Dynamic load electric energy metering optimization method and system

Abstract

The invention provides a dynamic load electric energy metering optimization method and system. Belongs to the technical field of data processing, and comprises the following steps: acquiring voltage and current signals in a dynamic load environment in real time by using a high-precision sensor, and discretizing the acquired signals to obtain a voltage value and a current value of a sampling point; and performing statistical analysis on the voltage and current values of the sampling points, extracting key parameters, and constructing a dynamic load feature vector based on the extracted key parameters. By collecting voltage and current signals in real time and carrying out accurate statistical analysis and feature extraction, the electric energy error value can be calculated more accurately, so that the accuracy of electric energy metering is improved.

Description

Technical Field

[0001] The invention proposes a dynamic load electric energy metering optimization method and system, belonging to the technical field of data processing. Background Art

[0002] Traditional energy metering methods often fail to accurately calculate energy errors in dynamic load environments, resulting in large deviations in metering results. Dynamic loads are diverse and random, and their types and sizes may vary greatly in different time periods and regions. In addition, dynamic loads are also affected by factors such as weather, seasons, and the popularity of smart devices and electric vehicles, making load changes more complex and difficult to predict. Therefore, an optimized method is needed to accurately measure energy in dynamic load environments. Summary of the invention

[0003] The present invention provides a dynamic load electric energy metering optimization method and system to solve the problems mentioned in the above background technology: The present invention provides a method for optimizing dynamic load electric energy metering, the method comprising: S1. Use high-precision sensors to collect voltage and current signals in a dynamic load environment in real time, and discretize the collected signals to obtain voltage and current values ​​at the sampling points; S2. Statistically analyze the voltage and current values ​​of the sampling points, extract key parameters, and construct a dynamic load feature vector based on the extracted key parameters; S3. Calculate the probability density value of the current amplitude according to the dynamic load characteristic vector, and obtain the metering parameters through measurement and calculation, wherein the metering parameters include power factor and phase difference; S4. Based on the pre-configured operation rules, the probability density value of the current amplitude and the metering parameters are combined to calculate the electric energy error value; and the operation rules are adaptively adjusted according to the dynamic load characteristic vector; S5. According to the electric energy error value, the power generation and transmission strategies are adjusted in real time.

[0004] The present invention provides a dynamic load electric energy metering optimization system, the system comprising: Data acquisition module: Use high-precision sensors to collect voltage and current signals in real time under dynamic load conditions, and discretize the collected signals to obtain the voltage and current values ​​at the sampling points; Parameter extraction module: performs statistical analysis on the voltage and current values ​​of the sampling points, extracts key parameters, and constructs a dynamic load feature vector based on the extracted key parameters; Parameter acquisition module: calculate the probability density value of the current amplitude according to the dynamic load characteristic vector, and obtain the metering parameters through measurement and calculation, the metering parameters include power factor and phase difference; Adaptive adjustment module: Based on the pre-configured operation rules, combined with the probability density value of the current amplitude and the metering parameters, the electric energy error value is calculated; and the operation rules are adaptively adjusted according to the dynamic load characteristic vector; Real-time adjustment module: According to the electric energy error value, the power generation and transmission strategies are adjusted in real time.

[0005] The invention has the following beneficial effects: by collecting voltage and current signals in real time and performing accurate statistical analysis and feature extraction, the electric energy error value can be calculated more accurately, thereby improving the accuracy of electric energy metering; by constructing dynamic load feature vectors and adaptively adjusting the operation rules according to the feature vectors, the electric energy metering system can adapt to the changing load conditions; by calculating the probability density value of the current amplitude, the abnormal current situation can be identified, and the fault detection and diagnosis capabilities of the system can be enhanced; by adjusting the power generation and transmission strategies in real time according to the electric energy error value, the operation efficiency and stability of the power system can be improved; by continuously monitoring the operation status of the power system, updating the dynamic load feature vector, and recalculating the electric energy error value, a closed-loop control system is formed, and continuous optimization of the power system can be achieved; by accurately measuring and optimizing the operation of the power system, unnecessary energy waste can be reduced and energy utilization efficiency can be improved; by real-time adjustment and optimization, the fluctuation of the power system caused by load changes can be reduced, and the reliability and stability of the system can be improved; by reducing energy waste and improving system efficiency, the operation cost of the power system can be reduced in the long run; improving energy utilization efficiency helps to reduce the consumption of fossil energy, thereby reducing greenhouse gas emissions, and has a positive effect on environmental protection. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 It is a step diagram of the method of the present invention; Figure 2 This is a system module diagram of the present invention. DETAILED DESCRIPTION

[0007] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0008] One embodiment of the present invention, as Figure 1 As shown, a dynamic load electric energy metering optimization method comprises: S1. Use high-precision sensors to collect voltage and current signals in a dynamic load environment in real time, and discretize the collected signals to obtain voltage and current values ​​at the sampling points; S2. Statistically analyze the voltage and current values ​​of the sampling points, extract key parameters, and construct a dynamic load feature vector based on the extracted key parameters; S3. Calculate the probability density value of the current amplitude according to the dynamic load characteristic vector, and obtain the metering parameters through measurement and calculation, wherein the metering parameters include power factor and phase difference; S4. Based on the pre-configured operation rules, the probability density value of the current amplitude and the metering parameters are combined to calculate the electric energy error value; and the operation rules are adaptively adjusted according to the dynamic load characteristic vector; S5. According to the electric energy error value, the power generation and transmission strategies are adjusted in real time.

[0009] The working principle of the above technical solution is as follows: first, the system deploys high-precision sensors, which can monitor the voltage and current signals under dynamic load environment in real time. The signals are usually discontinuous and change rapidly, so the sensors need to have sufficient sensitivity and sampling rate to capture these changes; the collected continuous voltage and current signals are then discretized, that is, the continuous signals are converted into voltage and current values ​​of a series of sampling points. This step is achieved by sampling the signals at regular intervals, and each sampling point represents the value of the signal at a certain moment; the voltage and current values ​​of the sampling points are statistically analyzed, including calculating statistical quantities such as the average, standard deviation, and peak value to understand the basic characteristics of the signal; key parameters such as voltage fluctuation range and current peak time are extracted from the statistical analysis results. These parameters can reflect the main characteristics of the dynamic load; based on the extracted key parameters, a dynamic load feature vector is constructed. A vector is a mathematical representation used to summarize and distinguish different types of dynamic loads; using the dynamic load feature vector, the probability density value of the current amplitude is calculated, which helps to understand the distribution of the current at different amplitudes; at the same time, the power factor (i.e., the ratio of active power to apparent power) and the phase difference (the phase angle difference between voltage and current) are obtained through measurement and calculation. These parameters are crucial for accurate metering of electric energy. Based on pre-configured operation rules (such as electric energy metering algorithm), the electric energy error value is calculated by combining the probability density value of the current amplitude with the metering parameters (power factor and phase difference). The error value reflects the difference between the actual metering result and the theoretical value. According to the dynamic load characteristic vector, the operation rules are adaptively adjusted, which means that the system can automatically adjust the metering algorithm according to the load changes to reduce errors and improve the accuracy of metering. Finally, according to the calculated electric energy error value, the power generation and transmission strategies are adjusted in real time. If the error value is large, it means that the current strategy may not be suitable for the current load conditions, so adjustments are needed to optimize the operation of the power system. By adjusting the output power of the generator, the allocation of the transmission line and other strategies, the stable operation of the power system under dynamic load conditions can be ensured, while reducing energy waste and power loss.

[0010] The effects of the above technical scheme are as follows: by using high-precision sensors for real-time acquisition, it is possible to capture subtle changes in voltage and current signals under dynamic load environments, thereby ensuring that the collected data is more accurate; the dynamic load feature vector constructed based on key parameters extracted by statistical analysis can fully reflect the dynamic characteristics of the load, providing a solid foundation for subsequent metering and error calculation; adaptively adjusting the operation rules according to the dynamic load feature vector can optimize the metering algorithm for different load conditions, further reduce errors, and improve the accuracy of electric energy metering; real-time adjustment of power generation and transmission strategies according to the electric energy error value can ensure that the power system maintains efficient operation under dynamic load conditions. It helps to reduce energy waste and improve energy utilization efficiency; through accurate metering and real-time adjustment strategies, it can reduce losses in the process of power transmission and distribution and improve the overall efficiency of the power system; the method can monitor and respond to load changes in real time, and respond to various load conditions through adaptive adjustment strategies, thereby enhancing the stability and reliability of the power system; through the analysis of dynamic load feature vectors, potential faults or abnormal conditions can also be discovered in time, providing early warning information for the maintenance and fault handling of the power system; the method combines high-precision sensors, data analysis, adaptive adjustment and other technical means to achieve intelligent management of the power system. It helps to promote the development of smart grids and improve the intelligence level of power systems; this method has good scalability and can be combined with other smart grid technologies, such as distributed energy management and demand response, to jointly build a more intelligent and efficient power system.

