Electric parameter sampling rate adjustment system for intelligent acquisition terminal
By introducing an electrical parameter measurement module, a transient disturbance measurement module, a measurement noise suppression module, and a sampling clock adaptive module into the intelligent acquisition terminal, and dynamically adjusting the sampling rate, the measurement stability and accuracy problems of the intelligent acquisition terminal under power grid nonlinearity and transient phenomena are solved, and high-precision electrical parameter measurement is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- YANGZHOU WANTAI ELECTRIC TECH CO LTD
- Filing Date
- 2026-05-22
- Publication Date
- 2026-06-19
AI Technical Summary
Existing intelligent data acquisition terminals suffer from data redundancy or distortion due to fixed sampling rates in electrical parameter measurements. This is especially true under nonlinear and transient conditions in the power grid, which affects measurement stability and accuracy.
The system employs an electrical parameter measurement module, a transient disturbance measurement module, a measurement noise suppression module, and a sampling clock adaptive module. By processing the first-order time difference quotient, using Kalman filtering, and employing a nonlinear compression mapping function, the sampling rate is dynamically adjusted to ensure automatic adjustment of the sampling rate when the power grid electrical parameters change, thereby improving measurement accuracy and stability.
It effectively eliminates interference caused by changes in sampling interval, improves transient capture accuracy, avoids measurement distortion, achieves a balance between measurement accuracy and data efficiency, and adapts to changes in different power grid environments.
Smart Images
Figure CN122238698A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of electrical variable measurement technology. In particular, it relates to an electrical parameter sampling rate adjustment system for intelligent data acquisition terminals. Background Technology
[0002] With the continuous advancement of smart grid interconnection and the construction of new power systems, a large number of distributed energy sources and nonlinear loads have been connected to the modern power grid. As a result, electrical parameters in the power grid, such as voltage and current, are no longer ideal steady-state waveforms, but are filled with high-frequency harmonics, transient voltage spikes and drops, inrush currents and other phenomena.
[0003] To accurately capture the aforementioned nonlinear and transient phenomena, intelligent acquisition terminals must perform analog-to-digital conversion (ADC) sampling on electrical parameters. Traditional electrical parameter measurement methods often employ a fixed sampling rate. However, the majority of electrical parameter measurements acquired at high sampling rates correspond to periodic steady-state signals during stable grid operation, leading to data redundancy; conversely, low sampling rates fail to accurately capture the high-frequency characteristics of transient electrical parameters, resulting in measurement distortion.
[0004] In existing dynamic sampling measurement methods, when the sampling time step changes dynamically, traditional electrical parameter feature extraction methods with a fixed sampling rate will generate spurious abrupt changes in measurement values due to the alteration of the sampling interval, severely affecting the stability and accuracy of electrical parameter measurements. Therefore, there is a need in this field for an electrical parameter sampling rate adjustment system for intelligent acquisition terminals to address the aforementioned problems affecting the stability and accuracy of electrical parameter measurements. Summary of the Invention
[0005] To address the technical problem of the above sampling methods affecting the stability and accuracy of electrical parameter measurements, the present invention provides the following solution.
[0006] An electrical parameter sampling rate adjustment system for intelligent data acquisition terminals includes: Electrical parameter measurement module: synchronously acquires the instantaneous phase voltage value output by the voltage transformer and the instantaneous phase current value output by the current transformer in the intelligent acquisition terminal, and calculates the instantaneous apparent power at the current moment; Transient disturbance measurement module: Performs first-order time difference quotient processing on instantaneous apparent power to obtain the differential energy at the current moment, calculates the mean of differential energy within a preset time window, obtains the local background mean at the current moment, and outputs transient change characteristic value based on the differential energy at the current moment and the local background mean. Measurement noise suppression module: Input transient change feature values as observation values into Kalman filter, combine with the posterior state estimate of the previous time step to obtain the current time step innovation value; obtain the innovation value sliding buffer queue of preset length and calculate its temporal variance, combine with the maximum local background mean within the preset history window to construct adaptive measurement noise covariance, dynamically adjust Kalman gain, and output posterior state estimate; Sampling clock adaptive module: Based on the absolute deviation of the posterior state estimate from the normal electrical parameter measurement baseline, the target sampling rate for the next moment is calculated through a nonlinear compression mapping function, and electrical parameters are collected based on the target sampling rate; The target sampling rate is between the system's allowed base sampling rate and maximum sampling rate, and it increases monotonically with the increase of absolute deviation, so as to automatically increase the sampling rate when there are transient disturbances in the power grid parameters and automatically decrease the sampling rate when the parameters are stable.
