Electric energy metering error compensation method, device and system

Abstract

The invention relates to the technical field of electric energy metering error compensation, in particular to an electric energy metering error compensation method, device and system, and the method comprises the steps: collecting a single-phase electric energy signal of a secondary side of a power grid transformer, and uniformly dividing the single-phase electric energy signal into each section of metering signal sequence; calculating high-frequency unsteady-state energy to quantify distortion by extracting first-order and second-order quasi-steady-state components of a metering signal sequence; meanwhile, impact pulses in the dynamic components are recognized, impact characteristics are analyzed through clustering, and the nonlinear impact disturbance degree is comprehensively calculated; the fundamental wave separation strength of each section of metering signal sequence is calculated, the initial step length factor of the LMS filtering algorithm is dynamically adjusted, effective separation of fundamental waves and distortion signals is achieved, and therefore electric energy metering error compensation is completed. The method aims at solving the problem that the metering precision of a traditional fixed step length factor is reduced under the power grid nonlinear load high permeability working condition, and the accuracy and reliability of electric energy metering are ensured.

Description

Technical Field

[0001] This application relates to the field of smart electricity technology, specifically to a method, device, and system for compensating for electricity metering errors. Background Technology

[0002] Load flexibility regulation of smart energy units refers to the process of real-time monitoring, analysis, and optimization of loads in a power system using a smart energy management system. The goal of this regulation is to achieve efficient energy utilization and demand-side management while ensuring stable power system operation. Currently, load regulation for smart units mainly includes real-time monitoring, load forecasting, demand response, optimized scheduling, remote control, user interaction, and energy efficiency management. Real-time monitoring and load forecasting provide the smart energy unit with fundamental data for load regulation, thereby enabling flexible load regulation and improving energy utilization efficiency and power system operational stability.

[0003] With the expansion of the energy market and the rise in peak loads, the risks of supply and demand gaps in specific time periods and local areas have become more prominent. Therefore, the accuracy of analysis for collecting electricity consumption data on the demand side and conducting real-time monitoring and forecasting to cope with complex electricity consumption scenarios is greatly affected. Traditional data monitoring and forecasting methods for analyzing data on electricity consumption change scenarios with time-varying characteristics, such as traditional time series forecasting, may lead to significant deviations in the analysis of electricity consumption status and load demand on the demand side when dealing with variable electricity consumption scenarios. This reduces the accuracy of load analysis for smart energy units and consequently reduces the load flexibility adjustment effect of smart energy units. Summary of the Invention

[0004] To address the aforementioned technical problems, the purpose of this application is to provide a method, device, and system for compensating for electricity metering errors. The specific technical solution adopted is as follows:

[0005] In a first aspect, embodiments of this application provide a method for compensating for electrical energy metering errors, the method comprising:

[0006] Collect single-phase power signals from the secondary side of the power grid transformer and divide them evenly into various metering signal sequences;

[0007] First-order and second-order quasi-steady-state components of the metrological signal sequence are extracted and their squared differences are analyzed to calculate high-frequency unsteady-state energy;

[0008] Extract the dynamic components of the metering signal sequence; obtain any maximum point on the dynamic component and the two minimum points closest to its left and right sides; use the time interval between the two minimum points to determine the impact period of the maximum point as the impact peak point; use the difference between the mean of all data points between the two minimum points and the numerical difference between the impact valley point and the impact peak point to determine the impact pulse intensity of the maximum point as the impact peak point.

[0009] The impact characterization vector of each impact peak in the dynamic component is constructed using the impact period and impact pulse intensity. The number of clusters obtained by clustering based on the similarity between the impact characterization vectors of any two impact peaks is multiplied by the average impact pulse intensity of all impact peaks in the dynamic component to determine the nonlinear impact disturbance degree of the measurement signal sequence.

[0010] The initial step size factor of the LMS filtering algorithm is dynamically adjusted based on the high-frequency unsteady energy and nonlinear impact disturbance degree to separate the distorted signal in the metering signal sequence, thereby completing the power metering error compensation.

[0011] Preferably, the single-phase electrical energy signal includes a single-phase current signal and a single-phase voltage signal.

