Modeling analysis method for influence of random electromagnetic signals on metering error of intelligent electric energy meter

Abstract

The invention discloses a modeling analysis method for the influence of random electromagnetic signals on the metering error of an intelligent electric energy meter, and the method comprises the steps: modulating a sine steady-state current signal i0 (t) through an m-sequence binary waveform, and building a parameter model # imgabs0 # of a current signal i1 (t) with a random dynamic characteristic. Secondly, according to the m-sequence dynamic test signal model, establishing a mathematical model of the influence of the electromagnetic signal on the metering error of the intelligent electric energy meter, and further simulating the influence of the electromagnetic signal on the random interference of the electric energy meter; and finally, obtaining the metering error of the electric energy meter through simulation and experimental test, and further analyzing the influence of the electromagnetic signal on the metering error of the intelligent electric energy meter.

Description

Technical Field

[0001] The present invention relates to a modeling and analysis method for the influence of random electromagnetic signals on the measurement error of smart energy meters. Background Art

[0002] With the rapid development of smart grids and user information acquisition systems, a modern power grid integrating advanced sensor measurement technology, communication technology, and control technology has taken initial shape. However, the sharp increase in the number of smart energy meters has also brought challenges to the reliability assessment of energy meters. Currently, many research results have been achieved in the study of the influence of environmental factors such as temperature and humidity on the measurement of energy meters. However, since smart energy meters all belong to electronic energy meters, a large number of electronic components contained therein are extremely vulnerable to the interference of the surrounding complex electromagnetic fields.

[0003] Currently, domestic and foreign scholars have conducted a large number of studies on the influence of smart energy meters in an electromagnetic environment. Some literature has proposed an automatic verification method for the measurement error of three-phase energy meters based on clustering optimization, thereby improving the accuracy of verification; some literature has combined cases where energy meters frequently restart automatically in a radio frequency electromagnetic field, and by connecting a bypass capacitor between the reference pin of the voltage reference source chip and the ground, effectively filtered the interference voltage introduced into the loop by the spatial radio frequency electromagnetic field, and solved the problem of the energy meter automatically resetting and restarting; some literature has mainly studied the design and improvement of power frequency electromagnetic field immunity tests; some literature has used a combination of multiple algorithms to establish a prediction model for the line loss rate of the distribution transformer area and estimate the error of the energy meter (the literature is: Wang Hao, Yang Peng, Li Chong, etc. Research on the energy meter error evaluation model combining improved SVR and fading memory recursive least squares algorithm [J]. Journal of Electric Power Science and Technology, 2023, 38(5): 206-215.); some literature has designed a wide-frequency voltage decoupling network, significantly improving the EMS anti-interference ability of the energy meter verification device; some literature has established an on-off keying (OOK) test dynamic current model and an OOK test dynamic load energy model, improving the problem of the dynamic accuracy of signal testing. However, the above research results have not been analyzed from the perspective of the randomness of electromagnetic signals.

[0004] In summary, in the study of the influence of electromagnetic signals on the measurement error of energy meters, the problem of random interference of electromagnetic signals remains a challenging problem. Therefore, it is necessary to design a new modeling and analysis method for the influence of random electromagnetic signals on the measurement error of smart energy meters. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a modeling and analysis method for the influence of random electromagnetic signals on the measurement error of smart energy meters. This modeling and analysis method for the influence of random electromagnetic signals on the measurement error of smart energy meters can well simulate the interference of random signals on the energy meter, so as to accurately reflect the error caused by the interference to the energy meter.

