A method for monitoring amplitude and phase difference of three-phase circuit based on variable step size LMS

Abstract

The present invention discloses a method for monitoring amplitude and phase differences in a three-phase circuit based on a variable-step-size LMS. The method comprises the following steps: converting three groups of large voltage signals in the three-phase circuit into three groups of small voltage signals based on a preset proportional coefficient; sampling the three groups of small voltage signals based on a preset sampling frequency to obtain three groups of digital voltage signals; performing a Hilbert transform on the three groups of digital voltage signals to obtain three groups of complex voltage signals; calculating the phase difference and amplitude ratio between the input complex voltage signal and the expected complex voltage signal based on the expected complex voltage signal, the input complex voltage signal, and the variable-step-size LMS; and determining whether the voltage signals in the three-phase circuit have achieved three-phase balance and whether the phase sequence is correct based on the phase difference and amplitude ratio. The purpose of the present invention is to provide a method for monitoring amplitude and phase differences in a three-phase circuit based on a variable-step-size LMS, thereby resolving the problem that existing monitoring methods cannot quickly and accurately monitor the amplitude and phase differences of three-phase voltage / current signals in a distribution network.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of industrial power system detection, and specifically to a three-phase circuit amplitude and phase difference monitoring method based on variable step size LMS. Background Art

[0002] Currently, most power generation, transmission, and distribution, as well as high-power electrical equipment, utilize a three-phase system. Asymmetric three-phase components and / or line parameters can lead to three-phase imbalance or a phase sequence error. This imbalance can reduce motor efficiency, increase transformer losses or even damage, shorten the life of electrical equipment, reduce distribution transformer output, and cause frequent overloads or short circuits. When the three phases are balanced but the phase sequence is incorrect, it can damage equipment that requires a phase sequence and make line fault repair difficult. Therefore, to ensure the normal operation of the distribution system, it is necessary to monitor the amplitude and phase differences (phase difference and amplitude ratio) of the three-phase voltage and current signals in the distribution network.

[0003] Currently, the commonly used measurement method is the discrete Fourier transform method. Since the discrete Fourier transform method has high requirements for the sampling frequency, it is easy to cause the system to overload. At the same time, it has high requirements for the number of sampling points, resulting in low real-time monitoring. Based on the above defects, the monitoring method in the existing technology cannot quickly and accurately monitor the amplitude and phase difference of the three-phase voltage / current signals in the distribution network. Summary of the Invention

[0004] The embodiment of the present application provides a three-phase circuit amplitude and phase difference monitoring method based on variable step size LMS, which aims to solve the problem that the existing monitoring method cannot quickly and accurately monitor the amplitude and phase difference of the three-phase voltage / current signals in the distribution network due to defects such as low real-time performance and excessively high sampling frequency requirements.

[0005] The present invention is achieved through the following technical solutions:

[0006] The present invention provides a method for monitoring the amplitude and phase difference of a three-phase circuit based on a variable step size LMS, including the following steps:

[0007] Based on a preset proportional coefficient, the three groups of large voltage signals in the three-phase circuit are converted into three groups of small voltage signals respectively;

[0008] The three groups of small voltage signals are sampled respectively based on a preset sampling frequency to obtain three groups of digital voltage signals;

[0009] performing Hilbert transform on the three groups of digital voltage signals respectively to obtain three groups of complex voltage signals;

[0010] Calculating a phase difference and an amplitude ratio between the input complex voltage signal and the expected complex voltage signal based on the expected complex voltage signal, the input complex voltage signal, and a variable step-size LMS algorithm;

[0011] The expected complex voltage signal is any one of the three groups of complex voltage signals, and the input complex voltage signal is the remaining two groups of complex voltage signals;

[0012] Based on the phase difference and the amplitude ratio, it is determined whether the voltage signal in the three-phase circuit has achieved three-phase balance and whether the phase sequence is correct.

[0013] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0014] (1) The variable step-size LMS algorithm is used to iteratively calculate the phase difference and amplitude ratio, which has the characteristics of fast convergence speed and small calculation amount, and can effectively improve the real-time performance and accuracy of monitoring;

[0015] (2) Filtering useless harmonic signals in the voltage signal can improve the anti-interference ability of monitoring and thus improve the accuracy of monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments of the present application. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0017] Figure 1 This is a flowchart of a specific implementation of an embodiment of the present application;

[0018] Figure 2 These are the measurement results of the amplitude ratio and phase difference of the two-phase voltage signals when the sampling frequency is 300 Hz and 128 time points are taken in the embodiment of the present application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0020] Example 1

[0021] A three-phase circuit amplitude and phase difference monitoring method based on variable step size LMS, such as Figure 1 As shown, the following steps are included:

[0022] S1: Converting three groups of large voltage signals in the three-phase circuit into three groups of small voltage signals based on a preset proportional coefficient;

[0023] The large voltage signal mentioned in the embodiment of the present application refers to a voltage signal of hundreds of volts in a three-phase circuit, generally a 220V voltage signal; the small voltage signal refers to a voltage signal of several volts.

