A universal ac-dc current sensor and a current measurement method

Abstract

The application discloses a kind of AC-DC universal current sensor and current measurement method, the current sensor includes Hall sampling unit, signal conditioning unit, microcontroller unit, temperature detection unit and analog output unit.Hall sampling unit senses measured current and exports Hall voltage signal, signal conditioning unit carries out biasing and exports conditioning signal, microcontroller unit carries out sampling to conditioning signal, according to temperature information, sampling value is carried out zero point bias compensation, and the true effective value of measured current is calculated by combining gain coefficient, is converted into standard analog signal by analog output unit.The application can be compatible with alternating current, direct current and AC-DC superimposed current measurement, through unified software processing pre-zero compensation and true effective value calculation, reduce hardware cost and improve the measurement accuracy and temperature stability under complex waveform.

Description

Technical Field

[0001] This invention relates to the field of industrial automation current detection and sensor technology, specifically to a universal AC / DC current sensor and current measurement method. Background Technology

[0002] In industrial automation, power monitoring, frequency converter drives, motor control, energy storage systems, and new energy equipment, it is often necessary to detect the current in conductors and convert the detection results into standard analog signals for subsequent PLCs, DCSs, display instruments, or monitoring systems to acquire. Common standard outputs include 0-5V, 0-10V, 1-5V, 2-10V, 0-20mA, and 4-20mA.

[0003] Existing current sensing products typically include current transformer sensors, Hall effect current sensors, and true RMS transmitters. Among them, current transformers are generally suitable for AC current sensing and are difficult to directly measure DC current; although ordinary open-loop Hall effect sensors can detect AC and DC currents, if only average value, peak value, or simple filtering is performed at the back end, it is difficult to accurately reflect the true RMS value with thermal effects in scenarios such as non-sinusoidal waves, pulsating currents, PWM currents, AC and DC superimposed currents, and inverter output currents.

[0004] Some existing products use dedicated true RMS conversion chips to calculate RMS values, but this method suffers from high costs, inflexible parameter configuration, limited accessibility to domestic alternatives, and constraints in temperature compensation and output mapping capabilities. For industrial applications with complex waveforms, wide frequency ranges, and significant ambient temperature variations, relying solely on fixed hardware conversion circuits makes it difficult to balance cost, accuracy, and versatility.

[0005] Furthermore, open-loop Hall effect devices are prone to zero-point drift and sensitivity changes due to temperature. Without temperature correction, this can lead to increased deviation in the low-current range, decreased full-scale accuracy, and poor long-term stability. Existing low-cost current sensors generally lack a unified software processing mechanism for sampling windows, digital filtering, zero-point compensation, gain compensation, and analog output mapping.

[0006] Therefore, there is an urgent need for a general-purpose current sensor that can be compatible with AC, DC, AC / DC superposition, and complex non-sinusoidal waveforms, while reducing hardware costs and ensuring measurement accuracy, temperature compensation capability, and versatility of output interface under complex waveforms. Summary of the Invention

[0007] The purpose of this invention is to overcome the defects in the prior art and provide a universal AC / DC current sensor and current measurement method.

[0008] To achieve the above objectives, a first aspect of the present invention provides an AC / DC universal current sensor, comprising: The Hall sampling unit is used to sense the current in the conductor being measured and output a Hall voltage signal; A signal conditioning unit, connected to the Hall sampling unit, is used to bias the Hall voltage signal and output a conditioning signal so that the conditioning signal is adapted to the sampling input range of the microcontroller unit. A temperature detection unit is used to collect temperature information from the Hall sampling unit; A microcontroller unit is connected to the signal conditioning unit and the temperature detection unit; An analog output unit, connected to the microcontroller unit, is used to convert the true RMS value of the current output by the microcontroller unit into a standard analog signal; The microcontroller unit is configured as follows: The conditioning signal is sampled within a preset sampling window to obtain multiple raw sample values; The current zero-point correction value is obtained based on the temperature information. The current zero-point correction value is used to perform bias subtraction on the multiple original sampled values ​​one by one to obtain multiple first sampled values ​​that have compensated for the zero-point drift error. The current gain coefficient is obtained based on the temperature information. The true RMS value of the current in the conductor under test is calculated by combining the current gain coefficient with the multiple first sample values.

[0009] Furthermore, when the microcontroller unit calculates the true RMS value of the current in the conductor under test by combining the current gain coefficient and the plurality of first sampled values, it is specifically configured to: convert the plurality of first sampled values ​​into a plurality of current sampled values ​​using the current gain coefficient, perform a square operation on the plurality of current sampled values ​​respectively, calculate the arithmetic mean of all the square operation results, and then perform a square root operation on the calculated arithmetic mean to obtain the true RMS value of the current; or, perform a square operation on the plurality of first sampled values ​​respectively, calculate the arithmetic mean of all the square operation results, and then perform a square root operation on the calculated arithmetic mean to obtain an initial result of the RMS value, and multiply the initial result of the RMS value by the current gain coefficient to obtain the true RMS value of the current.

[0010] Furthermore, when the current in the conductor under test is a superimposed AC / DC signal, the microcontroller unit, when calculating the true RMS value of the current in the conductor under test by combining the current gain coefficient and the plurality of first sample values, is specifically configured to: convert the plurality of first sample values ​​into a plurality of current sample values ​​using the current gain coefficient; calculate the arithmetic mean of the plurality of current sample values ​​as the DC component; calculate the difference between each current sample value and the DC component to obtain a plurality of AC fluctuation sample values; perform root mean square operation on the plurality of AC fluctuation sample values ​​to obtain the RMS value of the AC component; and calculate the square root of the sum of the square of the DC component and the square of the RMS value of the AC component as the true RMS value of the current in the conductor under test.

