High-precision electric power measurement method based on intelligent electric meter

Abstract

The invention discloses a high-precision electric power measurement method based on an intelligent electric meter, and relates to the technical field of electric power measurement. According to the invention, a plurality of high-precision voltage and current sensors with temperature compensation are uniformly arranged in a circuit, a low-noise front-end conditioning and multi-channel high-resolution ADC synchronous sampling technology is adopted, and a digital phase-locked loop is introduced to realize accurate correction of a sampling clock. And performing multi-stage weighted fusion, time domain instantaneous power calculation and FFT frequency domain decomposition compensation on the acquired data, and finally accurately calculating the real power in a power grid period. According to the method, measurement errors caused by temperature excursion, phase deviation and harmonic interference in the prior art are effectively solved, the measurement precision and stability of the system are remarkably improved, and the method is suitable for high-precision power monitoring in a complex power grid environment.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power measurement, and in particular to a high-precision electric power measurement method based on a smart meter. Background Art

[0002] Existing electric power measurement technology usually relies on single-path sampling, often using a single voltage or current sensor with a simple front-end conditioning circuit for signal amplification and filtering. However, under conditions such as temperature changes, component aging, and local interference, large drift and noise problems are prone to occur, resulting in low measurement accuracy. At the same time, most traditional sampling systems fail to achieve strict synchronous sampling of voltage and current signals, lack the correction of digital phase-locked loop technology, and are prone to instantaneous power calculation errors due to clock deviation and phase drift. More prominently, in nonlinear load or switching power supply environments, harmonic component interference is serious, and existing methods often only use time domain integral calculations, lack effective separation and compensation of harmonics, and cannot restore the true power component. In addition, the lack of multi-sensor redundant sampling and data intelligent fusion mechanism makes the overall measurement data less reliable when local signals are abnormal or drift for a long time, and cannot meet the requirements of high-precision, long-term stable monitoring. Summary of the invention

[0003] In order to overcome the defects of the above-mentioned prior art, the present invention provides the following technical solutions: a high-precision electric power measurement method based on a smart meter, comprising the following steps: S1, arranging multiple voltage sensors and current sensors in the circuit, and obtaining multiple independent measurement data of the same circuit segment through redundant sampling; S2, using a multi-channel analog-to-digital converter to synchronously sample the analog signal output by the sensor; S3, cross-checking, weighted fusion, FFT frequency domain analysis and error compensation are performed on the collected multi-channel data, and finally the accurate electric power is calculated; S4, calculating the instantaneous power at each moment; S5, performing FFT transformation on the time domain data within a power grid cycle to separate the fundamental wave and each harmonic component; S6, using a pre-calibrated compensation coefficient to correct each frequency component to eliminate the measurement error introduced by the harmonic; S7, integrating the direct calculation result of the time domain with the compensation result of the frequency domain to obtain a more accurate power value; S8, outputting the corrected electric power data to a display device, a remote monitoring system or a memory for real-time monitoring and subsequent analysis.

[0004] Preferably, temperature sensors are provided at the voltage sensor and the current sensor arranged in S1, and the data collected by the voltage sensor and the current sensor are compensated by a temperature compensation circuit to minimize the errors obtained.

[0005] Preferably, a digital phase-locked loop is used in S2 to synchronously sample with the grid fundamental wave, so that all sampling points are aligned within the grid cycle, thereby eliminating errors caused by clock offset and phase drift.

[0006] Preferably, a multi-stage signal processing strategy is adopted in S3 to correct the measurement error from two aspects: the time domain part uses the redundancy correction of multi-sensor data; the frequency domain part uses FFT decomposition to compensate for harmonic interference.

[0007] Preferably, multiple voltage sensors and current sensors are evenly distributed along the measured line or both ends of the load; each sensor front-end circuit is configured with a low-noise amplifier, a bandpass filter and a temperature detection element to achieve preliminary analog signal conditioning and temperature compensation.

[0008] Preferably, the multi-channel analog-to-digital converter in S2 has a synchronous sampling function, wherein the multi-channel analog-to-digital converter is 16 bits or 24 bits to ensure sampling accuracy and synchronization.

