Industrial sensor electromagnetic sensitivity self-adaptive calibration system

Abstract

The invention relates to the technical field of sensor electromagnetic sensitivity calibration, in particular to an industrial sensor electromagnetic sensitivity self-adaptive calibration system, which comprises a data acquisition module used for acquiring an output value of each sensor under different test conditions so as to obtain a data point corresponding to each sensor and the electromagnetic sensitivity of each sensor; the electromagnetic sensitivity characterization degree calculation module is used for calculating the electromagnetic sensitivity characterization degree of each dimension; the clustering effect calculation module is used for calculating the clustering effect of each dimension; the clustering module is used for acquiring weights of dimensions and clustering all the data points to obtain different multi-dimensional clusters; and the electromagnetic sensitivity calibration module is used for acquiring the electromagnetic sensitivity of each multi-dimensional cluster so as to obtain the multi-dimensional cluster to which the data point corresponding to the current sensor belongs and the electromagnetic sensitivity of the current sensor. According to the invention, the efficiency and accuracy of sensor electromagnetic sensitivity calibration can be ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sensor electromagnetic sensitivity calibration, and in particular to an industrial sensor electromagnetic sensitivity adaptive calibration system. BACKGROUND

[0002] Industrial sensor electromagnetic sensitivity calibration refers to the process of quantifying and calibrating the performance response capability of industrial sensors under different electromagnetic environments, aiming to evaluate the anti-interference capability and measurement stability of the sensors when subjected to electromagnetic interference, so as to ensure their accurate and reliable operation in complex industrial environments. Because industrial sensors are often used in high-voltage, high-current, high-frequency or strong magnetic field environments, electromagnetic interference may cause sensor data fluctuations or distortion, thereby affecting the safety and accuracy of industrial control systems, so the electromagnetic sensitivity calibration of industrial sensors is of great significance.

[0003] The existing technology for calibrating the electromagnetic sensitivity of sensors usually adopts a manual calibration method for test data, which is time-consuming and laborious and prone to errors, and cannot meet the needs of automated system production. Therefore, an automated adaptive calibration system is needed to improve the calibration efficiency of the electromagnetic sensitivity of industrial sensors through automated calibration. SUMMARY

[0004] To solve the above technical problems, the purpose of the present application is to provide an industrial sensor electromagnetic sensitivity adaptive calibration system, and the technical solution adopted is as follows:

[0005] One embodiment of the present application provides an industrial sensor electromagnetic sensitivity adaptive calibration system, which comprises:

[0006] A data acquisition module for acquiring the output values of each sensor under different test conditions and taking each test condition as a dimension; forming a data point from the output values of a sensor under each dimension, and obtaining the electromagnetic sensitivity of each sensor;

[0007] An electromagnetic sensitivity representation degree calculation module for calculating the electromagnetic sensitivity representation degree of a dimension according to the distribution of the output values of different data points under the dimension;

[0008] A clustering effect calculation module for clustering the output values under a dimension to obtain different one-dimensional clustering clusters; obtaining the clustering effect of the dimension according to the distribution of the one-dimensional clustering clusters corresponding to the dimension and the distribution of the output values within the one-dimensional clustering clusters;

[0009] A clustering module for obtaining the weight of a dimension according to the electromagnetic sensitivity representation degree and the clustering effect of the dimension; clustering all data points in combination with the weight of each dimension to obtain different multi-dimensional clustering clusters;

[0010] The electromagnetic sensitivity calibration module is configured to obtain electromagnetic sensitivity of each multi-dimensional clustering cluster according to electromagnetic sensitivity of sensors corresponding to each data point in each multi-dimensional clustering cluster; and determine a multi-dimensional clustering cluster to which a data point corresponding to a current sensor belongs, and the electromagnetic sensitivity of the multi-dimensional clustering cluster is the electromagnetic sensitivity of the current sensor.

