A multi-loop electric energy metering error compensation method suitable for extreme environment

Abstract

This invention discloses a method for compensating metering errors in multi-loop energy meters applicable to extreme environments, relating to the field of energy metering and error compensation technology. The method includes: synchronously sampling multiple channels using a multi-loop energy meter to obtain sampled data, and then performing dimensionless normalization processing; calculating the comprehensive disturbance intensity based on the dimensionless normalized sampled data; constructing a dynamic residual state based on the comprehensive disturbance intensity through recursion with self-feedback, mutual feedback, and rate of change suppression; mapping the comprehensive disturbance intensity and the dynamic residual state to generate a predicted energy error; and using the predicted energy error as an error correction amount, combined with historical compensated cumulative energy values ​​for compensation calculation to obtain the compensated cumulative energy value. This method solves the problems of significantly increased errors during multi-load parallel operation, and the tendency for compensation failure or overcompensation to occur during sudden environmental changes, making it difficult to achieve long-term stable and high-precision metering under extreme and complex operating conditions.

Description

Technical Field

[0001] This invention relates to the field of electricity metering and error compensation technology, and in particular to a method for compensating metering errors in multi-loop electricity meters suitable for extreme environments. Background Technology

[0002] Multi-circuit energy meters, as key metering devices capable of independently metering multiple electricity consumption types within a single device, have been widely used in various combined scenarios such as residential electricity consumption, electric heating electricity consumption, industrial and commercial electricity consumption, and three-phase power loads. Through multi-circuit synchronous sampling and centralized management, these meters achieve a "one meter, multiple uses" metering mode, which is of great significance in reducing power grid construction costs, simplifying wiring structures, and improving the efficiency of metering asset management.

[0003] In real-world operating environments, multi-loop energy meters are typically deployed in outdoor distribution boxes, building metering boxes, and complex industrial environments. Their operating conditions often exhibit characteristics such as wide temperature ranges, high humidity, air pressure fluctuations, and frequent load changes. In such environments, voltage and current sampling links and metering references are easily affected by environmental factors, leading to drift. Simultaneously, multiple metering channels within a multi-loop energy meter operate in parallel within a compact structure. Heat conduction and electromagnetic coupling effects may exist between different loops, meaning that the metering status of a single loop is not only related to its own load but also to the operating status of other loops. This multi-loop collaborative operation characteristic makes the metering process exhibit more complex dynamic changes.

[0004] Therefore, in the application of multi-loop energy meters, how to effectively characterize and process the deviations generated during the metering process under complex environmental conditions and multi-loop coupled operation, so as to maintain the stability and consistency of energy metering results, has become an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a method for compensating metering errors in multi-loop energy meters suitable for extreme environments. It addresses the problems of existing technologies lacking unified modeling capabilities for the synergistic effects of multiple factors such as temperature, humidity, reference source drift, and clock offset, and failing to consider the impact of inter-loop thermal coupling and electromagnetic interference on metering accuracy, leading to a significant increase in errors during parallel operation of multiple loads. It also solves the problems of failing to reflect the time lag and cumulative characteristics of energy meter metering errors, easily resulting in compensation failure or overcompensation during sudden environmental changes, and difficulty in achieving long-term stable and high-precision metering under extreme and complex operating conditions.

[0006] The present invention provides a method for compensating metering errors in multi-loop energy meters applicable to extreme environments, comprising the following steps:

[0007] S1. Multi-channel synchronous sampling is performed through a multi-circuit energy meter to obtain sampling data, and dimensionless normalization is performed; based on the dimensionless normalized sampling data, the comprehensive disturbance intensity is calculated; based on the comprehensive disturbance intensity, the dynamic residual state is constructed through recursion with self-feedback, mutual feedback and rate of change suppression.

[0008] S2. Map the comprehensive disturbance intensity and dynamic residual state to generate the predicted power error; use the predicted power error as the error correction amount, and combine it with the historical compensation cumulative power value to perform compensation calculation to obtain the compensation cumulative power value.