[0006] The technical solution of the invention is as follows:

[0007] A modeling and analysis method for the influence of random electromagnetic signals on the metering error of intelligent electricity meters

[0008] Using the binary waveform of the m-sequence to modulate the sinusoidal steady-state current signal i 0 (t) to establish a parameter model of the current signal i 1 (t) with random dynamic characteristics:

[0009]

[0010] Mod2 refers to addition modulo 2; the rules are 0 + 0 = 0, 0 + 1 = 1, 1 + 0 = 1, 1 + 1 = 0;

[0011] Mod2 is used for addition modulo 2 when summing up the whole of C i m(k - i). For a 4th-order sequence: (x 15 + 1), there are only two primitive polynomials (a primitive polynomial is a polynomial over a unique factorization domain whose greatest common divisor of all coefficients is 1. This means that if all coefficients of a polynomial are relatively prime, then this polynomial is a primitive polynomial.), which are f(x) = x 4 + x + 1 and f(x) = x 4 + x 3 + 1. Select one of them, f(x) = x 4 + x + 1. At this time, c 0 = c 1 = c 4 = 1, c 2 = c 3 = 0. Assume the initial m -4 = m -3 = m -2 = m -1 = 1, m 0 = c 1 m -1 modulo 2 plus c 4 m -4 = 0, and so on.

[0012] In the finite field GF(p), p is a prime number. Specifically, GF(p) is just mod p because after a number is modulo p, the result is between [0, p - 1]. For elements a and b, then (a + b) mod p and (a * b) mod p, and their results are both elements in the field. The addition and multiplication in GF(p) are similar to general addition and multiplication, except that the result needs to be mod p to ensure that the result is an element in the field.

[0013] In the present invention, the mod2 operation is adopted, so it is GF(2).

[0014] Factorization is carried out within the finite field (not ordinary factorization), and there is:

[0015] x 15 +1 = (x + 1)(x 2 + x + 1)(x 4 + x 3 + x 2 + x + 1)(x 4 + x 3 + 1)(x 4 + x + 1);

[0016] First of all, the prerequisite for f(x) that constitutes the m-sequence must be a primitive polynomial. Therefore, for a 4th-order feedback shift register, n = 4, so the length of the m-sequence generated by it is 15. Since its characteristic polynomial f(x) should be divisible by (x m + 1) = x 15 + 1, so factorize it and find f(x) from it. x 15 + 1 = (x + 1)(x 2 + x + 1)(x 4 + x 3 + x 2 + x + 1)(x 4 + x 3 + 1)(x 4 + x + 1);

[0017] Among them, f(x) = x 4 + x + 1 and f(x) = x 4 + x 3 + 1 meet the conditions of the primitive polynomial, and any one of them can generate the m-sequence. For f(x) = x 4 + x + 1 = c 4 x 4 + c 3 x 3 + c 2 x 2 + c 1 x 1 + c 0 , the corresponding linear homogeneous difference equation c 4 x 4 + c 3 x 3 + c 2 x 2 + c 1 x 1 + c 0 = c 4 m i-4 + c3 m i-3 +c 2 m i-2 +c 1 m i-1 +c 0 m i ,After rearrangement, it can be obtained that For the binary case, the recurrence equation can be simplified. Therefore, c 0 = c 1 = c 4 = 1, c 2 = c 3 = 0. Assume the initial values are given (m 3 , m 2 , m 1 , m 0 ) = (1, 0, 0, 0). And so on. Finally, we get Figure 4 a sequence of m 15 ... m 3 m 2 m 1 m 0 , which is the binary m-sequence. m k also corresponds to m(k).

[0018] GF(P) represents the finite field. In the present invention, the mod2 operation is adopted, so p = 2

[0019] For polynomial operations:

[0020] There are some differences between the polynomials here and the usual polynomial operations. Although their representation forms are the same: f(x) = x^6 + x^4 + x^2 + x + 1. The following are some of its characteristics:

[0021] (1). The coefficients of the polynomial can only be 0 or 1. Of course, for GF(p^n), if p equals 2, then the coefficients can take: 0, 1;

[0022] (2). When combining like terms, the coefficients perform an exclusive OR operation, not the usual addition operation. For example, x^4 + x^4 equals 0 * x^4. Because both coefficients are 1, after performing the exclusive OR, it equals 0

[0023] (3). Since it is an addition modulo 2, there is no subtraction (subtraction is equal to addition), or negative coefficients. Therefore, x^4 – x^4 is equal to x^4 + x^4. -x^3 is x^3.