[0011] Preferably, the electromagnetic sensitivity of each sensor is obtained by:

[0012] The electromagnetic sensitivity of each sensor is obtained by manual evaluation according to output values of each sensor under each test condition.

[0013] Preferably, the electromagnetic sensitivity representation degree of one dimension is calculated according to a distribution of output values of different data points in the one dimension, and the electromagnetic sensitivity representation degree of the one dimension is obtained by:

[0014] The output values of different data points in the one dimension are arranged in ascending order to obtain an output value sequence corresponding to the one dimension; absolute values of differences between each two adjacent output values in the output value sequence are obtained, and the absolute values of the differences are denoted as data intervals between the each two adjacent output values; a difference between a maximum value and a minimum value in the output value sequence is obtained, and the electromagnetic sensitivity representation degree of the one dimension is obtained by multiplying the difference by a reciprocal of a standard deviation of the data intervals between the each two adjacent output values.

[0015] Preferably, the clustering effect of one dimension is obtained according to a distribution of each one-dimensional clustering cluster corresponding to the one dimension and a distribution of output values in each one-dimensional clustering cluster, and the clustering effect of the one dimension is obtained by:

[0016] Each one-dimensional clustering cluster corresponding to the one dimension is grouped into a one-dimensional clustering cluster set, and the one-dimensional clustering clusters in the set are arranged in ascending order of output values; an absolute value of a difference between a maximum value in one one-dimensional clustering cluster and a minimum value in a next one-dimensional clustering cluster of the one-dimensional clustering cluster is obtained, and the absolute value of the difference is denoted as an interval distance; a Euclidean distance between one output value and a nearest output value to the one output value is calculated in the one-dimensional clustering cluster, and the Euclidean distance is denoted as a minimum distance of the one output value, and a mean value of minimum distances of all output values in the one-dimensional clustering cluster is obtained, and the mean value of the minimum distances is denoted as a minimum distance mean value of the one-dimensional clustering cluster; a mapping result is obtained by negatively correlating mapping of the mean value of the minimum distance mean values of all one-dimensional clustering clusters corresponding to the one dimension by using an exponential function with a natural constant as a base, and the mapping result is multiplied by a mean value of the interval distances between each two adjacent one-dimensional clustering clusters in the one-dimensional clustering cluster set corresponding to the one dimension to obtain the clustering effect of the one dimension.

[0017] Preferably, the weight of one dimension is obtained according to the electromagnetic sensitivity representation degree and the clustering effect of the one dimension, and the weight of the one dimension is obtained by:

[0018] The normalized value of the electromagnetic sensitivity representation degree of one dimension and the normalized value of the clustering effect are added and averaged to obtain the weight of the dimension.

[0019] Preferably, the electromagnetic sensitivity of each multi-dimensional clustering cluster is obtained according to the electromagnetic sensitivity of the sensors corresponding to each data point in each multi-dimensional clustering cluster, and the electromagnetic sensitivity of each multi-dimensional clustering cluster is obtained by:

[0020] The mode of the electromagnetic sensitivity of the sensors corresponding to each data point in one multi-dimensional clustering cluster is taken as the electromagnetic sensitivity of the multi-dimensional clustering cluster.

[0021] Preferably, the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs is determined by:

[0022] The reciprocal of the absolute value of the difference between the output value of one data point in one multi-dimensional clustering cluster in one dimension and the output value of the data point corresponding to the current sensor in the dimension is taken as the similarity degree corresponding to the dimension of the data point; the average value of the similarity degrees corresponding to all data points in the dimension in the multi-dimensional clustering cluster is taken as the dimension similarity degree of the dimension; the sum of the dimension similarity degrees of all dimensions is taken as the overall similarity degree; the reciprocal of the distance between the data point corresponding to the current sensor and the clustering center of the multi-dimensional clustering cluster is normalized and added to the normalized value of the overall similarity degree to obtain the belonging possibility corresponding to the multi-dimensional clustering cluster; the multi-dimensional clustering cluster with the maximum belonging possibility is the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs.