[0012] Preferably, the method for extracting the first-order and second-order quasi-steady-state components of the metering signal sequence is as follows:

[0013] Obtain all maximum and minimum points in the measurement signal sequence. Using all maximum points in the measurement signal sequence as input, fit the upper envelope of the measurement signal sequence using linear spline interpolation.

[0014] Using all the minimum points in the measurement signal sequence as input, the lower envelope of the measurement signal sequence is fitted using linear spline interpolation.

[0015] The mean of the upper and lower envelopes is taken as the first-order quasi-steady-state component of the measurement signal sequence;

[0016] Using the same method as that used to obtain the first-order quasi-steady-state component of the measurement signal sequence, the first-order quasi-steady-state component of the first-order quasi-steady-state component is obtained and used as the second-order quasi-steady-state component of the measurement signal sequence.

[0017] Preferably, the dynamic component is obtained by subtracting the second-order quasi-steady-state component from the measurement signal sequence as the dynamic component of the measurement signal sequence.

[0018] Preferably, the formula for calculating the impact pulse intensity at the maximum point, which is used as the impact peak point, is as follows:

[0019]

[0020] In the formula, The impact pulse intensity is the impact peak point p, which is the maximum point p on the dynamic component. It is the electrical energy value at the maximum point p. It is the electrical energy value of the impact valley point q within the impact cycle where the maximum point p is the impact peak point. It is the local average electrical energy at the maximum point p;

[0021] Among them, the impact valley point q is the minimum point corresponding to the minimum time interval between the maximum point p and the two minimum points closest to it on its left and right sides;

[0022] The local average electrical energy of the maximum point p is the average of all data points between the maximum point p and the two minimum points on its left and right sides that are closest to it.

[0023] Preferably, before constructing the impact characterization vector of each impact peak in the dynamic component using the impact period and impact pulse intensity, the impact period and impact pulse intensity of all impact peaks in the dynamic component are first subjected to maximum and minimum normalization processing.

[0024] Preferably, the dynamic adjustment formula for the initial step size factor is:

[0025]

[0026] In the formula, , These are the initial step size factors when processing the x-th and (x-1)-th measurement signal sequences using the LMS filtering algorithm, respectively. When x=1... Set as a fixed step size factor for the LMS filtering algorithm. It is the fundamental frequency separation strength of the x-th measurement signal sequence. It is a preset adjustment coefficient with a value range of [0,1].

[0027] The fundamental wave separation strength of the x-th measurement signal sequence is the normalized product of the high-frequency unsteady energy and the nonlinear impulse disturbance degree of the x-th measurement signal sequence.

[0028] Preferably, the method for compensating for electricity metering errors is as follows:

[0029] Calculate the product of the distorted voltage and distorted current of a single phase at any given time, and use it as the distorted power of the single phase.

[0030] Calculate the product of the measured voltage and measured current of a single phase at any given time, and use it as the measured power of the single phase.

[0031] The difference between the measured power and the distorted power of a single phase is calculated and used as the power calibration value for that single phase, which is then used to achieve power metering error compensation calibration.

[0032] Secondly, embodiments of this application provide an energy metering error compensation device, the device storing a computer program, which, when executed by a processor, implements the energy metering error compensation method described above.

[0033] Thirdly, embodiments of this application also provide an energy metering error compensation system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the energy metering error compensation method described above.

[0034] As can be seen from the above embodiments, the power metering error compensation method, device, and system provided in this application have at least the following beneficial effects:

[0035] 1. This application segments the AC power metering signal sequence and extracts quasi-steady-state and dynamic components, which can effectively identify various distortion types such as harmonics, pulse transients, and voltage flicker. It is applicable to complex scenarios such as industrial loads and electric vehicle charging, and improves the robustness of the power metering system under non-steady-state conditions.

[0036] 2. This application constructs nonlinear impact disturbance degree by impact pulse intensity and impact cluster number, comprehensively evaluates the cumulative distortion effect of nonlinear impact loads (such as industrial equipment such as electric arc furnaces and rolling mills) on power signals, improves the recognition rate of "event-based" impact pulses by intelligent metering equipment, and improves the applicability of power metering in industrial parks under conditions of high penetration of nonlinear loads.