[0009] Compared with the prior art, the present invention has the following beneficial effects: (1) The present invention adopts a multi-channel redundant sampling scheme, and evenly arranges multiple high-precision voltage and current sensors at both ends of the measured circuit. Each sensor uses a low-drift component and is equipped with a temperature compensation circuit, which effectively eliminates the error caused by temperature change or aging of a single sensor. The signal conditioning quality is guaranteed by front-end low-noise amplification and bandpass filtering, and then multi-channel ADC synchronous sampling is used to obtain the data of each sensor at the same time, ensuring the accurate correspondence between the data. After weighted fusion, statistical outlier elimination and multi-level cross-checking, the final electric power measurement result has extremely high accuracy, which significantly overcomes the defects of low measurement accuracy and large error in traditional single-point sampling; (2) The present invention introduces FFT frequency domain decomposition technology to solve the harmonic interference problem caused by nonlinear loads and switching power supplies, separates the collected voltage and current time domain signals into fundamental and harmonic components, and corrects each frequency component according to the pre-calibrated compensation coefficient. By establishing a harmonic compensation model, the fusion of the instantaneous product in the time domain and the correction result in the frequency domain is realized, the influence of harmonics on power calculation is effectively eliminated, and the real power level can still be accurately reflected under complex working conditions with severe harmonics, which greatly improves the applicability and accuracy of the system; (3) The present invention adopts multi-point arranged sensors in the measurement system, obtains multiple independent data through synchronous sampling, and then uses statistical methods such as weighted average, median filtering and abnormal data removal to comprehensively fuse the data of each channel. This multi-level data fusion technology can effectively reduce the errors caused by local interference, sensor drift and component aging, and ensure that the fusion result is closer to the true value. When the system detects that a certain data channel is abnormal, it automatically reduces its weight or even removes the data to ensure the robustness and long-term stability of the overall measurement result; (4) The present invention adopts digital phase-locked loop (PLL) technology to strictly synchronize the ADC sampling clock of each sensor with the grid fundamental wave. Through the real-time adjustment of the phase detector, loop filter and digital control oscillator, it is ensured that the voltage and current signals are sampled at the same phase point, effectively avoiding the instantaneous power calculation error introduced by clock deviation and phase drift. This dynamic correction mechanism not only adapts to the slight fluctuations of the grid frequency and phase, but also automatically corrects the deviation during the sampling process, ensuring the accuracy of the calculation results within each grid cycle, greatly improving the real-time response capability and precision of the measurement system; (5) The present invention constructs a full-cycle, multi-angle closed-loop correction mechanism by combining the power calculation methods in the time domain and the frequency domain. First, the instantaneous power is calculated using the synchronous sampling data, and then the data within a complete grid cycle is transformed by FFT. After the fundamental wave and harmonics are separated, the errors of each component are corrected, and finally the results of the two are combined to obtain the final power value. This closed-loop compensation technology combines the advantages of time-frequency analysis, not only corrects the errors in the instantaneous product calculation, but also overcomes the problem of insufficient accuracy of the traditional single calculation method under harmonic interference, making the entire measurement process more robust and stable. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 It is the overall flow chart of the present invention. DETAILED DESCRIPTION

[0011] The following is combined with Figure 1 The technical solution of the present invention is further illustrated by specific implementation methods.

[0012] The present invention provides a high-precision electric power measurement method based on a smart meter, comprising the following steps: S1, multiple voltage sensors and current sensors are arranged in the circuit, and multiple independent measurement data of the same circuit segment are obtained through redundant sampling; each sensor uses high-precision, low-drift components and is equipped with a dedicated temperature compensation circuit to ensure that the error is minimized when the temperature environment changes. S2, the analog signal output by the sensor is synchronously sampled using a high-resolution multi-channel analog-to-digital converter (ADC); S3, the collected multi-channel data is cross-checked, weighted fused, FFT frequency domain analysis and error compensation are performed, and the accurate electric power is finally calculated; S4, the instantaneous power at each moment is calculated; S5, the time domain data within a power grid cycle is FFT transformed to separate the fundamental wave and each harmonic component; S6, each frequency component is corrected using a pre-calibrated compensation coefficient to eliminate the measurement error introduced by the harmonic; S7, the time domain direct calculation result is integrated with the frequency domain compensation result to obtain a more accurate power value; S8, the corrected electric power data is output to a display device, a remote monitoring system or a memory for real-time monitoring and subsequent analysis.