[0009] Preferably, S1 specifically includes:

[0010] In the process of calculating the overall disturbance intensity, the local disturbances directly related to the current circuit are calculated based on the rated operating conditions of the circuit; then, the indirect disturbances caused by thermal coupling and electromagnetic coupling to the current circuit are calculated by using the load intensity of circuits other than the current circuit.

[0011] Preferably, S1 specifically includes:

[0012] The total input disturbance intensity is obtained based on local and indirect disturbances; the total input disturbance intensity is then normalized by environmental constraints to obtain the comprehensive disturbance intensity.

[0013] Preferably, S1 specifically includes:

[0014] Based on the historical dynamic residual state, a self-feedback term is constructed to represent the effect of the historical dynamic residual state on the current dynamic residual state; based on the historical dynamic residual states of loops other than the current loop, a mutual feedback term is constructed to represent the coupling effect of the historical dynamic residual states of loops other than the current loop on the current loop.

[0015] Preferably, S1 specifically includes:

[0016] Based on load intensity and comprehensive disturbance intensity, a rate of change suppression term is constructed to constrain the dynamic residual state update amplitude.

[0017] Preferably, S2 specifically includes:

[0018] In the calculation of predicted power error, based on the dynamic residual state, the comprehensive disturbance intensity is introduced to modulate the error amplitude, and a normalization constraint is introduced. Combined with the original power increment, the predicted power error is calculated.

[0019] Preferably, S2 specifically includes:

[0020] The historical cumulative energy value, the original energy increment, the predicted energy error, and the measured historical reference cumulative energy value are incorporated into the same energy domain, and a reference deviation term is introduced. By adjusting the correction strength of the reference deviation term to the compensation result, the compensated cumulative energy value is generated.

[0021] Preferably, S2 specifically includes:

[0022] The difference between the historical cumulative energy value and the historical reference cumulative energy value is used as the reference deviation item.

[0023] The beneficial effects of the technical solution of the present invention are:

[0024] 1. This invention addresses the thermal and electromagnetic coupling issues caused by the simultaneous operation of multiple metering circuits in multi-circuit energy meters. It incorporates the impact of the operating status of other circuits on the metering results of the current circuit into the compensation process, thus more realistically reflecting the error formation mechanism under complex power consumption scenarios. It is applicable to complex power consumption scenarios involving residential, heating, commercial, and three-phase power connections.

[0025] 2. By constructing a comprehensive disturbance intensity, further calculating the dynamic residual state, and finally generating the predicted power error, the entire process conforms to the actual formation law of measurement error from disturbance accumulation, state evolution to result offset. It can more accurately describe the gradual influence of factors such as low temperature hysteresis, high humidity slow change, reference source drift and clock offset on measurement error.

[0026] 3. The predicted power error maintains a strict energy domain correlation with the original power increment of the current sampling window, so that the error correction amount will not deviate from the current actual measurement scale. This avoids distortion caused by excessive compensation amount and insufficient correction caused by insufficient compensation amount, thereby improving the physical rationality and controllability of the compensation result.

[0027] 4. During the process of updating the cumulative energy compensation, the historical cumulative energy compensation value, the current original energy increment, the predicted energy error, and the reference cumulative energy are all incorporated into the same energy domain for processing. This ensures that the compensation results can not only respond quickly to changes in the environment and operating conditions, but also converge stably to the reference value during long-term operation, reducing cumulative error drift and improving long-term metering stability. Attached Figure Description

[0028] Figure 1 This is a flowchart of a multi-loop energy meter metering error compensation method suitable for extreme environments, as described in this invention. Detailed Implementation

[0029] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Unless otherwise defined, 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 invention pertains.

[0031] The following description, in conjunction with the accompanying drawings, details a specific scheme for a multi-loop energy meter metering error compensation method suitable for extreme environments provided by the present invention.