[0037] 3. This application reflects the difficulty of separating the fundamental wave from the power signal by the fundamental wave separation strength, and dynamically adjusts the step size factor of the LMS adaptive filtering algorithm. When the power signal distortion is severe, the step size is increased to accelerate convergence, and when the distortion is small, the step size is decreased to ensure accuracy. This overcomes the problem of decreased metering accuracy under the condition of high penetration of nonlinear loads in the power grid by the traditional fixed step size factor. By responding to the degree of power signal distortion in real time, the filtering convergence process is optimized, which can accelerate convergence when nonlinear impact loads occur frequently and the distortion is severe, and ensure that the power metering data does not diverge when the nonlinear impact load is single and the distortion is small, thus ensuring the accuracy and reliability of power metering. Attached Figure Description

[0038] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart illustrating the steps of an electricity metering error compensation method provided in one embodiment of this application. Detailed Implementation

[0040] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an electricity metering error compensation method, device, and system proposed according to this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0041] Unless otherwise specified and limited, terms such as “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items. All technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0042] The following description, in conjunction with the accompanying drawings, details a specific scheme for an electricity metering error compensation method, device, and system provided in this application.

[0043] Please see Figure 1 The diagram illustrates a flowchart of a method for compensating for electrical energy metering errors according to an embodiment of this application. The method includes the following steps:

[0044] Step 1: Collect the single-phase power signal from the secondary side of the power grid transformer and divide it evenly into various metering signal sequences.

[0045] This application involves installing smart metering equipment on the secondary side of voltage transformers / current transformers in the power grid to collect real-time data on single-phase electrical energy signals under AC operating conditions. These single-phase electrical energy signals include single-phase current signals and single-phase voltage signals. Specifically, the smart metering equipment can be a smart energy meter or a power analyzer.

[0046] In this embodiment, the acquisition frequency of the smart metering device is set to 20kHz, and the acquisition duration Q is 60s, in order to acquire single-phase power signals.

[0047] Furthermore, this application segments the single-phase power signal, dividing it evenly into several metering signal sequences. In this embodiment, it is set to 6 segments, with each segment having a duration of 10 seconds.

[0048] Step 2: By extracting the first-order and second-order quasi-steady-state components of the measurement signal sequence, the high-frequency unsteady-state energy is calculated to quantify the distortion; at the same time, the impact pulses in the dynamic components are identified, and the impact characteristics are analyzed by clustering to comprehensively calculate the nonlinear impact disturbance degree.

[0049] Correction and error compensation of electricity metering data can increase the stability of power system operation and facilitate fair trading in the electricity market. However, due to various factors such as equipment aging, changes in environmental factors and sudden changes in grid operating conditions, electricity metering data contains distorted signals such as harmonics and pulse transients. If not corrected and compensated in time, it will cause uncontrollable obstacles to the normal operation of the power system. The key to accurate electricity metering under the distortion conditions of smart grid is to separate the distorted power from the measured power.

[0050] The complex dynamic electrical energy signals collected by smart metering devices in the power grid under AC operating conditions can be decomposed into slowly fluctuating quasi-steady-state components and rapidly changing dynamic components. The quasi-steady-state components change slowly and generally reflect slow fluctuations on a large time scale, while the dynamic components reflect rapid fluctuations on a small time scale.

[0051] This application obtains a measurement signal sequence with arbitrary segmentation, acquires all maximum and minimum points in the measurement signal sequence, uses all maximum points in the measurement signal sequence as input, and fits the upper envelope of the measurement signal sequence using linear spline interpolation. Similarly, all local minima within the measurement signal sequence are used as input, and the lower envelope of the measurement signal sequence is fitted using linear spline interpolation. Through the formula The average envelope of the measurement signal sequence is obtained, which is the first-order quasi-steady-state component of the measurement signal sequence. This method is used to capture the slowly fluctuating trend components in a measurement signal sequence. Linear spline interpolation is a well-known technique and will not be elaborated upon further.

[0052] Because complex dynamic electrical signals contain many non-steady-state distorted signal components such as pulse transients, voltage flicker, and power interruptions, they have the characteristics of fast time-varying, large fluctuations, and non-stationary randomness. They exhibit large amplitude fluctuations on the millisecond time scale, and the first-order quasi-steady-state components manifest as many high-frequency "glitch" and "bump" points.

