Electric field detection method and system based on micro electric field sensor

Abstract

The present application relates to the technical field of intelligent sensor, in particular to an electric field detection method and system based on a micro electric field sensor. The present application obtains the voltage vector collected by three orthogonal sensitive units in the electric field sensor at each moment, and then obtains the voltage direction change coefficient at each moment based on the voltage vector at adjacent moments and the spatial geometric relationship of the voltage vector in a preset historical period. Meanwhile, all posture parameters of the electric field sensor at each moment are collected to determine the sensing posture change coefficient. Then, based on the change similarity characteristics of the voltage direction change coefficient and the sensing posture change coefficient, the sensing posture offset discrimination coefficient is obtained and the electric field detection result is output. The present application constructs the comprehensive discrimination coefficient of the voltage vector geometric characteristics and the posture change intensity, accurately decouples the sensor offset and the electric field fluctuation, and improves the accuracy of electric field detection in complex environments.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensor technology, and specifically to an electric field detection method and system based on a micro-sized electric field sensor. Background Technology

[0002] In scenarios such as smart grid equipment condition monitoring and industrial electrostatic discharge (ESD) protection, accurate measurement of electric field strength is crucial. Micro-electromechanical systems (MEMS) based miniature intelligent electric field sensors are widely used for electric field detection due to their small size, low power consumption, high frequency response, and ease of integration; their working principle is to measure the projected component of the electric field along the sensor's sensitive axis. Current technology directly converts the sensor's output voltage into an electric field strength value based on a pre-stored calibration curve.

[0003] However, the sensor output voltage depends not only on the strength of the electric field being measured, but also on the relative orientation of the sensitive axis and the actual electric field direction. When the sensor experiences a slight offset (such as loose installation or vibration), the sensitive axis deviates from its original direction, and the output voltage changes accordingly, resulting in a mismatch between the measured output and the actual electric field state. At the same time, existing methods cannot identify whether this mismatch originates from a change in the electric field being measured or from a sensor attitude offset, which will seriously affect the accuracy and reliability of electric field detection. Summary of the Invention

[0004] To address the low accuracy of existing electric field detection methods based on micro-miniature electric field sensors, this invention aims to provide an electric field detection method and system based on micro-miniature electric field sensors. The specific technical solution adopted is as follows: An electric field detection method based on micro-miniature electric field sensors, the method comprising: The voltage vector of the electric field at each moment during the electric field detection process is obtained. The voltage vector is composed of the single-axis voltages collected and calculated by three orthogonal sensing elements in the electric field sensor. All attitude parameters of the electric field sensor at each moment are collected. Based on the spatial geometric relationship between voltage vectors of adjacent time points at each time point, and the spatial geometric relationship of voltage vectors within a preset historical period at each time point, the voltage direction change coefficient at each time point is obtained; the sensing attitude change coefficient is determined based on all attitude parameters at each time point. Based on the similarity between the voltage direction change coefficient and the sensor attitude change coefficient during electric field detection, the sensor attitude deviation discrimination coefficient is obtained; the electric field detection result is output based on the sensor attitude deviation discrimination coefficient.

[0005] Furthermore, the attitude parameters include at least pitch angle, roll angle, and yaw angle.

[0006] Furthermore, the method for obtaining the voltage direction change coefficient includes: At each time point, the two nearest adjacent time points are extracted to construct an instantaneous analysis period. Based on the consistent direction of the voltage vectors between two adjacent time points within the instantaneous analysis period, the instantaneous voltage deviation coefficient is obtained. A preset historical time period is constructed with each moment as the endpoint. Based on the consistent direction of all voltage vectors within the preset historical time period, the voltage deviation coefficient at each moment is obtained. By combining the instantaneous voltage deviation coefficient and the continuous voltage deviation coefficient, the voltage direction change coefficient at each moment is obtained.

[0007] Furthermore, the method for obtaining the instantaneous voltage deviation coefficient includes: Within the instantaneous analysis period at each moment, the unit vector of the voltage vector at each moment is determined. The rotation vector is determined based on the cross product of the unit vectors at two adjacent moments. The instantaneous voltage deviation coefficient is determined based on the difference in magnitude and the dot product between the rotation vectors at each two adjacent moments.

[0008] Furthermore, the method for obtaining the voltage sustained deviation coefficient includes: Within a preset historical time period at each moment, the voltage vectors at all moments are summed. The magnitude of the cross product of the voltage vectors at all moments within the preset historical time period and the sum of the vectors is used to determine the voltage persistence deviation coefficient.

[0009] Furthermore, the method for obtaining the sensing attitude change coefficient includes: The sum of the absolute values ​​of pitch angle, roll angle, and yaw angle is used as the sensing attitude change coefficient.

[0010] Furthermore, the method for obtaining the sensing attitude offset discrimination coefficient includes: Based on the variation characteristics of voltage direction change coefficient, sensor attitude change coefficient, and voltage vector magnitude, the steady-state period in the electric field detection process is determined; the voltage change threshold is determined based on the distribution characteristics of the voltage direction change coefficient within the steady-state period, and the attitude change threshold is determined based on the distribution characteristics of the sensor attitude change coefficient within the steady-state period. When the voltage direction change coefficient is greater than the voltage change threshold or the sensor attitude change coefficient is greater than the attitude change threshold, anomaly analysis is triggered: the voltage direction change coefficient sequence and the sensor attitude change coefficient sequence are collected for a preset duration, and the sensor attitude offset discrimination coefficient is obtained based on the change synchronization characteristics of the two sequences.