[0007] Preferably, the first-order time difference quotient processing of the instantaneous apparent power includes: calculating the first difference between the instantaneous apparent power at the current moment and the instantaneous apparent power at the previous moment; calculating the time difference between the current moment and the previous moment; and calculating the square of the ratio of the first difference to the time difference to obtain the differential energy at the current moment.
[0008] Preferably, the step of outputting transient change feature values based on the differential energy and local background mean at the current moment includes: taking the moment before the current moment as the endpoint, obtaining the differential energy of all sampling points within the time window and calculating the mean of the differential energy, using the mean as the local background mean at the current moment; calculating the ratio of the differential energy at the current moment to the local background mean to obtain the transient change feature value at the current moment.
[0009] Preferably, constructing the adaptive measurement noise covariance includes: obtaining the posterior state estimate of the previous time step, calculating the difference between the transient change feature value of the current time step and the posterior state estimate of the previous time step, and obtaining the innovation value of the current time step; maintaining a sliding buffer queue of innovation values of a preset length, and calculating the temporal variance of the sliding buffer queue of innovation values; obtaining the local background mean of the previous time step, selecting the maximum local background mean within a preset historical window with the previous time step as the endpoint, calculating the ratio of the local background mean of the previous time step to the maximum local background mean, and multiplying the square of the ratio by the temporal variance to obtain the adaptive measurement noise covariance.
[0010] Preferably, the dynamic adjustment of the Kalman gain includes: obtaining the posterior covariance of the previous time step; calculating the sum of the posterior covariance of the previous time step and the preset process noise covariance to obtain the predicted covariance; calculating the sum of the predicted covariance and the adaptive measurement noise covariance, and using it as the first sum; and calculating the ratio of the posterior covariance of the previous time step to the first sum to obtain the Kalman gain.
[0011] Preferably, the output posterior state estimation includes: calculating the product of the Kalman gain and the adaptive measurement noise covariance, and using it as the first product; calculating the sum of the first product and the posterior state estimation at the previous time step to obtain the posterior state estimation at the current time step.
[0012] Preferably, the absolute deviation of the posterior state estimate relative to the normal electrical parameter measurement baseline includes: calculating the absolute difference between the posterior state estimate at the current moment and 1 to obtain the absolute deviation.
[0013] Preferably, the step of calculating the target sampling rate at the next moment using the nonlinear compression mapping function includes: calculating the square root of the sum of the absolute deviation and 1; calculating the ratio of 1 to the square root; calculating the difference between 1 and the ratio; multiplying the difference by a preset maximum sampling rate bandwidth to obtain a second product; and calculating the sum of the second product and a preset base sampling rate to obtain the target sampling rate.
[0014] Preferably, the maximum sampling rate bandwidth is 24.6 kHz, and the basic sampling rate is 1 kHz.
[0015] Preferably, the instantaneous apparent power is obtained by multiplying the instantaneous phase voltage value and the instantaneous phase current value.
[0016] The present invention has the following effects: 1. This invention, by setting up a transient disturbance measurement module, performs first-order time difference quotient processing on the instantaneous apparent power, eliminating the interference of time step distortion caused by dynamic frequency conversion sampling on electrical parameter measurement, so that the differential energy truly reflects the power change rate per unit time, effectively avoiding false measurement value abrupt changes caused by changes in sampling interval, and significantly improving the transient capture accuracy of electrical parameter measurement under frequency conversion sampling conditions.
