculated by integrating different moment orders. The following corresponds to the generalized Hurst exponent. Coupling is performed to centrally reflect the contribution of the time scale to the fractal energy, where... The normalization constant satisfies: like Figure 5 The figure shows the time scale. Power fractal energy distribution under .

[0073] Step A06: Calculate the time scale as follows Time-power fractal energy dynamic fluctuation factor By introducing a fractal periodic term into the exponentially weighted moving average algorithm, the power fractal energy distribution is incorporated. The time scale can be obtained as follows Time-power fractal energy dynamic fluctuation factor The calculation expression is as follows: In the formula The decay factor in the exponentially weighted moving average algorithm is usually taken as... By calculating the moving weighted average process of power fractal energy, multi-scale fluctuation characteristics were extracted, describing the degree to which the overall trend of the current power fractal energy distribution is affected by time, and enabling the interpolation function to adapt to the long-term memory and short-term burstiness of the load.

[0074] Step A07: Based on the time scale Time-power fractal energy dynamic fluctuation factor Construct time-varying Hermitian piecewise interpolation basis functions For each time interval The standard cubic Hermitian basis functions are defined as follows: In the formula For normalized time variables, For the collection interval, To ensure that the Hermitian basis functions can respond to the corresponding historical power fluctuation characteristics, for each basis function segment, based on its properties with respect to the final interpolation function, a power fractal energy dynamic fluctuation factor is used. By introducing different dynamic corrections, the time-varying Hermite piecewise interpolation basis functions are finally obtained. as follows: In the formula for The maximum volatility factor occurring within the time period, its initial value is The maximum allowable fluctuation threshold of the time system. The exponential decay factor describes the dynamic fluctuation factor of power fractal energy. When exceeding historical norms, basis functions The amplitude decays exponentially to suppress overshoot. for The minimum volatility factor occurring within the time period, its initial value is The minimum allowable fluctuation threshold of the time system. The gradient gain factor describes the dynamic fluctuation factor of power fractal energy. When the value is small, the basis functions are adjusted according to the historical power gradient factor. Provide gain so that the basis functions It is capable of responding quickly to weak volatility factors; and They are respectively exist Mean and variance over a period of time for The global mean parameter of the power data at any given time reflects the ratio of the local to the global mean, dynamically scaled by the basis function. The process of adapting to load baseline drift; for The global variance parameter of the power data at any given time reflects the fitting ability of the interpolation function to high variance intervals; that is, when the local volatility increases, the basis function... The process should involve correspondingly increasing its own curvature.

[0075] Step A08: Based on time-varying Hermitian piecewise interpolation basis functions Obtain the discrete power sequence in the time interval Interpolation function within : in The standard coefficient weights are calculated using the following expression: Time-varying Hermitian piecewise interpolation basis function Obtain the discrete power sequence in the time interval Interpolation function within It can extract feature parameters that characterize the nature of fluctuations in real time, while capturing global nonlinear trends and local dynamic details, thus solving the problem of loss of key dynamic features caused by traditional fixed-interval sampling.

[0076] like Figure 6 The image shows the result of the FEWMA interpolation function.

[0077] This invention provides a power data interpolation compensation system for grid-connected power plants during power transmission. It dynamically adjusts core parameters of the interpolation function, such as the scaling factor and derivative weights, to form an adaptive hybrid interpolation model. The core of the system lies in capturing global nonlinear trends and local dynamic details, establishing a compatible and complementary relationship between fractal interpolation and cubic Hermitian interpolation, and achieving dynamic compensation of power analysis data interpolation in grid-connected power transmission scenarios. This system can accurately interpolate and compensate power acquisition data with low power consumption and time complexity.

[0078] In this case, the evaluation functions selected are mean squared error (MSE) and mean absolute error (MAE). After normalizing the original data to a mean of 0 and a variance of 1, FEWMA interpolation is compared with other interpolation methods such as linear interpolation, Hermitian interpolation, and fractal interpolation. The results are as follows: Table 1: Comparison of FEWMA interpolation with other interpolation algorithms

[0079] The results show that the mean square error (MSE) of FEWMA interpolation is 0.290, which is better than the traditional Hermitian interpolation by 26.40% and the traditional fractal interpolation by 47.65%. The mean absolute error (MAE) of FEWMA interpolation is 0.100, which is better than the traditional Hermitian interpolation by 37.50% and the traditional fractal interpolation by 30.56%. This fully demonstrates the accuracy and superiority of the data interpolation dynamic compensation method proposed in this invention for power analysis of new energy grid connection.

[0080] Finally, it should be noted that the methods and devices described in detail above are merely embodiments, and those skilled in the art can modify these embodiments in different ways as long as they do not depart from the scope of the present invention.