[0023] The embodiment of the application has at least the following beneficial effects: the output values of each sensor under different test conditions in history are collected, one test condition is taken as one dimension, the output values of one sensor in different dimensions are taken as one data point, the data points corresponding to multiple sensors and the electromagnetic sensitivity of each sensor are obtained; then the electromagnetic sensitivity representation degree and the clustering effect of each dimension are obtained respectively, the electromagnetic sensitivity representation degree and the clustering effect are combined to obtain the weight of each dimension, and finally all data points are clustered to obtain different multi-dimensional clustering clusters by combining the weight of each dimension, so that the stronger the influence degree of the dimension itself, the higher the influence degree in the clustering process of the multi-dimensional data points, so that the obtained multi-dimensional clustering clusters are more accurate in the division result of the electromagnetic sensitivity; the electromagnetic sensitivity of each multi-dimensional clustering cluster is obtained, then the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs is determined, and the electromagnetic sensitivity of the multi-dimensional clustering cluster to which the current sensor belongs is the electromagnetic sensitivity of the current sensor, so that the electromagnetic sensitivity of the sensor is automatically calibrated, and the calibration efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort.

[0025] Figure 1 The system block diagram of the industrial sensor electromagnetic sensitivity adaptive calibration system provided by the embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined purpose, the specific embodiments, structure, features and effects of the industrial sensor electromagnetic sensitivity adaptive calibration system according to the present application are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0027] 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 the present application belongs.

[0028] The specific scheme of the industrial sensor electromagnetic sensitivity adaptive calibration system provided by the present application is specifically described below with reference to the drawings.

[0029] Embodiment: The main application scenario of the present application is that when the electromagnetic sensitivity of the industrial sensor is calibrated, the electromagnetic interference environment needs to be set and the electromagnetic sensitivity of the sensor is calibrated according to the sensor data collected under various interference environments. The manual calibration process is repetitive and redundant, the calibration efficiency is low, and it is not suitable for the overall integration and automation of the system. Therefore, the electromagnetic sensitivity of the sensor needs to be automatically calibrated.

[0030] Please refer to Figure 1 which shows the system block diagram of the industrial sensor electromagnetic sensitivity adaptive calibration system provided by the embodiment of the present application. The system includes the following modules:

[0031] The data acquisition module is used to acquire the output value of each sensor under different test conditions and take each test condition as a dimension; the output value of each dimension of a sensor is composed into a data point, and the electromagnetic sensitivity of each sensor is obtained.

[0032] The test of electromagnetic sensitivity of the industrial sensor is performed by building an electromagnetic interference test environment, using a radio frequency signal source to perform directional radiation on the sensor, injecting interference signals at different frequencies and different field strengths, and monitoring and collecting the output values of the sensor under the corresponding interference signals. The electromagnetic interference test environment includes different test conditions, as shown in the following table:

[0033] Table 1

[0034]

[0035] Among them, A1 and B1, A2 and B2, A3 and B3, A4 and B4, A5 and B5 are five different test conditions, and C1, C2, C3, C4 and C5 are the output values of a sensor under five test conditions. In the historical test, the output values of the sensor under different frequency bands and field strengths (different test conditions) are manually evaluated multiple times to obtain the electromagnetic sensitivity of the sensor.

[0036] Since subsequent clustering analysis is required for each sensor to cluster sensors with similar electromagnetic sensitivity in the historical test of electromagnetic sensitivity into a cluster, and the degree of interference of the sensor under different test conditions is not the same, some sensors are particularly sensitive to specific frequencies or field strengths, so it is necessary to distinguish the output values of the same sensor under each test condition; here, each test condition is taken as a dimension, such as frequency band A1 and field strength B1 corresponding to a dimension, the data value under this dimension is the output value of the sensor, frequency band A2 and field strength B2 corresponding to the second dimension, and so on. Thus, a plurality of dimensions composed of all test conditions is obtained. In the space composed of multiple dimensions, a data point is composed of the output value of each dimension (each test condition) of a sensor, and a data point corresponds to a sensor. All data points corresponding to the sensors in history are put into the space for subsequent analysis.