[0017] 2. This invention constructs a transient change characteristic value that follows the real-time changes in the power grid environment by dividing the differential energy at the current moment by the average value of the differential energy within a sliding local time window. This enables the system to adaptively adjust the judgment benchmark regardless of whether the power grid is under heavy load, light load, or strong harmonic interference. This effectively avoids the measurement distortion problem of minor faults being submerged under heavy load and normal load switching being misjudged as faults under light load.
[0018] 3. This invention sets up a measurement noise suppression module and constructs an adaptive measurement noise covariance using the maximum local background mean within the historical window. When the system is in an unloaded or lightly loaded state, the adaptive measurement noise covariance is significantly reduced, effectively suppressing the pseudo-variance oscillations caused by the underlying quantization white noise. When a real transient change occurs, the adaptive measurement noise covariance increases accordingly, the Kalman gain decreases, and the filter quickly tracks the changes in the real electrical parameters, outputting a stable posterior state estimate.
[0019] 4. This invention employs a sampling clock adaptive module and a nonlinear compression mapping mechanism to calculate the target sampling rate for the next moment, ensuring that the sampling rate is always limited between the base sampling rate and the maximum sampling rate, and monotonically increases with the increase of electrical parameter deviation. When transient disturbances occur in the electrical parameters, the sampling rate is automatically increased to capture high-frequency measurement characteristics; when the electrical parameters are stable, the sampling rate is automatically decreased to reduce redundant measurement data, thus achieving a balance between measurement accuracy and data efficiency. Attached Figure Description
[0020] Figure 1 This is a schematic diagram of the connection of each module in the electrical parameter sampling rate adjustment system for an intelligent acquisition terminal according to an embodiment of the present invention. Detailed Implementation
[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.
[0022] Reference Figure 1 The proposed solution presents an electrical parameter sampling rate adjustment system for intelligent data acquisition terminals. Deployed within the intelligent data acquisition terminal, its core function is to perform high-precision measurement of electrical parameters such as voltage, current, and power in the power grid. This system includes an electrical parameter measurement module, a transient disturbance measurement module, a measurement noise suppression module, and a sampling clock adaptive module. These modules are connected in series to form a data flow channel.
[0023] The following is a detailed description of the specific implementation methods of each module.
[0024] Electrical Parameter Measurement Module: This module synchronously acquires the instantaneous phase voltage value output by the voltage transformer and the instantaneous phase current value output by the current transformer in the intelligent acquisition terminal, and calculates the instantaneous apparent power at the current moment. The instantaneous apparent power is obtained by multiplying the instantaneous phase voltage value and the instantaneous phase current value. The specific formula is as follows: In the formula, Indicates the current time Instantaneous apparent power; Indicates the current moment The instantaneous phase voltage values collected; Indicates the current moment The instantaneous phase current value collected.
[0025] The electrical parameter measurement module packages the calculated instantaneous apparent power and its corresponding timestamp and transmits it to the transient disturbance measurement module.
[0026] Transient disturbance measurement module: Performs first-order time difference quotient processing on the instantaneous apparent power to obtain the differential energy at the current moment, calculates the mean of the differential energy within a preset time window, obtains the local background mean at the current moment, and outputs transient change characteristic value based on the differential energy at the current moment and the local background mean.
[0027] The method for calculating the differential energy at the current moment includes: calculating the first difference between the instantaneous apparent power at the current moment and the instantaneous apparent power at the previous moment; calculating the time difference between the current moment and the previous moment; and calculating the square of the ratio of the first difference to the time difference to obtain the differential energy at the current moment. This can be expressed by the following formula: In the formula, Indicates the current time The differential energy; Indicates the current time Instantaneous apparent power; Indicates the previous moment Instantaneous apparent power; Indicates the current time timestamp; Indicates the previous moment Timestamp.
[0028] For situations where the sampling interval of electrical parameters is not fixed, using the time difference as the denominator to quantify the differential energy can effectively avoid interference caused by the jump in the sampling time step.