[0053] Accordingly, this application obtains the first-order quasi-steady-state component. Based on the steps described above, obtain the first-order quasi-steady-state components from all maxima and minima. The upper and lower envelopes are calculated, and the first-order quasi-steady-state components are computed. The average envelope, i.e., obtaining the first-order quasi-steady-state component. The first-order quasi-steady-state component is denoted as the second-order quasi-steady-state component of the measurement signal sequence. Second-order quasi-steady-state components This further smooths out high-frequency fluctuations in the first-order quasi-steady-state component, reflecting a slower and more stable trend component in the electricity metering data.

[0054] This application obtains the high-frequency unsteady-state energy of an arbitrary segmented metering signal sequence using the following formula. :

[0055]

[0056] In the formula, , These are the first-order quasi-steady-state component and the second-order quasi-steady-state component of the metering signal sequence, respectively, where T is the sequence length of the metering signal sequence. Since non-steady-state distortions in the power grid signal can cause metering errors, calculating the sum of squares of the first-order and second-order quasi-steady-state components is to highlight the severity of the non-steady-state components. The difference between the first-order and second-order quasi-steady-state components represents the high-frequency dynamic components removed from the first-order quasi-steady-state component. This quantifies the energy of these high-frequency non-steady-state signals. (High-frequency non-steady-state energy...) The larger.

[0057] This application obtains a measurement signal sequence and its second-order quasi-steady-state component. The result of subtracting the second-order quasi-steady-state component from the measurement signal sequence is used as the dynamic component of the measurement signal sequence, which is used to characterize the rapid fluctuation and change of the measurement signal sequence on a small time scale, and comprehensively reflects the fast time-varying nature, large volatility and non-stationary randomness of the measurement signal sequence.

[0058] As the proportion of high-power dynamic loads connected to the grid continues to rise, the current, voltage, and power signals of the power grid exhibit complex dynamic characteristics. Common impact load devices include electric arc furnaces and rolling mills. Nonlinear impact loads easily generate unsteady and distorted electrical energy signals in the power grid. The amplitude of these unsteady and distorted electrical energy signals fluctuates rapidly and over a wide range on a very small time scale. In the dynamic components of the metering signal sequence, this manifests as brief, sudden impact pulses. The higher the intensity of the nonlinear impact load, the greater the difference in amplitude between the peak and trough of the impact pulse.

[0059] Accordingly, this application obtains the maximum and minimum points of the measurement signal sequence on the dynamic component, and obtains the order value on the dynamic component corresponding to each extreme point. Each maximum point may represent a nonlinear impulse peak in the dynamic component. For any maximum point p on the dynamic component, within the dynamic component, taking the maximum point p as the center, the nearest minimum point q1 to the left of the maximum point p is obtained, and the nearest minimum point q2 to the right of the maximum point p is obtained.

[0060] It should be noted that: if there is no minimum point to the left of the maximum point p, the first data point of the dynamic component is regarded as the minimum point q1; if there is no minimum point to the right of the maximum point p, the last data point of the dynamic component is regarded as the minimum point q2.

[0061] This application denotes the maximum point p as the impact peak point p, calculates the time interval between the dynamic component and the minimum point q1 to the minimum point q2, and denotes it as the impact period with the maximum point p as the impact peak point. The average of the electrical energy values ​​of all data points within the dynamic component from the minimum point q1 to the minimum point q2 is statistically calculated and denoted as the local average electrical energy value at the maximum point p. Based on the fast time-varying nature of the impact pulse, the time intervals between the impact peak and the minimum points q1 and q2 are calculated. The minimum point corresponding to the minimum time interval is selected as the maximum point p and the impact valley point q within the impact period of the impact peak.