[0037] The electromagnetic sensitivity representation degree calculation module is configured to calculate the electromagnetic sensitivity representation degree of a dimension according to the distribution of the output value of the data point under the dimension.

[0038] Because the distribution of the sensitivity of the sensor to electromagnetic interference under each dimension is different, some dimensions may have a higher representation of electromagnetic sensitivity, and the proportion weight of such dimensions in clustering needs to be expanded, so that the clustering result of all data points obtained is more accurate for the division of sensors with different electromagnetic sensitivity.

[0039] This analysis examines the distribution of output values ​​from multiple sensors within each dimension. A larger distribution range of output values ​​from multiple sensors within a dimension indicates greater susceptibility to fluctuations in that dimension's data. This suggests that the sensors are generally more susceptible to electromagnetic interference in that dimension, making it a more volatile and electromagnetically sensitive dimension, thus providing a better representation of electromagnetic sensitivity. However, the distribution range may be expanded from individual data points and cannot reflect the overall general distribution within that dimension. Therefore, considering the uniformity of the distribution between output values ​​within that dimension, a more uniform distribution interval suggests a more likely uniform distribution across the entire range. Consequently, a larger distribution range coupled with a more uniform distribution interval indicates a better representation of the overall electromagnetic sensitivity of the output values ​​within the current dimension.

[0040] Therefore, the electromagnetic susceptibility characterization level of a dimension is calculated based on the distribution of output values ​​for different data points in a single dimension. Specifically, the output values ​​for a single dimension of different data points are arranged in ascending order to obtain the output value sequence corresponding to that dimension; the absolute value of the difference between every two adjacent output values ​​in the output value sequence is obtained and denoted as the data interval between every two adjacent output values; the difference between the maximum and minimum values ​​in the output value sequence is calculated and multiplied by the reciprocal of the standard deviation of the data interval between every two adjacent output values ​​to obtain the electromagnetic susceptibility characterization level of that dimension.

[0041] The specific calculation model for the degree of electromagnetic susceptibility characterization is as follows:

[0042] ,

[0043] in, This represents the degree of electromagnetic susceptibility characterization in the q-th dimension; and These represent the maximum and minimum values ​​of the output values ​​for different data points (different sensors) in this dimension, that is, the maximum and minimum values ​​in the output value sequence. The larger the data distribution range in this dimension, the more likely the output value of this dimension is to fluctuate. This indicates that the sensor is generally more susceptible to electromagnetic interference in this dimension, making it a variable dimension of electromagnetic sensitivity, and the better the characterization of electromagnetic sensitivity. This represents the set of absolute differences between any two adjacent output values ​​in the output value sequence; that is, the set of data intervals between any two adjacent output values. The standard deviation of all data intervals is denoted as . The smaller the standard deviation, the more uniform the overall distribution of the data within the distribution range. On the basis of a larger distribution range, this represents a better degree of representation of the overall electromagnetic susceptibility of the output values ​​in the current dimension.

[0044] This allows us to obtain the degree of electromagnetic susceptibility characterization for each dimension.

[0045] a clustering effect calculation module, configured to cluster the output values in each dimension to obtain different one-dimensional clustering clusters, and obtain the clustering effect of the dimension according to the distribution of the one-dimensional clustering clusters corresponding to the dimension and the distribution of the output values in the one-dimensional clustering clusters.

[0046] The electromagnetic sensitivity representation degree of each dimension is obtained as described above, but when the output values are in a concentrated block distribution in the overall range, the uniformity of the data may decrease, resulting in a decrease in the electromagnetic sensitivity representation degree obtained, but in fact, the concentrated block distribution mode does not affect the overall distribution of the output values in the distribution range, and the output values are more easily clustered in multi-dimensional clustering, so the clustering effect in the current dimension is further obtained in combination with the concentrated block distribution characteristics of the output values.

