Drive output current detection method

Abstract

The application provides a driver output current detection method, which comprises the following steps: recording data and calculating average value and waveform sinusoidal degree according to the above method after the system reaches steady state; if the difference between the calculation result of the measured driver output current and the reference data of the detection equipment is within a reasonable range, the test is qualified, otherwise the test is unqualified. The application adopts a phase-locked loop method to determine the current zero-crossing point, and adopts an equivalent approximate method to measure the harmonic distortion rate and compare the transient error of the reference waveform to judge the waveform sinusoidal degree, so that the calculation efficiency and accuracy are improved from the above aspects.

Description

Technical Field

[0001] This invention belongs to the field of motor control technology, and in particular relates to a method for detecting the output current of a driver. Background Technology

[0002] As a precision electronic device, the manufacturing quality of motor drives not only affects system efficiency and product performance, but can even lead to serious safety accidents. Currently, the mainstream quality inspection method is power aging to screen for defective products. However, this method can only detect serious process defects such as short circuits that cause controller malfunctions, but it cannot detect faults caused by incorrect soldering or poor soldering of peripheral components. Therefore, a more precise testing method is needed to ensure the normal operation of the drive. Since the quality of the output current directly assesses the drive's performance, to improve testing accuracy, the quality of the sinusoidal current signal output by the drive can be tested simultaneously with power aging testing, helping to further improve the manufacturing quality of the drives.

[0003] The aforementioned method for detecting the sinusoidal property of the output current waveform warrants further investigation. There are many methods for detecting sinusoidal property. A traditional method involves using Fourier transform to convert the time-domain signal into a frequency-domain signal to analyze its sinusoidal property. However, achieving accurate results requires a large amount of data, is highly complex, computationally intensive, and lacks real-time performance. Furthermore, it is insensitive to occasional transient distortions. In addition, to address the issue of synchronization between the reference standard waveform and the actual waveform, zero-crossing detection is generally used. Traditional zero-crossing detection determines the position of the zero-crossing point by observing the sign of the current near it. Since actual output current measurements often have deviations, this method introduces a significant error, potentially affecting the detection results. Summary of the Invention

[0004] This invention aims to solve the following technical problems:

[0005] 1. The traditional method for detecting sinusoidal intensity is to convert the time-domain signal into a frequency-domain signal using Fourier transform to analyze the sinusoidal intensity of the signal. However, if the accuracy of the results is desired, a large amount of data is required, and the complexity and computational load are high. The real-time performance is not high, and it is not sensitive to some occasional transient distortions.

[0006] 2. To address the issue of whether the reference standard waveform is synchronized with the actual waveform, zero-crossing detection is generally used. Traditional zero-crossing detection determines the position of the zero-crossing point by observing the sign of the current near the zero-crossing point. However, since the measurement of the actual output current often has deviations, this method introduces a large error, which may affect the detection results.

[0007] Therefore, the present invention provides a method for detecting the output current of a driver to solve the above-mentioned technical problems.

[0008] The present invention adopts the following technical solution:

[0009] A driver output current detection method, characterized in that the detection device detects the current output by the driver in real time during operation, and evaluates whether the driver output current meets the requirements through the driver output current detection method, the driver output current detection method including the following steps:

[0010] S1, perform current zero-crossing detection. Assume the real-time sinusoidal signal is y(k), and the predicted sinusoidal signal at time k is...

[0011] ω is the frequency of the real sine wave signal. To estimate the amplitude of the sinusoidal signal, To estimate the phase of the signal;

[0012] Determine the zero-crossing point and use this point as the starting point for counting. Divide one cycle into N equal parts according to the N data points of each waveform.

[0013] S2, Standard reference data recording and point-by-point average calculation of the testing equipment;

[0014] S3, Standard reference data current sinusoidal calculation, uses an equivalent approximation measurement of harmonic distortion rate and a real-time calculation of transient error by comparing with the reference waveform to evaluate the current waveform sinusoidalness. Step 3 includes:

[0015] S3a: Real-time transient error is calculated for each data point of the test data and reference data, and the maximum value of the transient error calculation for N data points is taken as the evaluation criterion for the detection result.