[0011] Further, based on the synchronization characteristics of the changes in the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence, a sensing attitude offset discrimination coefficient is obtained, including: After standardizing the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence respectively, first-order difference is performed and the absolute value is taken. Then, the numbers are sorted in sequence. Based on the concentrated characteristics of the difference between the absolute values ​​of the difference under the same sort number in the two sequences, the asynchronous parameter is determined. The asynchronous parameter is negatively correlated and normalized to obtain the first synchronization parameter. The rising abrupt change points in the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence are extracted respectively. The rising abrupt change points in the two sequences are sorted and matched. The time interval between the corresponding times of the matched rising abrupt change points is negatively correlated and normalized to obtain the second synchronization parameter. By fusing the first synchronization parameter and the second synchronization parameter, the sensing attitude offset discrimination coefficient is obtained.

[0012] Furthermore, the output electric field detection result based on the sensing attitude offset discrimination coefficient includes: When the sensor attitude offset discrimination coefficient is greater than the preset discrimination threshold, it is determined and warned that there is a deviation in the electric field detection result and it is caused by the attitude offset of the electric field sensor. Based on the deviation of each attitude parameter relative to the preset reference attitude, the voltage vector at each moment after the triggering anomaly analysis is spatially rotated to obtain the corresponding corrected voltage vector. When the sensing attitude offset discrimination coefficient is less than or equal to the preset discrimination threshold, and the rate of change of the magnitude of the voltage vector acquired after triggering is greater than the preset rate of change threshold, it is determined that the electric field strength has changed, and the original voltage vector is retained. When the sensing attitude offset discrimination coefficient is less than or equal to a preset discrimination threshold, and the rate of change of the magnitude of the voltage vector acquired after triggering is less than or equal to a preset rate of change threshold, it is determined that the electric field direction has changed, the original voltage vector is retained, and an early warning is issued. If no anomaly analysis is triggered, the collected voltage vector will be output directly as the normal electric field detection result.

[0013] An electric field detection system based on a micro-miniature electric field sensor is provided. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the electric field detection method based on the micro-miniature electric field sensor.

[0014] The present invention has the following beneficial effects: This invention acquires voltage vectors collected by three orthogonal sensing units in an electric field sensor at each moment during electric field detection. Then, based on the spatial geometric relationship between voltage vectors at adjacent moments, it analyzes the instantaneous change characteristics of the voltage vector direction and the continuous change characteristics of the voltage vector direction within a preset historical time period at each moment. This yields the voltage direction change coefficient at each moment, providing a reliable quantitative basis for distinguishing between electric field changes and sensor attitude shifts. Furthermore, it collects all attitude parameters of the electric field sensor at each moment and determines the sensing attitude change coefficient to quantify the attitude change of the sensor body based on these parameters. Finally, based on the similarity between the voltage direction change coefficient and the sensing attitude change coefficient during electric field detection, it obtains a sensing attitude shift discrimination coefficient characterizing the probability of sensor attitude shift dominating. The electric field detection result is then output based on this discrimination coefficient. This invention achieves precise decoupling of sensor shift and electric field fluctuations by constructing a comprehensive judgment coefficient based on the geometric features of the voltage vector and the intensity of attitude change. Without increasing hardware costs, it performs correction and warning of detection data and sensor attitude correction, improving the accuracy of electric field detection in complex environments. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 A flowchart illustrating an electric field detection method based on a micro-sized electric field sensor, provided as an embodiment of the present invention; Figure 2 This is a flowchart illustrating a method for obtaining the voltage direction change coefficient according to an embodiment of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an electric field detection method and system based on a micro-sized electric field sensor proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0019] The following description, in conjunction with the accompanying drawings, details a specific scheme for an electric field detection method and system based on a micro-sized electric field sensor provided by this invention.

[0020] Please see Figure 1 The diagram illustrates a flowchart of an electric field detection method based on a micro-sized electric field sensor, according to an embodiment of the present invention, specifically including: Step S1: Obtain the voltage vector of the electric field at each moment during the electric field detection process. The voltage vector is composed of the single-axis voltages collected and calculated by the three orthogonal sensing units in the electric field sensor. Collect all attitude parameters of the electric field sensor at each moment.

[0021] An electric field is a vector field, possessing both magnitude and direction. If a miniature electric field sensor contains only one sensing element, it can only measure the projected component of the electric field in its sensing direction. When the sensor undergoes an attitude shift, changes in the output voltage may be misinterpreted as changes in the electric field strength, leading to unreliable measurement results. However, three orthogonal sensing elements can completely record the three components of the electric field in the sensor coordinate system. When the sensor's attitude undergoes a slight deflection, the voltages on the three axes will be redistributed according to the rotation matrix, thus uniquely determining the electric field vector. Based on this, one embodiment of the present invention first selects an electric field sensor integrating three orthogonal micro electric field sensing units. Each sensing unit can sense the electric field intensity in its sensing direction and output an analog voltage. The sensing direction axes of the sensing units are perpendicular to each other and are rigidly connected to the housing of the electric field sensor to form a spatial rectangular coordinate system (X, Y, Z axes). This spatial rectangular coordinate system does not require an external absolute reference and characterizes the initial attitude reference of the electric field sensor, preparing for subsequent evaluation of attitude deviation. At each (sampling) moment during the electric field detection process, the single-axis voltage (i.e., analog voltage, unit: V) collected by the three orthogonal sensing units is used as a sequence element to construct a three-dimensional column vector, which is denoted as the voltage vector. In this example, the sampling frequency is set to 200Hz, but the implementer can adjust it as needed.