[0029] During adaptive sampling rate operation, when the power grid electrical parameters are highly stable, the time interval between adjacent sampling points... It may approach 0, resulting in differential energy. Calculate overflow or severe oscillation. Therefore, the system sets a minimum time interval threshold. If the adjacent sampling time interval If the time interval is less than the minimum threshold, then a forced timeout will be taken. Simultaneously, the differential energy calculation results are smoothed using a first-order low-pass filter: Initial time This mechanism effectively suppresses numerical instability when the sampling interval approaches zero.
[0030] The method for calculating the transient change characteristic value includes: setting the time window length to 5ms (the time window length is a hyperparameter, which can be selected according to the fundamental frequency period of the power grid; for example, the value should cover 1 / 4 to 1 / 2 of the fundamental frequency period of the power grid to capture the leading edge characteristics of transient disturbances, while avoiding excessively long windows that suppress high-frequency responses); taking the time before the current time as the endpoint, obtaining the differential energy of all sampling points within this time window and calculating the mean of the differential energy; using this mean as the local background mean at the current time; calculating the ratio of the differential energy at the current time to the local background mean to obtain the transient change characteristic value at the current time. The specific formula is as follows: In the formula, Indicates the current time transient mutation characteristic values; Indicates the current time The differential energy; Indicates a previous moment The total number of sampling points within the time window ending at the endpoint; Indicates the time within the time window The differential energy; It represents the mean of all differential energies within the time window, that is, the mean of the local background at the current moment.
[0031] This scheme uses the differential energy at the current moment to divide by the mean of the differential energy within a sliding local time window. This mean represents the average activity level of the power grid over a very short period of time prior to the current moment, i.e., the current environmental background. Through this division operation, the absolute differential energy, which was originally strongly coupled with the power grid load state, is mapped to a dimensionless relative ratio, i.e., a transient change characteristic value.
[0032] When the transient change characteristic value is close to 1, it indicates that the change in the current electrical parameters is comparable to the recent background fluctuation level, which is a normal environmental fluctuation. When the transient change characteristic value is significantly greater than 1, it indicates that the current change level far exceeds the background level, which is a transient event that requires attention. By using its own mean as the denominator, the system can adaptively adjust the judgment benchmark regardless of whether the power grid is under heavy load, light load, or strong harmonic interference. This effectively avoids the problem of minor faults being overwhelmed under heavy load and normal load switching under light load being misjudged as faults, providing stable, reliable, and environmentally decoupled input characteristics for subsequent Kalman filtering.
[0033] Measurement noise suppression module: The transient change feature value is used as the observation value and input into the Kalman filter. Combined with the posterior state estimate of the previous time step, the current time step innovation value is obtained. The innovation value sliding buffer queue of preset length is obtained and its temporal variance is calculated. Combined with the maximum local background mean within the preset historical window, the adaptive measurement noise covariance is constructed. The Kalman gain is dynamically adjusted and the posterior state estimate is output.
[0034] At the first sampling moment after system power-on or reset, the Kalman filter needs to be initialized. The observations of the Kalman filter are transient change eigenvalues, and the estimated state variable is the filtered and smoothed transient change eigenvalue, i.e., the posterior state estimate. Let the initial time be... posterior state estimation initial value This indicates that during the system startup phase, the default grid electrical parameters are at the steady-state baseline (i.e., the differential energy at the current moment equals the local background mean). Initial posterior covariance The initial posterior covariance (PCO) represents the uncertainty of the initial value of the posterior state estimate: a smaller value indicates lower uncertainty (i.e., more certainty about the initial value of the posterior state estimate), while a larger value indicates higher uncertainty (i.e., more reliance on observations for correction). In the absence of prior information, the initial posterior covariance should not be too small to avoid slow filter convergence, nor should it be too large to cause severe oscillations in the state estimate during the startup phase.