[0011] Furthermore, when the current in the conductor under test contains a periodic AC component, the microcontroller unit is also configured to dynamically adjust the preset sampling window in the following manner: calculate the arithmetic mean as the DC component based on the plurality of first sample values, or based on the plurality of current sample values ​​converted from the plurality of first sample values; subtract the corresponding DC component from each of the first sample values ​​or each of the current sample values ​​to obtain a plurality of AC fluctuation sample values ​​fluctuating around a zero reference; record the time interval between two consecutive crosses of the zero reference by the AC fluctuation sample values ​​in the same direction as a single crossover interval; determine the current fundamental period of the current under test based on at least one of the single crossover intervals; and dynamically adjust the time length of the preset sampling window so that the difference between the time length of the preset sampling window and a positive integer multiple of the current fundamental period does not exceed one sampling period.

[0012] Furthermore, when the microcontroller unit records the time interval between two consecutive crossovers of the zero reference in the same direction, it is specifically configured to: filter the AC fluctuation sample value; set a hysteresis threshold; and determine that a crossover of the zero reference in the same direction has occurred when the filtered AC fluctuation sample value changes from below the negative hysteresis threshold to above the positive hysteresis threshold.

[0013] Furthermore, when determining the current fundamental frequency period, the microcontroller unit is specifically configured to: record multiple single-crossing intervals; and determine the average, median, or smoothing result of the multiple single-crossing intervals as the current fundamental frequency period.

[0014] Furthermore, the microcontroller unit is also configured to perform anomaly rollback processing: when preset rollback conditions are met, the preset sampling window is set to a fixed number of points window, a fixed time window, or the previously valid preset sampling window is reused; the preset rollback conditions include at least one of the following: the current in the conductor being measured is pure DC, the amplitude of the AC fluctuation sampling value is lower than a set threshold, no crossover of the zero value reference in the same direction is detected within a preset time, or the current fundamental frequency period is determined to be abnormal.

[0015] Furthermore, it also includes a parameter storage unit for storing a reference temperature, zero-point calibration parameters at the reference temperature, a current conversion factor, a zero-point temperature drift factor, and a gain temperature coefficient. When the microcontroller unit obtains the current zero-point correction value, it is specifically configured to: add the zero-point calibration parameters to the product of the zero-point temperature drift factor and the temperature difference to obtain the current zero-point correction value, where the temperature difference is the difference between the current temperature and the reference temperature. When the microcontroller unit obtains the current gain coefficient, it is specifically configured to: multiply the current conversion factor by a compensation factor to obtain the current gain coefficient, where the compensation factor is a product of the gain temperature coefficient and the temperature difference.

[0016] Furthermore, the microcontroller unit is also configured to perform digital filtering and / or outlier removal processing on the plurality of original sampled values; the digital filtering includes at least one of moving average filtering, median filtering, first-order low-pass filtering and multi-window averaging; the outlier removal processing includes: when it is determined that the absolute value of the difference between a certain original sampled value and an adjacent original sampled value exceeds a preset threshold, replacing the original sampled value with the average value of the adjacent original sampled values.

[0017] A second aspect of the present invention provides a current measurement method, comprising: It senses the current in the conductor being measured and outputs a Hall voltage signal; The Hall voltage signal is biased to obtain a conditioning signal, which is then adapted to the sampling input range. Collect temperature information; The conditioning signal is sampled within a preset sampling window to obtain multiple raw sample values; The current zero-point correction value is obtained based on the temperature information. The current zero-point correction value is used to perform bias subtraction on the multiple original sampled values ​​one by one to obtain multiple first sampled values ​​that have compensated for the zero-point drift error. The current gain coefficient is obtained based on the temperature information, and the true RMS value of the current in the conductor under test is calculated by combining the current gain coefficient with the multiple first sample values. The true RMS value of the current is converted into a standard analog signal.

[0018] The advantages and beneficial effects of this invention are as follows: The AC / DC universal current sensor of this invention senses the measured current through a Hall sampling unit and outputs a Hall voltage signal. After bias processing by a signal conditioning unit, a conditioning signal adapted to the sampling input range of a microcontroller is output. The microcontroller unit samples the conditioning signal within a preset sampling window to obtain the original sampled value. Based on the temperature information collected by the temperature detection unit, the current zero-point correction value is obtained, and the original sampled value is biased and subtracted one by one to obtain a first sampled value that compensates for the zero-point drift error. Then, the true RMS value of the measured current is calculated by combining the current gain coefficient and the first sampled value. Finally, the analog output unit converts the true RMS value of the current into a standard analog signal. Thus, the same sensor can be compatible with measuring AC, DC, and AC / DC superimposed currents. Furthermore, by combining the software-based true RMS value calculation of the microcontroller with the pre-zero-point bias compensation, the measurement accuracy and temperature stability under complex waveforms are improved while reducing hardware costs. The measurement results can be output through various standard analog signals, and both sampling and compensation parameters can be configured by software, facilitating batch calibration and flexible field deployment. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the AC / DC universal current sensor of the present invention; Figure 2 This is a circuit block diagram of the Hall sampling and signal conditioning unit of the present invention; Figure 3 This is a flowchart of the microcontroller sampling and true RMS value calculation process of the present invention; Figure 4 This is a flowchart of the temperature compensation process of the present invention; Figure 5 This is a flowchart of the analog output mapping process of the present invention; Figure 6 This is a flowchart of the current measurement method of the present invention.

[0020] Figure 7 The flowchart for determining the adaptive sampling window of the present invention is shown below. Detailed Implementation

[0021] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and examples. The following examples are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0022] like Figure 1 As shown, the AC / DC universal current sensor of this embodiment includes a Hall sampling unit 11, a signal conditioning unit 12, a microcontroller unit 13, a temperature detection unit 14, an analog output unit 15, and a parameter storage unit 16. Each unit is powered by a power supply unit. Figure 1The power supply unit shown can be a power module built into the sensor, or it can be connected to an external power source through an interface to meet the power supply requirements of different installation environments.

[0023] The conductor under test passes through the sensor detection hole or is close to the Hall detection area. When an alternating current, a direct current, or a superimposed alternating current or direct current flows through the conductor under test, the Hall sampling unit 11 senses the magnetic field generated by the current and outputs a Hall voltage signal related to the measured current. The Hall sampling unit 11 can adopt an open-loop Hall sampling structure, and its output Hall voltage signal can contain a DC component, an AC component, or a superimposed alternating current or direct current component, which can adapt to various measured current waveforms.