[0024] Principle for solving primitive polynomials:

[0025] (1) Factorize until it cannot be further factorized;

[0026] (2) Exclude all factors with a degree less than n from the set of obtained factors;

[0027] If the remaining factors cannot divide any x q - 1, q < m, then this factor is a primitive polynomial;

[0028] l is the order of the m-sequence, N is its corresponding maximum period length, N = 2 l - 1; The value of l mainly depends on the requirements of model calculation. In the present invention, the value of the order ranges from 10 to 20. If the value of l is too large, data overflow will occur.

[0029] U and I are the amplitudes of voltage and current respectively, are the phases of voltage and current respectively, and ω is the angular frequency.

[0030] C i is the feedback coefficient. If it participates in feedback, it is 1; if it does not participate in feedback, it is 0;

[0031] In, if a i exists, then c i participates in feedback, c i = 1, otherwise, c i is 0. Therefore, the value of c i determines the connection of feedback. The value of c i is mainly given by the initial state. However, there is only a fixed shift between the m-sequences generated by different initial states. As long as the part with a length of 2 l - 1 is intercepted, it is a period of the sequence.

[0032] Where:

[0033] {Here, N is the period length, and m(k) is equivalent to a value, which is more Figure 4 In, the initial a 0 , a 1 , a 2 , a 3 , is (0, 0, 0, 1), which is equivalent to m(0) = 0, m(1) = 0, m(2) = 0, m(3) = 1; After shifting, a complete periodic sequence a 0 a 1 a 2 a 3 ... a 14 = 000111101011001, and m(k) also corresponds one by one. After multiplying by the window function, m(t) can be obtained. Note: m(0) corresponds to t ∈ [0, T], and m(1) corresponds to t ∈ [T, 2T]}

[0034] 4. The modeling and analysis method for the influence of random electromagnetic signals on the measurement error of an intelligent electric energy meter according to claim 1, characterized in that:

[0035] The measurement error containing radio frequency electromagnetic signals is:

[0036] Where E is the electric energy containing radio frequency electromagnetic signals, and E 0 is the electric energy of the input signal;

[0037] There is:

[0038]

[0039] p 0 (k) is the active power of the input signal, p(k) is the active power of the signal containing radio frequency electromagnetic interference. The voltage and current need to be sampled and then discretized. n is equivalent to the number of sampling points.

[0040] Beneficial effects:

[0041] For the modeling and analysis method of the influence of random electromagnetic signals on the measurement error of an intelligent electric energy meter in the present invention, there are various types of electromagnetic interference signals and different modulation methods in a complex electromagnetic environment, resulting in a decline in the performance of surrounding electronic device systems. With the wide application of intelligent electric energy meters in the field of electric energy measurement, the influence of complex electromagnetic signals on the measurement error of electric energy meters has gradually attracted attention. Aiming at the characteristics of randomness and dynamics of complex electromagnetic signals, the present invention first establishes a parameter model of an m-sequence dynamic test signal; secondly, based on the m-sequence dynamic test signal model, a mathematical model of the influence of electromagnetic signals on the measurement error of an intelligent electric energy meter is established, and then the influence of electromagnetic signals on the random interference of the electric energy meter is simulated; finally, the measurement error of the electric energy meter is obtained through simulation and experimental tests, and then the influence of electromagnetic signals on the measurement error of the intelligent electric energy meter is analyzed.

