Current harmonic detection device and method
By optimizing the current harmonic detection algorithm through the linear array structure of the TMR sensor and the fast Fourier transform of the Hamming window function, the problem of insufficient detection performance of existing devices is solved, and accurate current harmonic detection and analysis are achieved, while reducing errors.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-16
- Publication Date
- 2026-04-03
AI Technical Summary
Existing Hamming window current harmonic detection devices and methods have insufficient detection performance and accuracy, and the error increases as the measurement range increases, resulting in unreliable data.
A linear array structure of TMR sensors is adopted, and the fast Fourier transform of differential processing and Hamming window function is combined to optimize the current harmonic detection algorithm. The theoretical value of current is inverted through Biot-Savart law, and the weighted averaging method is used to improve data reliability.
It enables accurate detection and analysis of current harmonics, reduces noise interference, improves detection performance and data reliability, and reduces errors.
Smart Images
Figure CN121784367A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power equipment information detection, and in particular to a current harmonic detection device and method. Background Technology
[0002] In power systems, current harmonics have always been a major concern for power engineers. With the widespread use of power electronic devices and the increase in nonlinear loads, harmonic pollution in power systems is becoming increasingly serious. Harmonics not only lead to a decline in power quality but can also damage power equipment and even affect the stable operation of the power system.
[0003] Traditional methods for detecting current harmonics mainly include analog filters and Fourier transforms. While analog filters are simple in structure, their filtering effect is easily affected by changes in component parameters and temperature, and they can only detect harmonics of specific frequencies, making them poorly adaptable to complex and variable harmonic environments. Although Fourier transforms can perform frequency domain analysis of signals, they suffer from spectral leakage and picket-fence effects when processing aperiodic or time-varying signals, thus limiting the accuracy and resolution of harmonic detection.
[0004] To overcome the limitations of traditional methods, researchers have begun exploring current harmonic detection methods based on digital signal processing (DSP) technology. Among these, the FFT transform method based on window functions has gained popularity due to its efficiency and accuracy. Window functions can effectively reduce the effects of spectral leakage and the picket fence effect, thereby improving the accuracy of harmonic detection. The Hamming window function, as a commonly used window function, has smooth transition characteristics and low sidelobe levels, which can further improve the performance of harmonic detection.
[0005] Therefore, a current harmonic detection device based on the Hamming window algorithm has emerged. This device combines the advantages of DSP technology with the characteristics of the Hamming window function, enabling real-time and accurate harmonic detection and analysis of current signals in power systems. The application of this device will help improve power quality, ensure the normal operation of power equipment, and promote the stability and development of the power system.
[0006] However, the existing Hamming window current harmonic detection devices have insufficient detection performance, and the error increases with the increase of the measurement range, making the detected data unreliable. Summary of the Invention
[0007] The purpose of this invention is to propose a current harmonic detection method to solve the technical problem of insufficient detection performance and accuracy of existing Hamming window current harmonic detection devices and methods.
[0008] Specifically, the present invention provides a current harmonic detection device, comprising: the device adopts a TMR sensor linear array structure.
[0009] A current harmonic detection method, applied to the aforementioned detection device, includes the following steps: S1. Use the detection device to acquire the magnetic field generated by the current signal and output a voltage signal; S2. Based on the magnetic field and voltage signals generated by the current signal, the theoretical value of the current is obtained by inversion using the Biot-Savart law; S3. Perform differential processing on the voltage signal to obtain the actual value of the current after differential processing; S4. Perform a Fast Fourier Transform of the Hamming window function on the actual current value after differential processing to complete the current harmonic detection.
[0010] The beneficial effects provided by this invention are: (1) Compared with traditional current harmonic detection methods such as analog filters and Fourier transform, the FFT algorithm based on Hamming window can achieve accurate detection and analysis of current harmonics. At the same time, the FFT algorithm with added Hamming window has smooth transition characteristics and lower sidelobe level compared with sine window and rectangular window, which can further improve the performance of harmonic detection.