[0013] The voltage sensors and current sensors arranged in S1 are equipped with temperature sensors. The data collected by the voltage sensors and current sensors are compensated by the temperature compensation circuit to minimize the errors. In S2, a digital phase-locked loop is used to synchronize sampling with the grid fundamental wave, so that all sampling points are aligned within the grid cycle, thereby eliminating errors caused by clock offset and phase drift. In S3, a multi-level signal processing strategy is used to correct measurement errors in both the time domain and the frequency domain: the time domain part uses redundant correction of multi-sensor data; the frequency domain part uses FFT decomposition to compensate for harmonic interference.

[0014] Multiple voltage sensors and current sensors are evenly distributed along the measured line or at both ends of the load; each sensor front-end circuit is equipped with a low-noise amplifier, a bandpass filter (mainly for power frequency signals) and a temperature detection element to achieve preliminary analog signal conditioning and temperature compensation. The output of each sensor is sampled simultaneously using a multi-channel high-resolution ADC to ensure that the values ​​of all sensors are obtained at the same time. The weighted average method is used to fuse the data of each sensor, and statistical methods (such as median filtering or eliminating extreme values) are used to detect and exclude abnormal data to ensure the reliability of the fusion result. Through cross-comparison of multi-point data, the single-point error caused by local interference or component aging can be significantly reduced, making the fused data closer to the true value. When a sensor undergoes a sudden change or long-term drift, the system can automatically identify and reduce its weight, or even eliminate it, to ensure the robustness of the overall system.

[0015] In AC power systems, accurate calculation of instantaneous power requires that voltage and current signals must be sampled strictly synchronously. If there is a phase deviation, the calculated instantaneous product will produce a large error. The digital phase-locked loop mainly consists of three parts: a phase detector (detects the phase difference between the sampling signal and the reference fundamental wave), a loop filter (filters the phase difference signal and stabilizes the control signal), and a digitally controlled oscillator (adjusts the sampling clock according to the control signal to lock it to the grid fundamental wave); Workflow: The system has a built-in reference fundamental wave (usually obtained from a highly calibrated sensor or provided through a dedicated reference circuit); the digital PLL compares the phase difference between each sensor signal and the fundamental wave through the phase detector; the loop filter smoothes the phase error and sends the adjustment signal to the digitally controlled oscillator; the digitally controlled oscillator adjusts the sampling clock to ensure that all ADC sampling points are strictly synchronized with the grid fundamental wave. Real-time dynamic correction: The digital PLL can respond to small fluctuations in grid frequency in real time (such as small changes in 50Hz or 60Hz grids), so that the sampling moment is always locked on a stable fundamental wave phase, thereby avoiding power calculation errors caused by sampling moment deviations. Correction effect: ensure that the voltage and current signals are sampled at the same time, so that the calculation of instantaneous power p(t)=v(t)⋅i(t) is based on the correct phase relationship, greatly reducing the error caused by phase misalignment and eliminating phase deviation; it can dynamically track small changes in grid frequency and phase, ensuring that the error in long-term measurement is always kept at an extremely low level and adapting to grid fluctuations.

[0016] Nonlinear loads, switching power supplies, etc. often introduce a large number of harmonics, making the voltage and current signals no longer pure sine waves, thus affecting the calculation of instantaneous power. Using FFT for frequency domain decomposition, the fundamental wave and each harmonic can be separated, and then each harmonic component can be quantitatively compensated. Achievement purpose: Accurately identify the amplitude and phase of the fundamental wave and each harmonic; establish a compensation model based on the sensor response and system characteristics, and quantitatively correct the harmonic interference to restore the real power component. Specifically, FFT frequency domain decomposition (fast Fourier transform (FFT) is performed on the synchronized voltage and current time domain data to decompose the signal into different frequency components), harmonic compensation model establishment, compensation calculation, full-cycle reconstruction and time-frequency joint compensation (the compensation algorithm not only corrects the error in the frequency domain, but also integrates it with the instantaneous product result in the time domain to form a full-cycle, multi-angle closed-loop correction process to ensure that the power measurement in each power grid cycle reaches the theoretical accuracy limit).