[0042] By analyzing the coupling mechanism of electromagnetic interference and the influence mechanism of electromagnetic signals on the electric energy meter, starting from the characteristic of uncertainty in the propagation process of electromagnetic signals, the present invention establishes a dynamic test signal model based on m-sequence modulation and a mathematical model of the influence of electromagnetic signals on the measurement of the electric energy meter, and conducts simulation and experimental analysis to obtain the influence of complex electromagnetic signals on the measurement error of the electric energy meter. Description of the drawings

[0043] Figure 1 is a schematic diagram of the interference path of radio frequency signals on the measurement of the electric energy meter;

[0044] Figure 2 is a general structure diagram of an m-sequence;

[0045] Figure 3 is a schematic diagram of the principle of a 4-stage feedback shift register;

[0046] Figure 4 It is a schematic diagram of the generation of a 4-level m-sequence;

[0047] Figure 5 It is a schematic diagram of the test current signal model of the m-sequence;

[0048] Figure 6 It is the influence curve of m-sequence modulation on the metering error of the electricity meter;

[0049] Figure 7 It is a schematic diagram of the influence of the m-sequence order on the metering error of the electricity meter;

[0050] Figure 8 It is a schematic diagram of the influence of the input current amplitude on the metering error;

[0051] Figure 9 It is a schematic diagram of the division of the interference area at the experimental site;

[0052] Figure 10 It is a three-sigma criterion diagram;

[0053] Figure 11 It is a diagram of the data analysis results of the first electricity meter;

[0054] Figure 12 It is a diagram of the data analysis results of the second electricity meter;

[0055] Figure 13 It is a diagram of the data analysis results of the third electricity meter. Specific implementation manners

[0056] The following will further elaborate on the present invention in conjunction with the accompanying drawings and specific embodiments: Embodiment 1:

[0057] 1. Electromagnetic interference coupling mechanism and its influence mechanism on the electricity meter

[0058] The three elements of electromagnetic interference include the interference source (harassment source), the coupling path, and the sensitive body. To solve the electromagnetic interference problem, it is necessary to conduct research on the interference source and the propagation path to find corresponding solutions.

[0059] Electromagnetic interference can be divided into radiation mode and conduction mode according to the coupling method. In the near field, the radiation coupling may be mainly magnetic field coupling or mainly electric field coupling; in the far field, it is mainly coupled through the form of electromagnetic waves, and the ratio of the electric field to the magnetic field is fixed.

[0060] Electromagnetic interference can be classified according to the spectrum as follows: Audio noise (0 - 20 kHz) is mainly generated by capacitors and high-frequency transformers, which affects the normal metering of the electricity meter and the phase sequence judgment; Radio frequency interference (20 kHz - 50 MHz) affects the normal operation and accuracy of the electricity meter; and Radiation interference (> 50 MHz) affects the accuracy of the electricity meter. Most of the common smart electricity meters currently in use adopt ordinary filters, and the effective filtering range is often less than 20 kHz, and the influence of radio frequency signals and radiation interference cannot be effectively filtered out. Therefore, radio frequency signals and radiation interference will cause large errors in the metering of smart electricity meters.

[0061] The influence of radio frequency signals on smart electricity meters is mainly reflected in two aspects: On the one hand, radio frequency signals can penetrate the shell of the electricity meter and directly enter its internal structure; on the other hand, radio frequency signals can also enter its interior through each wiring port of the electricity meter, thereby generating interference effects. The interference path of radio frequency signals on the metering of electricity meters is as Figure 1 shown.

[0062] The electricity meter obtains voltage signals and current signals through voltage sampling and current sampling, and converts the collected analog signals into digital signals for processing. Since the external analog quantity interface of the high-precision metering chip inside the electricity meter is directly connected to the wiring of the external PCB circuit board, when these analog quantity signal lines and reference voltage and current signal lines are coupled with high-frequency interference signals, the low-pass filter in the subsequent digital circuit may not be able to effectively perform filtering processing. If the intensity of the interference signal exceeds the noise threshold of the internal circuit of the electricity meter, it may have a significant impact on the metering accuracy of the electricity meter.

[0063] Modulation and Modeling of 2m Sequence Dynamic Test Signals

[0064] Since in the actual working environment of the electricity meter, the regional electromagnetic signals will undergo abnormal changes in terms of time, frequency domain, energy, etc., and the surrounding electromagnetic signals have characteristics such as randomness and time-variation [16-17] , therefore, in order to more accurately characterize radio frequency electromagnetic signals, this section adopts the method of converting steady-state signals into dynamic signals to establish an m sequence dynamic test signal model.