[0011] (2) The TMR linear array probe structure is adopted. This structure can accurately measure the gradient change of the magnetic field and reduce the error while increasing the measurement range.
[0012] (3) The algorithm for current harmonics in linear arrays is optimized by using differential method to reduce noise interference and weighted average method to improve data reliability. Attached Figure Description
[0013] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 It is the original signal for current harmonic detection; Figure 3 This is a block diagram of the Hamming window FFT harmonic analysis algorithm implementation; Figure 4 This is a comparison chart showing the effects of conventional FFT analysis and FFT data extraction with the addition of the Hamming window function; Figure 5 This is a comparison chart of the results of conventional FFT analysis and FFT analysis with the addition of the Hamming window function. Detailed Implementation
[0014] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described below with reference to the accompanying drawings.
[0015] Before formally describing the present invention, a general description of the solution of the present invention will be given first to facilitate understanding.
[0016] This invention provides a current harmonic detection device, specifically employing a TMR sensor linear array structure.
[0017] It should be noted that a TMR sensor is a magnetically sensitive element that utilizes the tunneling magnetoresistance effect. Changing the external magnetic field can cause a change in the resistance of the TMR sensor. Its core principle is the tunneling magnetoresistance effect, that is, the tunnel structure in the material causes a change in resistance under the action of an external magnetic field, thereby realizing the measurement of the magnetic field.
[0018] It should be noted that the TMR sensor linear array structure includes four single-axis TMR sensors, namely the first TMR sensor, the second TMR sensor, the third TMR sensor, and the fourth TMR sensor, which are arranged in a straight line.
[0019] It should be noted that the first TMR sensor is located at the very front of the array, while the other three sensors are arranged sequentially at the rear of the array.
[0020] For ease of description, the present invention will be referred to as TMR1-TMR4 in the following text.
[0021] Please refer to Figure 1 , Figure 1 This is a schematic diagram of the method flow of the present invention.
[0022] A current harmonic detection method, applied to the aforementioned detection device, specifically includes the following steps: S1. Use the detection device to acquire the magnetic field generated by the current signal and output a voltage signal; S2. Based on the magnetic field and voltage signals generated by the current signal, the theoretical value of the current is obtained by inversion using the Biot-Savart law; It should be noted that step S2 is a calculated value in the theoretical process, and the whole process is a theoretical calculation process, mainly used to fully explain the principle of the present invention.
[0023] As one example, according to the Biot-Savart theorem, the magnetic induction intensity of the magnetic field generated by a current element on a current-carrying conductor at any point P in space under ideal conditions is calculated as follows:
[0024]
[0025] in, The permeability of free space, For the magnitude of the current, For the integration path, It is a tiny line element of the current element. Let be the unit vector pointing from the current element to the point of the field to be determined; This represents the distance vector from the current element to point P. Let P be the distance from the current element to point P. for and The angle.
[0026] Extending the current element on the current-carrying conductor to an infinitely long straight conductor, we can obtain the formula for the distribution of the magnetic field around an infinitely long current-carrying straight conductor:
[0027] Therefore, the magnetic field strengths at TMR2, TMR3, and TMR4 can be obtained as follows:
[0028] The formulas for calculating the TMR output voltage and the applied magnetic field strength are as follows:
[0029] In the formula, B The magnetic flux density of the sensor, S For the sensitivity of the sensor, i For sensor serial number, The sensor output voltage value. The power supply voltage for the sensor, These represent the distances of the TMR2, TMR3, and TMR4 sensors from the top of the TMR sensor linear array structure, respectively.
[0030] Therefore, the currents measured by TMR2, TMR3, and TMR4 can be obtained as follows:
[0031] Where G is the amplification factor of the preamplifier circuit.
[0032]
[0033] The current This refers to the power frequency current signal to be measured, which is also the current value.