[0022] Although the voltage vector constructed by the three orthogonal sensing units can reflect the change in the electric field direction, the voltage vector alone cannot distinguish whether this change in direction is caused by a change in the electric field direction or by the attitude shift (such as rotation or tilt) of the electric field sensor body in space. In order to independently obtain the motion information of the electric field sensor body, one embodiment of the present invention integrates an attitude sensor in the electric field sensor. The attitude sensor includes a three-axis gyroscope and a three-axis accelerometer to collect the attitude parameters of the electric field sensor body relative to the spatial rectangular coordinate system at each moment.

[0023] Preferably, in one embodiment of the present invention, considering that any rotation in space can be decomposed into three independent rotations about the X, Y, and Z axes, the pitch angle, roll angle, and yaw angle can completely describe the rotation information of the electric field sensor body relative to the initial attitude, thereby identifying attitude changes; based on this, the attitude parameters include at least the pitch angle, roll angle, and yaw angle.

[0024] In this example, the pitch angle can be defined as the angle at which the attitude sensor rotates around its own Y-axis, the roll angle as the angle at which it rotates around its own X-axis, and the yaw angle as the angle at which it rotates around its own Z-axis.

[0025] Step S2: Based on the spatial geometric relationship between voltage vectors of adjacent times and the spatial geometric relationship of voltage vectors within a preset historical period at each time, obtain the voltage direction change coefficient at each time; determine the sensing attitude change coefficient based on all attitude parameters at each time.

[0026] Since changes in electric field intensity do not change the direction of voltage vector, but when the sensitive direction axis is offset relative to the electric field direction, the direction of voltage vector will change regardless of whether the electric field sensor body rotates or the electric field direction changes. When the orientation of the electric field sensor body shifts rapidly in a very short time or the direction of the electric field changes rapidly, the voltage vectors at adjacent moments will exhibit non-collinear geometric characteristics, and the degree of non-collinearity is independent of the magnitude of the electric field strength. By analyzing the spatial geometric relationship of the voltage vectors at adjacent moments, it is possible to capture the instantaneous change characteristics of the voltage vector direction. However, if the electric field sensor deflects continuously at an extremely slow speed, the directional difference between the voltage vectors at adjacent moments is very small, even smaller than the measurement noise. Relying solely on adjacent moments will miss this slow change. By analyzing the spatial geometric relationship of the voltage vectors within a continuous preset historical period, it is possible to assess whether the direction of the voltage vector is continuously changing. To comprehensively cover all possible forms of electric field sensor attitude offset (rapid jumps or slow creep), the instantaneous geometric relationship between adjacent moments and the cumulative geometric relationship over a preset historical period are combined to extract instantaneous change features and continuous change features to evaluate the voltage direction change coefficient. This provides a reliable quantitative basis for subsequent judgment of the source of mismatch, i.e., distinguishing between electric field changes and sensor attitude offset.

[0027] Preferably, in one embodiment of the present invention, please refer to Figure 2 The flowchart illustrates a method for obtaining the voltage direction change coefficient according to an embodiment of the present invention, specifically including: Step S201: At each time point, extract the two nearest adjacent time points to construct an instantaneous analysis period. Based on the consistent direction characteristics between the voltage vectors of every two adjacent time points within the instantaneous analysis period, obtain the instantaneous voltage deviation coefficient.

[0028] To facilitate the analysis of the instantaneous change characteristics of the voltage vector direction, one embodiment of the present invention takes any given moment as an example and first constructs an instantaneous analysis period; specifically, the time period corresponding to this moment and its two nearest adjacent moments (a total of three consecutive moments) is taken as the instantaneous analysis period.

[0029] When the electric field sensor is in a stable attitude and the electric field direction remains unchanged, the directions of the voltage vectors at three consecutive moments should be consistent (i.e., collinear in space). However, once a momentary jitter or rapid deflection occurs, the directions of the three voltage vectors will deviate from collinearity. By analyzing the consistency of the directions among these three voltage vectors, the instantaneous voltage deviation coefficient, which characterizes the degree of instantaneous change in voltage direction, can be quantified.

[0030] In a preferred embodiment of the present invention, considering that the cross product of voltage vectors at any two adjacent moments can determine a rotation vector that is perpendicular to both voltage vectors, the direction of the rotation vector indicates the instantaneous axis of rotation around which the voltage rotates from the previous moment to the next moment, and its magnitude is proportional to the magnitude of the rotational angular velocity; normalizing the voltage vectors to obtain unit vectors not only removes the dimensions but also eliminates the influence of the magnitude of the voltage vector on the magnitude of the cross product; then, by calculating two consecutive cross product vectors (rotation vectors), on the one hand, comparing the difference in their magnitudes can measure the degree of change in the rotational angular velocity, and on the other hand, calculating their dot product can measure whether the rotational axis directions of the two rotations are consistent, thereby quantifying the rotational deviation to obtain the instantaneous voltage deviation coefficient, which is only related to the change in voltage direction and is independent of the magnitude of the electric field strength; the method for obtaining the instantaneous voltage deviation coefficient includes: Within the instantaneous analysis period at each moment, the unit vector of the voltage vector at each moment is determined. The rotation vector is determined based on the cross product of the unit vectors at two adjacent moments. The instantaneous voltage deviation coefficient is determined based on the difference in magnitude and the dot product between the rotation vectors at each two adjacent moments.