[0024] like Figure 2 As shown, the signal conditioning unit 12 includes at least one of a bias voltage generation circuit, a low-pass filter circuit, a limiting protection circuit, and an impedance matching circuit. Since the microcontroller's analog-to-digital converter port can typically only acquire unipolar voltage signals, the signal conditioning unit 12 can superimpose bidirectional current signals onto a preset midpoint bias voltage, ensuring that both forward and reverse currents are within the sampling range allowed by the analog-to-digital converter, and output a conditioning signal that adapts to the sampling input range of the microcontroller unit 13. The low-pass filter circuit suppresses high-frequency interference, the limiting protection circuit prevents overcurrent or overvoltage damage to the microcontroller's sampling port, and the impedance matching circuit ensures the integrity of signal transmission.

[0025] The microcontroller unit 13 can be an industrial-grade microcontroller with an ADC, timer, memory, and communication interface, such as a domestically produced low-cost microcontroller. The microcontroller unit 13 is connected to the signal conditioning unit 12 and the temperature detection unit 14, receiving the conditioning signal output from the signal conditioning unit 12 and the temperature information collected by the temperature detection unit 14. The microcontroller unit 13 periodically samples the conditioning signal according to a preset sampling frequency and performs digital filtering, zero-point subtraction, true RMS calculation, temperature compensation, and output mapping on the sampled data based on the temperature information. The temperature detection unit 14 can be located near the Hall effect device or on the sensor circuit board, used to collect the temperature information from the Hall effect sampling unit 11, providing a basis for temperature compensation calculation. The analog output unit 15 is connected to the microcontroller unit 13 and can use DAC output, voltage buffer output, voltage-to-current conversion output, or PWM filtering output to convert the true RMS value of the current output by the microcontroller unit 13 into a standard analog signal. The parameter storage unit 16 is connected to the microcontroller unit 13. It can use non-volatile memory or utilize the internal Flash of the microcontroller to store calibration parameters, ensuring that the parameters are not lost after power failure.

[0026] During the operation of the current sensor, the microcontroller unit 13 periodically samples the conditioning signal output by the signal conditioning unit 12 according to a preset sampling frequency. Specifically, the microcontroller unit 13 samples the conditioning signal within a preset sampling window to obtain multiple raw sample values. This preset sampling window can be a fixed-point window, a fixed-time window, or an adaptive dynamic window. Subsequently, the microcontroller unit 13 obtains the current zero-point correction value based on the temperature information collected by the temperature detection unit 14, and uses the current zero-point correction value to perform bias subtraction on each of the multiple raw sample values ​​to obtain multiple first sample values ​​that compensate for the zero-point drift error. This bias subtraction step is performed before the calculation of the true RMS value, which can avoid the zero-point offset being carried into the subsequent squaring operation and generating irreversible cross-term errors. Afterwards, the microcontroller unit 13 obtains the current gain coefficient based on the temperature information, and calculates the true RMS value of the current in the measured conductor by combining the current gain coefficient and the multiple first sample values. Finally, the microcontroller unit 13 sends the calculated true RMS value of the current to the analog output unit 15, which converts it into a standard analog signal output. Through the above processing flow, the current sensor realizes a complete processing link for signal sampling, temperature compensation, true RMS value calculation and standard analog output of the measured current, and can be compatible with the measurement of AC, DC and AC-DC superimposed current in the same sensor.

[0027] The AC / DC universal current sensor of this invention senses the measured current through a Hall sampling unit and outputs a Hall voltage signal. After bias processing by a signal conditioning unit, a conditioning signal adapted to the sampling input range of a microcontroller is output. The microcontroller unit samples the conditioning signal within a preset sampling window to obtain the original sampled value. Based on the temperature information collected by the temperature detection unit, the current zero-point correction value is obtained, and the original sampled value is biased and subtracted one by one to obtain a first sampled value that compensates for the zero-point drift error. Then, the true RMS value of the measured current is calculated by combining the current gain coefficient and the first sampled value. Finally, the analog output unit converts the true RMS value of the current into a standard analog signal. Thus, the same sensor can be used to measure AC, DC, and AC / DC superimposed currents. By combining the software-based true RMS value calculation of the microcontroller with the pre-zero bias compensation, the measurement accuracy and temperature stability under complex waveforms are improved while reducing hardware costs. The measurement results can be output through various standard analog signals, and both sampling and compensation parameters can be configured by software, facilitating batch calibration and flexible field deployment.

[0028] To flexibly achieve gain compensation and adapt to different engineering implementation requirements, a preferred embodiment of the present invention is that, when the microcontroller unit 13 calculates the true RMS value of the current in the conductor under test by combining the current gain coefficient and the plurality of first sampled values, it is specifically configured to: convert the plurality of first sampled values ​​into a plurality of current sampled values ​​using the current gain coefficient; perform a square operation on each of the plurality of current sampled values; calculate the arithmetic mean of all the square operation results; and then perform a square root operation on the calculated arithmetic mean to obtain the true RMS value of the current; or, perform a square operation on each of the plurality of first sampled values; calculate the arithmetic mean of all the square operation results; then perform a square root operation on the calculated arithmetic mean to obtain an initial RMS value; and multiply the initial RMS value by the current gain coefficient to obtain the true RMS value of the current.

[0029] Combination Figure 3 The illustrated true RMS calculation process involves microcontroller unit 13 acquiring N sampling points within a sampling window, denoted as ADC1, ADC2, ADC3...ADCN. Microcontroller unit 13 performs bias subtraction on the raw sampled data to be processed based on the current zero-point correction value ADC0 to obtain the first sampled value. , i=1,2,…,N. Microcontroller unit 13 is configured to execute one of the following two equivalent computation paths.

[0030] In the first calculation path, the first sampled value Xi is converted into a current sampled value Ii = K × Xi based on the gain coefficient K. Then, the square, sum, arithmetic mean, and square root operations are performed sequentially to obtain the true RMS value of the current. .