[0024] During specific implementation, the ratio of the primary winding and the secondary winding of the transformer can be adjusted according to actual needs, so that the three groups of large voltage signals in the three-phase circuit (A-phase large voltage signal, B-phase large voltage signal and C-phase large voltage signal) are respectively converted into three groups of small voltage signals (A-phase small voltage signal, B-phase small voltage signal and C-phase small voltage signal).

[0025] S2: sampling the three groups of small voltage signals based on a preset sampling frequency to obtain three groups of digital voltage signals;

[0026] In this step, the analog signal is converted into a digital signal by sampling for subsequent processing. In specific implementation, synchronous sampling can be performed by an A / D sampler according to a preset sampling frequency.

[0027] S3: performing Hilbert transform on the three sets of digital voltage signals respectively, thereby converting the three sets of real voltage signals into three sets of complex voltage signals;

[0028] S4: Calculate the phase difference and amplitude ratio between the input complex voltage signal and the expected complex voltage signal based on the expected complex voltage signal, the input complex voltage signal, and the variable step size LMS algorithm;

[0029] Among them, the expected complex voltage signal is any one of the three groups of complex voltage signals, and the input complex voltage signal is the remaining two groups of complex voltage signals; that is, if the A-phase complex voltage signal is taken as the expected complex voltage signal, then the B-phase complex voltage signal and the C-phase complex voltage signal are the input complex voltage signals.

[0030] Specifically, step S4 includes the following sub-steps:

[0031] Step 1: Let the desired complex voltage signal be d(k), and any input complex voltage signal be u(k) (the other input complex voltage signal is processed similarly, and synchronous or asynchronous processing is adopted according to the actual situation). The difference between the desired complex voltage signal and the input complex voltage signal is e(k), k is the sampling time point, and the iteration step size μ at each sampling time point is:

[0032]

[0033] Among them, β and α represent coefficients, which are constants, and α>0. λ max It is the maximum eigenvalue of the autocorrelation matrix of the input complex voltage signal. During its implementation, its specific value is set according to the initial error (actual situation).

[0034] Assume that the initial time k = 1, then the estimation error and iteration step size at the initial time point are:

[0035] e(1)=d(1)-u(1);

[0036]

[0037] Step 2: Update the weight vector w(k+1) at the next sampling time k+1 as follows:

[0038]

[0039] in, is the complex conjugate of e(k);

[0040] Step 3: Calculate the estimated error e(k+1) as follows:

[0041] e(k+1)=d(k+1)–w(k+1) T u(k+1);

[0042] Among them, w(k+1) T is the conjugate transpose of w(k+1).

[0043] Step 4: Update the step size factor as follows:

[0044]

[0045] Step 5: Repeat steps 2 to 4 until the sampling point loop iteration is completed, and obtain the convergence process curve of the weight vector w(k). Select an appropriate steady-state time t, and read the amplitude and phase angle of w(t), which are the amplitude and phase difference between the desired complex voltage signal and the input complex voltage signal. Among them, the amplitude of w(t) represents the ratio of the desired complex voltage signal to the input complex voltage signal, and the phase angle of w(t) represents the phase difference between the desired complex voltage signal and the input complex voltage signal.

[0046] S5: Determine whether the voltage signal in the three-phase circuit has achieved three-phase balance and whether the phase sequence is correct based on the phase difference and amplitude ratio.

[0047] Specifically, if the phase difference is within the preset phase difference range and the amplitude ratio is within the preset amplitude ratio range, it is judged that the voltage signal in the three-phase circuit reaches three-phase balance; if the phase difference is within the preset phase difference range, it is judged that the phase sequence in the three-phase circuit is correct.