[0031] Specifically, within the instantaneous analysis period (i.e., three consecutive moments), the voltage vector at each moment is first divided by its magnitude to obtain a unit vector; then, the cross product of the unit vectors at every two adjacent moments is calculated to obtain two rotation vectors; then, the difference in magnitude between the two rotation vectors is compared (specifically, the absolute value of the magnitude difference is calculated) to measure the degree of change in the rotational angular velocity, and the consistency of the rotation axis direction is measured by calculating the dot product of the two rotation vectors and taking its absolute value; the absolute value of the magnitude difference is added to the absolute value of the dot product to obtain the instantaneous voltage deviation coefficient; the instantaneous voltage deviation coefficient depends only on the change in voltage direction and is independent of the absolute voltage amplitude (i.e., the magnitude of the electric field strength), and can sensitively reflect instantaneous changes in direction.

[0032] Step S202: Construct a preset historical time period with each time point as the endpoint, and obtain the voltage continuous deviation coefficient at each time point based on the consistent direction of all voltage vectors within the preset historical time period.

[0033] To facilitate the analysis of the continuous change characteristics of the voltage vector direction, one embodiment of the present invention takes any given moment as an example and obtains a historical period of a preset length along the reverse direction of the time sequence with that moment as the endpoint, that is, constructs a preset historical period; the preset length is recommended to be 10-50, and this example is 50.

[0034] By pre-setting the directional consistency characteristics of all voltage vectors within a historical period, a feature quantity sensitive to continuous slow changes can be obtained, namely the voltage continuous deviation coefficient, which characterizes whether the direction of the voltage vector has undergone cumulative shift. This complements the voltage instantaneous deviation coefficient, fully covering attitude changes at various speeds.

[0035] In a preferred embodiment of the present invention, considering that when the directions of all voltage vectors are consistent within a preset historical period, their vector sum direction will be the same as the direction of each voltage vector, and the cross product magnitude of each voltage vector and its vector sum will also be zero; conversely, if the directions of voltage vectors are dispersed within the preset historical period (i.e., continuous direction changes occur), the cross product magnitude of each voltage vector and its vector sum will be greater than zero, and the greater the degree of dispersion, the greater the average value of the cross product magnitude; therefore, by calculating the cross product magnitude of each voltage vector and its vector sum within the preset historical period, the direction consistency can be quantified, thereby obtaining the voltage continuous deviation coefficient; this coefficient is sensitive to slow direction drift and is not affected by the magnitude of the absolute voltage value; the method for obtaining the voltage continuous deviation coefficient includes: Within a preset historical time period at each moment, the voltage vectors at all moments are summed. The magnitude of the cross product of the voltage vectors and the sum of the vectors at all moments within the preset historical time period is used to determine the voltage persistence deviation coefficient.

[0036] Specifically, the voltage vectors at all times within a preset historical period are summed. The direction of the sum represents the average direction of all voltage vectors, and the magnitude reflects the consistency of direction (the more consistent the direction, the closer the magnitude of the sum is to the sum of the magnitudes of each voltage vector). Then, the cross product of the voltage vector and the vector sum at each moment within the preset historical period is calculated, and the magnitude of the cross product is taken. When the direction of a voltage vector is exactly the same as the direction of the vector sum, the magnitude of the cross product is zero; conversely, the greater the deviation between the two directions, the larger the magnitude of the cross product. The average value of the magnitude of the cross product of all voltage vectors and the vector sum within the preset historical period is calculated. In order to eliminate the influence of the voltage amplitude (i.e., electric field strength) on the average value, a very small constant, such as 0.01, is added to the square of the average magnitude of all voltage vectors within the preset historical period to prevent division by zero. The average value is divided by this sum to obtain the voltage persistence deviation coefficient. The voltage persistence deviation coefficient can capture slow but continuous attitude deviation or electric field direction change. The larger the value, the more obvious the continuous change in the direction of the voltage vector.

[0037] Step S203: Combine the instantaneous voltage deviation coefficient and the continuous voltage deviation coefficient to obtain the voltage direction change coefficient at each moment.

[0038] Specifically, for each time point, the instantaneous voltage deviation coefficient and the sustained voltage deviation coefficient are weighted and summed with weights of 0.5 and 0.5 respectively, and the weighted summation result is used as the voltage direction change coefficient. In other examples, the weights can be adjusted, and other fusion methods can be used, such as directly multiplying and fusing the instantaneous voltage deviation coefficient and the sustained voltage deviation coefficient.

[0039] Since electric field sensors only output voltage signals related to electric field strength, it is impossible to know whether the electric field sensor body has rotated or tilted; when the measurement results are abnormal, it is impossible to determine whether the electric field itself has changed or the sensor attitude has shifted; while the attitude parameters of electric field sensors can quantify the attitude changes of the sensor body in real time. When the electric field detection result (voltage signal) is abnormal, the change in voltage direction is compared with the change in the attitude of the sensor body. If the two change trends are consistent, it indicates that the abnormality is likely caused by the sensor attitude deviation; otherwise, it may be that the electric field itself has changed. Therefore, after obtaining the voltage direction change coefficient at each moment, this embodiment of the invention further determines the sensor attitude change coefficient based on all attitude parameters at each moment, providing a reliable independent reference for distinguishing the source of error in the future.

[0040] Preferably, in one embodiment of the present invention, considering that the pitch angle, roll angle, and yaw angle output by the attitude sensor each reflect the rotation angle of the sensor body around different axes, and their changes can be positive or negative (positive and negative indicate the direction of rotation); taking the absolute value can eliminate the influence of the direction sign, retaining only the magnitude of the rotation; by summing and combining the absolute values ​​of the rotation amplitude angles in the three directions into a scalar, the sensing attitude change coefficient, which quantifies the overall severity of the sensor attitude change, can be obtained; then the method for obtaining the sensing attitude change coefficient includes: The sum of the absolute values ​​of pitch angle, roll angle, and yaw angle is used as the sensing attitude change coefficient.