[0011] In one embodiment of the present invention, the S1 includes: S11. Real-time capture of voltage and current signals under dynamic load conditions by deploying high-precision voltage and current sensors at key locations in the power system; S12, filtering the continuous voltage and current signals output by the sensor through digital filtering technology, and dynamically adjusting the filtering parameters according to the signal characteristics; S13, sampling the preprocessed signal at equal intervals to obtain discrete sampling points.

[0012] The working principle of the above technical solution is as follows: high-precision voltage and current sensors are deployed at key locations of the power system, such as substations, distribution stations, and user terminals. The sensors use advanced sensing technology and can capture voltage and current signals in real time and accurately under dynamic load environments. The signals are a direct reflection of the operating status of the power system and contain rich load information and grid status information; the continuous voltage and current signals output by the sensors may contain noise and interference, which may come from the power system itself or the external environment. In order to obtain a purer signal, digital filtering technology is used to filter the signal output by the sensor. Digital filtering technology uses algorithms to perform mathematical processing on signals, which can remove or reduce noise and interference while retaining the main characteristics of the signal; the selection of filtering parameters (such as filter type, cutoff frequency, etc.) is crucial to the filtering effect. In this solution, the filtering parameters are dynamically adjusted according to the signal characteristics. The system can analyze the frequency components, noise levels and other characteristics of the signal in real time, and adjust the filtering parameters accordingly to ensure the best filtering effect. This dynamic adjustment method improves the adaptability and flexibility of the system; the signal after filtering is continuous, but it needs to be converted into a discrete signal when processed by the computer. Therefore, the pre-processed signal is sampled at equal intervals. The selection of sampling interval should be determined based on the frequency content of the signal and the sampling rate requirements of the system. Equally spaced sampling means that at each sampling moment, the system extracts a sample value from the continuous signal; through equally spaced sampling, the system obtains a series of discrete sampling points. These sampling points contain the main information of the signal and are easy for the computer to process and analyze. The number and quality of discrete sampling points depend on the selection of sampling interval and sampling rate.

[0013] The effect of the above technical solution is: by deploying high-precision voltage and current sensors at key locations in the power system, the voltage and current signals under dynamic load environments can be captured in real time. The sensors have high precision and high sensitivity, and can capture subtle changes in the signals, providing an accurate data basis for subsequent data processing and analysis; the continuous voltage and current signals output by the sensors are filtered using digital filtering technology, which can effectively remove noise and interference in the signals. The signal-to-noise ratio of the signal is improved, making the signal clearer and more accurate; the filtering parameters are dynamically adjusted according to the signal characteristics, so that the filtering effect is more adaptable to the changes of different signals. This adaptive filtering technology can further improve the quality of the signal and reduce errors and distortion; the pre-processed signal is sampled at equal intervals to obtain a series of discrete sampling points. These sampling points contain the main information of the signal and are convenient for subsequent data processing and analysis. Equal-interval sampling ensures the uniformity and consistency of sampling and improves the efficiency of data processing; through equal-interval sampling, it can be ensured that every important change of the signal is recorded under dynamic load environments. It helps to maintain the integrity and continuity of the data and provide reliable data support for subsequent analysis and decision-making; the technical solution can adapt to changes in dynamic load environments and capture and process voltage and current signals in real time. It helps the power system to maintain stability and reliability under dynamic load conditions and improve the operating efficiency and safety of the power system; this technical solution has good scalability and can be combined with other power system monitoring and control technologies to jointly build a more intelligent and efficient power system.

[0014] In one embodiment of the present invention, the S12 includes: Evaluate and select the adaptive filtering algorithm based on the characteristics of voltage and current signals under dynamic load conditions; Based on the selected adaptive filtering algorithm, corresponding parameter configuration is performed, wherein the corresponding parameters include filter order, learning rate and forgetting factor; During the filtering process, the statistical characteristics of the voltage and current signals, as well as the output signal quality of the filter, are monitored in real time; Based on the results of real-time monitoring, formulate rules for dynamic adjustment of filter parameters; feed back the quality of the filter output signal after parameter adjustment into the parameter adjustment rules for iterative optimization; After the filtering process is completed, the quality of the filter's output signal is evaluated; based on the evaluation results, the filter's parameters and algorithms are further optimized.

[0015] The working principle of the above technical solution is as follows: First, different adaptive filtering algorithms are evaluated according to the characteristics of voltage and current signals under dynamic load conditions (such as frequency range, noise type, dynamic changes of signals, etc.). These characteristics determine the selection of filtering algorithms to ensure that the algorithm can effectively remove noise and retain the main characteristics of the signal; based on the results of characteristic analysis, the most suitable adaptive filtering algorithm for the current signal characteristics is selected. Common adaptive filtering algorithms include the least mean square error (LMS) algorithm, the recursive least squares (RLS) algorithm, etc.; according to the requirements of the selected algorithm, the corresponding parameters are configured. These parameters include the filter order, learning rate, and forgetting factor, etc. The selection of these parameters is crucial to the filtering effect and needs to be weighed according to the signal characteristics and algorithm characteristics; before the filtering begins, these parameters are initially set. These settings will serve as the starting point of the filtering process and will be dynamically adjusted according to the results of real-time monitoring in the subsequent process; during the filtering process, the statistical characteristics of the voltage and current signals (such as mean, variance, spectral distribution, etc.) are monitored in real time. These characteristics reflect the basic characteristics and changing trends of the signal; at the same time, the output signal quality of the filter (such as signal-to-noise ratio, error sum of squares, etc.) is monitored. These indicators measure the quality of the filtering effect; based on the results of real-time monitoring, rules for dynamic adjustment of filtering parameters are formulated. For example, when the signal noise level increases, the filter order can be appropriately increased or the learning rate can be reduced to improve the filtering effect; when the signal frequency range changes, the filter cutoff frequency can be adjusted to adapt to the new signal characteristics; the quality of the filter output signal after parameter adjustment is fed back to the parameter adjustment rules for iterative optimization. Through continuous learning and adjustment, the filter can adaptively respond to signal changes under dynamic load environments; after the filtering process is completed, the quality of the filter output signal is evaluated. Including checking key indicators such as the degree of signal distortion and phase delay; based on the evaluation results, the filter parameters and algorithms are further optimized. For example, if the filtered signal still has obvious distortion or phase delay, you can try to change the filtering algorithm or adjust the parameter configuration to further improve the filtering effect.

[0016] The effect of the above technical solution is: according to the characteristics of the voltage and current signals under the dynamic load environment, the most suitable adaptive filtering algorithm can be flexibly evaluated and selected. This flexibility ensures the best match between the filtering effect and the signal characteristics; by real-time monitoring of the statistical characteristics of the signal and the output signal quality of the filter, the filtering parameters (such as filter order, learning rate, forgetting factor, etc.) can be dynamically adjusted. This dynamic adjustment mechanism enables the filter to adaptively respond to changes in the signal and always maintain the best filtering effect; through the dynamic adjustment of the adaptive filtering algorithm and parameters, the noise in the signal can be effectively suppressed and the signal-to-noise ratio of the signal can be improved. This is of great significance for improving the stability and reliability of the power system; while filtering out the noise, the main features of the signal, such as frequency components, phase information, etc., can be retained as much as possible. It ensures that the filtered signal still has sufficient fidelity, providing a reliable basis for subsequent data processing and analysis; the quality of the filter output signal after parameter adjustment is fed back to the parameter adjustment rule for iterative optimization. This feedback mechanism enables the filter to continuously learn and adjust to adapt to the ever-changing signal environment; through continuous iteration and optimization, the filter parameters and algorithms can gradually approach the optimal solution, further improving the filtering effect and performance; after the filtering process is completed, the quality of the filter output signal is evaluated. It helps to timely discover and correct possible problems and defects in the filtering process; based on the evaluation results, the filter parameters and algorithms are further optimized. This continuous optimization mechanism enables the filter to always maintain the best working state and improve the overall performance and stability of the power system; through precise filtering processing, the quality of voltage and current signals can be improved, and signal distortion and phase delay can be reduced. It is of great significance for the stable operation and fault detection of the power system; the application of adaptive filtering algorithms enables the filter to cope with various complex dynamic load environments and improve the adaptability and reliability of the power system.