[0062] Furthermore, the impact pulse intensity, with the maximum point p on the dynamic component as the impact peak, is obtained using the following formula. :

[0063]

[0064] In the formula, It is the electrical energy value at the maximum point p. It is the electrical energy value of the impact trough q within the impact period where the maximum point p is the impact peak. This application is approved. It reflects the average local electrical energy level of the impact pulse corresponding to the maximum point p. , These respectively reflect the peak and trough values ​​of the electrical signal when the impulse pulse occurs; the power grid equipment generates huge changes in electrical energy in a very short time, causing the electrical signal to fluctuate drastically from trough to peak, i.e. The larger the value, the stronger the impact and nonlinear characteristics of the nonlinear load equipment, and the greater the impact pulse intensity. The larger.

[0065] Furthermore, this application obtains the impact period and impact pulse intensity of all impact peaks in the dynamic component of the measurement signal sequence, performs Max-Min normalization on each, and then constructs the impact characterization vector for each impact peak. This application constructs a similarity matrix corresponding to all impact peaks by calculating the cosine similarity of the impact characterization vectors corresponding to any two impact peaks as the similarity calculation method, and uses this matrix as the input to the spectral clustering algorithm. In this embodiment, the number of clusters is obtained using the elbow method, outputting k clusters. Since the impact characterization vectors corresponding to the impact peaks within each cluster are similar and the impact pulse characteristics are similar, each cluster is considered an impact cluster. All impact peaks within the same impact cluster may be caused by the same nonlinear impact load.

[0066] Cosine similarity, spectral clustering algorithm, and elbow method are all well-known techniques and will not be elaborated further.

[0067] Furthermore, this application obtains the nonlinear impulse disturbance degree of the measurement signal sequence using the following formula. :

[0068]

[0069] In the formula, It is the number of clusters within the dynamic component. The impact pulse intensity is the impact peak point p, which is the maximum point p on the dynamic component. It is the total number of maxima within the dynamic component.

[0070] Used to reflect the frequency of different types of impact pulses occurring under nonlinear impact loads within a metering signal sequence. The larger the value, the more different nonlinear impact load sources are generated in a short period of time, and the more impact pulses with different characteristics are mixed in a short period of time. The more complex the shape of the electrical signal, the higher the interference degree of the nonlinear impact load on the electrical signal. The larger; It is used to reflect the cumulative distortion of the electrical signal caused by nonlinear impact loads within the metering signal sequence. The larger the value, the greater the severity of the unsteady-state distortion of the electrical signal caused in a short period of time, and the more likely it is to lead to power grid faults such as damage to power equipment and malfunction of protection systems. The higher the interference of nonlinear impact loads on the electrical signal, the greater the interference. The larger.

[0071] Step 3: By calculating the fundamental wave separation strength of each segment of the metering signal sequence, the initial step size factor of the LMS filtering algorithm is dynamically adjusted to achieve effective separation of the fundamental wave and the distorted signal, thereby completing the compensation for power metering errors.

[0072] Based on the above steps, this application can obtain the high-frequency unsteady-state energy and nonlinear impulse disturbance degree of the metering signal sequence. The high-frequency unsteady-state energy is used to capture the faster random fluctuations superimposed on the slow changing trend of the power signal, and is used to evaluate the overall stability of the power grid's power quality. The larger the high-frequency unsteady-state energy, the worse the power quality stability, and the more effectively the fundamental current and fundamental voltage can be separated from the measured power signal (i.e., the metering signal sequence). The nonlinear impulse disturbance degree focuses on significant "event-based" impulse pulses. With the increasing penetration rate of nonlinear loads in the power grid, attention is paid to the impact of the start-up, shutdown, and violent fluctuations of impact loads such as electric arc furnaces, rolling mills, and welding machines in industrial parks on the distortion of the power signal. The nonlinear impulse loads of these industrial equipment cause more severe distortion of the power signal, increasing the difficulty of effectively separating the fundamental current and fundamental voltage from the measured power signal (i.e., the metering signal sequence).

[0073] This application calculates the product of high-frequency unsteady energy and nonlinear impulse disturbance degree of any metering signal sequence, and performs Max-Min normalization on the above product results of all metering signal sequences. The normalized product result is recorded as the fundamental wave separation strength, which is used to reflect the difficulty of effectively separating the fundamental wave current and fundamental wave voltage from the metering signal sequence.