[0047] All the output values in a dimension are one-dimensionally clustered, the clustering algorithm is a K-means clustering algorithm, and a clustering cluster is obtained, which is denoted as a one-dimensional clustering cluster. After one-dimensional clustering, the one-dimensional clustering clusters are arranged according to the numerical values of the output values, so that a set of one-dimensional clustering clusters is obtained, and the one-dimensional clustering clusters in the set are arranged according to the numerical values of the output values. When the distribution interval between every two adjacent one-dimensional clustering clusters is larger, the distribution between the one-dimensional clustering clusters is more discrete, indicating that the block distribution characteristics of the current one-dimensional clustering cluster are more obvious, and the cluster division effect of the clustering result is better. On the other hand, the data distribution concentration degree in the one-dimensional clustering cluster is obtained. When the data distribution in each one-dimensional clustering cluster is more concentrated, the block distribution characteristics of the corresponding one-dimensional clustering cluster are more obvious, the one-dimensional clustering cluster effect of the dimension is better, and the electromagnetic sensitivity in the cluster is more concentrated and similar, and the electromagnetic sensitivity reliability in the cluster is higher.

[0048] Each one-dimensional clustering cluster corresponding to a dimension is grouped into a set of one-dimensional clustering clusters, and the one-dimensional clustering clusters in the set are arranged according to the numerical values of the output values. In the set of one-dimensional clustering clusters, the absolute value of the difference between the maximum value in a one-dimensional clustering cluster and the minimum value in the next one-dimensional clustering cluster is obtained, which is denoted as an interval distance. In a one-dimensional clustering cluster, the Euclidean distance between an output value and the output value closest to the output value is calculated, which is denoted as the minimum distance of the output value, and the mean value of the minimum distances of all the output values in the one-dimensional clustering cluster is calculated, which is denoted as the minimum distance mean value of the one-dimensional clustering cluster. The mean value of the minimum distance mean values of all the one-dimensional clustering clusters corresponding to the dimension is negatively correlated by using an exponential function with a natural constant as a base to obtain a mapping result, and the mapping result is multiplied by the mean value of the interval distances between every two adjacent one-dimensional clustering clusters in the set of one-dimensional clustering clusters corresponding to the dimension to obtain the clustering effect of the dimension.

[0049] The calculation model of the clustering effect of a dimension is specifically as follows:

[0050] ,

[0051] wherein, represents the clustering effect of the qth dimension, represents the number of one-dimensional clustering clusters obtained after one-dimensional clustering in this dimension, is the interval distance between the ith one-dimensional clustering cluster and the adjacent next one-dimensional clustering cluster, and is the absolute value of the difference between the maximum value of the output value in the ith one-dimensional clustering cluster and the maximum value of the output value in the adjacent next one-dimensional clustering cluster, is the mean value of the interval distance between every two adjacent one-dimensional clustering clusters in the one-dimensional clustering cluster set corresponding to this dimension. The greater the mean value of the interval distance, the more discrete the distribution between the one-dimensional clustering clusters corresponding to this dimension, the more obvious the block distribution characteristics of the current dimension, and the better the cluster class division effect of the clustering result.

[0052] represents the number of output values in the ith one-dimensional clustering cluster, represents the Euclidean distance between the jth output value in the ith one-dimensional clustering cluster and the output value closest to it, that is, the minimum distance of the jth output value, is the minimum distance mean value of the ith one-dimensional clustering cluster. The smaller the mean value, the more concentrated the distribution of the output values in the one-dimensional clustering cluster, and the more obvious the block distribution characteristics of the current dimension, is the mean value of the minimum distance mean values of all one-dimensional clustering clusters. The smaller the mean value, the stronger the concentration degree of the intra-cluster distribution of all one-dimensional clustering clusters in this dimension, and the better the clustering effect degree of this dimension; exp represents the exponential function with the natural constant as the base, which is used for inverse proportional normalization, that is, negative correlation mapping.