[0017] In one embodiment of the present invention, the S2 includes: S21, performing statistical analysis on the voltage and current values ​​of the sampling points, performing spectrum analysis on the signals using fast Fourier transform, and extracting the main frequency components and their energy distribution; S22. Based on the above statistical characteristics and spectrum analysis results, key parameters are extracted and combined into a dynamic load feature vector.

[0018] The working principle of the above technical solution is as follows: first, the voltage and current values ​​of discrete sampling points obtained from the S1 stage are preprocessed, including removing outliers, smoothing data, etc., to ensure the accuracy of the analysis results; then, these preprocessed data are analyzed in depth using statistical methods. Including calculating the mean, standard deviation, skewness, kurtosis and other statistical quantities of voltage and current to reveal the distribution characteristics, discreteness and possible asymmetry of the data; in order to obtain the spectrum information of the signal, the voltage and current signals are transformed by fast Fourier transform. FFT can convert time domain signals into frequency domain signals, thereby revealing the energy distribution of the signal on different frequency components; by analyzing the results after FFT transformation, the main frequency components of the signal and their energy distribution are extracted. These characteristics are crucial for identifying periodic changes and emergencies in dynamic loads. For example, the sudden increase of certain frequency components may indicate equipment failure or a sharp change in load; based on the statistical analysis and spectrum analysis results of the S21 stage, a series of key parameters are defined to describe the characteristics of dynamic loads. These parameters include the maximum value, minimum value, peak time (i.e., the time point at which the maximum or minimum value is reached) and frequency offset (i.e., the deviation of the main frequency component of the signal from the reference frequency) of the voltage and current; these key parameters are extracted from the preprocessed data using mathematical methods and algorithms. For example, the maximum and minimum values ​​and their corresponding time points are found by traversing the data sequence; the main frequency components and their deviation from the reference frequency are calculated through the FFT transformation results; and the extracted key parameters are combined into a dynamic load feature vector. This vector is a multidimensional data structure in which each dimension corresponds to a key parameter. The construction of the feature vector enables subsequent systems to easily use these parameters for monitoring, fault diagnosis, and load management.

[0019] The effect of the above technical solution is: by statistically analyzing the voltage and current values ​​at the sampling points, the distribution characteristics, discreteness and possible change trends of the data can be revealed. It helps to deeply understand the operating status and potential problems of dynamic loads; using fast Fourier transform (FFT) to perform spectrum analysis on the signal can accurately extract the main frequency components and their energy distribution. This information is crucial for identifying periodic changes and emergencies in dynamic loads, and helps to timely discover and deal with potential faults or anomalies; the extracted key parameters such as the maximum value, minimum value, peak time and frequency offset of voltage and current fully reflect the state characteristics and frequency characteristics of the dynamic load at different time points. These parameters provide strong data support for subsequent monitoring, control and fault diagnosis; combining key parameters into dynamic load feature vectors enables the system to conveniently use these features for real-time monitoring, pattern recognition and decision support. The construction of feature vectors not only improves the efficiency of data processing, but also enhances the intelligence and automation level of the system; through the analysis of dynamic load feature vectors, the system can timely discover and warn of potential faults or anomalies, thereby avoiding or reducing the impact of faults on the power system. At the same time, the fault diagnosis method based on feature vectors can locate the fault source more accurately and improve the efficiency and accuracy of fault handling; the dynamic load feature vector also provides important reference information for the operation and management of the power system. By real-time monitoring and analyzing feature vectors, the system can optimize load distribution and adjust operation strategies to improve the overall performance and reliability of the power system; the construction of dynamic load feature vectors enables the system to better adapt to dynamic changes in load. Whether it is periodic changes or emergencies, the system can monitor and respond in real time through feature vectors to ensure the stable operation of the power system; this technical solution is not only suitable for the current dynamic load environment, but also has good flexibility and scalability. As the power system develops and changes, the dimensions and parameters of the feature vectors can be easily adjusted and expanded to meet new needs and challenges.

[0020] In one embodiment of the present invention, the S22 includes: Calculate basic parameters according to the voltage and current values ​​of the sampling points, wherein the basic parameters include maximum value, minimum value, average value and variance; Based on the peak detection algorithm, the peak values ​​in the voltage and current signals and their corresponding time points are identified, where the peak time is used to reflect the speed and trend of signal changes; Based on step S21, the signal is subjected to spectrum analysis again; spectrum feature extraction is performed again through resolution bandwidth and filter type; Based on the spectrum analysis results, the offset between the actual frequency of the signal and the theoretical frequency is calculated; the offset is calculated using the following formula:

[0021] in, Indicates signal The spectrum amplitude of Indicates the frequency pull corresponding to the maximum value of the spectrum amplitude; N indicates the number of FFT points; represents the sampling interval; Indicates the theoretical frequency of the signal; On the basis of preliminary parameter extraction and spectrum feature extraction, the elements for constructing feature vectors are selected according to the actual application requirements and the purpose of data analysis; these features should be able to fully reflect the characteristics and changing trends of dynamic loads.

[0022] The selected features are integrated to construct a dynamic load feature vector; in the process of constructing the feature vector, continuous optimization and adjustment are performed.

[0023] The working principle of the above technical solution is as follows: first, the voltage and current values ​​obtained from the sampling device are preprocessed, including removing outliers and smoothing data to ensure the accuracy of subsequent analysis; then, the basic parameters of these preprocessed data are calculated using mathematical methods. Including maximum value, minimum value, average value and variance, these parameters can preliminarily reflect the fluctuation range and discreteness of the signal; the peak detection algorithm is used to identify the peak value in the voltage and current signals. The algorithm can accurately find the local maximum value in the signal and determine its corresponding time point; the peak time reflects the speed and trend of the signal change. By analyzing the peak time, we can understand the changing characteristics of the dynamic load in different time periods, such as the start-up, stable operation and unloading of the load; based on the S21 stage, the signal is analyzed again by spectrum. This analysis will select the appropriate resolution bandwidth and filter type to more accurately extract the spectrum features; through spectrum analysis, the frequency value, energy proportion and phase information of the main frequency components of the signal are extracted. These characteristics can reflect the energy distribution and phase relationship of the dynamic load at different frequencies; the frequency offset refers to the deviation between the actual frequency of the signal and the theoretical frequency. This parameter can reflect the impact of dynamic changes in load on the frequency stability of the power system; based on the results of spectrum analysis, the offset between the actual frequency of the signal and the theoretical frequency is calculated. Based on the preliminary parameter extraction and spectrum feature extraction, the elements for constructing the feature vector are selected according to the actual application requirements and the purpose of data analysis. The feature elements comprehensively reflect the characteristics and changing trends of the dynamic load, such as basic parameters, peak time, spectrum characteristics, and frequency offset; the selected feature elements are fused to construct a dynamic load feature vector. This vector is a multidimensional data structure, in which each dimension corresponds to a feature element; in the process of constructing the feature vector, it is continuously optimized and adjusted. This includes removing redundant features, improving the independence between features, and enhancing the representativeness of features, to ensure that the feature vector can more effectively reflect the characteristics and changing trends of the dynamic load.