[0017] The multi-channel analog-to-digital converter in S2 has a synchronous sampling function, wherein the multi-channel analog-to-digital converter is 16-bit or 24-bit to ensure sampling accuracy and synchronization.

[0018] At each moment, each sensor collects voltage and current signals respectively. After front-end conditioning, digital data is obtained by synchronous sampling using a multi-channel ADC. The data of each channel is weighted averaged according to the weight to obtain the fused voltage and current values. The PLL module detects the phase difference between the reference fundamental wave and the signal of each channel, adjusts the ADC sampling clock, and ensures that all data are sampled at the same phase point; after synchronization, it ensures that the sampled data meets the strict timing requirements and eliminates the instantaneous power calculation error caused by clock deviation. For each moment, the instantaneous power is calculated, and the data within a power grid cycle is integrated or averaged to obtain the preliminary average power. The sum within a complete power grid cycle is FFT (fast Fourier transform (FFT) is performed on the synchronized voltage and current time domain data); for each frequency component (including the fundamental wave and harmonics), the corrected power contribution is calculated using a predetermined compensation coefficient, the corrected values ​​of each component are accumulated, and the time domain and frequency domain calculation results are fused to obtain the final corrected average power value.

Claims

1. A high-precision electric power measurement method based on a smart meter, characterized in that: The following steps are involved: S1. Arrange multiple voltage sensors and current sensors in the circuit to obtain multiple independent measurement data of the same circuit segment through redundant sampling; S2, using a multi-channel analog-to-digital converter to synchronously sample the analog signal output by the sensor; S3, cross-check, weighted fusion, FFT frequency domain analysis and error compensation are performed on the collected multi-channel data, and finally the accurate electric power is calculated; S4, calculate the instantaneous power at each moment; S5, performing FFT transformation on the time domain data within a power grid cycle to separate the fundamental wave and each harmonic component; S6. Correct each frequency component using a pre-calibrated compensation coefficient to eliminate the measurement error introduced by harmonics; S7, fusing the direct calculation result in the time domain with the compensation result in the frequency domain to obtain a more accurate power value; S8. Output the corrected electric power data to a display device, a remote monitoring system or a memory for real-time monitoring and subsequent analysis.

2. The high-precision electric power measurement method based on a smart meter according to claim 1 is characterized in that: The voltage sensor and the current sensor arranged in S1 are both provided with temperature sensors, and the data collected by the voltage sensor and the current sensor are compensated by the temperature compensation circuit to minimize the errors obtained.

3. The high-precision electric power measurement method based on a smart meter according to claim 2 is characterized in that: In S2, a digital phase-locked loop is used to synchronize sampling with the grid fundamental wave, so that all sampling points are aligned within the grid cycle, thereby eliminating errors caused by clock offset and phase drift.

4. The high-precision electric power measurement method based on a smart meter according to claim 3 is characterized in that: S3 adopts a multi-level signal processing strategy to correct measurement errors in both time and frequency domains. The time domain part uses the redundancy correction of multi-sensor data; the frequency domain part uses FFT decomposition to compensate for harmonic interference.

5. The high-precision electric power measurement method based on a smart meter according to claim 4 is characterized in that: Multiple voltage sensors and current sensors are evenly distributed along the measured line or at both ends of the load; each sensor front-end circuit is equipped with a low-noise amplifier, a bandpass filter and a temperature detection element to achieve preliminary analog signal conditioning and temperature compensation.

6. The high-precision electric power measurement method based on a smart meter according to claim 5 is characterized in that: The multi-channel analog-to-digital converter in S2 has a synchronous sampling function, wherein the multi-channel analog-to-digital converter is 16-bit or 24-bit to ensure sampling accuracy and synchronization.

Images