[0053] Thus, the clustering effect of each dimension can be obtained.

[0054] The clustering module is configured to obtain the weight of each dimension according to the electromagnetic sensitivity representation degree and the clustering effect of the dimension; and all data points are clustered to obtain different multi-dimensional clustering clusters in combination with the weight of each dimension.

[0055] The electromagnetic sensitivity representation degree and the clustering effect of each dimension are obtained as described above, and the weights of each dimension can be obtained by comprehensively combining the normalized values of the electromagnetic sensitivity representation degree and the normalized values of the clustering effect of each dimension. The stronger the electromagnetic sensitivity representation degree, the better the clustering effect, and the more suitable a dimension is for obtaining a higher weight in multi-dimensional clustering, so that the clustering result obtained is more accurate in the division of sensors with different electromagnetic sensitivities.

[0056] The calculation formula of the weight is specifically:

[0057] ,

[0058] wherein, represents the weight of the qth dimension, represents the control value domain magnitude is normalized by linear, further, the different multi-dimensional clustering clusters are obtained by clustering all data points combined with the weight of each dimension, specifically, when calculating the Euclidean distance between each two data points, the weight of different dimensions is multiplied on the calculation component of different dimensions, and the final Euclidean distance is obtained, so that the obtained clustering result is more accurate for the division of sensors with different electromagnetic sensitivity. For example, two-dimensional data points, the weights of two dimensions are a and b respectively, and the two data points are (x1, y1) and (x2, y2), then the Euclidean distance of the two data points is . Thus, the multi-dimensional clustering of the data points corresponding to each sensor is completed, wherein the clustering algorithm is K-means clustering algorithm. Thus, the clustering of the sensors tested in history is completed.

[0059] The electromagnetic sensitivity calibration module is used to obtain the electromagnetic sensitivity of each multi-dimensional clustering cluster according to the electromagnetic sensitivity of the sensors corresponding to each data point in each multi-dimensional clustering cluster; determine the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs, and the electromagnetic sensitivity of the multi-dimensional clustering cluster is the electromagnetic sensitivity of the current sensor.

[0060] The clustering result of the sensors tested in history is obtained, the clustering result divides the sensors with different electromagnetic sensitivity characteristics into different types through different multi-dimensional clustering clusters. Here, the output values of the newly tested sensor under multiple dimensions (multiple test conditions) are obtained, and the newly tested sensor is clustered and classified according to the similarity of the output values of the newly tested sensor and the output values in the multi-dimensional clustering cluster, and the latest tested sensor is recorded as the current sensor; in addition, the electromagnetic sensitivity of each multi-dimensional clustering cluster needs to be obtained, specifically, the mode of the electromagnetic sensitivity of the sensors corresponding to each data point in a multi-dimensional clustering cluster is taken as the electromagnetic sensitivity of the multi-dimensional clustering cluster.

[0061] The similarity degree of the output values under multiple dimensions in the data point corresponding to the current sensor and the data in each multi-dimensional clustering cluster is obtained, if the similarity degree of the output value of each dimension of the data point corresponding to the current sensor and the output value of each dimension in the multi-dimensional clustering cluster is higher, the possibility of belonging to the multi-dimensional clustering cluster is stronger; on the other hand, the Euclidean distance between the data point corresponding to the current sensor and the clustering center of each multi-dimensional clustering cluster in the sample space is obtained, the closer to the distance of the clustering center, the greater the possibility of belonging to the multi-dimensional clustering cluster.