[0024] The effects of the above technical solution are: by calculating the basic parameters of the voltage and current values ​​at the sampling points, such as the maximum value, minimum value, average value and variance, the fluctuation range, discreteness and overall trend of the signal can be preliminarily understood, providing a basis for subsequent in-depth analysis; by accurately identifying the peak values ​​and corresponding time points in the voltage and current signals through the peak detection algorithm, it is not only possible to capture the extreme values ​​of the signal, but also to reflect the speed and trend of the signal change through the peak time, providing an important basis for real-time monitoring and early warning of dynamic loads; on the basis of step S21, the signal is analyzed again by spectrum, and the appropriate resolution bandwidth and filter type are selected, so that the spectrum features can be more accurately extracted, such as the frequency value, energy proportion and phase information of the main frequency components. These features are of great significance for in-depth understanding of the spectrum characteristics of dynamic loads, identification of potential faults and optimization of power system operation; based on the spectrum analysis results, the offset between the actual frequency and the theoretical frequency of the signal is calculated, which can accurately reflect the impact of the dynamic changes of the load on the frequency stability of the power system. This indicator plays a key role in the stable operation, fault warning and load management of the power system. On the basis of preliminary parameter extraction and spectrum feature extraction, the elements for constructing the feature vector are carefully selected according to the actual application requirements and the purpose of data analysis. These feature elements should be able to fully reflect the characteristics and change trends of the dynamic load, and provide strong support for subsequent monitoring, diagnosis, optimization, etc. The selected feature elements are integrated to construct a dynamic load feature vector, and it is continuously optimized and adjusted during the construction process. This feature vector is not only highly representative and independent, but also can adapt to different application scenarios and data requirements, and improve the accuracy and efficiency of data analysis. Through the constructed dynamic load feature vector, the system can monitor the state and change trend of the dynamic load in real time, and timely discover and warn potential faults or anomalies, thereby improving the reliability and stability of the power system. The feature vector also provides strong data support for the optimization decision-making in the operation management, load scheduling and fault handling of the power system, which helps to improve the overall performance and operation efficiency of the power system. The above formula can accurately capture the frequency characteristics of the actual signal, especially when the signal has frequency drift or offset. This method can improve the accuracy of frequency detection, and is particularly suitable for scenarios that require high-precision frequency tracking, such as dynamic load monitoring, wireless communications, and other applications. The formula is based on spectrum analysis of actual signals, so it can adaptively adapt to changes in different signals, especially when there is noise, interference, or non-ideal signals in the signal environment. This improves the robustness of the formula in complex signal environments and can adapt to different sampling conditions and actual environments. By using the frequency corresponding to the maximum value of the spectrum amplitude, the formula can capture the change of signal frequency over time, thereby revealing the trend of signal changes.If the frequency offset is large, it may mean that the signal is interfered by external factors or the system has a frequency change, which can help the system respond quickly and make adjustments. The formula provides a simple and efficient way to evaluate the frequency error of the signal, that is, the difference between the actual frequency and the theoretical frequency. This error evaluation is very important for the performance adjustment and optimization of the system, especially in applications where the signal frequency needs to be precisely controlled (such as modems, wireless transmission systems, etc.). Since the offset is calculated by the maximum value of the spectrum amplitude and the known theoretical frequency, the formula is relatively simple to implement and the calculation amount is moderate, which is suitable for application in real-time signal processing systems. By calculating the spectrum through Fourier transform (FFT), the signal frequency can be efficiently extracted, avoiding the complex frequency measurement process. In the signal processing of dynamic loads, the frequency offset usually means the fluctuation of the system load. Through this calculation of the frequency offset, the dynamic behavior of the load change can be further understood on the basis of spectrum analysis, helping to build a more accurate load prediction and optimization model, especially for power systems, communication systems and other fields with high requirements for dynamic loads. This offset formula is not only applicable to the analysis of voltage and current signals, but can also be extended to other signal types that require spectrum analysis, such as calibrating the signal frequency offset in wireless communications or detecting the frequency change trend in audio signal processing. Because its calculation method is simple and intuitive, it has potential application value in many fields.

[0025] In one embodiment of the present invention, S3 includes: S31. Calculate the probability density value of the sampling point according to the current amplitude data of the sampling point by using the histogram method; wherein the probability density value reflects the distribution characteristics of the current amplitude and is used to identify abnormal current conditions; the probability density value is calculated by the following formula:

[0026] in, represents the probability density estimate at point x; n represents the number of samples; h represents the bandwidth; represents the current amplitude of the i-th sample point; x represents the point where you want to estimate the probability density value.

[0027] S32. The power factor (PF) and phase difference (θ) are measured and calculated through a measurement algorithm. The power factor reflects the phase relationship between current and voltage and is used to measure the efficiency of electric energy utilization. The phase difference is used to reflect the lag or lead of current relative to voltage.

[0028] The working principle of the above technical solution is as follows: first, a series of current amplitude data points are collected from the current sensor of the power system. These data points reflect the current fluctuation of the power system in a specific time period; the collected current amplitude data are grouped according to predetermined intervals, and the number of data points in each interval is counted. In this way, a histogram with the current amplitude as the horizontal axis and the number of data points as the vertical axis is formed; based on the histogram, the probability density value of each interval can be calculated. The probability density value reflects the distribution characteristics of the current amplitude in the interval, that is, the probability that the current amplitude falls in the interval. It is obtained by dividing the number of data points in the interval by the total number of data points and multiplying it by the width of the interval; by comparing the probability density values ​​of different intervals, abnormal current conditions can be identified. For example, if the probability density value of a certain interval is much higher or much lower than other intervals, it may indicate abnormal conditions such as current overload, short circuit or open circuit; at the same time, the voltage and current signals of the power system are collected. These signals are continuous waveform data, reflecting the changes of voltage and current over time; the collected voltage and current signals are preprocessed, including filtering and denoising, to improve the accuracy and reliability of the signals; the power factor is calculated by measuring the effective values ​​of voltage and current and the phase relationship between them using the power factor calculation formula (PF = active power / apparent power). The power factor reflects the phase relationship between current and voltage and is an important indicator for measuring the efficiency of electric energy utilization; by measuring the phase difference (θ) between the voltage and current waveforms, the lag or lead of the current relative to the voltage can be reflected.

[0029] The effect of the above technical solution is: by processing the current amplitude data of the sampling points through the histogram method, the probability density value of each current amplitude interval can be accurately calculated. These probability density values ​​intuitively reflect the distribution characteristics of the current amplitude, so that technicians can clearly understand the distribution of current in different amplitude intervals; based on the calculated probability density values, the system can automatically identify abnormal current conditions, such as current overload, short circuit or open circuit. This automatic identification function greatly improves the safety and stability of the power system and reduces faults and accidents caused by abnormal current; the power factor (PF) is accurately measured by the measurement algorithm, which can accurately reflect the phase relationship between current and voltage. The power factor is an important indicator for measuring the efficiency of electric energy utilization. The closer its value is to 1, the higher the efficiency of electric energy utilization, and vice versa; based on the measured power factor, technicians can optimize the energy efficiency of the power system, such as adjusting the load, improving the circuit design or installing a power factor correction device, so as to improve the efficiency of electric energy utilization and reduce energy consumption and cost; the phase difference (θ) is accurately calculated by the measurement algorithm, which can accurately reflect the lag or lead of the current relative to the voltage. Phase difference is an important parameter in the power system, and changes in its value will affect the stability of the system and the quality of power. Based on the calculated phase difference, technicians can analyze and evaluate the stability of the power system. When the phase difference exceeds the normal range, it may mean that there are unstable factors or potential faults in the system, and timely measures need to be taken to intervene and adjust. By simultaneously monitoring multiple aspects such as current distribution characteristics, power utilization efficiency and phase relationship, comprehensive monitoring of the power system is achieved. This comprehensive monitoring method can more comprehensively reflect the operating status and performance of the power system. Based on the monitoring results, the system can automatically issue early warning signals to remind technicians to deal with potential faults or abnormal conditions in a timely manner. This early warning function can greatly reduce the probability and loss of faults and improve the reliability and safety of the power system. Among them; Compared with the traditional histogram, kernel density estimation (KDE) provides a smoother probability density function, which helps to better understand the true distribution of data, especially when there are fewer data points or uneven distribution. By observing the shape of the probability density curve, it is easy to identify which current amplitudes are normal and which may be outliers. For example, if the probability density corresponding to a certain current amplitude is significantly lower than other values, then this value may represent an abnormal situation. The KDE method allows users to choose different kernel functions (such as Gaussian kernel, triangular kernel, etc.), as well as adjust the bandwidth hh to change the smoothness of the estimate. This flexibility enables the method to adapt to different types of distribution data. It provides a continuous probability density curve that is more intuitive, easier to understand and interpret than a discrete histogram. This is a very useful tool for analysts. There is no need to assume that the data follows a specific distribution form, which makes it very useful when dealing with data with unknown or complex distributions.

[0030] In one embodiment of the present invention, the S4 includes: S41. The operation rules are preset according to the basic principles of electric energy metering and dynamic load characteristics; the rules should take into account the comprehensive influence of multiple factors such as current amplitude probability density, power factor, phase difference, etc.

[0031] S42, calculating the electric energy error value based on the preset operation rules and the extracted dynamic load characteristic vector, wherein the error value is used to reflect the deviation between the actual measurement value and the theoretical value; S43. According to the change of the dynamic load characteristic vector, the parameters or weights in the operation rules are adjusted in real time through the neural network, and the operation rules are automatically optimized based on the learning historical data and current characteristics.