[0065] 2.1 Generation Principle of m Sequence

[0066] The m sequence is also called a linear feedback shift register sequence, and its general structure diagram is as Figure 2 shown. It is the sequence with the longest period generated by a shift register with linear feedback.

[0067] Let the initial state of the m sequence be (a 0 , a 1 ,..., a n-2 , a n-1), After one - shift linear feedback, the input to the first stage of the shift register is:

[0068]

[0069] where c i is the feedback coefficient. After k shifts, the input to the first stage is:

[0070]

[0071] where, l=n + k - 1, (n, k = 1, 2, 3...). Thus, it can be seen that the input to the first stage of the shift register is determined by the feedback logic and the original state of the shift register. Taking a 4 - stage linear feedback shift register as an example, as Figure 3 shown, its period is p = 2 4 - 1 = 15, and its characteristic polynomial is a 4 - degree primitive polynomial that can divide (x 15 + 1). First, factorize (x 15 + 1) so that each factor is an irreducible polynomial (a polynomial that cannot be factored into the product of two non - constant polynomials within the domain), and then find f(x), as shown in Equation (3).

[0072]

[0073] Among them, there are 3 4 - degree irreducible polynomials, but only two can generate m - sequences. The m - sequence formed by f(x)=x 4 + x + 1 is as Figure 4 shown.

[0074] Its initial state is (a 3 , a 2 , a 1 , a 0 )=(1, 0, 0, 0). When shifted once, a new input is generated by adding a 3 and a 0 modulo 2 and placed in register a 3 . Register a 2 is updated to the original value of a 3 , a 1 is updated to the original value of a 2 , a 0 is updated to the original value of a 1 . Therefore, the state at this time becomes (1, 1, 0, 0). Shifting 15 times in this way returns to the initial state. From the above shifting method, it can be seen that if the initial state is (0, 0, 0, 0), it remains in the all - zero state after shifting. Therefore, this feedback shift register should avoid the all - zero state.

[0075] Based on the above analysis of the m-sequence principle, it is not difficult to see that within one period, it can effectively reflect the random changes of dynamic signals and has strong autocorrelation. However, after exceeding the period length, it will enter the next period cycle. Therefore, by utilizing the pseudo-randomness characteristics of the m-sequence, the radio frequency electromagnetic signal model under complex electromagnetic environments can be simulated.

[0076] 2.2 Dynamic Test Current Signal Model with m-Sequence Modulation

[0077] Based on the m-sequence modulation principle described in 2.1, the steady-state signal can be converted into a dynamic signal, and the m-sequence test current signal model is constructed as Figure 5 shown below.

[0078] This method uses the binary waveform of the m-sequence to modulate the sinusoidal steady-state current signal i 0 (t) to establish a current signal i 1 (t) with random dynamic characteristics. Among them, let the sinusoidal steady-state voltage signal and the sinusoidal steady-state current signal be respectively:

[0079]

[0080] where U and I are the amplitudes of the voltage and current respectively, are the phases of the voltage and current respectively, and ω is the angular frequency.

[0081] The binary waveform function m(t) is expressed in the form of the product of a matrix window function and a numerical m-sequence:

[0082]

[0083] where N is 2 l -1, l is the order of the m-sequence; the function g(t) is the window function, T is the period of the sinusoidal steady-state current, and k = 1, 2, 3..., N.

[0084] Multiplying the function m(t) with the sinusoidal steady-state current i 0 (t) for amplitude modulation, we can obtain:

[0085]

[0086] where I is the maximum amplitude of the modulated current i d (t), and |m(t)| ≤ 1.

[0087] According to the m-sequence generation principle in 2.1, the m(k) sequence can be obtained as:

[0088]

[0089] Substituting it into the above formula (7), we can obtain:

[0090]

[0091] In summary, the parameter model of the test current signal modulated by the m-sequence can be given by Equation (8), where the model parameters of i 1 (t) are determined by C i (i = 1, 2, 3..., n), I, U, ω. Among them, i 1 (t) is a current signal with random dynamic characteristics.