[0032] See attached document Figure 1 The diagram illustrates a flowchart of a multi-loop energy meter metering error compensation method suitable for extreme environments, provided by an embodiment of the present invention. The method includes the following steps:

[0033] S1. Multi-channel synchronous sampling is performed through a multi-circuit energy meter to obtain sampling data, and dimensionless normalization is performed. Based on the dimensionless normalized sampling data, the comprehensive disturbance intensity is calculated. Based on the comprehensive disturbance intensity, the dynamic residual state is constructed through recursion with self-feedback, mutual feedback and rate of change suppression.

[0034] Multi-loop energy meters trigger multi-channel synchronous sampling through a unified hardware timer. The voltage and current signals of each loop are synchronously converted from analog to digital by the front-end metering chip, ensuring strict time consistency in the instantaneous power calculation of each loop. The sampling period is set to milliseconds based on metering accuracy requirements and is determined by the processing power of the main control chip. Simultaneously, the outputs of the temperature, humidity, and barometric pressure sensors are read, and the reference voltage offset and metering clock frequency offset information are obtained from the internal registers of the metering chip. All sampled data, including voltage signals, current signals, reference voltage offset, metering clock frequency offset information, temperature, humidity, and barometric pressure, are encapsulated into structured data frames according to the unified timestamp of the energy meter's internal time base system and stored in chronological order in the local buffer to avoid data misalignment due to communication or processing delays.

[0035] Because different sampled data have different dimensions, they cannot be directly mixed and calculated without constraints in linear or nonlinear expressions. Therefore, before performing measurement error compensation, dimensionless normalization is first performed using the minimum-maximum normalization method. Then, the comprehensive disturbance intensity is calculated, condensing the dispersed disturbances caused by extreme environments on the measurement link into a transitive, accumulative, and interpretable intermediate state. This allows subsequent models to no longer deal with independent temperature, humidity, reference voltage offset, and harmonic variables, but rather with a unified disturbance input that has been reorganized by measurement physical constraints. Harmonic variables refer to total harmonic distortion (THD) and harmonic components. The measurement chip performs frequency domain analysis on current and voltage signals using Fourier transform to obtain the THD. The measurement chip itself has harmonic or power quality calculation functions. After completing analog-to-digital conversion, it obtains the fundamental and harmonic components internally through digital filtering and frequency domain decomposition, and provides the THD as readable register data. First, using the rated operating conditions of the circuit provided by the energy meter's design parameters as a reference, construct local disturbance terms directly related to the current circuit. Then, calculate the indirect disturbances caused by thermal coupling and electromagnetic coupling to the current circuit using the load strength of other circuits. Multiply the dimensionless normalized voltage by the current and then by the power factor to obtain the instantaneous active power of the circuit. Divide the instantaneous active power by the rated power of the circuit provided by the energy meter's design parameters to obtain the dimensionless load strength. The power factor is the ratio of instantaneous active power to apparent power (the apparent power of the product of voltage and current).

[0036] The combined disturbance intensity is obtained by weighting and aggregating the local and indirect disturbances. Its calculation formula is defined as follows:

[0037]