[0031] In the second calculation path, the root mean square operation is first performed on the first sampled value Xi to obtain the initial result of the effective value. Then multiply it by the gain factor K to obtain the true RMS value of the current. .

[0032] Based on the associative law of mathematics, the two calculation paths described above are equivalent in numerical results. The second path moves the multiplication operation of the gain coefficient to after the square root, reducing the number of multiplication instruction calls within the loop body.

[0033] To facilitate the analysis and processing of AC / DC superimposed signals, a preferred embodiment of the present invention is as follows: when the current in the conductor under test is an AC / DC superimposed signal, the microcontroller unit 13, in calculating the true RMS value of the current in the conductor under test by combining the current gain coefficient and the plurality of first sample values, is specifically configured to: convert the plurality of first sample values ​​into a plurality of current sample values ​​using the current gain coefficient; calculate the arithmetic mean of the plurality of current sample values ​​as the DC component; calculate the difference between each current sample value and the DC component to obtain a plurality of AC fluctuation sample values; perform root mean square operation on the plurality of AC fluctuation sample values ​​to obtain the RMS value of the AC component; and calculate the square root of the sum of the square of the DC component and the square of the RMS value of the AC component as the true RMS value of the current in the conductor under test.

[0034] In practice, the microcontroller unit 13 first calculates the average value of the current samples within the sampling window to obtain the DC component. Then, the difference between each current sample value and the DC component is calculated to obtain the AC fluctuation sample value. , and by The effective value of the AC component is obtained. The total true effective value of the AC / DC superimposed signal can be expressed as: This method can distinguish between DC and AC fluctuation components, which is of great value in applications such as inverter output current monitoring and energy storage system charging and discharging current analysis. For typical AC-DC superimposed signals such as PWM waves with DC bias and energy storage system charging and discharging current, this component separation algorithm can accurately give the magnitude of DC bias and the intensity of AC fluctuations, providing richer measurement information for system diagnosis and control.

[0035] To reduce waveform truncation error, a preferred embodiment of the present invention is as follows: Figure 7 As shown, when the current in the conductor being measured contains a periodic AC component, the microcontroller unit 13 is further configured to dynamically adjust the preset sampling window in the following ways: calculate the arithmetic mean as the DC component based on multiple first sampling values, or multiple current sampling values ​​converted from multiple first sampling values; subtract the corresponding DC component from each first sampling value or each current sampling value to obtain multiple AC fluctuation sampling values ​​that fluctuate around a zero reference; record the time interval between two consecutive crossovers of the AC fluctuation sampling value across the zero reference in the same direction as a single crossover interval; determine the current fundamental period of the current being measured based on at least one single crossover interval; and dynamically adjust the time length of the preset sampling window so that the difference between the time length of the preset sampling window and a positive integer multiple of the current fundamental period does not exceed one sampling period, thereby reducing the waveform truncation error when calculating the true RMS value.

[0036] In practice, the DC component is obtained by calculating the arithmetic mean of multiple first sampled values ​​or multiple current sampled values ​​within the sampling window. For pure AC signals, this arithmetic mean is usually close to zero; for periodic signals containing DC bias, this arithmetic mean represents the DC component of the signal. The microcontroller unit 13 subtracts the corresponding DC component from each first sampled value or each current sampled value, causing the resulting AC fluctuation sampled value to fluctuate around a zero-value reference, and then performs a same-direction crossing detection based on the zero-value reference. Thus, regardless of whether the measured current is a pure AC signal or an AC / DC superimposed signal, a unified period detection logic can be used; for signals with a DC bias greater than the AC amplitude, the period detection will not fail because the original signal has not crossed the absolute zero point.

[0037] The microcontroller unit 13 detects the time interval between two consecutive crossovers of the zero-value reference in the same direction of the AC fluctuation sample value, and takes this as a single-crossing interval, which corresponds to one fundamental frequency cycle of the measured current. Based on at least one single-crossing interval, the microcontroller unit 13 determines the current fundamental frequency cycle and dynamically adjusts the sampling window duration accordingly. The goal of dynamic adjustment is to ensure that the difference between the sampling window duration and a positive integer multiple (1, 2, 3, etc.) of the current fundamental frequency cycle does not exceed one sampling cycle. In discrete sampling systems, since the sampling frequency is fixed, it cannot be guaranteed that the sampling window duration is numerically equal to a positive integer multiple of the fundamental frequency cycle; therefore, a rounding error not exceeding one sampling cycle is allowed. In specific implementations, the number of sampling window points can be determined according to... The sampling frequency (fs) and the estimated fundamental period (T) are defined, where M is a positive integer, fs is the sampling frequency, and T is the estimated fundamental period. This tolerance design balances engineering feasibility and measurement accuracy. Because the sampling window closely follows the variation of the fundamental period, it effectively avoids the waveform truncation problem caused by a fixed window at different frequencies, thereby reducing the truncation error when calculating the true RMS value.

[0038] To suppress noise interference with the detection of crossover in the same direction, a preferred embodiment of the present invention is that, when the microcontroller unit 13 records the time interval between two consecutive crossovers of the zero-value reference in the same direction, it is specifically configured to: filter the AC fluctuation sample value; set a hysteresis threshold; and determine that a crossover of the zero-value reference in the same direction has occurred when the filtered AC fluctuation sample value changes from below the negative hysteresis threshold to above the positive hysteresis threshold.

[0039] Filtering of AC fluctuation samples can employ methods such as moving average filtering and low-pass filtering. The aim is to suppress high-frequency components generated by PWM switching ripple or waveform distortion, preventing multiple false crossovers due to noise jitter within a single fundamental cycle. A detection interval is formed on both sides of the zero reference value. Only when the filtered AC fluctuation sample value continuously rises from below the negative hysteresis threshold and crosses the positive hysteresis threshold is it considered a valid crossover in the same direction. If the signal only jitters slightly near the zero reference value and does not completely cross the detection interval, it will not be falsely judged as a valid crossover. This mechanism effectively suppresses noise interference and improves the reliability of cycle detection.