[0048] Compared with the existing discrete Fourier transform method, the variable step size LMS algorithm is used in the embodiment of the present application to calculate the phase difference and amplitude ratio between the input complex voltage signal and the expected complex voltage signal. Since the variable step size LMS algorithm uses an adaptive step size factor (a large step size factor is used to make it converge quickly when the error is large, and a small step size factor is used to obtain a smaller steady-state error when the error is small), it has the characteristics of fast convergence speed and small amount of calculation when calculating the phase difference and amplitude ratio. The phase difference and amplitude ratio can be obtained at a faster speed, thereby effectively improving the real-time and accuracy of monitoring; in addition, compared with the discrete Fourier transform method, the variable step size LMS algorithm in the embodiment of the present application has low requirements on the sampling frequency. While meeting the same accuracy, the sampling frequency required by the variable step size LMS algorithm is significantly lower than that of the discrete Fourier transform method, which can effectively avoid the system from being in an overloaded working state and further improve the real-time performance of monitoring.

[0049] Furthermore, considering that voltage signals often contain useless harmonic signals, and to prevent harmonic signals from affecting subsequent identification, the present embodiment also includes a filtering step, which is used to filter the digital voltage signal to remove harmonic signals and suppress noise, thereby further improving monitoring accuracy. Specifically, in this embodiment, a bandpass filter is used to filter the digital voltage signal.

[0050] Furthermore, in order to facilitate informing the staff when the voltage signal in the three-phase circuit is in a three-phase imbalance or a phase sequence disorder, an alarm step is also included, which is used to alarm when the voltage signal in the three-phase circuit is in a three-phase imbalance or a phase sequence disorder.

[0051] Example 2

[0052] This embodiment takes the A and B phase voltage signals in the three-phase voltage as an example to illustrate the solution provided by the embodiment of the present application as follows:

[0053] Step 1: Assume that the B-phase voltage phase leads the A-phase voltage phase by 120 degrees. Under normal conditions, the lower threshold of the amplitude ratio of phases B and A is 0.95, the upper threshold is 1.05, the lower threshold of the phase difference between phases A and B is -122 degrees, and the upper threshold is -118 degrees. Set the A-phase voltage signal and the B-phase voltage signal to be composed of a fundamental frequency of 50Hz, a second harmonic frequency of 100Hz, and a third harmonic frequency of 150Hz. Set the transformer strong-weak current conversion ratio to 220. After the A-phase voltage signal is converted by the transformer strong-weak voltage, the amplitudes of the fundamental frequency component, the second harmonic component, and the third harmonic component are 1, 0.3, and 0.2, respectively, and the initial phases are all 0. After the B-phase voltage signal is converted by the transformer strong-weak voltage, the amplitudes of the fundamental frequency component, the second harmonic component, and the third harmonic component are 0.8, 0.3, and 0.2, respectively, and the initial phases are 0. The two-phase voltage signals are doped with Gaussian white noise with a variance of 0.01 and a mean of 0. At this time, the amplitude ratio of the fundamental frequency component of phase B to the fundamental frequency component of phase A is 0.8, and the phase difference between the fundamental frequency component of phase A and the fundamental frequency component of phase B is That is -132 degrees.

[0054] Step 2: Set the sampling frequency of the A / D sampler to 300 Hz and the number of sampling points to 128, and synchronously sample the A-phase small voltage signal and the B-phase small voltage signal to obtain the A-phase digital voltage signal and the B-phase digital voltage signal;

[0055] Step 3: Set the bandpass filter order to 10 and the passband frequency to 40-60 Hz, filter the A-phase digital voltage signal and the B-phase digital voltage signal, and perform Hilbert transform on them to obtain the A-phase complex voltage signal and the B-phase complex voltage signal;

[0056] Step 4: Set the parameters α and β in the variable step size LMS iterative algorithm to 30 and 0.5, and use the B-phase complex voltage signal as the desired complex voltage signal d(k), and the A-phase complex voltage signal as the input complex voltage signal u(k). The difference between the desired complex voltage signal and the input complex voltage signal is e(k), where k is the sampling time point, and k = 1, 2, 3, ..., 128.

[0057] Let the initial weight vector w(1) be 0, then the initial estimation error is as shown in formula (1):

[0058] e(1)=d(1)-u(1); (1)

[0059] At this time, the initial step size μ is as shown in formula (2):

[0060]

[0061] Step 5: Update the weight vector w(k+1) at the next sampling moment according to formula (3):

[0062]

[0063] in, is the complex conjugate of e(k);

[0064] Step 6: Calculate the estimated error e(k+1) according to formula (4):

[0065] e(k+1)=d(k+1)–w(k+1) T u(k+1); (4)

[0066] Step 7: Update the step size factor according to formula (5):

[0067]

[0068] Step 8: Repeat steps 5 to 7 until the sampling point k = 128, that is, the loop iteration is completed, and the weight vector w(k) convergence process curve is obtained, as shown in Figure 2 As shown by Figure 2 It can be seen that when it iterates to the 20th sampling point, or 0.066s, it reaches a stable state; at the 60th sampling time point, or 0.2s, it converges to a relatively accurate amplitude ratio and phase difference (the vertical axis values ​​in the figure). Moreover, from 60 to 128 sampling time points, the maximum estimated error of the amplitude ratio is only 0.0046, and the maximum estimated error of the phase difference is only 0.1949 degrees. Therefore, the weight vector at the 60th sampling point is selected, and its amplitude and phase angle are read to obtain the amplitude and phase difference between the input signal and the desired signal.