[0035] The method for calculating the adaptive measurement noise covariance includes: obtaining the posterior state estimate of the previous time step; calculating the difference between the transient change feature value of the current time step and the posterior state estimate of the previous time step to obtain the innovation value of the current time step; maintaining a sliding buffer queue of innovation values of a preset length and calculating the temporal variance of the sliding buffer queue; obtaining the local background mean of the previous time step; selecting the maximum local background mean within a preset historical window with the previous time step as the endpoint; calculating the ratio of the local background mean of the previous time step to the maximum local background mean; and multiplying the square of this ratio by the temporal variance to obtain the adaptive measurement noise covariance. The specific formula is as follows: In the formula, Indicates the current time Adaptive measurement noise covariance; ,in This represents the temporal variance of the moving buffer queue of information values. This represents the total number of new values in the new value sliding buffer queue, in this embodiment. That is, the latest 10 innovation values are cached in the sliding cache queue of innovation values. Indicates the time in the moving buffer queue of the new information value The new interest value; Indicates the previous moment The local background mean; In this embodiment, the length of the history window is indicated. This length covers typical transient processes in the power grid (such as voltage spikes and drops lasting 0.5 to 1 second), and is particularly relevant in scenarios with frequent load fluctuations. This can be reduced to 0.5 seconds to improve adaptive speed; Indicates the time within the historical window The local background mean.
[0036] The Kalman gain adjustment method includes: obtaining the posterior covariance of the previous time step; calculating the sum of the posterior covariance of the previous time step and the preset process noise covariance to obtain the predicted covariance; calculating the sum of the predicted covariance and the adaptive measurement noise covariance, and using this as the first sum; and calculating the ratio of the posterior covariance of the previous time step to the first sum to obtain the Kalman gain. The specific formula is as follows: In the formula, Indicates the current time The corresponding Kalman gain; This represents the forecast covariance at the current moment. , Indicates the previous moment The posterior covariance, This represents the preset process noise covariance, used to describe the real random fluctuations that may occur in the power grid parameters within adjacent sampling intervals. In this embodiment... ; This represents the adaptive measurement noise covariance.
[0037] The value of is related to the rated voltage level of the power grid, the degree of load fluctuation, and the sampling rate. In this embodiment, Dynamically set according to the following principles: when the system's current sampling rate... hour, ;when hour, This principle is based on the premise that at high sampling rates, the electrical parameters between adjacent sampling points change less, and the process noise covariance should be correspondingly reduced. If no prior information is available, the default value is used. In environments with highly fluctuating power grids (such as industrial microgrids), it is possible to... Increased to 0.01 to enhance filter tracking capability.
[0038] The method for calculating the posterior state estimate includes: calculating the product of the Kalman gain and the adaptive measurement noise covariance, and using this as the first product; then calculating the sum of the first product and the posterior state estimate from the previous time step to obtain the posterior state estimate for the current time step. This is specifically expressed by the following formula: In the formula, Indicates the current time Posterior state estimation; This represents the posterior state estimate from the previous time step. Indicates the current time The corresponding Kalman gain; Indicates the current time The new interest value.
[0039] When the system is in an unloaded or lightly loaded state The value is very small. Approaching 0, making This significantly reduces the variance, thereby effectively suppressing the oscillations caused by the underlying quantization white noise; when a true transient change occurs, Increase The Kalman gain is increased accordingly, while the Kalman gain is decreased, enabling the filter to quickly track changes in the actual electrical parameters and output a stable posterior state estimate.
[0040] The measurement noise suppression module outputs the obtained posterior state estimate for the current time step to the sampling clock adaptive module, and calculates the posterior covariance for the current time step based on the Kalman gain and prediction covariance at the current time step. Specifically, it calculates the difference between 1 and the Kalman gain at the current time step, and multiplies this difference with the prediction covariance at the current time step as the posterior covariance for the current time step, which is used for inference calculation at the next time step. This is expressed by the following formula: In the formula, Indicates the current time The posterior covariance; Indicates the current time The corresponding Kalman gain; Indicates the current time The corresponding prediction covariance. Sampling clock adaptive module: Based on the absolute deviation of the posterior state estimate from the normal electrical parameter measurement baseline, it calculates the target sampling rate for the next time step using a nonlinear compression mapping function, and collects electrical parameters based on the target sampling rate.