[0040] To improve the stability of fundamental frequency period determination, a preferred embodiment of the present invention is that, when determining the current fundamental frequency period, the microcontroller unit 13 is specifically configured to: record multiple single-crossing intervals; and determine the average, median, or smoothing result of the multiple single-crossing intervals as the current fundamental frequency period.

[0041] In practice, the microcontroller unit 13 records the time interval for each crossing of the reference zero point in the same direction within multiple consecutive fundamental frequency periods, obtaining multiple single-crossing intervals. Statistical analysis is performed on these single-crossing intervals using methods such as the average method (dividing the sum of all intervals by the number of intervals), the median method (taking the middle value after sorting), or smoothing (such as exponential smoothing or moving average) to obtain smoothed time intervals. The smoothed result is determined as the current fundamental frequency period and used to adjust the subsequent sampling window length. This multi-period statistical processing method effectively filters out random errors from single measurements, making the period estimate more stable and avoiding drastic fluctuations in the sampling window length due to single abnormal measurements, thereby improving the overall stability of the true RMS value calculation.

[0042] In order to maintain basic sampling function and provide an abnormal rollback mechanism under specific operating conditions, a preferred embodiment of the present invention is that the microcontroller unit 13 is further configured to perform abnormal rollback processing: when the preset rollback conditions are met, the preset sampling window is set to a fixed number of points window, a fixed time window, or the previously valid preset sampling window is reused; the preset rollback conditions include at least one of the following: the current in the conductor being measured is pure DC, the amplitude of the AC fluctuation sampling value is lower than a set threshold, no crossover of the zero value reference in the same direction is detected within a preset time, or the current fundamental frequency period is determined to be abnormal.

[0043] In practice, the microcontroller unit 13 continuously monitors whether preset backoff conditions are met during the same-direction crossing detection process. When the current in the conductor being measured is pure DC, the amplitude of the AC fluctuation sampling value is close to zero, and there is no periodic same-direction crossing event in the signal, making period estimation impossible. In this case, it returns to the fixed window mode. When the amplitude of the AC fluctuation sampling value is lower than the set threshold, the signal-to-noise ratio is too low, and the reliability of the same-direction crossing detection is insufficient, also triggering backoff. When no valid same-direction crossing zero-value reference event is detected within a preset time, the period detection is judged to have timed out, and it returns to the fixed window. Backoff is also triggered when the period estimation result shows obvious abnormalities. The sampling window after backoff can be a preset fixed-point window, a fixed-time window, or the adaptive window parameters that were previously confirmed to be valid, so that the sensor can maintain basic measurement functions under different operating conditions.

[0044] To compensate for zero-point drift and gain changes caused by temperature variations, a preferred embodiment of the present invention further includes a parameter storage unit 16 for storing a reference temperature, zero-point calibration parameters and current conversion coefficients at the reference temperature, a zero-point temperature drift coefficient, and a gain temperature coefficient. When the microcontroller unit 13 obtains the current zero-point correction value, it is specifically configured to: add the zero-point calibration parameters to the product of the zero-point temperature drift coefficient and the temperature difference to obtain the current zero-point correction value, where the temperature difference is the difference between the current temperature and the reference temperature. When the microcontroller unit 13 obtains the current gain coefficient, it is specifically configured to: multiply the current conversion coefficient by a compensation factor to obtain the current gain coefficient, where the compensation factor is a product of the gain temperature coefficient and the temperature difference.

[0045] Combination Figure 4 The temperature compensation process shown involves the microcontroller unit 13 periodically reading the temperature value T output by the temperature detection unit 14 and performing zero-point compensation and gain compensation based on the temperature compensation parameters pre-stored in the parameter storage unit 16. Zero-point compensation employs a linear compensation method. ,in This is the zero-point correction value at the current temperature. Let A be the zero-point value at the reference temperature, and let A be the zero-point temperature drift coefficient. The reference temperature is used. Gain compensation employs a linear compensation method. ,in The gain coefficient at the current temperature. B is the gain coefficient at the reference temperature, and B is the gain temperature coefficient. In another implementation, the temperature compensation parameters can also be implemented using a lookup table: during the production calibration phase, the zero-point value and gain coefficient are recorded at multiple temperature points to form a temperature compensation table stored in the parameter storage unit 16. During runtime, the microcontroller unit 13 looks up the table based on the current temperature value and performs linear interpolation to obtain the current zero-point correction value and gain coefficient. It should be noted that zero-point compensation must be completed before the root mean square (RMS) calculation because the zero-point offset will be introduced into the square operation, generating cross-term errors that cannot be eliminated by simple linear correction after the RMS calculation; while gain compensation can be performed before or after the RMS calculation, and the two are mathematically equivalent, allowing for flexible selection based on actual engineering needs.

[0046] To suppress sampling noise and transient interference and improve the quality of sampling data, a preferred embodiment of the present invention is that the microcontroller unit 13 is further configured to perform digital filtering and / or outlier removal processing on the plurality of original sampled values; the digital filtering includes at least one of moving average filtering, median filtering, first-order low-pass filtering and multi-window averaging; the outlier removal processing includes: when it is determined that the absolute value of the difference between a certain original sampled value and an adjacent original sampled value exceeds a preset threshold, the original sampled value is replaced with the average value of the adjacent original sampled values.

[0047] Digital filtering can be applied to the original sampled values ​​before the true RMS value calculation to remove random noise and impulse interference at the source. Moving average filtering suppresses random noise by averaging the most recent sampled values; median filtering effectively suppresses impulse interference by taking the median value of a set of sampled values; first-order low-pass filtering implements software filtering through a recursive method, allowing adjustment of the filter coefficients to balance response speed and filtering effect; multi-window averaging further improves measurement stability by averaging the calculation results of multiple sampling windows. For outlier removal, when the difference between a sampled point and its adjacent sampled points exceeds a preset threshold, the microcontroller unit 13 determines that the sampled point is an outlier and replaces it with the average of adjacent valid sampled values, thereby reducing the impact of transient interference on subsequent zero-point correction and true RMS value calculation results.