[0041] Step S3: Based on the similarity between the voltage direction change coefficient and the sensor attitude change coefficient during the electric field detection process, obtain the sensor attitude offset discrimination coefficient; output the electric field detection result based on the sensor attitude offset discrimination coefficient.

[0042] Since the sensing attitude change coefficient only reflects the rotation of the sensor body, while the voltage direction change coefficient only reflects the angular change of the sensitive axis relative to the electric field direction (regardless of whether the cause is a change in the electric field direction or sensor rotation); if the voltage direction change and the sensor body attitude change are highly similar in waveform shape and time response (i.e., similar or synchronous changes), it can be said that the main reason for the voltage direction change is that the sensor itself is rotating; based on this similarity feature, the embodiments of the present invention can construct a sensing attitude offset discrimination coefficient to quantify the possibility that the mismatch is dominated by attitude offset.

[0043] Preferably, in one embodiment of the present invention, considering that the voltage direction and attitude of the electric field sensor may have slight random fluctuations (such as noise or slight vibration) during normal operation, analyzing all fluctuations would lead to a large number of false triggers; therefore, it is first necessary to determine a steady-state period, that is, a period in which the sensor attitude is stable, the electric field direction is stable, and the electric field strength does not change significantly. Within this steady-state period, the normal fluctuation range (i.e., distribution characteristics) of the voltage direction change coefficient and the sensor attitude change coefficient can be statistically determined, and reasonable trigger thresholds (voltage change threshold and attitude change threshold) can be set accordingly to determine whether a valid voltage direction change or sensor attitude change has occurred, and to trigger analysis to determine whether the voltage direction change and attitude change are truly synchronized, so as to accurately determine the sensor attitude offset discrimination coefficient; the method for obtaining the sensor attitude offset discrimination coefficient includes: Based on the variation characteristics of voltage direction change coefficient, sensor attitude change coefficient, and voltage vector magnitude, the steady-state period in the electric field detection process is determined; the voltage change threshold is determined based on the distribution characteristics of the voltage direction change coefficient within the steady-state period, and the attitude change threshold is determined based on the distribution characteristics of the sensor attitude change coefficient within the steady-state period. When the voltage direction change coefficient is greater than the voltage change threshold or the sensor attitude change coefficient is greater than the attitude change threshold, anomaly analysis is triggered: the voltage direction change coefficient sequence and the sensor attitude change coefficient sequence are collected for a preset duration respectively; based on the change synchronization characteristics of the two sequences, the sensor attitude offset discrimination coefficient is obtained.

[0044] As an example, the voltage direction change coefficient, sensing attitude change coefficient, and voltage vector magnitude are calculated in real time at each moment. A fixed-length analysis window is set. In this example, the analysis window contains 100 moments (0.5 seconds when the sampling frequency is 200Hz). Implementers can also adjust it themselves. After the detection begins, several analysis windows are constructed sequentially. Within each analysis window, it is checked whether the following three conditions are met simultaneously: 1. The range of the voltage direction change coefficient within the analysis window is less than the preset voltage direction change threshold; 2. The range of the sensor attitude change coefficient within the analysis window is less than the preset sensor attitude change threshold; 3. The rate of change of the magnitude of the voltage vector within the analysis window is less than a preset rate of change threshold, such as 0.1. It should be noted that the preset voltage direction change threshold is based on the sensor's factory noise level calibration. It characterizes the critical value of apparent direction jitter caused by circuit thermal noise and mechanical Brownian motion of MEMS structure under a standard electrostatic field in the laboratory. Specifically, the voltage direction change coefficient of 1000 sampling points in the static state is statistically analyzed, and 3-6 times the standard deviation (3 times in this example) is taken as the preset voltage direction change threshold. The preset sensor attitude change threshold is based on the zero bias instability and angle random walk parameters of the attitude sensor and can be obtained from the datasheet. When the sensor is physically stationary, there will still be a small random drift in the attitude sensor output. Setting the preset sensor attitude change threshold above the noise floor given in the datasheet (usually twice the root mean square noise value) can effectively distinguish between the actual small sensor deflection and the sensor zero drift. The aforementioned rate of change is the average of the rate of change of the magnitude of adjacent voltage vectors within the analysis window. The calculation of the rate of change of the magnitude is a conventional technique in this field and will not be elaborated further.

[0045] When all three conditions above are met, the analysis window is recorded as a steady-state window. During the detection process, when the cumulative duration of multiple consecutive steady-state windows reaches a preset duration, such as 3 seconds (i.e., 6 consecutive non-overlapping steady-state windows), it is considered to have entered a continuous stable state. This 3-second period is taken as the steady-state period.

[0046] The mean μ1 and standard deviation σ1 of the voltage direction change coefficient during the steady-state period are calculated; the mean μ2 and standard deviation σ2 of the sensor attitude change coefficient during the steady-state period are calculated; according to the Raida criterion (3σ principle) in statistics, under normal conditions, 99.7% of the data should fall within the range of mean ± 3 times standard deviation σ; therefore, μ1 + 3 × σ1 is used as the voltage change threshold, and μ2 + 3 × σ2 is used as the attitude change threshold.