[0034] S3. Perform differential processing on the voltage signal to obtain the differentially processed current value; It should be noted that step S3 is the actual calculation process of this invention, belonging to the optimization process. It is parallel or substitute for step S2, rather than being processed sequentially. Both are calculations of current values, but the methods are different. Step S2 calculates the theoretical value using the traditional Biot-Savart law, while step S3 is the optimized value specific to this invention.
[0035] It should be noted that the signal received by each sensor in the TMR sensor array can be equivalently regarded as the superposition of the magnetic field generated by the current being measured and the external interference magnetic field. At the same time, the circuit will also generate a certain amount of internal noise signal, as shown in the following formula:
[0036] in U The voltage value output by the sensor. Y The voltage signal is generated by the detection system after the magnetic field generated by the current to be measured is converted into a detection system. N The voltage noise signal is converted from the noise magnetic field. From equation (5), we can obtain:
[0037] Where B1, B2, B3, and B4 represent the magnetic field strength generated by the current to be measured at each sensor measurement position along the x-axis, and B0 represents the external interference magnetic field at... x Components in the axial direction, S For the sensitivity of each sensor, V cc Let be the power supply voltage on the TMR. From equation (3), we can obtain:
[0038] in, I For the magnitude of the current, μ 0 is the permeability of free space. , r i The distance from the TMR to the current-carrying conductor ( i =1, 2, 3, 4).
[0039] As one embodiment, to reduce the influence of external magnetic field noise on the measurement results, the output of the TMR array is differentially processed. The value of U is divided by the corresponding sensor sensitivity S and Vcc, and then subtracted from each other, as shown in the following formula:
[0040] In the formula, C is the difference value, and I is the current value to be obtained from each difference value. Then we have:
[0041] Because the sensitivity and accuracy of each sensor in this differential method are different, it is a measurement with unequal precision. In unequal precision measurement, in order to make the measurement results more accurate, a weighted average method is needed to process the data so that the more reliable data can be given a higher weight.
[0042]
[0043] In the formula p The weights are used to determine the accuracy of the final result. Two main factors influence the detection result: one is the sensor's sensitivity; higher sensitivity leads to higher detection accuracy, meaning the measured data is closer to the true value. The other is the distance between the sensor and the current-carrying conductor; the closer the distance, the closer the detected magnetic field data is to the true value. Therefore, the weights must be determined based on sensitivity and distance, as shown in the following formula:
[0044]
[0045] In the formula The weights corresponding to I (i=1,2,3) are used, and the sensitivity S is the lower of the two differentially expressed sensor values, with the distance being... The sensor that is closer to the other one in the differential calculation r value, The distance between the two sensors is when differential is used.
[0046] S4. Perform a Fast Fourier Transform of the Hamming window function on the actual current value after differential processing to complete the current harmonic detection.
[0047] It should be noted that current harmonic detection actually involves detecting the content of the fundamental frequency and each harmonic in the power frequency current signal. The frequencies of the fundamental frequency and each harmonic are fixed, that is, integer multiples of 50Hz of the fundamental frequency. FFT is a fast algorithm of discrete Fourier transform, which can transform a signal to the frequency domain and extract the spectrum of a signal for spectrum analysis. Therefore, FFT is used to perform amplitude-frequency analysis of the signal to obtain the content of each harmonic in the original signal.
[0048] In spectral analysis, many factors influence the results. Sampling frequency, data window type, window length, and the number of FFT sampling points all affect the detection results to varying degrees. In reality, the acquired signals are not ideally periodic, which contradicts the essence of Fourier detection. In actual detection, discrete signals must be truncated to periodize non-periodic signals, but this causes spectral leakage, which is the most significant source of error in FFT.
[0049] As one embodiment, this invention simulates and compares the effects of different window types on harmonic detection, including rectangular window, sinusoidal window, and Hamming window. An input sinusoidal signal containing 150Hz, 550Hz, and 950Hz is shown in the attached diagram. Figure 2 As shown.
[0050] A rectangular window sequence is a sequence of all 1s, a sine window sequence is a periodic sine function, and a Hamming window sequence is actually a cosine window. The algorithm implementation diagram is attached. Figure 3 As shown. Its window function is:
[0051] in =0.53836 is the Hamming window. w ( n ) is the signal number n The value of each point, N This is the number of sampling points; the simulation is set to 1024.