[0092] U and I are the amplitudes of voltage and current respectively, are the phases of voltage and current respectively, and ω is the angular frequency.

[0093] C i is the feedback coefficient, which is 1 if it participates in the feedback and 0 if it does not participate in the feedback;

[0094] 3 Modeling of the Influence of Radio Frequency Electromagnetic Signals on Smart Energy Meters

[0095] Since radio frequency interference is an approximately uniformly distributed noise, this noise is independent of the input current and voltage of the energy meter, and can be superimposed on the input current and voltage, thus affecting the metering accuracy of the energy meter. [18-19] . Based on this, this chapter will establish a mathematical model for the influence of radio frequency electromagnetic signals on the metering error of smart energy meters based on m-sequence modulation.

[0096] Establish the input sine signals u 0 (t) and i 0 (t) respectively as:

[0097]

[0098] In the formula, U 0 , I 0 respectively represent the voltage and current amplitudes of the input signal, are the phase differences of voltage and current respectively, and ω 0 is the angular frequency of the input signal.

[0099] Establish the mathematical models of the unmodulated radio frequency current signal and the radio frequency voltage signal as follows:

[0100] e i (t) = I e cos(ω 1 t + ψ i ) - I e sin(ω 1 t + ψ i ) + σ i (10)

[0101] eu i(t) = U e cos(ω 1 t + ψ u ) - U e sin(ω 1 t + ψ u ) + σ u (11)

[0102] Wherein, U e , I e are respectively the voltage amplitude and current amplitude of the RF signal. Among them, ω 1 = 2πf 1 is the angular frequency of the RF signal, f 1 = 1.84 GHz is the angular frequency of the RF signal, ψ i and ψ u are respectively the phases of the RF current signal and the RF voltage signal, σ i and σ u are respectively the current noise error and voltage noise error of the RF signal, ψ i , ψ u , σ i and σ u all follow a Gaussian distribution with a mean of 0 and a variance of 1.

[0103] In order to simulate the randomness characteristics of the RF signal in a complex electromagnetic environment, the RF current signal can be obtained as e id (t) after being modulated by the m-sequence:

[0104]

[0105] Wherein, N is the period length of the m-sequence, l is the number of stages of the m-sequence, N = 2 l - 1.

[0106] Adding the modulated input current to the RF current can obtain i e (t):

[0107] i e (t) = i 0 (t) + e id (t) (13)

[0108] Adding the input voltage to the RF voltage can obtain u e (t):

[0109] u e (t) = u 0 (t) + e u (t) (14)

[0110] Multiplying i e (t) by the gain coefficient ki , during the modulus conversion, discretization can obtain i e (n). Similarly, multiply u e (t) by the gain coefficient k u and then perform discretization to obtain u e (n):

[0111]

[0112]

[0113] In the formula, t n is the symbol width of the m-sequence, t s is the time interval of the sampling point, β is the zero-crossing radian of the first sampling point, l is the order of the m-sequence.

[0114] The discrete signals i e (n), u e (n) are multiplied to obtain the instantaneous power signal p i (n) containing radio frequency interference:

[0115] p i (n) = u e (n)i e (n) (17)

[0116] The instantaneous power signal can effectively filter out high-order harmonics after passing through a low-pass filter. Suppose the effective sampling response of the active power low-pass filter is {r(k) = 1 / L: 0 ≤ k ≤ L - 1}. The discretized instantaneous active power signal can be decomposed into the active power p 0 (k) of the input signal and the active power p(k) of the signal containing radio frequency electromagnetic interference:

[0117]

[0118]

[0119] By performing cumulative summation on the active power, the cumulative electric energy can be obtained. Therefore, the electric energy of the input signal is E 0 , and the electric energy of the signal containing radio frequency electromagnetic signals is E:

[0120]

[0121]

[0122] In the formula, n is the number of sampling points. According to formulas (20) and (21), the measurement error of the signal containing radio frequency electromagnetic signals is:

[0123]

[0124] Experimental and Simulation Analysis of 4 Errors

[0125] In order to better obtain the influence results of electromagnetic signals on the metering error of electric energy meters, first, based on the mathematical model of the influence of radio frequency electromagnetic signals on the electric energy metering error of intelligent electric energy meters established in the previous section, a program is written using MATLAB to analyze the error magnitude. Secondly, by building a dynamic error test system for electric energy meters, dynamic error experiments are carried out on three different brands of electric energy meters respectively, and the experimental data are analyzed to further determine the influence of metering error.