[0038] in, Let be the combined disturbance intensity of the i-th loop at the k-th sampling time; and The weighting coefficients, obtained through training with calibration samples, were acquired via a combined standard electrical energy calibration test in an extreme environment chamber. For example, the specific method is as follows: under conditions of fixed humidity, fixed load, and stable reference voltage, only the temperature variable is scanned, and the actual measurement error changes at different temperature points are recorded. Then, the temperature term is inversely calculated using least squares fitting, weighted least squares fitting, or nonlinear regression with physical constraints. Contribution to the intensity of the disturbance , The value range is [-5, 5]. As a specific example, Take 1.8, Take 1.2, , , The value range is [-10, 10]. As a specific example, Take 3.5, Take 2.7, Take 4.2, , The value range is [0,5]. As a specific embodiment, Take 1.6, Take 2.3; This represents the coupling strength of the j-th loop to the i-th loop, with a value range of [-5, 5]. As a specific example, Specifically, under extreme environmental conditions and standard electrical energy conditions, the load and environmental variables of the i-th loop are fixed, and only the j-th loop is subjected to graded loading or dynamic changes. At the same time, the change in the comprehensive disturbance intensity of the i-th loop is recorded. Then, the mapping relationship between "load change of the j-th loop - error change of the i-th loop" is established by the least squares fitting method, so as to obtain the corresponding coupling influence intensity. This represents the normalized temperature deviation at the k-th sampling time. This represents the normalized humidity deviation at the k-th sampling time. This represents the reference voltage offset at the k-th sampling time. This represents the frequency offset information of the metering clock at the k-th sampling time; This represents the load intensity of the i-th loop at the k-th sampling time; This represents the load intensity of the j-th circuit at the k-th sampling time; This represents the total harmonic distortion rate of the i-th circuit at the k-th sampling time; This represents the normalized air pressure deviation at the k-th sampling time. The normalized temperature deviation, normalized humidity deviation, and normalized air pressure deviation are obtained as follows: using the calibration reference value as the denominator, the absolute value of the difference between the temperature, humidity, and air pressure in the sampled data and the corresponding calibration reference value is the numerator, which is then normalized. The calibration reference value is the data calibrated at the factory of the electricity meter. The first term in the numerator... This reflects the direct metering chain disturbance in the same loop under the combined effects of temperature, humidity, reference voltage offset, and clock frequency offset. The second term in the numerator... The third term in the numerator reflects the amplification effect of the total harmonic distortion rate on the sampling nonlinearity. This is the inter-loop coupling term, i.e., indirect disturbance, reflecting the thermal and electromagnetic coupling introduced when other loops are working simultaneously. This represents a local disturbance term. This represents the total input disturbance intensity obtained at the k-th sampling time after uniformly weighting and superimposing the local disturbance of the current loop with the coupling effects introduced by the operation of other loops. The denominator contains... and It serves as an environmental constraint normalization mechanism, used to limit the numerical divergence of the overall disturbance intensity under conditions of low pressure and high temperature and humidity.

[0039] In the actual operation of multi-loop energy meters, resistance drift caused by temperature changes, insulation conductance changes caused by humidity changes, and reference source and clock offsets all exhibit time delay and asymptotic characteristics. A disturbance input at a single moment cannot be directly equated to a metering error. Therefore, it is necessary to construct a dynamic residual state that can describe the error evolution process, to receive the comprehensive disturbance intensity and transform it into a time-continuous error precursor. Specifically, at each sampling moment, the residual state from the previous moment is read, then the current comprehensive disturbance intensity is called, and the residual state from the previous moment of the adjacent loop is introduced to express the hysteresis coupling effect of the multi-loop. Then, a recursive relationship with self-feedback, mutual feedback, and rate-of-change suppression terms is constructed. The specific calculation is as follows:

[0040]

[0041] in, Let be the dynamic residual state of the i-th loop at the k-th sampling time; This represents the historical residual inheritance coefficient of the current loop, used to measure the dynamic residual state of the i-th loop at the previous time step. The historical dynamic residual state, representing the degree to which the current dynamic residual state is preserved, is obtained through offline calibration and constrained optimization methods. Specifically, a large number of samples are collected during the experiment. The goal is to minimize the mean square loss between the dynamic residual state and the target residual state. A sequential quadratic programming method with boundary constraints of physical stability and numerical boundedness is used for identification. The value range is [0,1]. As a specific example, Take 0.74, where the sample is the synchronous measurement sample of each loop at each sampling time, including voltage, current, power factor, total harmonic distortion, temperature, humidity, air pressure, reference voltage offset, metering clock frequency offset information, raw energy increment and standard energy increment; the target residual state is the ratio of the difference between the standard energy increment measured by the standard energy meter and the raw energy increment of the multi-loop energy meter to the raw energy increment within the same sampling window, i.e. from the (k-1)th sampling time to the kth sampling time. This represents the current integrated perturbation injection coefficient, used to describe the direct driving force of the current integrated perturbation intensity on the residual state. The training method is similar to... The values ​​are the same, ranging from [0,2]. As a specific example, Take 0.83; It is the rate of change response coefficient of the disturbance, used to amplify or suppress the difference term. The impact on the current residual state is used to capture rapidly changing scenarios such as sudden temperature rises and falls, and sudden load changes. It is obtained using variance estimation based on differences, with values ​​ranging from [-0.5, 0.5]. As a specific example, Take 0.11; Let be the combined disturbance intensity of the i-th loop at the (k-1)-th sampling time; This represents the degree of influence of the dynamic residual state of the j-th loop at the previous moment on the current dynamic residual state of the i-th loop. It is obtained through multi-loop joint calibration and has a value range of [-0.3, 0.3]. As a specific embodiment, Take 0.06; Let J represent the dynamic residual state of the j-th loop at the (k-1)-th sampling time. This is the comprehensive perturbation suppression coefficient, used to suppress the state update amplitude when the current comprehensive perturbation is too large. It is obtained through offline training under discrete Lyapunov stability constraints, and its value ranges from [0,3]. As a specific example... Take 1.32; This is the multi-loop load mutation suppression coefficient, used to limit state bursts under strong disturbances and multi-loop synchronous transitions. It is obtained by combining the least squares method with stability verification, and its value ranges from [0,2]. As a specific embodiment... Take 0.79; This represents the load intensity of the j-th circuit at the (k-1)-th sampling time.