[0048] During the production calibration phase, the calibration process is as follows: First, input zero current and record the average value of the ADC sampling at this time as the zero-point calibration value ADC0; then input the rated current and calculate the current conversion factor K based on the ratio of the actual current value to the sampled value; record the zero-point value and gain coefficient under different temperature conditions (such as low temperature, normal temperature, and high temperature), calculate the zero-point temperature drift coefficient A and the gain temperature coefficient B through linear fitting, or directly generate temperature compensation lookup table data; write all the above parameters into the parameter storage unit 16. During product operation, the microcontroller unit 13 reads the calibration parameters from the parameter storage unit 16 each time it is powered on, and calls these parameters in real time during operation to correct them, ensuring the long-term stability of measurement accuracy.

[0049] In one implementation, to balance the computational load of the low-cost microcontroller unit with the accuracy of true RMS calculation, the sampling frequency of the analog-to-digital converter is preferably configured to be between 1kHz and 20kHz. The most typical frequency range for the measured current in industrial settings is 50 / 60Hz (power frequency), while the output frequency of frequency converters is generally in the hundreds of Hz, and the PWM carrier frequency is in the several kHz range. According to the Nyquist sampling theorem, the sampling frequency must be greater than twice the highest frequency of the signal. Therefore, setting the sampling frequency in the range of 1kHz to 20kHz covers the measurement requirements of typical application scenarios and allows for smooth operation on a low-cost microcontroller. Simultaneously, the number of sampling points N within the sampling window is preferably configured as an integer power of 2, such as 64, 128, 256, or 512. A significant advantage of using an integer power of 2 is that when the microcontroller unit 13 performs the step of "calculating the arithmetic mean of the squared result (i.e., dividing the sum by N)" in the root mean square operation, it can directly call the underlying binary right shift instruction to replace the extremely resource-intensive hardware division instruction. For example, when N=64, shifting 6 bits to the right achieves the division by 64 operation. This parameter configuration reduces the CPU overhead of true RMS calculation, improves the system's real-time response speed, and provides a feasible basis for replacing dedicated hardware true RMS chips with extremely low-cost microcontrollers. It should be noted that the above-mentioned scheme where the number of sampling points N is preferably an integer power of 2 applies to fixed-point windows; when using an adaptive dynamic window, the number of sampling points is calculated based on the current fundamental period, sampling frequency, and the selected period multiple, and is not limited by an integer power of 2.

[0050] Combination Figure 5 The output mapping process shown is that the microcontroller unit 13 maps the compensated true RMS current value Irms to the full-scale current set by the user. Mapped to a standard analog output. When the output is a 0-10V voltage signal, the output voltage can be expressed as... When the output is a 4-20mA current signal, the output current can be expressed as: When the output is 0-5V, 1-5V, 2-10V, or 0-20mA, the corresponding linear mapping method can be used. The output range can be configured via DIP switches, communication interface, programmed parameters, or internally stored parameters to meet the connection requirements of different downstream devices.

[0051] In summary, the beneficial effects of the embodiments of the present invention include: (1) AC / DC universal measurement: The same sensor can measure AC current, DC current and AC / DC superimposed current, which expands the application range; (2) Strong adaptability to complex waveforms: The effective value calculation of non-sinusoidal waves, PWM waves, pulsating current and inverter output current is performed by the microcontroller true effective value calculation processing, avoiding the error caused by estimating only the average value or peak value; (3) Reduce reliance on dedicated RMS chips: True RMS measurement is achieved by using microcontroller calculation, which facilitates domestic substitution and cost control; (4) Flexible compensation: Zero drift compensation and gain compensation are achieved through temperature detection and parameter storage, which improves the measurement stability of the open-loop Hall effect scheme; (5) Universal output: The measurement results can be mapped to various standard voltage or current outputs, which is convenient for connecting to different downstream devices; (6) Facilitates production calibration and function expansion: The sampling window, filtering parameters, temperature compensation parameters and output range parameters can all be adjusted through software configuration or stored parameters; (7) Adaptive window adjustment: For the measured current containing periodic AC components, the AC fluctuation sampling value fluctuating around the zero reference is obtained by subtracting the DC component, and the fundamental period is determined based on multi-cycle smoothing. The sampling window is dynamically adjusted so that the difference between it and the positive integer multiple of the fundamental period does not exceed one sampling period, which reduces the waveform truncation error caused by non-integer period sampling and improves the accuracy of true RMS measurement under pure AC and AC-DC superimposed signals. (8) Strong robustness: It automatically reverts to a fixed window when the signal is pure DC, low amplitude, or when the effective zero crossing point cannot be detected, which improves the working stability of the sensor under various working conditions.

[0052] like Figure 6 As shown, based on the same inventive concept as the aforementioned AC / DC universal current sensor, this embodiment of the invention also provides a current measurement method, which can be applied to the aforementioned sensor. The current measurement method includes the following steps: S601, senses the current in the conductor being measured and outputs a Hall voltage signal.

[0053] S602. The Hall voltage signal is biased to obtain a conditioning signal, which is then adapted to the sampling input range.

[0054] S603, Collect temperature information.

[0055] S604. Sample the conditioning signal within a preset sampling window to obtain multiple raw sample values.

[0056] S605. Obtain the current zero-point correction value based on the temperature information, and use the current zero-point correction value to perform bias subtraction on the multiple original sampled values ​​one by one to obtain multiple first sampled values ​​that have compensated for the zero-point drift error.

[0057] S606. Obtain the current gain coefficient based on the temperature information, and calculate the true RMS value of the current in the conductor under test by combining the current gain coefficient and the multiple first sample values.

[0058] S607. Convert the true RMS value of the current into a standard analog signal.

[0059] In one implementation, calculating the true RMS value of the current in the conductor under test by combining the current gain coefficient and the plurality of first sampled values ​​includes: converting the plurality of first sampled values ​​into a plurality of current sampled values ​​using the current gain coefficient; squaring each of the plurality of current sampled values; calculating the arithmetic mean of all the squaring results; and then taking the square root of the calculated arithmetic mean to obtain the true RMS value of the current; or, squaring each of the plurality of first sampled values; calculating the arithmetic mean of all the squaring results; then taking the square root of the calculated arithmetic mean to obtain an initial RMS value; and multiplying the initial RMS value by the current gain coefficient to obtain the true RMS value of the current.