[0047] In another example of the present invention, considering that the steady-state statistical characteristics of the sensor may drift slowly due to factors such as temperature and aging during long-term operation, an exponentially weighted moving average can be used to periodically update the two changing thresholds: after each new steady-state period data is obtained, the mean and standard deviation are updated by retaining the old statistics with 90% weight and introducing the new statistics with 10% weight, thereby achieving adaptive tracking of the thresholds.

[0048] When the voltage direction change coefficient is greater than the voltage change threshold or the sensor attitude change coefficient is greater than the attitude change threshold, the acquisition and analysis are triggered: starting from the triggering time, the voltage direction change coefficient sequence and the sensor attitude change coefficient sequence are recorded within a short window of a preset duration (e.g., 0.05s) after the triggering; the synchronization characteristics of the two sequences are further analyzed to obtain the sensor attitude offset discrimination coefficient.

[0049] In a preferred embodiment of the present invention, considering that whether the sequence changes are synchronized is mainly reflected in the trend and magnitude of the changes, performing first-order difference on the sequence elements and taking the absolute value can help evaluate the drastic change in the trend and magnitude of the sequence elements (voltage direction change coefficient or sensing attitude change coefficient), and then comparing the difference in the absolute value of the difference with the same label to quantify the first synchronization parameter; also considering that whether the sequence changes are synchronized is also reflected in the change response (matching), by extracting significant rise points (i.e., abrupt change moments) in the two sequences respectively, and matching the time interval of these abrupt change points to evaluate the response delay, the smaller the time interval, the more synchronized the two are, thereby quantifying the second synchronization parameter; and then comprehensively evaluating the change synchronization characteristics to obtain the sensing attitude offset discrimination coefficient, including: After standardizing the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence respectively, first-order difference is performed and the absolute value is taken. Then, the numbers are sorted in sequence. Based on the concentrated characteristics of the difference between the absolute values ​​of the difference under the same sort number in the two sequences, the asynchronous parameter is determined. The asynchronous parameter is negatively correlated and normalized to obtain the first synchronization parameter. The rising abrupt change points in the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence are extracted respectively. The rising abrupt change points in the two sequences are sorted and matched. The time interval between the corresponding times of the matched rising abrupt change points is negatively correlated and normalized to obtain the second synchronization parameter. By fusing the first synchronization parameter and the second synchronization parameter, the sensor attitude offset discrimination coefficient is obtained.

[0050] Specifically, the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence are first standardized to remove dimensions. This can be done through Z-score standardization. The mean and standard deviation of each sequence are calculated from the corresponding sequence. When the standard deviation is 0, it indicates that the sequence elements are consistent, and the standardization result of each sequence element can be directly set to 1. In other examples, other standardization methods such as max-min normalization can also be used. These are all conventional techniques in this field and will not be elaborated further. Then, each standardized sequence is first-order (forward) differencing is performed, and the absolute value of each difference is taken and sorted according to the difference order in each sequence. The difference is measured by the absolute value of the difference and the mean is used to represent the central tendency. The absolute value of the difference between the absolute values ​​of the differences under the same sorting number in two sequences is calculated and averaged to obtain the changing asynchronous parameter. The changing asynchronous parameter is then negatively correlated and normalized. For example, the changing asynchronous parameter is mapped as the independent variable x to the exponential function exp(-x) with the natural constant as the base to obtain the first synchronization parameter. The rising abrupt change points in the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence are extracted respectively. The rising abrupt change point is defined as: the value of the sequence element is significantly increased compared to the two adjacent sequence elements and exceeds a preset threshold, such as 0.05. The rising abrupt change points in the two sequences are sorted respectively, and the rising abrupt change points with the same sort number in different sequences are used as a pair of matching points and the corresponding (acquisition) time is extracted. In other examples, the rising abrupt change point of one sequence can also be used as a benchmark, and the rising abrupt change point with the closest time interval to its corresponding time in the other sequence can be found to form a match. The time intervals between the corresponding times of the matching rising abrupt change points in different sequences are averaged and then negatively correlated and normalized, for example, as the independent variable x is mapped to the exponential function exp(-x) with the natural constant as the base to obtain the second synchronization parameter. It should be noted that the number of rising mutation points in different sequences may be inconsistent, which may cause some rising mutation points to be unable to participate in matching. Rising mutation points that do not participate in matching will not participate in subsequent time interval analysis. Alternatively, the number of unmatched rising mutation points can be used as a penalty weight for normalization, and then the time interval can be weighted and negatively correlated to obtain the second synchronization parameter.

[0051] In another embodiment of the present invention, the standardized voltage direction change coefficient sequence and the sensing attitude change coefficient sequence can be directly matched by DTW and the DTW distance can be calculated; the DTW distance can be negatively correlated and normalized to obtain the first synchronization parameter; the time interval between the corresponding (acquisition) times of the sequence elements in the DTW matching pair can be averaged and then negatively correlated and normalized to obtain the second synchronization parameter.

[0052] Finally, the first synchronization parameter and the second synchronization parameter are multiplied and fused to obtain the sensing attitude offset discrimination coefficient; in other examples, other fusion methods such as weighted summation can also be used.

[0053] After obtaining the sensor attitude shift discrimination coefficients that characterize the probability that attitude shift dominates, the electric field detection results can be further output based on the sensor attitude shift discrimination coefficients.