[0042] This is a self-feedback term, representing the continuation effect of the dynamic residual state at the previous sampling time on the current dynamic residual state. This indicates the direct driving effect of the current comprehensive disturbance intensity on the dynamic residual state, reflecting the immediate impact of environmental factors such as temperature and humidity on the metering link. It represents the moderating effect of the rate of change of disturbance on the dynamic residual state, and is used to characterize the dynamic response enhancement or suppression of error when the environment or load changes rapidly. This is a mutual feedback term, representing the coupling effect of the historical dynamic residual state of other loops on the current loop, reflecting the indirect effect of thermal and electromagnetic coupling generated when multiple loops are running simultaneously on error evolution. It represents the comprehensive driving force of the measurement error offset trend of the i-th loop at the current sampling time, that is, the error evolution source term determined by the historical dynamic residual state, the current environmental disturbance and the multi-loop coupling effect.

[0043] The constant term 1 in the denominator of the dynamic residual state calculation formula is used to provide a baseline scale, ensuring that the dynamic residual state update maintains a linear response under undisturbed or weakly disturbed conditions. This indicates the suppression effect of the current comprehensive disturbance intensity. When factors such as temperature and humidity combine to increase the disturbance, the nonlinear saturation characteristics of the metering link under strong disturbances are reflected by compressing the dynamic residual state update amplitude. This indicates the suppression effect of multi-loop load changes. When the load of other loops changes abruptly, it suppresses the transient coupling shock caused by the simultaneous change of multiple loops by limiting the dynamic residual state update amplitude. The rate of change suppression term represents the constraint and stabilization factor of the dynamic residual state update amplitude. It is used to normalize and suppress the error driving force generated by the molecule under environmental disturbances or multi-loop load abrupt changes.

[0044] S2. Map the comprehensive disturbance intensity and dynamic residual state to generate the predicted power error; use the predicted power error as the error correction amount, and combine it with the historical compensation cumulative power value to perform compensation calculation to obtain the compensation cumulative power value.

[0045] The dynamic residual state only reflects the degree and trend of deviation, and does not yet possess physical quantity attributes that can be directly used for power metering correction. Therefore, a mapping mechanism that strictly corresponds to the power metering process needs to be introduced to convert the dimensionless dynamic residual state into an error estimate with the dimension of electrical energy, i.e., to predict the power error, and ensure that the error magnitude does not deviate from the actual metered energy scale of the current sampling window. The sampling window is from the (k-1)th sampling time to the kth sampling time. In multi-loop power meters, the error essentially originates from the sampling deviation of instantaneous power and the cumulative deviation during the time integration process. Therefore, the power error at any time should maintain a dimensional correlation with the original power increment corresponding to the current time, and cannot be directly generated by environmental quantities such as temperature or humidity. Secondly, the dynamic residual state is used as the dominant input, and the current comprehensive disturbance intensity is introduced to modulate the error amplitude to reflect the error amplification or suppression effect under extreme environmental conditions. A normalization constraint term is introduced into the mapping relationship to ensure that the error estimate remains bounded under any operating condition. The specific calculation is as follows:

[0046]

[0047] In the formula, Let be the predicted power error of the i-th circuit at the k-th sampling time, in kilowatt-hours (kWh). The original energy increment of the i-th circuit at the k-th sampling time is expressed in kilowatt-hours (kWh). It is formed by integrating the voltage, current, and power factor obtained from synchronous sampling. Specifically, it is obtained within the sampling window using the rectangular method or the equivalent constant power approximation. This represents the primary influence weight of the dynamic residual state, which adjusts the contribution intensity of the dynamic residual state to the predicted power error, and its value range is [0,2]. This represents the weighting coefficient of the second-order residual, which reflects the nonlinear effect of the error being amplified with the residual state under extreme conditions, and its value ranges from [0, 1.5]. This represents the residual-perturbation coupling weighting coefficient, reflecting the modulation effect of the current comprehensive perturbation intensity on the existing residual state, with a value range of [-1, 1]. The denominator contains... and Used for normalization restrictions, express The normalized suppression coefficient has a value range of [0,3]. express The normalized suppression coefficient has a value range of [0,3]. The method of obtaining the error is as follows: the squared difference between the predicted energy error and the actual error is used as the objective function, and the optimal solution is obtained by batch gradient descent. The actual error is the difference between the energy value of the standard energy meter and the original energy increment. This is a normalization constraint term.

[0048] During the compensation calculation, the cumulative compensation energy value at the previous sampling time, the original energy increment of the current sampling window, the predicted energy error driven by the dynamic residual state and the comprehensive disturbance intensity, and the reference cumulative energy value at the previous sampling time are all incorporated into the same energy domain for calculation.

[0049] The compensated cumulative energy value represents the cumulative metering result after error correction across all sampling windows, serving as the baseline state for compensation calculation. The original energy increment reflects the uncompensated energy contribution within the current time interval. The predicted energy error is introduced as the error correction amount for the current sampling window, forming the update expression for the compensated cumulative energy:

[0050]

[0051] in, The cumulative electrical energy value of the i-th circuit after compensation at the k-th sampling time is expressed in kilowatt-hours (kWh). This represents the cumulative compensated energy value output at the previous sampling time, i.e., the historical cumulative compensated energy value, in kilowatt-hours (kWh). This is the reference cumulative energy value at the previous sampling time, i.e., the historical reference cumulative energy value, in kilowatt-hours (kWh), measured by a standard energy meter; The closed-loop correction gain coefficient of the i-th loop is used to adjust the reference deviation term. The correction strength of the current compensation result was collected under different temperature and humidity, load combinations, and loop coupling conditions. The optimal historical sequence was obtained by optimizing three objectives: cumulative error after compensation, short-term fluctuations, and convergence speed. The value range is [0,1]; the cumulative error after compensation is the cumulative difference between the compensated cumulative energy value and the standard energy value obtained through the standard energy meter; short-time fluctuation is the change amplitude of the compensation result within adjacent sampling windows; convergence speed refers to the reference deviation term. The speed at which the error decreases to a stable range over time. The optimization method involves constructing a multi-objective function that weights the cumulative error after compensation, short-term fluctuations, and convergence speed, with each factor having an equal weight. Parameter optimization is performed using offline data to minimize the cumulative difference, the fluctuation, and the convergence time. It is the suppression weight coefficient of the i-th loop for the dynamic residual state, which is obtained by grid search based on historical samples and has a value range of [0,5]. The historical samples are multi-loop operation data sequences, including the original power increment, compensated cumulative power, reference cumulative power, temperature, humidity, air pressure and load intensity, etc., which are formed by continuous sampling under different temperatures, humidity, load types and power to form time series data. It is the suppression weight coefficient of the i-th loop for the overall disturbance intensity, which is obtained based on the grid search of historical samples and has a value range of [0,6].