[0041] The method for calculating the absolute deviation includes: calculating the absolute difference between the posterior state estimate at the current time and 1, and obtaining the absolute deviation.
[0042] Posterior state estimation As a dimensionless quantity, its physical meaning is the transient change characteristic value after Kalman filtering smoothing. When the power grid is in steady-state operation and there are no transient disturbances, the current power change rate is approximately equal to the recent average change rate, that is, the current differential energy is approximately equal to the local background mean. Therefore, the transient change characteristic value is approximately equal to 1. Thus, 1 is used as the steady-state baseline for calculating the absolute deviation.
[0043] The target sampling rate is calculated as follows: Calculate the square root of the sum of the absolute deviation and 1; calculate the ratio of 1 to the square root; calculate the difference between 1 and the ratio; multiply this difference by the preset maximum sampling rate bandwidth to obtain a second product; and calculate the sum of the second product and the preset base sampling rate to obtain the target sampling rate. The specific formula is as follows: In the formula, Indicates the next moment The target sampling rate; This indicates the minimum allowed baseline sleep sampling rate of the system, i.e., the baseline sampling rate. In this embodiment... ; This indicates the maximum allowed sampling rate bandwidth in the system, as described in this embodiment. ; Indicates the absolute deviation. , Indicates the current time The posterior state estimate. When hour, ;when hour, Approaching Target sampling rate as a function of absolute deviation The increase is monotonically increasing, and the growth rate gradually slows down to avoid drastic changes in the sampling rate. The selection is based on the fact that the highest sampling rate of this intelligent acquisition terminal is 25.6kHz, and the basic sampling rate is... Therefore, the maximum bandwidth that can be increased is 25.6−1=24.6kHz.
[0044] Electrical parameters for the next time step are acquired based on the target sampling rate. The target sampling rate for the next time step is calculated using a nonlinear compression mapping function based on the absolute deviation of the posterior state estimate from the normal baseline. This ensures that the sampling rate is always limited between the base sampling rate and the maximum sampling rate, and monotonically increases with the increase of the absolute deviation. The sampling rate is automatically increased to capture high-frequency measurement characteristics when transient disturbances occur in the electrical parameters, and automatically decreased to reduce redundant measurement data when the electrical parameters are stable, achieving a balance between measurement accuracy and data efficiency. All modules of the system use a unified time base, with the timestamp of the sampling moment as the data alignment identifier. When the sampling rate changes, the sampling clock adaptive module switches the sampling interval at the next time step without interpolation or resampling to maintain the phase continuity of the original measurement data. The sliding buffer queues of each module continue to store data in chronological order after the sampling rate switch, without clearing or compensation, and use first-in-first-out queue management to ensure timing consistency. The system also includes other components well-known to those skilled in the art, such as communication buses and communication interfaces, whose settings and functions are known in the art and will not be described further here.
[0045] After the system powers on, it first enters the initialization phase, which lasts for 50 milliseconds. During this phase, electrical parameters are collected using the base sampling rate to populate the following data structures: the time window for differential energy (all sampling points within 5ms), and the moving buffer queue for innovation values (length...). ), the historical window (length) of the local background mean (seconds). During the initialization phase, no transient disturbance detection or sampling rate adjustment is performed. Once all queues are full, the system automatically switches to normal operation mode. If insufficient cache queue length is detected during the initialization phase, the mean or variance of the current data is used for the corresponding item.
[0046] It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept, and these all fall within the scope of protection of this invention. Therefore, the scope of protection of this patent should be determined by the appended claims.