[0054] Preferably, in one embodiment of the present invention, the output electric field detection result based on the sensing attitude offset discrimination coefficient includes: When the sensor attitude offset discrimination coefficient is greater than the preset discrimination threshold, it is determined and warned that there is a deviation in the electric field detection result and it is caused by the attitude offset of the electric field sensor. Based on the deviation of each attitude parameter relative to the preset reference attitude, the voltage vector at each moment after the triggering anomaly analysis is spatially rotated to obtain the corresponding correction voltage vector. When the sensor attitude offset discrimination coefficient is less than or equal to the preset discrimination threshold, or when the rate of change of the magnitude of the voltage vector acquired after triggering is greater than the preset rate of change threshold, it is determined that the electric field strength has changed, and the original voltage vector is retained. When the sensor attitude offset discrimination coefficient is less than or equal to the preset discrimination threshold, or when the rate of change of the magnitude of the voltage vector acquired after triggering is less than or equal to the preset rate of change threshold, it is determined that the electric field direction has changed, the original voltage vector is retained, and an early warning is issued. If no anomaly analysis is triggered, the collected voltage vector will be output directly as the normal electric field detection result.

[0055] Specifically, the preset discrimination threshold is recommended to be 0.6, but implementers can also adjust it themselves; the value is based on the minimum value of the sensor attitude deviation discrimination coefficient in the electric field detection events with known sensor attitude deviation in history; if there are no historical electric field detection events initially, the initial value can be set to 0.5. The preset change rate threshold is recommended to be 0.1 (i.e. 10%). The basis for this value is that in industrial monitoring or power grid environments, the normal background random fluctuation of electric field strength is usually within 5%. Setting a limit of 10% can accurately separate normal environmental white noise from substantial electric field strength jumps in a physical sense. Then, a categorized discussion is conducted. The following discussion takes place at each time point after the anomaly analysis is triggered, taking any given time point as an example: (1) When the sensor attitude offset discrimination coefficient is greater than the preset discrimination threshold, it is determined and warned that there is a deviation in the electric field detection result and it is caused by the attitude offset of the electric field sensor; extract and calculate the difference between each attitude parameter and the preset reference attitude corresponding to its axis to obtain the difference angle; where the preset reference attitude is the axis of each sensitive unit axis of the electric field sensor corresponding to the spatial rectangular coordinate system, which is the initial attitude reference, or can be regarded as the initial zero value; Based on the opposite values ​​of the difference angles along each axis, independent single-axis rotation matrices Rx (around the X-axis), Ry (around the Y-axis), and Rz (around the Z-axis) are constructed. Following the complete reversal of the factory physical rotation order (e.g., ZYX) (i.e., XYZ order), each single-axis rotation matrix is ​​multiplied sequentially to obtain the total reverse rotation matrix. The voltage vector at each moment after triggering anomaly analysis is multiplied by the total reverse rotation matrix to perform spatial rotation compensation, yielding the corresponding corrected voltage vector. Simultaneously, a warning is issued indicating that the sensor attitude needs correction to align the sensor attitude parameters with the preset reference attitude. (2) When the sensor attitude offset discrimination coefficient is less than or equal to the preset discrimination threshold, and the rate of change of the voltage vector magnitude acquired after triggering is greater than the preset rate of change threshold, it is determined that the electric field strength has changed, and the original voltage vector is retained as the electric field detection output result; where the rate of change of the voltage vector magnitude refers to the rate of change of the voltage vector magnitude at this moment relative to the magnitude of the voltage vector at the adjacent previous moment. The calculation of the rate of change is a conventional technical means in this field and will not be elaborated here. (3) When the sensor attitude offset discrimination coefficient is less than or equal to the preset discrimination threshold, and the rate of change of the voltage vector magnitude after triggering is less than or equal to the preset rate of change threshold, it is determined that the electric field direction has changed, but the sensor is not deviated and the electric field strength has not changed substantially. The original voltage vector is retained, and the warning information of "abnormal deflection of electric field direction" is output; where the rate of change of the voltage vector magnitude is the same as (2).

[0056] If the above-mentioned anomaly analysis is not triggered, i.e. the sensor attitude offset discrimination coefficient cannot be calculated, the collected voltage vector will be directly output as the normal electric field detection result.

[0057] Based on the same inventive concept, the present invention also proposes an electric field detection system based on a micro-miniature electric field sensor. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the electric field detection method based on the micro-miniature electric field sensor described in steps S1-S3 above.

[0058] In summary, this invention acquires the voltage vector collected by three orthogonal sensing units in the electric field sensor at each moment. Then, based on the voltage vectors at adjacent moments and the spatial geometric relationship of voltage vectors within a preset historical period, it comprehensively obtains the voltage direction change coefficient at each moment. Simultaneously, it collects all attitude parameters of the electric field sensor at each moment to determine the sensing attitude change coefficient. Furthermore, based on the similarity characteristics of the changes in the voltage direction change coefficient and the sensing attitude change coefficient, it obtains the sensing attitude offset discrimination coefficient and outputs the electric field detection result. This invention achieves precise decoupling of sensor offset and electric field fluctuations by constructing a comprehensive discrimination coefficient of voltage vector geometric features and attitude change intensity. Without increasing hardware costs, it performs correction and early warning of detection data and sensing attitude correction, improving the accuracy of electric field detection in complex environments.

[0059] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0060] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. An electric field detection method based on micro-miniature electric field sensors, characterized in that, The method includes: The voltage vector of the electric field at each moment during the electric field detection process is obtained. The voltage vector is composed of the single-axis voltages collected and calculated by three orthogonal sensing elements in the electric field sensor. All attitude parameters of the electric field sensor at each moment are collected. Based on the spatial geometric relationship between voltage vectors of adjacent time points at each time point, and the spatial geometric relationship of voltage vectors within a preset historical period at each time point, the voltage direction change coefficient at each time point is obtained; the sensing attitude change coefficient is determined based on all attitude parameters at each time point. Based on the similarity between the voltage direction change coefficient and the sensor attitude change coefficient during electric field detection, the sensor attitude deviation discrimination coefficient is obtained; the electric field detection result is output based on the sensor attitude deviation discrimination coefficient.