[0126] 4.1 Error Simulation Analysis

[0127] Based on the mathematical model established in Chapter 3, with the input voltage set at 220V and the input current at 5A, the influence of radio frequency electromagnetic signals on the metering of electric energy meters is studied from the following three aspects.

[0128] (1) When both phase noise and amplitude noise exist, compare the influence of the radio frequency signal obtained by m-sequence modulation on the metering error of the electric energy meter with that of the radio frequency signal without m-sequence modulation. The simulation results are as Figure 6 shown.

[0129] (2) When ensuring the simultaneous existence of phase noise and amplitude noise, set the order of the m-sequence to increase from 1 to 12, and set 20 simulation experiments for each order. Analyze and compare the scatter plots of the metering errors of the electric energy meters at different orders. The results are as Figure 7 shown.

[0130] (3) When the power factors are 0.5, 0.8, and 1.0 respectively, after m-sequence modulation, analyze and compare the influence of the input current amplitude on the metering error of the electric energy meter. The results are as Figure 8 shown.

[0131] As Figure 6 can be seen, when ensuring the existence of both phase noise and amplitude noise, the metering error of the mathematical model obtained after m-sequence modulation is about 50% of that in the unmodulated case; as Figure 7 can be seen, with the increase of the order of the m-sequence, the change range of the metering error is not obvious. Generally, the metering error is less than 0.3%, and in the worst case it is 0.79%, meeting the metering standard of the electric energy meter; as Figure 8 can be seen, with the increase of the input current amplitude, the metering error of the electric energy meter gradually decreases. When the input current is greater than 5A, the metering error is lower than 0.3%. When the input current is greater than 15A, the metering error is lower than 0.1%, and this rule exists under different power factors.

[0132] 4.2 Experimental Test Scheme

[0133] To further study the influence of radio frequency signals on the metering error of electric energy meters through experimental tests, the present invention constructs an error test platform, which consists of a single-phase electric energy meter, an electric energy meter error calibrator, a power frequency magnetic field generator, a high-frequency electric field generator, and a radio frequency electromagnetic field generator. The parameters are shown in Table 1.

[0134]

[0135]

[0136] Table 1 Experimental equipment and parameters

[0137] The smart electric energy meter is divided into 12 interference regions as shown in Figure 9 . Using a tuned dipole antenna at electromagnetic field frequencies from 400 MHz to 1 GHz (8 frequency points: 400 MHz, 450 MHz, 500 MHz, 600 MHz, 700 MHz, 800 MHz, 900 MHz, 1 GHz), and a horn antenna in the high-frequency band from 1.1 GHz to 2.5 GHz (4 frequency points: 1 GHz, 1.5 GHz, 2 GHz, 2.5 GHz). Approximately 20 minutes of testing are required for each frequency point, each position, and each polarization direction. Interference for 60 s, pause for 20 s, then continue interference. 100 tests are conducted for each frequency band.

[0138] 4.3 Analysis of experimental data errors

[0139] According to the above experimental design, dynamic error tests are respectively carried out on electric energy meters from three different manufacturers.

[0140] Due to the large amount of data, gross errors are inevitable during the testing process. To reduce the influence of this phenomenon on the experimental conclusion, the present invention uses the three-sigma criterion to clean the power values of each interference group to ensure the reliability and accuracy of subsequent data analysis. The calculation formulas for its mean and standard deviation are shown in Equation (23).