[0016]

[0017] The average value of the Nth point in the standard reference data, N N The instantaneous value of the Nth point in the standard reference data / test data;

[0018] S3b, an equivalent approximation for measuring harmonic distortion rate;

[0019] The standard method for calculating the harmonic distortion rate of testing equipment is as follows:

[0020]

[0021] I1 is the fundamental frequency amplitude, I rms This is the average value over one current cycle;

[0022] S4. Determine whether the tested driver passes the test:

[0023] The real-time transient error of the current data of the driver under test is calculated according to step 3a and compared with the real-time transient error of the standard reference data. If the error is within the allowable range, this item is qualified.

[0024] According to step 3b, the harmonic distortion rate of the sinusoidal current of the driver under test is calculated and compared with the harmonic distortion rate of the sinusoidal current of the standard reference data. If the error is within the allowable range, this item is qualified.

[0025] If both of the above criteria are met, the driver passes the test; otherwise, the test fails.

[0026] As a preferred technical solution, the method for determining the zero-crossing point in step 1 is as follows:

[0027] The estimated error value e0(k) is expressed as:

[0028]

[0029] Where ω is the frequency of the real sinusoidal signal. To estimate the amplitude of the sinusoidal signal, To estimate the phase of the signal;

[0030] Calculate amplitude and phase:

[0031]

[0032]

[0033] When the target value y(k) and the predicted value When the system approaches a stable value, equations (3) and (4) tend to reach a stable value. The moment is taken as the starting point for counting.

[0034] As a preferred technical solution, step 2 includes:

[0035] After the system stabilizes, data recording begins. m waveforms are collected, and N points are recorded for each waveform. The N data points for each waveform are then sorted. After recording m sets of data, the 1st, 2nd, 3rd...Nth points of each of the m waveforms are stored and their averages are calculated. For example, the average value is calculated from the Nth point N1 of the first waveform to the Nth point N of the mth waveform. m Summing and then averaging, we obtain the average value N of the Nth point. N By doing so, a sequence of N data points is finally obtained as a reference value for the point-by-point average of the data processing.

[0036]

[0037] This is the average value of the Nth point of the standard reference current data.

[0038] As a preferred technical solution, in step 2, if the non-ideal factors of the actual load can be ignored, the estimated sine wave obtained by equations (3) and (4) is used as the standard parameter data.

[0039] By adopting the above technical solution, the present invention has the following advantages:

[0040] 1) The traditional method for judging the sinusoidal nature of current is to use Fourier transform to calculate the proportion of each harmonic component. This method has a large amount of calculation, low real-time performance, and is not sensitive to some occasional transient distortions. This invention improves upon this by using an equivalent approximation to measure the harmonic distortion rate and a reference waveform to calculate the transient error in real time to evaluate the sinusoidal nature of the current waveform. This ensures data synchronization and improves the efficiency and accuracy of the calculation.

[0041] 2) The traditional method for determining the zero-crossing point uses the commutation near the zero-crossing point as the detection basis, and uses the positive or negative current value to determine the position of the zero-crossing point. Near the zero point, the current sample in the previous cycle is negative and the current sample in the next cycle is positive. This is used to determine the zero-crossing point and use this point as the starting point for counting. However, because the current may fluctuate near the zero-crossing point, there will be errors in the detection, resulting in inaccurate zero-crossing point judgment. Moreover, the accumulated error will have a significant impact on the test results. In order to avoid the above problems, this invention adopts a phase-locked loop to lock the phase to determine the counting starting point, which greatly reduces the error in zero-crossing point judgment. Attached Figure Description

[0042] Figure 1 A schematic diagram of the driver output current detection device;

[0043] Figure 2 A flowchart illustrating the steps of a driver output current detection method;

[0044] Figure 3 The waveform diagram is for transient distortion detection.