Claims
1. An electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal, characterized in that, include: Electrical parameter measurement module: synchronously acquires instantaneous phase voltage and instantaneous phase current values from the intelligent acquisition terminal, and calculates the instantaneous apparent power at the current moment; Transient disturbance measurement module: Performs first-order time difference quotient processing on instantaneous apparent power to obtain the differential energy at the current moment, calculates the mean of differential energy within a preset time window, obtains the local background mean at the current moment, and outputs transient change characteristic value based on the differential energy at the current moment and the local background mean. Measurement noise suppression module: Input transient change feature values as observation values into Kalman filter, combine with the posterior state estimate of the previous time step to obtain the current time step innovation value; obtain the innovation value sliding buffer queue of preset length and calculate its temporal variance, combine with the maximum local background mean within the preset history window to construct adaptive measurement noise covariance, dynamically adjust Kalman gain, and output posterior state estimate; Sampling clock adaptive module: Based on the absolute deviation of the posterior state estimate from the normal electrical parameter measurement baseline, the target sampling rate for the next moment is calculated through a nonlinear compression mapping function, and electrical parameters are collected based on the target sampling rate; The target sampling rate is between the system's allowed base sampling rate and maximum sampling rate, and it increases monotonically with the increase of absolute deviation, so as to automatically increase the sampling rate when there are transient disturbances in the power grid parameters and automatically decrease the sampling rate when the parameters are stable.
2. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The first-order time difference quotient processing of the instantaneous apparent power includes: calculating the first difference between the instantaneous apparent power at the current moment and the instantaneous apparent power at the previous moment; calculating the time difference between the current moment and the previous moment; and calculating the square of the ratio of the first difference to the time difference to obtain the differential energy at the current moment.
3. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The step of outputting transient change feature value based on the differential energy and local background mean at the current moment includes: taking the previous moment as the endpoint, obtaining the differential energy of all sampling points within the time window and calculating the mean of the differential energy, and using this mean as the local background mean at the current moment; calculating the ratio of the differential energy at the current moment to the local background mean to obtain the transient change feature value at the current moment.
4. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The construction of the adaptive measurement noise covariance includes: obtaining the posterior state estimate of the previous time step, calculating the difference between the transient change feature value of the current time step and the posterior state estimate of the previous time step to obtain the innovation value of the current time step; maintaining a sliding buffer queue of innovation values of a preset length and calculating the temporal variance of the sliding buffer queue of innovation values; obtaining the local background mean of the previous time step, selecting the maximum local background mean within a preset historical window with the previous time step as the endpoint, calculating the ratio of the local background mean of the previous time step to the maximum local background mean, and multiplying the square of the ratio by the temporal variance to obtain the adaptive measurement noise covariance.
5. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The dynamic adjustment of the Kalman gain includes: obtaining the posterior covariance of the previous time step; calculating the sum of the posterior covariance of the previous time step and the preset process noise covariance to obtain the predicted covariance; calculating the sum of the predicted covariance and the adaptive measurement noise covariance, and using it as the first sum; and calculating the ratio of the posterior covariance of the previous time step to the first sum to obtain the Kalman gain.
6. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The output posterior state estimation includes: calculating the product of the Kalman gain and the adaptive measurement noise covariance, and using it as the first product; calculating the sum of the first product and the posterior state estimation at the previous time step to obtain the posterior state estimation at the current time step.
7. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The absolute deviation of the posterior state estimate relative to the normal electrical parameter measurement baseline includes: calculating the absolute difference between the posterior state estimate at the current moment and 1 to obtain the absolute deviation.
8. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The calculation of the target sampling rate at the next moment using the nonlinear compression mapping function includes: calculating the square root of the sum of the absolute deviation and 1; calculating the ratio of 1 to the square root, calculating the difference between 1 and the ratio, multiplying the difference by the preset maximum sampling rate bandwidth to obtain a second product; and calculating the sum of the second product and the preset base sampling rate to obtain the target sampling rate.
9. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 8, characterized in that, The maximum sampling rate bandwidth is 24.6 kHz, and the basic sampling rate is 1 kHz.
10. The electrical parameter sampling rate adjustment system for an intelligent data acquisition terminal according to claim 1, characterized in that, The instantaneous apparent power is obtained by multiplying the instantaneous phase voltage value and the instantaneous phase current value.