[0141]

[0142] In the formula, μ is the mean of the metering error of each interference group; σ is the standard deviation of the metering error of each interference group. Then, the outliers of the interference group errors are removed using the rules in Equation (24).

[0143]

[0144] After the above processing, it can be known from Figure 10 that the probability of the numerical distribution within 3σ is as high as 99.73%. Those outside this range can be regarded as outliers and removed, which can effectively reduce the influence of gross errors on data analysis.

[0145] According to the measured experimental data, the measurement errors of three different electric meters are analyzed respectively using MATLAB software. When the signal source frequencies are 400 MHz, 450 MHz, 500 MHz, 600 MHz, 700 MHz, 800 MHz, 900 MHz, 1000 MHz, 1500 MHz, 2000 MHz, 2500 MHz; and the transmitter powers are -30 dB, -12 dB, -10 dB, -8 dB, -5 dB, the relationship between the signal source power and the dynamic error is studied. The test change curves of electric energy meters A, B, and C are as Figures 11 - 13 shown.

[0146] It can be seen from Figures 11 - 13 that the data analysis results of the three different electric meters are similar. When the signal source frequency is constant, the electric energy measurement error increases with the increase of the signal source power. When it increases to a certain extent, the error change range decreases. Since the magnitude of the signal source power mainly depends on the amplitude of the radio frequency signal, it can be considered that when the signal source frequency is constant, the electric energy measurement error increases with the increase of the signal source amplitude.

[0147] When the signal source power is constant, the electric energy measurement error will increase slowly with the increase of the signal source frequency. When it reaches a certain extent, it shows a decreasing trend, and the turning frequencies are similar under different transmitter powers.

[0148] 5 Conclusion

[0149] First, based on the characteristic that electromagnetic interference signals have random propagation in a complex electromagnetic environment, a parameter model of the m-sequence current test signal is established in the present invention. Secondly, based on the characteristics of the parameter model of the m-sequence current test signal, a mathematical model of the radio frequency electromagnetic signal on the measurement error of the electric energy meter is established, and a method for calculating the measurement error of the electric energy meter is obtained. Finally, through simulation analysis and experimental tests, the following conclusions are obtained:

[0150] (1) When the signal source frequency remains unchanged, the measurement error increases with the increase of the signal source amplitude; when the signal source power remains unchanged, with the increase of the signal source frequency, the measurement error of the electric energy meter shows a trend of increasing first and then decreasing;

[0151] (2) When the power factor remains unchanged, the electric energy measurement error caused by the radio frequency electromagnetic field decreases with the increase of the measured current amplitude; the measurement accuracy of the mathematical model established by using m-sequence modulation is higher; the change of the order of the m-sequence has no obvious influence on the measurement error of the electric energy meter.

[0152] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A modeling and analysis method for the influence of random electromagnetic signals on the measurement error of smart electric energy meters, characterized in that: The sinusoidal steady-state current signal i0(t) is modulated by the m-sequence binary waveform, and the parameter model of the current signal i1(t) with random dynamic characteristics is established: l is the series number of m sequence, N is the corresponding maximum period length, N = 2 l -1; U and I are the amplitudes of voltage and current respectively. are the phases of voltage and current respectively, and ω is the angular frequency. C i is the feedback coefficient, which is 1 if the user participates in the feedback and 0 if the user does not participate in the feedback; in:

2. The modeling and analysis method for the effect of random electromagnetic signals on the measurement error of smart electric energy meters according to claim 1 is characterized by: The value of l is between 10 and 20.

3. The modeling and analysis method for the influence of random electromagnetic signals on the measurement error of smart electric energy meters according to claim 1 or 2, characterized in that: The measurement error involving radio frequency electromagnetic signals is: Wherein, E is the electrical energy of the radio frequency electromagnetic signal, and E0 is the electrical energy of the input signal; have: Among them, p0(k) is the active power of the input signal, p(k) is the active power of the signal containing radio frequency electromagnetic interference, and n is equivalent to the number of sampling points.

Images