[0032] The working principle of the above technical solution is as follows: first, deeply understand the basic principles of electric energy metering, including the influence of key parameters such as current, voltage, power factor, phase difference, etc. on electric energy metering; based on the dynamic load characteristic vector extracted in the early stage, analyze the current amplitude probability density, power factor, phase difference and other characteristics of the load, as well as their changes with time and load; according to the principle of electric energy metering and dynamic load characteristics, preset a set of operation rules. This set of rules should comprehensively consider the comprehensive influence of multiple factors such as current amplitude probability density, power factor, phase difference, etc., to accurately calculate electric energy; for example, the operation rule can be set as: electric energy = current effective value × voltage effective value × power factor × time, and on this basis, make corrections according to the current amplitude probability density and phase difference; extract the current dynamic load characteristic vector from the monitoring system, including current amplitude probability density, power factor, phase difference, etc.; substitute the extracted characteristic vector into the preset operation rule to calculate the current electric energy value; compare the calculated electric energy value with the theoretical value (such as the electric energy value measured by a high-precision instrument) to calculate the electric energy error value. This error value reflects the deviation between the actual measurement value and the theoretical value; continuously monitor the changes in the dynamic load characteristic vector, including the fluctuations of parameters such as the current amplitude probability density, power factor, and phase difference; use the neural network to adjust the parameters or weights in the operation rules in real time. The neural network can automatically adjust its internal parameters based on the input characteristic vector and the output error value to reduce the error; the neural network not only considers the current characteristic vector and error value, but also learns the characteristics and error patterns in the historical data. This helps the neural network better understand and adapt to the dynamic changes of the load, thereby improving the accuracy and adaptability of the operation rules; based on the learning results of the neural network, the operation rules are automatically optimized. This optimization can be to adjust parameters, weights, or introduce new correction items to further improve the accuracy of electric energy metering.

[0033] The effect of the above technical solution is: the preset operation rules not only take into account the basic principles of electric energy metering, but also incorporate the comprehensive influence of various dynamic load characteristics such as current amplitude probability density, power factor, phase difference, etc. This comprehensive consideration makes the electric energy metering more accurate and can more accurately reflect the actual operating status of the power system; based on the preset operation rules and the extracted dynamic load characteristic vector, the electric energy error value is calculated. This step can monitor the accuracy of electric energy metering in real time, discover and correct deviations in time, thereby improving the accuracy of electric energy metering; by extracting the dynamic load characteristic vector, the load changes can be monitored in real time. It helps the system to adapt to load fluctuations in time and ensure the accuracy and stability of electric energy metering; neural networks are used to adjust the parameters or weights in the operation rules in real time, and the operation rules are automatically optimized according to the learning historical data and current characteristics. This dynamic adjustment mechanism enables the system to better adapt to changes in load and improve the adaptability and flexibility of electric energy metering; by introducing advanced technologies such as neural networks, the automation and intelligent management of electric energy metering is realized. It greatly reduces the need for manual intervention and improves work efficiency and accuracy. Based on the learning and optimization capabilities of neural networks, it can use historical data and current features to make intelligent decisions, optimize operation rules, and further improve the accuracy and efficiency of electric energy metering. Assume that in a certain power system, the probability density distribution of load current amplitude is uneven, and there are power factor fluctuations and phase difference changes. Traditional electric energy metering methods may not accurately reflect these dynamic changes, resulting in metering errors. After adopting the S4 solution, the system can comprehensively consider these factors, preset more accurate operation rules, and adjust and optimize rule parameters in real time through neural networks. In this way, even if the load fluctuates, the system can maintain high-accuracy electric energy metering to ensure stable operation and energy efficiency management of the power system.

[0034] In one embodiment of the present invention, the S43 includes: S431, selecting a neural network model according to the complexity of the dynamic load characteristic vector and the diversity of the operation rules; S432, constructing a neural network architecture according to the characteristics of the selected model, and determining the number of nodes in the input layer, the hidden layer, and the output layer, wherein the input layer receives the dynamic load feature vector, and the output layer outputs the electric energy error value or the adjustment parameter of the operation rule; S433. During the electric energy metering process, the change of the dynamic load characteristic vector is monitored in real time, and based on the forward propagation and back propagation mechanism of the neural network, the weight and bias of the neural network are dynamically adjusted according to the input of the current characteristic vector and the output of the electric energy error value; S434, based on the learning ability of the neural network, extract the mapping relationship between the feature vector and the electric energy error value from the historical data; and in the real-time metering process, use the newly collected feature vector and electric energy error value as new training samples, and continuously input them into the neural network for online learning; S435. Automatically optimize the parameters or weights in the operation rules according to the output results of the neural network and the changes in the current feature vector; and feed back the optimized operation rules to the electric energy metering system.

[0035] The working principle of the above technical solution is as follows: First, the required neural network model is evaluated according to the complexity of the dynamic load feature vector and the diversity of the operation rules. The complexity of the feature vector includes its dimension, distribution characteristics, and changes over time; according to the analysis results, a suitable neural network model is selected, such as a deep neural network (DNN), a convolutional neural network (CNN), a recurrent neural network (RNN), or a long short-term memory network (LSTM). For example, if the feature vector has significant time series characteristics, RNN or LSTM can be selected; if the feature vector contains image or spatial structure information, CNN may be more appropriate; according to the characteristics of the selected model, the architecture of the neural network is constructed, including the number of nodes in the input layer, hidden layer, and output layer. The number of nodes in the input layer should match the dimension of the feature vector, and the number of nodes in the output layer should be determined according to the specific task (such as predicting the power error value or adjusting the operation rule parameters); in the hidden layer, select a suitable activation function to introduce nonlinear characteristics and improve the network's expressiveness; according to the task objectives, select a suitable loss function to measure the difference between the predicted value and the actual value, and determine the optimization algorithm (such as gradient descent, Adam, etc.) to minimize the loss function; in the process of power metering, monitor the changes of dynamic load feature vectors in real time, especially those features that have a significant impact on the accuracy of power metering, such as current amplitude probability density, power factor and phase difference; based on the forward propagation and back propagation mechanism of the neural network, dynamically adjust the weight and bias of the neural network according to the input of the current feature vector and the output of the power error value. This step calculates the gradient through the back propagation algorithm and updates the network parameters to reduce the error; based on the learning ability of the neural network, extract the mapping relationship between the feature vector and the power error value from the historical data. This step enables the neural network to accurately predict the electric energy error value under a given eigenvector through the training process; in the real-time metering process, the newly collected eigenvectors and electric energy error values ​​are used as new training samples and continuously input into the neural network for online learning. This step enables the neural network to adapt to the changes in dynamic loads and improve metering accuracy by continuously updating weights and parameters; according to the output results of the neural network and the changes in the current eigenvectors, the parameters or weights in the operation rules are automatically optimized. This step adjusts the parameters or weights in the operation rules to reduce the electric energy error value and improve the metering accuracy; the optimized operation rules are fed back to the electric energy metering system to achieve continuous improvement in metering accuracy. By updating the operation rules in the electric energy metering system, it can be more adapted to the changes in dynamic loads and improve the overall performance of the system.

[0036] The effects of the above technical solution are as follows: by real-time monitoring of changes in dynamic load feature vectors, and dynamically adjusting network weights and biases based on the forward propagation and back propagation mechanisms of the neural network, it is possible to respond to load fluctuations in real time, thereby significantly improving the accuracy of electric energy metering; by extracting the mapping relationship between feature vectors and electric energy error values ​​from historical data, and continuously inputting newly collected feature vectors and electric energy error values ​​for online learning during the real-time metering process, the neural network can continuously adapt to changes in dynamic loads and further improve metering accuracy; according to the complexity of the dynamic load feature vectors and the diversity of operation rules, a suitable neural network model, such as DNN, CNN, RNN or LSTM, is flexibly selected to adapt to different application scenarios and load characteristics; according to the output results of the neural network and the changes in the current feature vectors, the parameters or weights in the operation rules are automatically optimized, and the optimized operation rules are fed back to the electric energy metering system, thereby enhancing the adaptability and flexibility of the system; by introducing advanced technologies such as neural networks, intelligent management of electric energy metering is realized. Neural networks can automatically learn, predict and optimize operation rules, reducing the need for manual intervention and improving work efficiency and accuracy. Based on the learning ability of neural networks, the mapping relationship between feature vectors and electric energy error values ​​is extracted from historical data, and weights and parameters are continuously updated through online learning. This data-driven optimization method enables the system to continuously learn and improve, improving the level of intelligent management. By improving the accuracy of electric energy metering, it can more accurately reflect the actual operating status of the power system and provide strong support for the stable operation of the power system. By real-time monitoring of changes in dynamic load feature vectors, S43 can also promptly detect potential faults or abnormal conditions, providing an important basis for fault warning and diagnosis of the power system.