[0060] In one embodiment, when the current in the conductor under test is a superimposed AC / DC signal, the true RMS value of the current in the conductor under test is calculated by combining the current gain coefficient and the plurality of first sample values, including: converting the plurality of first sample values ​​into a plurality of current sample values ​​using the current gain coefficient; calculating the arithmetic mean of the plurality of current sample values ​​as the DC component; calculating the difference between each current sample value and the DC component to obtain a plurality of AC fluctuation sample values; performing root mean square operation on the plurality of AC fluctuation sample values ​​to obtain the RMS value of the AC component; and calculating the square root of the sum of the square of the DC component and the square of the RMS value of the AC component as the true RMS value of the current in the conductor under test.

[0061] In one implementation, when the current in the conductor under test contains a periodic AC component, the preset sampling window is dynamically adjusted as follows: an arithmetic mean is calculated as the DC component based on multiple first sampling values, or multiple current sampling values ​​converted from multiple first sampling values; the corresponding DC component is subtracted from each first sampling value or each current sampling value to obtain multiple AC fluctuation sampling values ​​fluctuating around a zero reference; the time interval between two consecutive crossovers of the AC fluctuation sampling value across the zero reference in the same direction is recorded as a single crossover interval; the current fundamental period of the current under test is determined based on at least one single crossover interval; and the time length of the preset sampling window is dynamically adjusted so that the difference between the time length of the preset sampling window and a positive integer multiple of the current fundamental period does not exceed one sampling period.

[0062] In one implementation, recording the time interval between two consecutive crossovers of the zero-value reference in the same direction includes: filtering the AC fluctuation sample value; setting a hysteresis threshold; and determining that a crossover of the zero-value reference in the same direction has occurred when the filtered AC fluctuation sample value changes from below the negative hysteresis threshold to above the positive hysteresis threshold.

[0063] In one embodiment, determining the current fundamental frequency period includes: recording a plurality of single-crossing intervals; and determining the average, median, or smoothed result of the plurality of single-crossing intervals as the current fundamental frequency period.

[0064] In one embodiment, the method further includes: when a preset backoff condition is met, setting the preset sampling window to a fixed number of points window, a fixed time window, or using the previously valid preset sampling window; the preset backoff condition includes at least one of the following: the current in the conductor being measured is pure DC, the amplitude of the AC fluctuation sampling value is lower than a set threshold, no crossover of the zero value reference in the same direction is detected within a preset time, or the current fundamental frequency period is determined to be abnormal.

[0065] In one embodiment, obtaining the current zero-point correction value includes: adding the zero-point calibration parameter to the product of the zero-point temperature drift coefficient and the temperature difference to obtain the current zero-point correction value, wherein the temperature difference is the difference between the current temperature value and the reference temperature; obtaining the current gain coefficient includes: multiplying the current conversion coefficient by a compensation factor to obtain the current gain coefficient, wherein the compensation factor is a product of the gain temperature coefficient and the temperature difference.

[0066] In one embodiment, after obtaining the plurality of original sampled values ​​and before performing bias subtraction, the method further includes: performing digital filtering and / or outlier removal processing on the plurality of original sampled values; the digital filtering includes at least one of moving average filtering, median filtering, first-order low-pass filtering, and multi-window averaging; the outlier removal processing includes: when it is determined that the absolute value of the difference between a certain original sampled value and an adjacent original sampled value exceeds a preset threshold, replacing the original sampled value with the average value of the adjacent original sampled values.

[0067] The current measurement method in the above embodiments is based on the same technical concept as the AC / DC universal current sensor described above. Its implementation principle and technical effect are similar, and will not be repeated here.

[0068] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0069] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A universal AC / DC current sensor, characterized in that, include: The Hall sampling unit is used to sense the current in the conductor being measured and output a Hall voltage signal; A signal conditioning unit, connected to the Hall sampling unit, is used to bias the Hall voltage signal and output a conditioning signal so that the conditioning signal is adapted to the sampling input range of the microcontroller unit. A temperature detection unit is used to collect temperature information from the Hall sampling unit; A microcontroller unit is connected to the signal conditioning unit and the temperature detection unit; An analog output unit, connected to the microcontroller unit, is used to convert the true RMS value of the current output by the microcontroller unit into a standard analog signal; The microcontroller unit is configured as follows: The conditioning signal is sampled within a preset sampling window to obtain multiple raw sample values; The current zero-point correction value is obtained based on the temperature information. The current zero-point correction value is used to perform bias subtraction on the multiple original sampled values ​​one by one to obtain multiple first sampled values ​​that have compensated for the zero-point drift error. The current gain coefficient is obtained based on the temperature information. The true RMS value of the current in the conductor under test is calculated by combining the current gain coefficient with the multiple first sample values.

2. The sensor according to claim 1, characterized in that, When the microcontroller unit calculates the true RMS value of the current in the measured conductor by combining the current gain coefficient and the plurality of first sampled values, it is specifically configured as follows: The multiple first sample values ​​are converted into multiple current sample values ​​using the current gain coefficient. The multiple current sample values ​​are squared respectively. The arithmetic mean of all the squared results is calculated. The square root of the calculated arithmetic mean is then taken to obtain the true effective value of the current. or, The multiple first sampled values ​​are squared respectively, the arithmetic mean of all the squared results is calculated, and the square root of the calculated arithmetic mean is then performed to obtain the initial result of the effective value. The initial result of the effective value is multiplied by the current gain coefficient to obtain the true effective value of the current.