[0074] The LMS (Least Mean Squares) adaptive filtering algorithm is simple and efficient. It uses a unit sine wave in phase with the power supply voltage as a reference signal, filters out the fundamental component of the measured electrical energy signal, and directly obtains the distorted voltage and current signals, thereby achieving accurate measurement of real and distorted electrical energy. However, with the large-scale grid connection of distributed photovoltaic, electric vehicles and other new energy equipment, and the continuous increase in the penetration rate of nonlinear loads, the phenomenon of nonlinear impact loads in industrial parks is becoming increasingly serious. The fixed step size factor of LMS is difficult to adapt to the dynamic operating conditions of the power grid, resulting in a decrease in measurement accuracy.

[0075] This application obtains the initial step size factor for processing the x-th measurement signal sequence using the LMS filtering algorithm according to the following formula. :

[0076]

[0077] In the formula, It is the initial step size factor when processing the (x-1)th measurement signal sequence using the LMS filtering algorithm. When x=1, The fixed step size factor for the LMS filtering algorithm is set to 0.5 in this embodiment. It is the fundamental frequency separation strength of the x-th measurement signal sequence. It is a preset adjustment coefficient, and the value range of this application is [0,1]. In this embodiment, it is 0.5.

[0078] The higher the frequency of power grid phenomena such as pulse transients, voltage flicker, and power outages in the metering signal sequence, the worse the power quality stability. Furthermore, the richer the components of impact loads such as electric arc furnaces, rolling mills, and welding machines, the more severe the overall and local distortion of the power signal. To more effectively separate the fundamental voltage and current, the LMS filtering algorithm should have a larger initial step size factor to provide faster initial convergence and avoid reducing the timeliness of power metering calibration and compensation. Conversely, if the power signal distortion in the metering signal sequence is low and it is closer to the fundamental voltage and current, a smaller initial step size factor should be used to avoid signal divergence caused by a large step size factor in the LMS filtering algorithm, thereby improving the accuracy of power metering calibration and compensation.

[0079] Because the time transitions of the various segments of the AC power signal metering sequence are smooth, the difficulty of separating the fundamental voltage and fundamental current varies similarly. Therefore, the initial step size factor of the LMS filtering algorithm may also be approximate. This application... Calculate the initial step size factor for processing the x-th measurement signal sequence using the LMS filtering algorithm. That is, the initial step size factor of each measurement signal sequence is related to the initial step size factor of the previous measurement signal sequence.

[0080] Furthermore, the initial step size factor of each segment of the AC power signal sequence is obtained, and each segment of the signal sequence is used as the input to the LMS adaptive filtering algorithm. A unit sine wave in phase with the current and voltage is used as the reference signal, and the distorted signal of each segment of the signal sequence is output. Specifically, when the signal sequence is a voltage sequence, the distorted voltage is output, and when the signal sequence is a current sequence, the distorted current is output.

[0081] The product of the distorted voltage and distorted current of a single phase at any given time is calculated as the distorted power of the single phase. The product of the measured voltage and measured current of a single phase at any given time is also calculated as the measured power of the single phase. The difference between the measured power and the distorted power of a single phase is calculated as the power calibration value of the single phase. This method is used to implement a power metering error compensation calibration method in the scenario of a nonlinear load with high penetration rate in the power grid.

[0082] Based on the same inventive concept as the above method, this application provides an energy metering error compensation device. The device stores a computer program, which, when executed by a processor, implements the energy metering error compensation method described above.

[0083] Based on the same inventive concept as the above method, this application embodiment also provides an energy metering error compensation system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the energy metering error compensation method described above.

[0084] The various embodiments in this application are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0085] It should be noted that, unless otherwise specified and limited, terms such as “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a circuit structure, article, or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such article or device. Without further limitations, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes said element. Furthermore, the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0086] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not invented in this application.