[0037] In one embodiment of the present invention, the S433 includes: In the process of electric energy metering, based on sensors and data processing technology, the changes of dynamic load characteristic vectors are monitored in real time, and the characteristic vector data monitored in real time are preprocessed; Preset weight adjustment trigger conditions, and when the trigger conditions are met, start the weight adjustment process; Using the forward propagation mechanism of the neural network, the current feature vector is input into the neural network to calculate the electric energy error value of the output layer; by comparing the actual error value with the theoretical error value, the prediction performance of the neural network is evaluated; Based on the error evaluation results of the forward propagation, the gradient is calculated using the back-propagation mechanism, and the weights and biases of the neural network are updated according to the gradient descent method or other optimization algorithms; By continuously adjusting the weights, the prediction performance of the neural network gradually approaches the optimal solution. During the weight adjustment process, the learning rate is dynamically adjusted according to the speed and stability of the error reduction.

[0038] The working principle of the above technical solution is as follows: in the process of electric energy metering, the changes of dynamic load feature vectors are monitored in real time through sensors deployed in the power system. These feature vectors include key features such as current amplitude probability density, power factor, phase difference, etc., which have a significant impact on the accuracy of electric energy metering; the feature vector data obtained from the sensor is preprocessed, including data cleaning, normalization, feature selection and other steps to ensure the quality and consistency of the data and provide reliable input for subsequent neural network processing; in order to effectively control the frequency and timing of weight adjustment, a series of weight adjustment trigger conditions are preset. These conditions may include feature vector changes exceeding a preset threshold, electric energy error values ​​exceeding an acceptable range, etc.; in the real-time monitoring process, it is continuously judged whether the current feature vector and electric energy error value meet the trigger conditions. Once the conditions are met, the weight adjustment process is started; the current feature vector is input into the neural network using the forward propagation mechanism of the neural network, and the electric energy error value of the output layer is finally obtained after calculation of each layer; the prediction error of the neural network (i.e., the difference between the actual error value and the theoretical error value) is used as an evaluation indicator to measure the prediction performance of the neural network; according to the error evaluation result of the forward propagation, the gradient is calculated using the back propagation mechanism. The gradient represents the sensitivity of the error to the weights of each layer of the neural network and is the basis for weight adjustment. The weights and biases of the neural network are updated according to the gradient descent method or other optimization algorithms. This step makes the prediction performance of the neural network gradually approach the optimal solution by continuously reducing the error. During the weight adjustment process, the learning rate is dynamically adjusted according to the speed and stability of the error reduction. The learning rate is a hyperparameter that controls the amplitude of weight update and has an important impact on the training effect of the neural network. When the error decreases rapidly, the learning rate can be appropriately increased to accelerate convergence. When the error fluctuates greatly, the learning rate should be reduced to avoid overfitting. This dynamic adjustment strategy helps to improve the training efficiency and stability of the neural network. By repeatedly repeating the process of forward propagation, error evaluation, back propagation and weight update, the weights and biases of the neural network are continuously adjusted until the preset convergence conditions or performance requirements are met. During the weight adjustment process, the prediction performance of the neural network is continuously monitored, and the training strategy is adjusted according to the performance changes, such as adjusting the learning rate, increasing the training samples, etc., to further optimize the performance of the neural network.

[0039] The effects of the above technical solution are: the changes of dynamic load characteristic vectors are monitored in real time through sensors and data processing technology, and preprocessed to ensure the quality and accuracy of input data. It helps to improve the training effect of neural networks and the accuracy of electric energy metering; preset weight adjustment trigger conditions, and start the weight adjustment process when the conditions are met. This dynamic adjustment mechanism can optimize the neural network parameters in real time according to load changes, so as to more accurately predict the electric energy error value and improve the metering accuracy; the technical solution can monitor the changes of load characteristic vectors in real time and adjust the neural network parameters accordingly. This enables the system to flexibly respond to various load changes, including sudden load increases, load type switching, etc., so as to maintain stable metering performance; during the weight adjustment process, the learning rate is dynamically adjusted according to the speed and stability of error reduction. This adaptive learning rate adjustment strategy helps to speed up the convergence speed of the neural network while avoiding overfitting, improving the adaptability and robustness of the system; by introducing neural networks and data processing technology, intelligent monitoring and adjustment of the electric energy metering process is realized. This reduces the need for manual intervention and improves work efficiency and accuracy; the technical solution supports online learning and continuous optimization. By continuously inputting new feature vectors and electric energy error values ​​as training samples, the neural network can continuously learn and improve, further improving the metering accuracy and adaptability; improving the accuracy of electric energy metering helps to more accurately reflect the actual operating status of the power system and provide strong support for the stable operation of the power system; by real-time monitoring of changes in load feature vectors and electric energy error values, this technical solution can also promptly detect potential faults or abnormal conditions, providing an important basis for fault warning and diagnosis of the power system; the intelligent monitoring and adjustment mechanism reduces dependence on manual intervention, reduces the workload and operation and maintenance costs of operation and maintenance personnel; through precise metering and monitoring, equipment failures or abnormal conditions can be promptly detected and handled, improving equipment utilization and reliability.

[0040] In one embodiment of the present invention, S5 includes: S51. Formulate an adjustment plan for power generation and transmission strategy based on the electric energy error value; and adjust relevant parameters of the generator in real time according to the adjustment plan; S52, continuously monitoring and evaluating the adjusted power system operation status, acquiring new voltage and current data, updating the dynamic load characteristic vector, and recalculating the electric energy error value; S53. According to the evaluation results, the adjustment strategy is iteratively optimized to form a closed-loop control.

[0041] The working principle of the above technical solution is as follows: first, the electric energy error value obtained in the S433 stage is used as an important basis for formulating and adjusting the power generation and transmission strategies; the electric energy error value reflects the deviation of the current power system in electric energy metering. By analyzing it, the problems or deficiencies in the power system can be identified; when formulating the adjustment plan, it is necessary to comprehensively consider the stability, economy and environmental protection of the power system; in terms of stability, it is necessary to ensure that the adjusted power system can operate stably without problems such as voltage fluctuations and frequency deviations; in terms of economy, it is necessary to consider the impact of the adjustment plan on the cost of power production, power grid losses and other aspects, and strive to maximize economic benefits; in terms of environmental protection, it is necessary to pay attention to the impact of the adjustment plan on the environment, such as reducing emissions and improving energy utilization efficiency; according to the adjustment plan, the output power of the generator, the load distribution of the transmission line and other related parameters are adjusted in real time; by adjusting these parameters, the operating state of the power system can be changed, thereby reducing the electric energy error value and improving the accuracy of electric energy metering; after adjusting the power generation and transmission strategies, the operating state of the power system is continuously monitored to obtain new The voltage, current and other data are important bases for evaluating the operation status of the power system and the basis for the subsequent calculation of the electric energy error value. The dynamic load characteristic vector is updated according to the new voltage and current data obtained. The dynamic load characteristic vector reflects the load change of the power system and is an important reference for formulating and adjusting the power generation and transmission strategy. The electric energy error value is recalculated using the updated dynamic load characteristic vector and the new voltage and current data. By comparing the electric energy error values ​​before and after the adjustment, the effect of the adjustment strategy can be evaluated, providing a basis for subsequent iterative optimization. According to the continuous monitoring and evaluation results of the operation status of the power system, the adjustment strategy is iteratively optimized. The purpose of iterative optimization is to continuously improve the adjustment strategy to make it more in line with the actual situation of the power system, further improve the accuracy of electric energy metering and the operation efficiency of the power system. A closed-loop control is formed through the cyclic iteration of the three steps of S51, S52 and S53. In the closed-loop control, the power generation and transmission strategy is continuously adjusted and optimized, and the operation status of the power system is continuously monitored and evaluated to achieve the optimized operation of the power system and improve the accuracy of electric energy metering.

[0042] The effects of the above technical solutions are as follows: by adjusting the output power of the generator and the load distribution of the transmission line in real time, it is possible to ensure that the power system can maintain efficient operation under different operating conditions; this real-time adjustment can minimize energy waste and improve the overall efficiency of the power system; the closed-loop control mechanism means that the system can continuously adjust and optimize according to the current state, thereby ensuring that the system is always in the optimal operating state; this mechanism can significantly improve the response speed and stability of the power system and further improve the operating efficiency; when formulating the adjustment plan, the stability, economy and environmental protection of the power system are comprehensively considered; this comprehensive consideration can ensure that the adjustment plan will not sacrifice the stability of the system while improving efficiency; through the continuous updating of the dynamic load characteristic vector, the system can more accurately Reflect the actual load situation of the power system; by recalculating the electric energy error value, this technical solution can timely discover and correct the deviation in electric energy metering; it helps to improve the accuracy of electric energy metering and ensure the fairness and transparency of power trading; it continuously monitors and evaluates the operating status of the adjusted power system to timely discover potential problems and risks; it helps the system to take timely measures to correct and optimize, thereby further improving the accuracy of electric energy metering; when formulating the adjustment plan, environmental factors are fully considered; it helps to reduce environmental pollution and carbon emissions in the power production process and promote the green development of the power industry; by improving the operating efficiency and accuracy of the power system, it can reduce the cost of power production and transaction costs; it helps to enhance the overall competitiveness of the power industry and promote the healthy development of the power market.