3. The sensor according to claim 1, characterized in that, When the current in the conductor under test is a superimposed AC / DC signal, the microcontroller unit is specifically configured to calculate the true RMS value of the current in the conductor under test by combining the current gain coefficient and the multiple first sampled values: The plurality of first sample values ​​are converted into a plurality of current sample values ​​using the current gain coefficient; Calculate the arithmetic mean of the multiple current sample values ​​as the DC component; Calculate the difference between each current sample value and the DC component to obtain multiple AC fluctuation sample values; The root mean square operation is performed on the multiple AC fluctuation sample values ​​to obtain the effective value of the AC component; The square root of the sum of the square of the DC component and the square of the effective value of the AC component is taken as the true effective value of the current in the conductor being measured.

4. The sensor according to claim 1, characterized in that, When the current in the conductor under test contains a periodic alternating component, the microcontroller unit is also configured to dynamically adjust the preset sampling window in the following manner: The arithmetic mean of the multiple first sampled values, or the multiple current sampled values ​​converted from the multiple first sampled values, is calculated as the DC component. Subtract the corresponding DC component from each of the first sampled values ​​or each of the current sampled values ​​to obtain a plurality of AC fluctuation sampled values ​​that fluctuate around the zero reference. The time interval between two consecutive crossovers of the zero-value reference by the AC fluctuation sampling value in the same direction is recorded as the single crossover interval; Based on at least one single-cross interval, the current fundamental period of the measured current is determined, and the time length of the preset sampling window is dynamically adjusted so that the difference between the time length of the preset sampling window and a positive integer multiple of the current fundamental period does not exceed one sampling period.

5. The sensor according to claim 4, characterized in that, When recording the time interval between two consecutive crossings of the zero reference in the same direction, the microcontroller unit is specifically configured as follows: The AC fluctuation sampled values ​​are filtered; Set the hysteresis threshold; When the filtered AC fluctuation sample value changes from below the negative hysteresis threshold to above the positive hysteresis threshold, it is determined that a crossover of the zero value reference has occurred in the same direction.

6. The sensor according to claim 4, characterized in that, When determining the current fundamental frequency period, the microcontroller unit is specifically configured as follows: Record multiple single-cross intervals; The average, median, or smoothed result of multiple single-cross intervals is determined as the current fundamental frequency period.

7. The sensor according to claim 4, characterized in that, The microcontroller unit is also configured to perform exception rollback processing: When the preset rollback conditions are met, the preset sampling window is set to a fixed number of points window, a fixed time window, or the previously valid preset sampling window is reused. The preset backoff conditions include at least one of the following: the current in the conductor being measured is pure DC, the amplitude of the AC fluctuation sampling value is lower than a set threshold, no crossover of the zero value reference is detected in the same direction within a preset time, or the current fundamental frequency period is determined to be abnormal.

8. The sensor according to claim 1, characterized in that, It also includes a parameter storage unit for storing a reference temperature, zero-point calibration parameters at the reference temperature, current conversion factor, zero-point temperature drift factor, and gain temperature factor. When the microcontroller unit obtains the current zero-point correction value, it is specifically configured to: add the product of the zero-point temperature drift coefficient and the temperature difference to the zero-point calibration parameter to obtain the current zero-point correction value, wherein the temperature difference is the difference between the current temperature value and the reference temperature; When the microcontroller unit obtains the current gain coefficient, it is specifically configured to: multiply the current conversion factor by a compensation factor to obtain the current gain coefficient, wherein the compensation factor is a product of the gain temperature coefficient and the temperature difference.

9. The sensor according to claim 1, characterized in that, The microcontroller unit is also configured to perform digital filtering and / or outlier removal processing on the plurality of raw sampled values; The digital filtering includes at least one of moving average filtering, median filtering, first-order low-pass filtering, and multi-window averaging. The outlier removal process includes: when the absolute value of the difference between a certain original sample value and an adjacent original sample value exceeds a preset threshold, the original sample value is replaced with the average value of the adjacent original sample values.

10. A current measurement method, characterized in that, include: It senses the current in the conductor being measured and outputs a Hall voltage signal; The Hall voltage signal is biased to obtain a conditioning signal, which is then adapted to the sampling input range. Collect temperature information; The conditioning signal is sampled within a preset sampling window to obtain multiple raw sample values; The current zero-point correction value is obtained based on the temperature information. The current zero-point correction value is used to perform bias subtraction on the multiple original sampled values ​​one by one to obtain multiple first sampled values ​​that have compensated for the zero-point drift error. The current gain coefficient is obtained based on the temperature information, and the true RMS value of the current in the conductor under test is calculated by combining the current gain coefficient with the multiple first sample values. The true RMS value of the current is converted into a standard analog signal.

11. The current measurement method according to claim 10, characterized in that, When the current in the conductor under test contains a periodic alternating component, the preset sampling window is also dynamically adjusted in the following way: The arithmetic mean of the multiple first sampled values, or the multiple current sampled values ​​converted from the multiple first sampled values, is calculated as the DC component. Subtract the corresponding DC component from each of the first sampled values ​​or each of the current sampled values ​​to obtain a plurality of AC fluctuation sampled values ​​that fluctuate around the zero reference. The time interval between two consecutive crossovers of the zero-value reference by the AC fluctuation sampling value in the same direction is recorded as the single crossover interval; The current fundamental period of the measured current is determined based on at least one single-cross interval, and the time length of the preset sampling window is dynamically adjusted so that the difference between the time length of the preset sampling window and a positive integer multiple of the current fundamental period does not exceed one sampling period.

12. The current measurement method according to claim 11, characterized in that, Record the time interval between two consecutive crossings of the zero reference in the same direction, including: The AC fluctuation sample values ​​are filtered and a hysteresis threshold is set; When the filtered AC fluctuation sample value changes from below the negative hysteresis threshold to above the positive hysteresis threshold, it is determined that a crossover of the zero value reference has occurred in the same direction.

13. The current measurement method according to claim 11, characterized in that, Also includes: When the preset rollback conditions are met, the preset sampling window is set to a fixed number of points window, a fixed time window, or the previously valid preset sampling window is reused. The preset backoff conditions include at least one of the following: the current in the conductor being measured is pure DC, the amplitude of the AC fluctuation sampling value is lower than a set threshold, no crossover of the zero value reference is detected in the same direction within a preset time, or the current fundamental frequency period is determined to be abnormal.

Images