[0045] Figure 4 The waveform diagram is for waveform distortion detection;

[0046] Figure 5 This represents the harmonic distortion rate of the equivalent approximation measurement. Detailed Implementation

[0047] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] like Figure 1As shown, the driver output current detection device includes: a power supply, a driver, a load circuit, a detection device, and a fault indication unit. The detection device includes a current detection circuit. The power supply is connected to the driver, the driver is connected to the current detection circuit of the detection device, and the fault indication unit is connected to the detection device. The load circuit is connected in parallel between the driver and the current detection circuit. An inductor serves as the load of the driver under test. The detection device detects the output current of the driver in real time during operation and evaluates whether the driver output current meets the requirements according to a predetermined detection method. The detection method mainly evaluates whether the sinusoidal nature and error value of the driver output current are within the allowable range based on the reference data of the detection device.

[0049] like Figure 2 As shown, using the aforementioned driver output current detection device, the present invention provides a driver output current detection method, comprising the following steps:

[0050] A. Equipment calibration; B. Driver testing; C. Test result analysis.

[0051] The reference data for the following testing equipment is referred to as "reference data", and the actual output data of the driver is referred to as "test data".

[0052] Step A is as follows: If the testing equipment is being used for the first time or requires calibration, first use a standard driver to output a current with a specific frequency and amplitude. The testing equipment will then calibrate the standard reference data based on the output current. Otherwise, proceed directly to step B, driver testing.

[0053] In step A, the calibration data is divided into the following steps:

[0054] S1. Current zero-crossing detection:

[0055] Assuming the real-time sinusoidal signal is y(k), the predicted value at time k is... The predicted error value e0(k) can be expressed by equation (1), and can be expressed by equation (2):

[0056]

[0057]

[0058] Where ω is the frequency of the real sinusoidal signal. To estimate the amplitude of the sinusoidal signal, To estimate the phase of the signal.

[0059] The amplitude and phase are calculated using the RGN calculation method, and equations (3) and (4) are obtained:

[0060]

[0061]

[0062] When the target value y(k) and the predicted value When the system approaches a stable value, equations (3) and (4) tend to reach a stable value. At any given moment, a cycle is divided into N equal parts.

[0063] S2. Standard reference data recording and point-by-point average calculation:

[0064] As described in step 1, after the system state converges, begin recording data. Sort the N data points for each waveform, and record m sets of data. Then, store the 1st, 2nd, 3rd...Nth points recorded for each of the m waveforms and calculate the average value. For example, the average value can be calculated from the Nth point N1 of the first waveform to the Nth point N of the mth waveform. m Summing and then averaging, we obtain the average value N of the Nth point. N This process is repeated until a sequence of N data points is obtained, which serves as a reference value for the point-by-point average in data processing.

[0065]

[0066] This is the average value of the Nth point of the standard reference current data.

[0067] It is worth noting that the above solution is the preferred option. When considering that the actual load is not ideal, the output current itself may not be sinusoidal due to load reasons. However, if a standard sine wave is used for judgment in this case, misjudgment is very likely to occur.

[0068] If the non-ideal factors of the actual load can be ignored, the estimated sine waves obtained by equations (3) and (4) can be directly used as standard parameter data. It is worth noting that since the standard reference data obtained in step 1 and equation (1) only require ω, Three representative quantities can save a lot of data storage space.

[0069] S3. Standard reference data current sinusoidal calculation:

[0070] The traditional method for judging the sinusoidal nature of current is to use Fourier transform to calculate the proportion of each harmonic component. This method involves a large amount of calculation, has low real-time performance, and is not sensitive to some occasional transient distortions. This invention improves upon this by using an equivalent approximation to measure the harmonic distortion rate and a reference waveform to calculate the transient error in real time to evaluate the sinusoidal nature of the current waveform, thereby improving the efficiency and accuracy of the calculation.

[0071] S3a, real-time calculation of transient error;

[0072] The real-time transient error calculation method is as shown in equation (6). The real-time transient error is calculated for each data point of the test data and the reference data, and the maximum value of the transient error calculation of N data points is taken as the evaluation criterion of the detection result. This method has a fast response and simple processing for occasional transient distortions.

[0073]

[0074] The average value of the Nth point in the standard reference data, N N The instantaneous value of the Nth point in the standard reference data / test data.