2. The electric field detection method based on a micro-sized electric field sensor according to claim 1, characterized in that, The attitude parameters include at least pitch angle, roll angle, and yaw angle.

3. The electric field detection method based on a micro-sized electric field sensor according to claim 1, characterized in that, The method for obtaining the voltage direction change coefficient includes: At each time point, the two nearest adjacent time points are extracted to construct an instantaneous analysis period. Based on the consistent direction of the voltage vectors between two adjacent time points within the instantaneous analysis period, the instantaneous voltage deviation coefficient is obtained. A preset historical time period is constructed with each moment as the endpoint. Based on the consistent direction of all voltage vectors within the preset historical time period, the voltage deviation coefficient at each moment is obtained. By combining the instantaneous voltage deviation coefficient and the continuous voltage deviation coefficient, the voltage direction change coefficient at each moment is obtained.

4. The electric field detection method based on a micro-sized electric field sensor according to claim 3, characterized in that, The method for obtaining the instantaneous voltage deviation coefficient includes: Within the instantaneous analysis period at each moment, the unit vector of the voltage vector at each moment is determined. The rotation vector is determined based on the cross product of the unit vectors at two adjacent moments. The instantaneous voltage deviation coefficient is determined based on the difference in magnitude and the dot product between the rotation vectors at each two adjacent moments.

5. The electric field detection method based on a micro-sized electric field sensor according to claim 3, characterized in that, The method for obtaining the voltage persistent deviation coefficient includes: Within a preset historical time period at each moment, the voltage vectors at all moments are summed. The magnitude of the cross product of the voltage vectors at all moments within the preset historical time period and the sum of the vectors is used to determine the voltage persistence deviation coefficient.

6. The electric field detection method based on a micro-sized electric field sensor according to claim 2, characterized in that, The method for obtaining the sensing attitude change coefficient includes: The sum of the absolute values ​​of pitch angle, roll angle, and yaw angle is used as the sensing attitude change coefficient.

7. The electric field detection method based on a micro-sized electric field sensor according to claim 1, characterized in that, The method for obtaining the sensing attitude offset discrimination coefficient includes: Based on the variation characteristics of voltage direction change coefficient, sensor attitude change coefficient, and voltage vector magnitude, the steady-state period in the electric field detection process is determined; the voltage change threshold is determined based on the distribution characteristics of the voltage direction change coefficient within the steady-state period, and the attitude change threshold is determined based on the distribution characteristics of the sensor attitude change coefficient within the steady-state period. When the voltage direction change coefficient is greater than the voltage change threshold or the sensor attitude change coefficient is greater than the attitude change threshold, anomaly analysis is triggered: the voltage direction change coefficient sequence and the sensor attitude change coefficient sequence are collected for a preset duration, and the sensor attitude offset discrimination coefficient is obtained based on the change synchronization characteristics of the two sequences.

8. The electric field detection method based on a micro-sized electric field sensor according to claim 7, characterized in that, Based on the synchronization characteristics of the changes in the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence, the sensing attitude offset discrimination coefficient is obtained, including: After standardizing the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence respectively, first-order difference is performed and the absolute value is taken. Then, the numbers are sorted in sequence. Based on the concentrated characteristics of the difference between the absolute values ​​of the difference under the same sort number in the two sequences, the asynchronous parameter is determined. The asynchronous parameter is negatively correlated and normalized to obtain the first synchronization parameter. The rising abrupt change points in the voltage direction change coefficient sequence and the sensing attitude change coefficient sequence are extracted respectively. The rising abrupt change points in the two sequences are sorted and matched. The time interval between the corresponding times of the matched rising abrupt change points is negatively correlated and normalized to obtain the second synchronization parameter. By fusing the first synchronization parameter and the second synchronization parameter, the sensing attitude offset discrimination coefficient is obtained.

9. The electric field detection method based on a micro-sized electric field sensor according to claim 7, characterized in that, The electric field detection results based on the sensor attitude offset discrimination coefficient include: When the sensor attitude offset discrimination coefficient is greater than the preset discrimination threshold, it is determined and warned that there is a deviation in the electric field detection result and it is caused by the attitude offset of the electric field sensor. Based on the deviation of each attitude parameter relative to the preset reference attitude, the voltage vector at each moment after the triggering anomaly analysis is spatially rotated to obtain the corresponding corrected voltage vector. When the sensing attitude offset discrimination coefficient is less than or equal to the preset discrimination threshold, and the rate of change of the magnitude of the voltage vector acquired after triggering is greater than the preset rate of change threshold, it is determined that the electric field strength has changed, and the original voltage vector is retained. When the sensing attitude offset discrimination coefficient is less than or equal to a preset discrimination threshold, and the rate of change of the magnitude of the voltage vector acquired after triggering is less than or equal to a preset rate of change threshold, it is determined that the electric field direction has changed, the original voltage vector is retained, and an early warning is issued. If no anomaly analysis is triggered, the collected voltage vector will be output directly as the normal electric field detection result.

10. An electric field detection system based on a micro-sized electric field sensor, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the electric field detection method based on a micro-miniature electric field sensor as described in any one of claims 1 to 9.

Images