[0052] The data required for the Hamming window is stored in a register array. During windowing, the values in the array are simply retrieved, synchronized with the data acquired by the AD converter, and multiplied by 1024 points to obtain the corresponding data. w ( n ).
[0053] The sampling sequence for each window function is shown in the attached figure after simulation experiments. Figure 4 As shown. If a rectangular window is added, the effect is almost indistinguishable from directly extracting the data, while the sine window and Hamming window have obvious data weighting processing.
[0054] Subsequently, FFT amplitude-frequency analysis was performed on the sampled data after each windowing process, and the results are shown in the attached figure. Figure 5 As shown, the most significant difference between the FFT results with sinusoidal and Hamming windows and those with rectangular windows is the marked reduction in spectral tailing. Windowing also widens the bandwidth and significantly reduces peak values. Windowed FFT is a fixed-window truncation transform, which eliminates high-frequency interference, reduces spectral leakage, and improves accuracy.
[0055] The beneficial effects of this invention are: (1) Compared with traditional current harmonic detection methods such as analog filters and Fourier transform, the FFT algorithm based on Hamming window can achieve accurate detection and analysis of current harmonics. At the same time, the FFT algorithm with added Hamming window has smooth transition characteristics and lower sidelobe level compared with sine window and rectangular window, which can further improve the performance of harmonic detection.
[0056] (2) The TMR linear array probe structure is adopted. This structure can accurately measure the gradient change of the magnetic field and reduce the error while increasing the measurement range.
[0057] (3) The algorithm for current harmonics in linear arrays is optimized by using differential method to reduce noise interference and weighted average method to improve data reliability.
[0058] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A current harmonic detection method, applied to a current harmonic detection device, the device employing a TMR sensor linear array structure, the TMR sensor linear array structure comprising four single-axis TMR sensors, namely a first TMR sensor, a second TMR sensor, a third TMR sensor, and a fourth TMR sensor, the four sensors being arranged in a straight line, the first TMR sensor being located at the front end of the array, and the other three sensors being arranged sequentially at the rear end of the array, characterized in that: The method includes the following steps: S1. Use the detection device to acquire the magnetic field generated by the current signal and output a voltage signal; S2. Based on the magnetic field and voltage signals generated by the current signal, the theoretical value of the current is obtained by inversion using the Biot-Savart law; S3. Perform differential processing on the voltage signal to obtain the actual value of the current after differential processing; S4. Perform a Fast Fourier Transform of the Hamming window function on the actual current value after differential processing to complete the current harmonic detection.
2. The current harmonic detection method as described in claim 1, characterized in that: The formula for calculating the theoretical value of the current in step S2 is as follows: ，i=2，3，4； Among them, the serial number of the TMR sensor, The sensor output voltage value. The power supply voltage for the sensor; r i S represents the distance between the i-th TMR sensor and the first TMR sensor; i Let be the sensitivity of the i-th TMR sensor; ρ is the permeability of free space; G is the amplification factor of the preamplifier circuit.
3. The current harmonic detection method as described in claim 2, characterized in that: The actual current value after differential processing in step S3 is as follows: ， m =1，2，3； ， in, for I i The corresponding weights, U1-U4, are the output voltages of the first to fourth TMR sensors, respectively; It is a constant.
4. The current harmonic detection method as described in claim 3, characterized in that: weight p i The sensitivity of the TMR sensor itself and the distance between the sensors are determined as follows: Sensitivity Take the value with lower sensitivity from the two sensors that are compared, and the distance... Of the two sensors that are closer together, the one being analyzed by difference r value, This represents the distance between the two sensors when differential is used.
5. The current harmonic detection method as described in claim 1, characterized in that: The Hamming window function in step S4 is as follows: ， in, w ( n ) is the signal number n The value of each point, N It is the number of sampling points. This is the default value.