[0043] In one embodiment of the present invention, the S52 includes: Based on sensor networks and Internet of Things technologies, key parameters of the power system are collected in real time, including voltage, current, power factor, frequency, etc., and the collected multi-source data are pre-processed; Based on data fusion algorithms, data from different sources are integrated to form a comprehensive view of the power system operation status; Based on the real-time monitored voltage and current data, the key features of dynamic loads are extracted using signal processing technology and machine learning algorithms; Reconstruct the dynamic load feature vector according to the extracted features; update the electric energy metering model based on the updated dynamic load feature vector; process the voltage and current data monitored in real time based on the updated metering model, and recalculate the electric energy error value; Conduct statistical analysis on the error values ​​and evaluate the accuracy and stability of the measurement model; if the error value exceeds the preset threshold, further adjustments and optimizations will be made; Based on real-time monitoring data and recalculated electric energy error values, the operating status of the power system is comprehensively evaluated, and the evaluation results are fed back to the planners of the adjustment plan for power generation and transmission strategies, so as to perform iterative optimization according to the actual situation of the current power system; At the same time, the updated dynamic load characteristic vector and electric energy error value are used as new input data to enter the next closed-loop control cycle.

[0044] The working principle of the above technical solution is: using the sensor network deployed in the power system to collect key parameters such as voltage, current, power factor and frequency in real time. These sensors transmit the collected data to the data processing center through the Internet of Things technology; pre-processing the collected multi-source data, including data cleaning, denoising, outlier detection and processing, etc., to ensure the accuracy and reliability of the data; integrating data from different sensors and sources, and using data fusion algorithms to form a comprehensive view of the power system operation status, so as to more comprehensively understand the current status of the power system; based on the real-time monitored voltage and current data, using signal processing technology and machine learning algorithms to extract key features of dynamic loads, such as current amplitude probability density, power factor change rate, phase difference distribution, etc. These features can reflect the load characteristics and change trends of the power system; based on the extracted features, the dynamic load feature vector is reconstructed. Including updating the elements of the feature vector, adjusting the dimension of the feature vector, introducing new features, etc., to ensure that the feature vector can more accurately reflect the current load status of the power system; based on the updated dynamic load feature vector, the electric energy metering model is updated. This includes adjusting the parameters of the metering model, introducing new metering algorithms or methods, optimizing the metering process, etc., to better adapt to the load characteristics of the current power system; using the updated metering model, processing the voltage and current data monitored in real time, and recalculating the electric energy error value; statistically analyzing the calculated error value to evaluate the accuracy and stability of the metering model. If the error value exceeds the preset threshold, it indicates that the metering model needs further adjustment and optimization; based on the real-time monitoring data and the recalculated electric energy error value, a comprehensive evaluation of the operating status of the power system is conducted. This includes evaluating the stability, economy, and environmental protection of the power system; feeding back the evaluation results to the planners of the adjustment plan for power generation and transmission strategies, and iteratively optimizing according to the actual situation of the current power system. This includes adjusting the output power of the generator, the load distribution of the transmission line, etc., to further improve the operating status of the power system; using the updated dynamic load characteristic vector and the electric energy error value as new input data to enter the next round of closed-loop control cycle. This means that the system will continuously monitor, evaluate, optimize and feedback to achieve continuous optimization and stable operation of the power system.

[0045] The effects of the above technical solutions are as follows: through sensor networks and Internet of Things technology, key parameters of the power system, including voltage, current, power factor and frequency, are collected in real time, ensuring the timeliness and accuracy of the data; data fusion algorithms are used to integrate data from different sources to form a comprehensive view of the power system operation status, which improves the visibility and transparency of the system status and helps to discover potential problems in a timely manner; through signal processing technology and machine learning algorithms, key features of dynamic loads are extracted from real-time monitoring data, such as current amplitude probability density, power factor change rate, phase difference distribution, etc., to improve the accuracy of load characteristic recognition; dynamic load feature vectors are reconstructed based on the extracted features to more accurately reflect the current load status of the power system, providing a solid foundation for subsequent updates of the electric energy metering model; based on the updated dynamic load feature vectors, the electric energy metering model is parameter adjusted, the algorithm is introduced or the process is optimized to better adapt to the load of the current power system. characteristics, significantly improving the accuracy and stability of electricity metering; by recalculating the electricity error value and conducting statistical analysis, the accuracy and stability of the metering model are objectively evaluated, providing a scientific basis for subsequent adjustments and optimizations; based on real-time monitoring data and recalculated electricity error values, a comprehensive evaluation of the operating status of the power system is conducted, covering multiple aspects such as stability, economy, and environmental protection, providing comprehensive guidance for the adjustment of power generation and transmission strategies; the evaluation results are fed back to strategy makers, and iterative optimization is carried out according to the actual situation of the current power system, forming a closed-loop control mechanism, which promotes continuous improvement and performance enhancement of the system; the updated dynamic load characteristic vector and electricity error value are used as new input data to enter the next round of closed-loop control cycle, realizing continuous monitoring and optimization of the power system; by continuously optimizing power generation and transmission strategies, the accuracy and stability of electricity metering are improved, energy waste and environmental pollution are reduced, and the sustainable development of the power industry is promoted.

[0046] One embodiment of the present invention, as Figure 2 As shown, a dynamic load electric energy metering optimization system comprises: Data acquisition module: Use high-precision sensors to collect voltage and current signals in real time under dynamic load conditions, and discretize the collected signals to obtain the voltage and current values ​​at the sampling points; Parameter extraction module: performs statistical analysis on the voltage and current values ​​of the sampling points, extracts key parameters, and constructs a dynamic load feature vector based on the extracted key parameters; Parameter acquisition module: calculate the probability density value of the current amplitude according to the dynamic load characteristic vector, and obtain the metering parameters through measurement and calculation, the metering parameters include power factor and phase difference; Adaptive adjustment module: Based on the pre-configured operation rules, combined with the probability density value of the current amplitude and the metering parameters, the electric energy error value is calculated; and the operation rules are adaptively adjusted according to the dynamic load characteristic vector; Real-time adjustment module: According to the electric energy error value, the power generation and transmission strategies are adjusted in real time.

[0047] The working principle of the above technical solution is as follows: first, the system deploys high-precision sensors, which can monitor the voltage and current signals under dynamic load environment in real time. The signals are usually discontinuous and change rapidly, so the sensors need to have sufficient sensitivity and sampling rate to capture these changes; the collected continuous voltage and current signals are then discretized, that is, the continuous signals are converted into voltage and current values ​​of a series of sampling points. This step is achieved by sampling the signals at regular intervals, and each sampling point represents the value of the signal at a certain moment; the voltage and current values ​​of the sampling points are statistically analyzed, including calculating statistical quantities such as the average, standard deviation, and peak value to understand the basic characteristics of the signal; key parameters such as voltage fluctuation range and current peak time are extracted from the statistical analysis results. These parameters can reflect the main characteristics of the dynamic load; based on the extracted key parameters, a dynamic load feature vector is constructed. A vector is a mathematical representation used to summarize and distinguish different types of dynamic loads; using the dynamic load feature vector, the probability density value of the current amplitude is calculated, which helps to understand the distribution of the current at different amplitudes; at the same time, the power factor (i.e., the ratio of active power to apparent power) and the phase difference (the phase angle difference between voltage and current) are obtained through measurement and calculation. These parameters are crucial for accurate metering of electric energy. Based on pre-configured operation rules (such as electric energy metering algorithm), the electric energy error value is calculated by combining the probability density value of the current amplitude with the metering parameters (power factor and phase difference). The error value reflects the difference between the actual metering result and the theoretical value. According to the dynamic load characteristic vector, the operation rules are adaptively adjusted, which means that the system can automatically adjust the metering algorithm according to the load changes to reduce errors and improve the accuracy of metering. Finally, according to the calculated electric energy error value, the power generation and transmission strategies are adjusted in real time. If the error value is large, it means that the current strategy may not be suitable for the current load conditions, so adjustments are needed to optimize the operation of the power system. By adjusting the output power of the generator, the allocation of the transmission line and other strategies, the stable operation of the power system under dynamic load conditions can be ensured, while reducing energy waste and power loss.