[0062] Specifically, the reciprocal of the absolute value of the difference between the output value of a data point in a multi-dimensional clustering cluster in one dimension and the output value of the data point corresponding to the current sensor in the dimension is taken as the similarity degree corresponding to the dimension of the data point; the average of the similarity degrees corresponding to the dimension of all data points in the multi-dimensional clustering cluster is calculated and denoted as the dimension similarity degree of the dimension; the sum of the dimension similarity degrees of all dimensions is calculated and denoted as the overall similarity degree; the reciprocal of the distance between the data point corresponding to the current sensor and the cluster center of the multi-dimensional clustering cluster is normalized and added to the normalized value of the overall similarity degree to obtain the belonging possibility of the multi-dimensional clustering cluster; the multi-dimensional clustering cluster with the maximum belonging possibility is the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs.

[0063] The calculation model of the belonging possibility of the multi-dimensional clustering cluster is specifically:

[0064] ,

[0065] Among them, represents the belonging possibility of the zth multi-dimensional clustering cluster, that is, the possibility that the data point corresponding to the current sensor belongs to the zth multi-dimensional clustering cluster; is the number of dimensions, represents the number of data points in the zth multi-dimensional clustering cluster; represents the output value in the jth data point in the ith dimension, represents the output value in the data point corresponding to the current sensor in the ith dimension, represents the similarity degree corresponding to the ith dimension of the jth data point in the zth multi-dimensional clustering cluster. The greater the value, the more similar the output value of the ith dimension of the data point corresponding to the current sensor is to the output value of the ith dimension of the jth data point in the zth multi-dimensional clustering cluster, is the dimension similarity degree of the ith dimension. The stronger the dimension similarity degree, the more similar the output value of the ith dimension of the data point corresponding to the current sensor is to the output value of the ith dimension in the multi-dimensional clustering cluster. Then, the dimension similarity degrees corresponding to each dimension are summed to obtain the overall similarity degree . The greater the value, the greater the possibility that the data point corresponding to the current sensor belongs to the multi-dimensional clustering cluster;

[0066] represents the distance between the data point corresponding to the current sensor and the cluster center of the zth multi-dimensional clustering cluster. The distance is the Euclidean distance. The smaller the distance, the greater the possibility that the data point corresponding to the current sensor belongs to the zth multi-dimensional clustering cluster; and norm represents the normalization operation.

[0067] The electromagnetic sensitivity of the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs is the electromagnetic sensitivity of the current sensor, thereby realizing efficient adaptive calibration of the electromagnetic sensitivity of the automatic industrial sensor; in addition, the clustering results of the multi-dimensional clustering can be updated in real time by setting an update interval.

[0068] It should be noted that the above-mentioned order of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The above describes a specific embodiment of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or can be advantageous.

[0069] Each embodiment in the specification is described in a progressive manner, and the same or similar parts between each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments.

[0070] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. within the principles of the present application shall be included in the protection scope of the present application.

Claims

1. An industrial sensor electromagnetic sensitivity self-adaptive calibration system, characterized in that, The system comprises: a data acquisition module, configured to acquire output values of each sensor under different test conditions and take each test condition as a dimension; form a data point by the output values of each sensor under each dimension of a sensor, and acquire electromagnetic sensitivity of each sensor; an electromagnetic sensitivity representation degree calculation module, configured to calculate electromagnetic sensitivity representation degree of a dimension according to distribution of output values of different data points under the dimension; a clustering effect calculation module, configured to cluster the output values under the dimension to obtain different one-dimensional clustering clusters; and acquire clustering effect of the dimension according to distribution of the one-dimensional clustering clusters corresponding to the dimension and distribution of the output values in the one-dimensional clustering clusters; a clustering module, configured to acquire weight of the dimension according to electromagnetic sensitivity representation degree and clustering effect of the dimension; and cluster all data points by combining the weight of each dimension to obtain different multi-dimensional clustering clusters; an electromagnetic sensitivity calibration module, configured to acquire electromagnetic sensitivity of each multi-dimensional clustering cluster according to electromagnetic sensitivity of the sensors corresponding to the data points in the multi-dimensional clustering cluster; determine the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs, and electromagnetic sensitivity of the multi-dimensional clustering cluster is electromagnetic sensitivity of the current sensor.