[0075] like Figure 3 The diagram shows the waveform changes of the waveform reference data and the waveform test data. When the measured waveform is distorted due to noise interference, the distortion shown in the diagram will occur. If this is the Nth data point sampled, the transient error calculated at this point is relatively large, while the transient errors calculated at other points are all smaller than this value. Then, the maximum transient error calculated for this current cycle is the transient error at this point, which is the maximum transient error of the current test waveform. By comparing the measurement error with the reference error, it can be determined whether the data is within a reasonable range.

[0076] S3b, an equivalent approximation for measuring harmonic distortion rate;

[0077] The real-time transient error calculation described in equation (1) can satisfy whether the initial screening data is within the set threshold range, but it lacks theoretical guidance for further problem analysis. Therefore, this invention introduces harmonic distortion rate for further analysis. The standard harmonic distortion rate of the detection equipment is calculated as shown in equation (7).

[0078]

[0079] I1 is the fundamental frequency amplitude, I rms This is the average value over one current cycle.

[0080] like Figure 4 As shown, the waveforms of the standard reference waveform, abnormal waveform 1 (positive half-cycle waveform anomaly), and abnormal waveform 2 (noise interference) are displayed. After equivalent approximate measurement of the harmonic distortion rate, as shown... Figure 5 As shown, the harmonic distortion rate of the standard reference waveform < the harmonic distortion rate of the abnormal waveform 2 < the harmonic distortion rate of the abnormal waveform 1.

[0081] When the real-time transient error calculated in step 3a is large, it can be clearly seen that the test data is abnormal, but it cannot give a specific abnormality prompt. By measuring the harmonic distortion rate in step 3b, we can further analyze what kind of abnormality it is, screen out the cause of the abnormality of the driver under test, and help engineers find the problem.

[0082] Step B, the detection and calculation of the output current of the driver under test, includes:

[0083] according to Figure 1 Connect the testing equipment, the driver under test, and the load. If the testing equipment has been calibrated or calibrated, proceed to the driver testing stage. After starting the driver, output current in real time. The testing equipment detects the output current of the driver and calculates the zero-crossing point of the driver using the method in step 1. After the system reaches stability, start storing data and calculate the harmonic distortion rate and real-time transient error of the current sinusoidal data of the driver under test according to step 3.

[0084] In step C, the test results are calculated and analyzed to determine whether the tested driver passes the test.

[0085] Once the driver testing is complete, the test results analysis phase begins.

[0086] S4. Based on step 3a, calculate the real-time transient error of the current data of the driver under test and compare it with the real-time transient error of the standard reference data. If the error is within the allowable range, this item is qualified.

[0087] According to step 3b, the harmonic distortion rate of the sinusoidal current of the driver under test is calculated and compared with the harmonic distortion rate of the sinusoidal current of the standard reference data. If the error is within the allowable range, this item is qualified.

[0088] First, calculate whether the transient error meets the requirements. If it does not meet the requirements, the test is unqualified. At this time, the harmonic distortion rate can be further detected to determine whether the sine is within the required range, so as to filter whether the result belongs to abnormal waveform type 1 or abnormal waveform type 2.

[0089] If both of the above criteria are met, the driver passes the test; otherwise, the test fails.

[0090] When first using this method, since there is no actual test data as parameters, the error range can be initially set to 10%. That is, if the maximum transient error and harmonic distortion rate of the tested drive data are within 10% of the maximum transient error and harmonic distortion rate of the standard reference data, respectively, it is considered a pass. After a certain number of tests, all test data are statistically analyzed, and the error threshold is set according to the 6 Sigma principle.

[0091] The Six Sigma principle states that the tolerance limits of any product are pre-agreed (usually customer-specified), while the actual product specifications are affected by various random factors (minor errors). This causes the actual product specifications (most commonly geometrical indicators such as length, diameter, parallelism, and perpendicularity; other quantifiable indicators also follow this pattern) to follow a normal distribution. If the randomly generated product specifications fall outside the tolerance range, the product is considered defective, thus reducing the overall yield.

[0092] Errors are classified into systematic errors that cause deviations from the mean and random errors that cause an increase in the standard deviation.