[0087] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method of compensating for an error in electrical energy metering, characterized by, The method comprises: Collecting a single-phase power signal at a secondary side of a power grid transformer and uniformly dividing the single-phase power signal into each segment of a metering signal sequence; Extracting first-order and second-order quasi-steady components of the metering signal sequence and analyzing square differences thereof to calculate high-frequency non-steady energy; Extracting dynamic components of the metering signal sequence; obtaining an arbitrary maximum value point and two minimum value points closest to the maximum value point on left and right sides; determining the maximum value point as an impact cycle of an impact peak point by using a time interval between the two minimum value points; and determining an impact pulse intensity of the maximum value point as the impact peak point by using a numerical difference between a mean value of all data points between the two minimum value points and a value of an impact valley point and a numerical difference between the value of the impact valley point and the value of the impact peak point; Using the impact cycle and the impact pulse intensity to construct an impact feature vector of each impact peak point in the dynamic components; and determining a non-linear impact disturbance degree of the metering signal sequence by multiplying a number of clustering clusters obtained by clustering according to a similarity between impact feature vectors of any two impact peak points and an average value of impact pulse intensities of all impact peak points in the dynamic components. According to the high-frequency non-steady energy and the non-linear impact disturbance degree, an initial step factor in LMS filtering algorithm processing of the corresponding metering signal sequence is dynamically adjusted to separate a distortion signal in the metering signal sequence, so as to complete power metering error compensation.

2. The method of claim 1, wherein, The single-phase power signal comprises a single-phase current signal and a single-phase voltage signal.

3. The method of claim 1, wherein, The extraction method of the first-order and second-order quasi-steady components of the metering signal sequence comprises: Obtaining all maximum value points and minimum value points in the metering signal sequence; using linear spline interpolation to fit an upper envelope of the metering signal sequence by taking all maximum value points in the metering signal sequence as input; Using linear spline interpolation to fit a lower envelope of the metering signal sequence by taking all minimum value points in the metering signal sequence as input; Taking a mean value of the upper envelope and the lower envelope as the first-order quasi-steady component of the metering signal sequence; Using the same method of obtaining the first-order quasi-steady component of the metering signal sequence to obtain a first-order quasi-steady component of the first-order quasi-steady component as the second-order quasi-steady component of the metering signal sequence.

4. The method of claim 1, wherein, The dynamic component is obtained by subtracting the second-order quasi-steady component from the metering signal sequence.

5. The method of claim 4, wherein, The calculation formula of the impact pulse intensity of the maximum value point as the impact peak point is: wherein is the impact pulse intensity with the maximum point p of the dynamic component as the impact peak point, is the electric energy value of the maximum point p, is the electric energy value of the impact valley point q in the impact cycle in which the maximum point p is the impact peak point, is the local electric energy average value of the maximum point p; Wherein, the impact valley point q is the minimum value point corresponding to the minimum time interval between the maximum value point p and the minimum value points on the left and right sides of the maximum value point p; The local power mean value of the maximum value point p is the mean value of all data points between the maximum value point p and the minimum value points on the left and right sides of the maximum value point p.

6. The method of claim 5, wherein, Before constructing the impact feature vector of each impact peak point in the dynamic components by using the impact cycle and the impact pulse intensity, maximum and minimum normalization processing is performed on the impact cycle of all impact peak points and the impact pulse intensity of all impact peak points in the dynamic components, respectively.

7. The method of claim 1, wherein, The dynamic adjustment formula of the initial step factor is: In the formula, , are initial step factors when the xth and (x-1)th metering signal sequences are processed by the LMS filtering algorithm respectively, when x=1, is set as a fixed step factor of the LMS filtering algorithm, is a fundamental wave separation strength of the xth metering signal sequence, is a preset adjustment coefficient, and the value range is [0, 1]. Wherein, the fundamental separation degree of the xth metering signal sequence is a normalized product result of the high-frequency non-steady energy and the non-linear impact disturbance degree of the xth metering signal sequence.

8. The method of claim 2, wherein, The method for power metering error compensation is: The product of the distortion voltage and the distortion current of the single phase at any time is calculated as the distortion power of the single phase; The product of the measured voltage and the measured current of the single phase at any time is calculated as the measured power of the single phase; The difference between the measured power and the distortion power of the single phase is calculated as the power calibration value of the single phase, which is used to realize the compensation and calibration of the electric energy metering error.

9. An electrical energy metering error compensation device, said device having stored therein a computer program, characterized in that, The computer program is executed by the processor to realize the electric energy metering error compensation method according to any one of claims 1 to 8.

10. An electric energy metering error compensation system comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to realize the electric energy metering error compensation method according to any one of claims 1 to 8.

Images