2. The system of claim 1, wherein, The electromagnetic sensitivity of each sensor is acquired by: artificially evaluating the output values of each sensor under each test condition to acquire electromagnetic sensitivity of each sensor.

3. The system of claim 1, wherein the system is configured to: The electromagnetic sensitivity representation degree of the dimension is calculated according to the distribution of the output values of different data points under the dimension by: arranging the output values of different data points under the dimension in ascending order to obtain an output value sequence corresponding to the dimension; acquiring absolute values of differences between each two adjacent output values in the output value sequence, denoted as data intervals between the two adjacent output values; and multiplying a difference between a maximum value and a minimum value in the output value sequence and a reciprocal of a standard deviation of the data intervals between the two adjacent output values to obtain the electromagnetic sensitivity representation degree of the dimension.

4. The system of claim 1, wherein, The clustering effect of the dimension is acquired according to the distribution of the one-dimensional clustering clusters corresponding to the dimension and the distribution of the output values in the one-dimensional clustering clusters by: The one-dimensional clustering cluster corresponding to one dimension is grouped into a one-dimensional clustering cluster set, and the one-dimensional clustering clusters in the set are arranged according to the size of the output values; the absolute value of the difference between the maximum value in one one-dimensional clustering cluster and the minimum value in the one-dimensional clustering cluster after the one-dimensional clustering cluster in the one-dimensional clustering cluster set is obtained, and is recorded as the interval distance; the Euclidean distance between an output value and the output value closest to the output value in one one-dimensional clustering cluster is calculated, recorded as the minimum distance of the output value, and the mean value of the minimum distances of all output values in the one-dimensional clustering cluster is calculated, recorded as the minimum distance mean value of the one-dimensional clustering cluster; the mean value of the minimum distance mean values of all one-dimensional clustering clusters corresponding to the dimension is negatively correlated by using an exponential function with a natural constant as the base to obtain a mapping result, and the mapping result is multiplied by the mean value of the interval distances between every two adjacent one-dimensional clustering clusters in the one-dimensional clustering cluster set corresponding to the dimension to obtain the clustering effect of the dimension.

5. The system of claim 1, wherein, The weight of the dimension is obtained according to the electromagnetic sensitivity representation degree and the clustering effect of the dimension, and the weight of the dimension comprises: The weight of the dimension is obtained by adding and averaging the normalized value of the electromagnetic sensitivity representation degree and the normalized value of the clustering effect of the dimension.

6. The system of claim 1, wherein, The electromagnetic sensitivity of each multi-dimensional clustering cluster is obtained according to the electromagnetic sensitivity of the sensor corresponding to each data point in each multi-dimensional clustering cluster, and the electromagnetic sensitivity of the multi-dimensional clustering cluster comprises: The mode of the electromagnetic sensitivity of the sensor corresponding to each data point in one multi-dimensional clustering cluster is taken as the electromagnetic sensitivity of the multi-dimensional clustering cluster.

7. The system of claim 1, wherein the system is configured to: The multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs is determined, and the multi-dimensional clustering cluster comprises: The reciprocal of the absolute value of the difference between the output value of one data point in one multi-dimensional clustering cluster and the output value of the dimension corresponding to the current sensor is taken as the similarity degree of the dimension corresponding to the data point; the average value of the similarity degrees of all data points in the multi-dimensional clustering cluster is calculated, recorded as the dimension similarity degree of the dimension; the sum of the dimension similarity degrees of all dimensions is calculated, recorded as the overall similarity degree; the reciprocal of the distance between the data point corresponding to the current sensor and the clustering center of the multi-dimensional clustering cluster is normalized and added to the normalized value of the overall similarity degree to obtain the belonging possibility corresponding to the multi-dimensional clustering cluster; the multi-dimensional clustering cluster with the maximum belonging possibility is the multi-dimensional clustering cluster to which the data point corresponding to the current sensor belongs.

Images