[0069] Step 9: After reading the amplitude ratio and phase difference, determine whether the A-phase voltage signal and the B-phase voltage signal are balanced and whether the phase sequence of the A-phase voltage signal and the B-phase voltage signal are normal based on the set threshold value; if unbalanced or abnormal, an automatic alarm is issued. In this embodiment, the phase sequence of phases A and B is normal, and the three phases are unbalanced, so the device alarms.

[0070] It can be seen from this that when the solution provided by the embodiment of the present invention is adopted, the obtained three-phase voltage amplitude ratio and phase difference estimation results are accurate, the convergence speed is very fast, the estimation error is very small, and the real-time performance is high, which solves the shortcomings of traditional zero-crossing detection method, FFT method, etc., which are greatly affected by harmonics, have low real-time performance, and have high sampling frequency requirements.

[0071] The above is a detailed introduction to the three-phase circuit amplitude and phase difference monitoring method based on variable step size LMS provided by this application. This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method of this application and its core idea; at the same time, for general technical personnel in this field, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on this application.

Claims

1. A three-phase circuit amplitude and phase difference monitoring method based on variable step size LMS, characterized in that: The following steps are involved: Based on a preset proportional coefficient, the three groups of large voltage signals in the three-phase circuit are converted into three groups of small voltage signals respectively; The three groups of small voltage signals are sampled respectively based on a preset sampling frequency to obtain three groups of digital voltage signals; performing Hilbert transform on the three groups of digital voltage signals respectively to obtain three groups of complex voltage signals; Calculating a phase difference and an amplitude ratio between the input complex voltage signal and the expected complex voltage signal based on the expected complex voltage signal, the input complex voltage signal, and a variable step-size LMS algorithm; The expected complex voltage signal is any one of the three groups of complex voltage signals, and the input complex voltage signal is the remaining two groups of complex voltage signals, comprising the following steps: Step 1: Update the weight vector at the next sampling moment based on the weight vector at the current sampling moment and the iteration step size at the current sampling moment: e(k)=d(k)-u(k); Among them, w(k+1) represents the weight vector at the next sampling moment, w(k) represents the weight vector at the current sampling moment, the weight vector w(1)=0 at the initial moment, μ(k) represents the iteration step size at the current sampling moment, β and α represent coefficients, which are constants. is the conjugate complex number of e(k), e(k) represents the difference between the expected complex voltage signal and the input complex voltage signal at the current sampling time, d(k) represents the expected complex voltage signal, and u(k) represents the input complex voltage signal; Step 2: Calculate the difference between the expected complex voltage signal and the input complex voltage signal at the next sampling moment: e(k+1)=d(k+1)–w(k+1) T u(k+1); Where, e(k+1) represents the difference at the next sampling moment, d(k+1) represents the expected complex voltage signal at the next sampling moment, and w(k+1) T is the conjugate transpose of w(k+1), and u(k+1) represents the iterative step size of the next sampling moment; Step 3: Update the step size factor for the next sampling moment: Step 4: Repeat steps 1 to 3 until the sampling point loop iteration is completed and the weight vector convergence process curve is obtained; Step 5: Read the amplitude and phase angle corresponding to the steady-state time t in the weight vector convergence process curve to obtain the phase difference and the amplitude ratio; Based on the phase difference and the amplitude ratio, it is determined whether the voltage signal in the three-phase circuit has achieved three-phase balance and whether the phase sequence is correct.

2. The method for monitoring amplitude and phase difference of a three-phase circuit based on variable step size LMS according to claim 1, characterized in that: The method further includes a filtering step for filtering the digital voltage signal.

3. The method for monitoring amplitude and phase difference of a three-phase circuit based on variable step size LMS according to claim 1, characterized in that: The method further comprises an alarm step, wherein the alarm step is used to alarm when the voltage signal in the three-phase circuit is in three-phase imbalance or phase sequence disorder.

Images