[0048] The effects of the above technical scheme are as follows: by using high-precision sensors for real-time acquisition, it is possible to capture subtle changes in voltage and current signals under dynamic load environments, thereby ensuring that the collected data is more accurate; the dynamic load feature vector constructed based on key parameters extracted by statistical analysis can fully reflect the dynamic characteristics of the load, providing a solid foundation for subsequent metering and error calculation; adaptively adjusting the operation rules according to the dynamic load feature vector can optimize the metering algorithm for different load conditions, further reduce errors, and improve the accuracy of electric energy metering; real-time adjustment of power generation and transmission strategies according to the electric energy error value can ensure that the power system maintains efficient operation under dynamic load conditions. It helps to reduce energy waste and improve energy utilization efficiency; through accurate metering and real-time adjustment strategies, it can reduce losses in the process of power transmission and distribution and improve the overall efficiency of the power system; the method can monitor and respond to load changes in real time, and respond to various load conditions through adaptive adjustment strategies, thereby enhancing the stability and reliability of the power system; through the analysis of dynamic load feature vectors, potential faults or abnormal conditions can also be discovered in time, providing early warning information for the maintenance and fault handling of the power system; the method combines high-precision sensors, data analysis, adaptive adjustment and other technical means to achieve intelligent management of the power system. It helps to promote the development of smart grids and improve the intelligence level of power systems; this method has good scalability and can be combined with other smart grid technologies, such as distributed energy management and demand response, to jointly build a more intelligent and efficient power system.

[0049] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A dynamic load electric energy metering optimization method, characterized in that: The method comprises: S1. Use high-precision sensors to collect voltage and current signals in a dynamic load environment in real time, and discretize the collected signals to obtain voltage and current values ​​at the sampling points; S2. Statistically analyze the voltage and current values ​​of the sampling points, extract key parameters, and construct a dynamic load feature vector based on the extracted key parameters; S3. Calculate the probability density value of the current amplitude according to the dynamic load characteristic vector, and obtain the metering parameters through measurement and calculation, wherein the metering parameters include power factor and phase difference; S4. Based on the pre-configured operation rules, the probability density value of the current amplitude and the metering parameters are combined to calculate the electric energy error value; and the operation rules are adaptively adjusted according to the dynamic load characteristic vector; S5. According to the electric energy error value, the power generation and transmission strategies are adjusted in real time.

2. A dynamic load electric energy metering optimization method according to claim 1, characterized in that: Said S1 comprises: S11. Real-time capture of voltage and current signals under dynamic load conditions by deploying high-precision voltage and current sensors at key locations in the power system; S12, filtering the continuous voltage and current signals output by the sensor through digital filtering technology, and dynamically adjusting the filtering parameters according to the signal characteristics; S13, sampling the preprocessed signal at equal intervals to obtain discrete sampling points.

3. A dynamic load electric energy metering optimization method according to claim 2, characterized in that: The S12 comprises: Evaluate and select the adaptive filtering algorithm based on the characteristics of voltage and current signals under dynamic load conditions; Based on the selected adaptive filtering algorithm, corresponding parameter configuration is performed, wherein the corresponding parameters include filter order, learning rate and forgetting factor; During the filtering process, the statistical characteristics of the voltage and current signals, as well as the output signal quality of the filter, are monitored in real time; Based on the results of real-time monitoring, formulate rules for dynamic adjustment of filter parameters; feed back the quality of the filter output signal after parameter adjustment into the parameter adjustment rules for iterative optimization; After the filtering process is completed, the quality of the filter's output signal is evaluated; based on the evaluation results, the filter's parameters and algorithms are further optimized.

4. The method for optimizing dynamic load electric energy metering according to claim 1, characterized in that: The S2 comprises: S21, performing statistical analysis on the voltage and current values ​​of the sampling points, performing spectrum analysis on the signals using fast Fourier transform, and extracting frequency components and their energy distribution; S22. Based on the above statistical characteristics and spectrum analysis results, key parameters are extracted and combined into a dynamic load feature vector.

5. A dynamic load electric energy metering optimization method according to claim 4, characterized in that: The S22 comprises: Calculate basic parameters according to the voltage and current values ​​of the sampling points, wherein the basic parameters include maximum value, minimum value, average value and variance; Based on the peak detection algorithm, the peak values ​​in the voltage and current signals and their corresponding time points are identified, where the peak time is used to reflect the speed and trend of signal changes; Based on step S21, the signal is subjected to spectrum analysis again; spectrum feature extraction is performed again through resolution bandwidth and filter type; Based on the spectrum analysis results, calculate the offset between the actual frequency of the signal and the theoretical frequency; On the basis of preliminary parameter extraction and spectrum feature extraction, the elements for constructing feature vectors are selected according to actual application requirements and the purpose of data analysis; The selected features are integrated to construct a dynamic load feature vector; in the process of constructing the feature vector, continuous optimization and adjustment are performed.

6. The method for optimizing dynamic load electric energy metering according to claim 1, characterized in that: The S3 includes: S31, calculating the probability density value of the sampling point according to the current amplitude data of the sampling point by using the histogram method; wherein the probability density value reflects the distribution characteristics of the current amplitude and is used to identify abnormal current conditions; S32. Measure and calculate the power factor and phase difference through a measurement algorithm.

7. The method for optimizing dynamic load electric energy metering according to claim 1, characterized in that: The S4 comprises: S41. Preset the operation rules according to the basic principles of electric energy measurement and dynamic load characteristics; S42, calculating the electric energy error value based on the preset operation rules and the extracted dynamic load characteristic vector, wherein the error value is used to reflect the deviation between the actual measurement value and the theoretical value; S43. According to the change of the dynamic load characteristic vector, the parameters or weights in the operation rules are adjusted in real time through the neural network, and the operation rules are automatically optimized based on the learning historical data and current characteristics.

8. A dynamic load electric energy metering optimization method according to claim 7, characterized in that: The S43 includes: According to the complexity of dynamic load characteristic vector and the diversity of operation rules, the neural network model is selected; According to the selected model, a neural network architecture is constructed, and the number of nodes of the input layer, the hidden layer and the output layer is determined, wherein the input layer receives the dynamic load feature vector, and the output layer outputs the electric energy error value or the adjustment parameter of the operation rule; In the process of electric energy metering, the changes of dynamic load characteristic vectors are monitored in real time. Based on the forward propagation and back propagation mechanism of the neural network, the weight and bias of the neural network are dynamically adjusted according to the input of the current characteristic vector and the output of the electric energy error value. Based on the learning ability of neural network, the mapping relationship between feature vectors and electric energy error values ​​is extracted from historical data; and in the real-time metering process, the newly collected feature vectors and electric energy error values ​​are used as new training samples and continuously input into the neural network for online learning; According to the output results of the neural network and the changes of the current feature vector, the parameters or weights in the operation rules are automatically optimized; and the optimized operation rules are fed back to the electric energy metering system.

9. The method for optimizing dynamic load electric energy metering according to claim 1, characterized in that: The S5 comprises: S51. Formulate an adjustment plan for power generation and transmission strategy based on the electric energy error value; and adjust relevant parameters of the generator in real time according to the adjustment plan; S52, continuously monitoring and evaluating the adjusted power system operation status, acquiring new voltage and current data, updating the dynamic load characteristic vector, and recalculating the electric energy error value; S53. According to the evaluation results, the adjustment strategy is iteratively optimized to form a closed-loop control.

10. A dynamic load electric energy metering optimization system, characterized in that: The system comprises: Data acquisition module: Use high-precision sensors to collect voltage and current signals in real time under dynamic load conditions, and discretize the collected signals to obtain the voltage and current values ​​at the sampling points; Parameter extraction module: performs statistical analysis on the voltage and current values ​​of the sampling points, extracts key parameters, and constructs a dynamic load feature vector based on the extracted key parameters; Parameter acquisition module: calculate the probability density value of the current amplitude according to the dynamic load characteristic vector, and obtain the metering parameters through measurement and calculation, the metering parameters include power factor and phase difference; Adaptive adjustment module: Based on the pre-configured operation rules, combined with the probability density value of the current amplitude and the metering parameters, the electric energy error value is calculated; and the operation rules are adaptively adjusted according to the dynamic load characteristic vector; Real-time adjustment module: According to the electric energy error value, the power generation and transmission strategies are adjusted in real time.

Images