Claims

1. A power data interpolation compensation method for grid-connected power generation of a new energy power plant, characterized in that: include, Obtain discrete power data sequences under renewable energy grid-connected scenarios; The cumulative deviation after removing the mean is calculated for the power discrete data sequence, and the local linear trend of the cumulative deviation is removed to obtain the power fluctuation function; Analyze the multi-scale characteristics of the power fluctuation function, extract fluctuation feature characterization parameters, and quantify the fractal characteristics of the sequence under different fluctuation amplitudes; The power fractal energy distribution is calculated based on the fluctuation characteristic characterization parameters, and a dynamic fluctuation factor is generated by combining time-series weighted processing. Based on the dynamic fluctuation factor and combined with the local and global statistical characteristics of the discrete power data sequence, a multi-parameter dynamic correction is introduced into the standard interpolation basis function to construct a dynamic interpolation basis function that adapts to the local dynamic details of power fluctuations. Based on the dynamic interpolation basis function and the conventional coefficient weights, the power sequence interpolation function is calculated to achieve power data compensation.

2. The power data interpolation compensation method for grid-connected power generation of new energy power plants as described in claim 1, characterized in that: The analysis of the multi-scale characteristics of the power fluctuation function, and the extraction of fluctuation feature characterization parameters, include: Based on the power fluctuation function, q-order fluctuation functions with different moment orders are constructed to describe the impact of fluctuations with different amplitudes. Establish the proportional relationship between the q-order wave function and the time scale; By fitting the proportional relationship, the fluctuation characteristic characterization parameters that can quantify the fractal characteristics of the sequence are obtained.

3. The power data interpolation compensation method for grid-connected power generation of new energy power plants as described in claim 2, characterized in that: The process of calculating the power fractal energy distribution based on the fluctuation characteristic characterization parameters and generating a dynamic fluctuation factor by combining time-series weighted processing includes: By integrally coupling the wave characteristic characterization parameters under different moment orders, the power fractal energy distribution at different time scales is calculated. A time-weighted algorithm is introduced into the power fractal energy distribution to extract multi-scale fluctuation features and describe the overall trend of the power fractal energy distribution. The overall trend of change is used as the dynamic fluctuation factor.

4. The power data interpolation compensation method for grid-connected power generation of new energy power plants as described in claim 1, characterized in that: Based on the dynamic fluctuation factor and combined with the local and global statistical characteristics of the power discrete data sequence, the standard interpolation basis function is dynamically modified with multiple parameters to construct the dynamic interpolation basis function, which includes: Obtain the standard interpolation basis functions; Based on the dynamic fluctuation factor combined with the local and global mean and variance parameters of the power discrete data sequence; The multi-parameter dynamic correction is introduced into the standard interpolation basis function to generate the dynamic interpolation basis function that can adapt to power fluctuations.

5. The power data interpolation compensation method for grid-connected power generation of new energy power plants as described in claim 2, characterized in that: The fluctuation characteristic characterization parameter is the generalized Hurst exponent; The construction of q-order wave functions with different moment orders includes: The power fluctuation function is raised to the power of q. The result of summing and averaging the q-th power is then taken as 1 / q-th power to obtain the q-th order wave function; The process of obtaining the fluctuation characteristic characterization parameters that can quantify the fractal characteristics of the sequence by fitting the proportional relationship includes: Establish the power-law relationship between the average value of the q-order wave function on the time scale and the time scale; The generalized Hurst exponent is obtained by fitting the power-law relationship using the least squares method.

6. The power data interpolation compensation method for grid-connected power generation of new energy power plants as described in claim 3, characterized in that: The time-series weighted algorithm is an exponentially weighted moving average algorithm; The time-weighted algorithm introduced for the power fractal energy distribution includes: A fractal periodic term is introduced into the exponentially weighted moving average algorithm; By substituting the power fractal energy distribution into the calculation of the moving weighted average of the power fractal energy distribution, the dynamic fluctuation factor is obtained.

7. The power data interpolation compensation method for grid-connected power generation of new energy power plants as described in claim 4, characterized in that: The standard interpolation basis function is the standard cubic Hermit basis function; The dynamic interpolation basis function is a time-varying Hermitian piecewise interpolation basis function; The multi-parameter dynamic correction includes: Introducing the maximum volatility factor and exponential decay factor to suppress interpolation overshoot; Introducing a minimum volatility factor and a gradient gain factor for rapid response to weak volatility; By introducing local and global mean and variance parameters, the basis function is dynamically scaled to adapt to load baseline drift and volatility.

8. A power data interpolation compensation method for grid-connected power generation of a new energy power plant, using the method described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to acquire discrete power data sequences in the grid-connected scenario of new energy power plants; The feature extraction module is used to calculate the cumulative deviation after removing the mean from the power discrete data sequence, remove the local linear trend of the cumulative deviation to obtain the power fluctuation function, analyze the multi-scale characteristics of the power fluctuation function, and extract fluctuation feature characterization parameters. The factor generation module is used to calculate the power fractal energy distribution based on the fluctuation characteristic characterization parameters, and generate a dynamic fluctuation factor by combining time-series weighted processing. The basis function construction module is used to construct a dynamic interpolation basis function by introducing multi-parameter dynamic correction to the standard interpolation basis function based on the dynamic fluctuation factor and the local and global statistical characteristics of the power discrete data sequence. The interpolation calculation module is used to calculate the power sequence interpolation function based on the dynamic interpolation basis function and the conventional coefficient weights, so as to realize power data compensation.

9. An electronic device, comprising: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, they implement the steps of the power data interpolation compensation method for grid-connected power energy of new energy power plants as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the power data interpolation compensation method for grid-connected power energy of a new energy power plant as described in any one of claims 1 to 7.

Images