[0093] Yield is the value of an error function with standard deviation:

[0094]

[0095] The upper limit (USL) and lower limit (LSL) are predetermined. The management objective is to reduce the error factors (mean and standard deviation) σ so that the values ​​M and N are within the specified management target values. The 3.4 ppm defect rate required in management textbooks is equivalent to N = 4.5 and M = 1.5, meaning the sum of random error and systematic error equals 6σ. Random error and systematic error are the maximum transient error and harmonic distortion rate of the data in this invention.

[0096] In addition to the preferred embodiments described above, there are other embodiments of the present invention. Those skilled in the art can make various changes and modifications based on the present invention, and all such changes and modifications should fall within the scope defined by the appended claims, as long as they do not depart from the spirit of the present invention.

Claims

1. A method for detecting the output current of a driver, characterized in that, The detection equipment monitors the output current of the driver in real time during operation, and evaluates whether the driver output current meets the requirements using the driver output current detection method. The driver output current detection method includes the following steps: S1 performs current zero-crossing detection, assuming the real-time sinusoidal signal is... The sinusoidal signal predicted at time k is : ；（1） The frequency of the real sine wave signal. To estimate the amplitude of the sinusoidal signal, To estimate the phase of the signal; Determine the zero-crossing point and use this point as the starting point for counting. Divide one cycle into N equal parts according to the N data points of each waveform. S2, Standard reference data recording and point-by-point average calculation of the testing equipment; S3, Standard reference data current sinusoidal calculation, uses an equivalent approximation measurement of harmonic distortion rate and a real-time calculation of transient error by comparing with the reference waveform to evaluate the current waveform sinusoidalness. Step S3 includes: S3a: Real-time transient error is calculated for each data point of the test data and reference data, and the maximum value of the transient error calculation for N data points is taken as the evaluation criterion for the detection result. ；（6） The average value of the i-th point of the standard reference data. The instantaneous value of the i-th point in the standard reference data / test data; S3b, an equivalent approximation for measuring harmonic distortion rate; The standard method for calculating the harmonic distortion rate of testing equipment is as follows: ；（7） The fundamental amplitude, This is the average value over one current cycle; S4. Determine whether the tested driver passes the test: According to step S3a, the real-time transient error of the current data of the driver under test is calculated and compared with the real-time transient error of the standard reference data. If the error is within the allowable range, this item is qualified. According to step S3b, the harmonic distortion rate of the current sinusoidal value of the driver under test is calculated and compared with the harmonic distortion rate of the current sinusoidal value of the standard reference data. If the error is within the allowable range, this item is qualified. If both of the above criteria are met, the driver passes the test; otherwise, the test fails.

2. The driver output current detection method according to claim 1, characterized in that, In step S1, the method for determining the zero-crossing point is as follows: Predicted error value Expressed as: ；（2） in, The frequency of the real sine wave signal. To estimate the amplitude of the sinusoidal signal, To estimate the phase of the signal; Calculate amplitude and phase: ,0<c(k)<1;(3) ,0<c(k)<1;(4) When the target value and predicted value When the system approaches a certain value, it tends to stabilize, meaning equations (3) and (4) approach their stable values. The moment is taken as the starting point for counting.

3. The driver output current detection method according to claim 1 or 2, characterized in that, Step S2 includes: After the system stabilizes, data recording begins. m waveforms are collected, and N points are recorded for each waveform. The N data points for each waveform are sorted. After recording m sets of data, the 1st, 2nd, 3rd...Nth points of each of the m waveforms are stored and their averages are calculated. The i-th point of the first waveform... to the i-th point of the m-th waveform Summing and averaging yields the average value of the i-th point of the m waveforms. By doing so, a sequence of N data points is finally obtained as a reference value for the point-by-point average of the data processing. ；（5） This is the average value of the i-th point of the standard reference current data.

4. The driver output current detection method according to claim 2, characterized in that, In step S2, if the non-ideal factors of the actual load are ignored, the estimated sine wave obtained by equations (3) and (4) is used as the standard parameter data.

Images