[0052] Regarding the acquisition of algorithm parameters, in actual implementation, they are not set manually based on experience. Series parameters Series parameters Series parameters , , Instead of using parameters like these, calibration is performed through a combination of extreme environment chambers and standard electrical energy sources. The specific process involves applying different temperature ranges, humidity ranges, and load combinations to typical multi-loop scenarios such as residential + single-phase electric heating, residential + three-phase electric heating, residential + electric heating + commercial, residential + electric heating + three-phase power, and electric heating + industrial / commercial + residential shared power. For each operating condition, the original sampling sequence, standard source electrical energy sequence, and reference source and clock offset sequence are recorded. Then, the true error is calculated, and the above parameters are iterated with the objective of minimizing the squared difference between the predicted electrical energy error and the true error.

[0053] In summary, a method for compensating metering errors in multi-loop energy meters suitable for extreme environments has been developed.

[0054] The order of the embodiments is for illustrative purposes only and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0055] The various embodiments in this specification 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.

[0056] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for compensating metering errors in multi-loop energy meters suitable for extreme environments, characterized in that, Includes the following steps: S1. Multi-channel synchronous sampling is performed through a multi-circuit energy meter to obtain the sampling data, and then dimensionless normalization processing is performed. The overall disturbance intensity is calculated based on the dimensionless normalized sampled data. Based on the comprehensive disturbance intensity, a dynamic residual state is constructed through recursion with self-feedback, mutual feedback and rate of change suppression; S2. Map the comprehensive disturbance intensity and dynamic residual state to generate the predicted power error; use the predicted power error as the error correction amount, and combine it with the historical compensation cumulative power value to perform compensation calculation to obtain the compensation cumulative power value.

2. The method for compensating metering errors of multi-loop energy meters applicable to extreme environments according to claim 1, characterized in that, S1 specifically includes: In the process of calculating the overall disturbance intensity, the local disturbances directly related to the current circuit are calculated based on the rated operating conditions of the circuit; then, the indirect disturbances caused by thermal coupling and electromagnetic coupling to the current circuit are calculated by using the load intensity of circuits other than the current circuit.

3. The method for compensating metering errors of multi-loop energy meters suitable for extreme environments according to claim 2, characterized in that, S1 specifically includes: The total input disturbance intensity is obtained based on local and indirect disturbances; the total input disturbance intensity is then normalized by environmental constraints to obtain the comprehensive disturbance intensity.

4. The method for compensating metering errors of multi-loop energy meters suitable for extreme environments according to claim 1, characterized in that, S1 specifically includes: Based on the historical dynamic residual state, a self-feedback term is constructed to represent the effect of the historical dynamic residual state on the current dynamic residual state; based on the historical dynamic residual states of loops other than the current loop, a mutual feedback term is constructed to represent the coupling effect of the historical dynamic residual states of loops other than the current loop on the current loop.

5. The method for compensating metering errors of multi-loop energy meters suitable for extreme environments according to claim 1, characterized in that, S1 specifically includes: Based on load intensity and comprehensive disturbance intensity, a rate of change suppression term is constructed to constrain the dynamic residual state update amplitude.

6. The method for compensating metering errors of multi-loop energy meters applicable to extreme environments according to claim 1, characterized in that, S2 specifically includes: In the calculation of predicted power error, based on the dynamic residual state, the comprehensive disturbance intensity is introduced to modulate the error amplitude, and a normalization constraint is introduced. Combined with the original power increment, the predicted power error is calculated.

7. The method for compensating metering errors of multi-loop energy meters suitable for extreme environments according to claim 6, characterized in that, S2 specifically includes: The historical cumulative energy value, the original energy increment, the predicted energy error, and the measured historical reference cumulative energy value are incorporated into the same energy domain, and a reference deviation term is introduced. By adjusting the correction strength of the reference deviation term to the compensation result, the compensated cumulative energy value is generated.

8. The method for compensating metering errors of multi-loop energy meters suitable for extreme environments according to claim 7, characterized in that, S2 specifically includes: The difference between the historical cumulative energy value and the historical reference cumulative energy value is used as the reference deviation item.

Images