A method for optimizing the accuracy of an electromagnetic sensor

Abstract

The present invention relates to the technical field of sensor signal processing, and particularly to a method for optimizing the accuracy of electromagnetic sensors. The present invention aims to solve the problem that electromagnetic sensors are easily affected by various interference factors in the actual working environment, resulting in a decrease in the accuracy and reliability of measurement results. The present invention first obtains the original measurement signal of the electromagnetic sensor; then uses an adaptive noise suppression algorithm to suppress environmental noise; and the abnormal data identification model monitors and eliminates abnormal data in real time; analyzes the sensor data through the parameter adjustment model and adjusts the working parameters of the sensor; uses the built-in reference signal source to dynamically calibrate the electromagnetic sensor regularly; finally, obtains the monitoring data of multiple electromagnetic sensors through the cloud platform for remote monitoring and dynamic calibration. The present invention combines the dynamic calibration function of the electromagnetic sensor with the remote monitoring and dynamic calibration of the cloud platform, further improving the measurement accuracy of the electromagnetic sensor.

Description

Technical Field

[0001] The present invention relates to the technical field of sensor signal processing, and particularly relates to a method for optimizing the accuracy of an electromagnetic sensor. Background Art

[0002] An electromagnetic sensor is a sensor that uses electromagnetic principles for measurement and detection. Its working principle is mainly based on electromagnetic induction and electrical signal processing. By sensing the magnetic field change of the target object and converting it into an electrical signal, and then through the signal processing process to obtain the final output. Electromagnetic sensors are widely used in industrial control, aerospace, and national defense security and other fields to measure physical quantities such as electromagnetic field strength, magnetic field direction, and magnetic flux density. These technical parameters are crucial for many key applications. However, in the actual working environment, electromagnetic sensors are easily affected by various interference factors, resulting in a decline in the accuracy and reliability of measurement results.

[0003] Therefore, it is urgent to develop a method for optimizing the accuracy of an electromagnetic sensor to overcome the shortcomings in the prior art. Summary of the Invention

[0004] To solve the problem that the accuracy and reliability of measurement results of electromagnetic sensors in the prior art are affected by various interference factors, the present invention provides such a method for optimizing the accuracy of an electromagnetic sensor, including the following steps:

[0005] S100: Obtain the original measurement signal of the electromagnetic sensor and preprocess the original measurement signal;

[0006] S200: Use an adaptive noise suppression algorithm to dynamically adjust the parameters of the adaptive filter to suppress environmental noise;

[0007] S300: Establish an abnormal data recognition model by collecting historical measurement data, and use the abnormal data recognition model to monitor and eliminate abnormal data in real time;

[0008] S400: Analyze the electromagnetic sensor data and adjust the working parameters of the electromagnetic sensor;

[0009] S500: Use the built-in reference signal source to dynamically calibrate the electromagnetic sensor regularly and update the calibration coefficient.

[0010] Further, the specific steps of using an adaptive noise suppression algorithm to dynamically adjust the parameters of the adaptive filter to suppress environmental noise are:

[0011] S201: Real-time collect the real-time environmental noise data in the working environment of the electromagnetic sensor, and use spectrum analysis to obtain the frequency characteristics of the noise;

[0012] S202: Adjust the parameters of the adaptive filter according to the frequency characteristics of the noise, including the cut-off frequency and the order, using the least mean square (LMS) algorithm;

[0013] S203: Compare the output signal of the electromagnetic sensor after being processed by the adaptive filter with the original signal, and calculate the residual noise level;

[0014] S204: When the residual noise level exceeds the preset threshold, repeat S201 until the residual noise is effectively suppressed.

[0015] Furthermore, the abnormal data recognition model is constructed using a random forest. The specific construction steps are as follows:

[0016] S301: Collect the data of the electromagnetic sensor under normal working conditions as the training set ;

[0017] S302: Perform Bootstrap sampling on the data in the training set to obtain sample sets ;

[0018] S303: Train a decision tree classifier for each sample set using the random forest algorithm ;

[0019] S304: Integrate the decision tree classifiers into a random forest model, which is the abnormal data recognition model;

[0020] S305: For the newly input electromagnetic sensor data , perform abnormal monitoring using the abnormal data recognition model:

[0021] Calculate the output in each decision tree classifier ;

[0022] Calculate the average output value:

[0023] ;

[0024] where is the predicted value of the abnormal data recognition model, is the number of samples in the training set , is the th predicted value output by the decision tree classifier;

[0025] If is less than the preset threshold, then it is determined that It is abnormal data; otherwise, it is normal data.

[0026] Furthermore, analyze the data measured by the electromagnetic sensor to adjust the working parameters of the electromagnetic sensor. Calculate the statistical characteristics of the measured data, including the mean and variance, and use the statistical characteristics as the optimization objective, and use the parameter adjustment model to adjust the working parameters of the electromagnetic sensor.

[0027] Furthermore, the parameter adjustment model is constructed based on the gradient descent algorithm, and the specific calculation formula is:

[0028] ;

[0029] is the mean square error loss function, which is used to represent the difference between the measured value and the true value of the electromagnetic sensor. Specifically, it is:

[0030] ;

[0031] is the adaptive weight coefficient, which is used to balance the weights of the loss function and the regularization term . Specifically, it is:

[0032] ;

[0033] is the L2 regularization term, which is used to constrain the change range of the parameter to avoid overfitting caused by too large parameter changes. Specifically, it is:

[0034] ;

[0035] Among them, is the optimization objective function. By solving , obtain when it is the smallest ; is the value of the electromagnetic sensor parameter that needs to be updated; is the number of samples of the electromagnetic sensor measurement data; is the th sample's standard value; is the predicted value of the parameter adjustment model for the th sample; is the th sample's predicted value; is the number of iterations of the parameter adjustment model, which is used to adjust the speed of weight change; is the initial working parameter of the electromagnetic sensor.

[0036] Furthermore, a reference signal source and a programmatically controllable switching circuit are built into the electromagnetic sensor;

[0037] The reference signal source is used to generate a stable reference signal;

[0038] The switching circuit is used to automatically switch the operating mode of the electromagnetic sensor, switching from the normal operating mode to the reference signal source or switching the reference signal source to the normal operating mode.

[0039] Furthermore, the electromagnetic sensor is dynamically calibrated periodically by the built-in reference signal source, which specifically includes the following steps:

[0040] S501: Periodically switch the operating mode of the electromagnetic sensor, connect the electromagnetic sensor to the built-in reference signal source, and the electromagnetic sensor is converted from the normal operating mode to the calibration mode;

[0041] S502: On the premise that the output signal of the electromagnetic sensor and the output signal of the reference signal source are synchronously collected, compare the output signal of the electromagnetic sensor and the output signal of the reference signal source at the same time, and calculate the error value between the two. The error value can be calculated using absolute error, relative error, and root mean square error;

[0042] S503: Calculate the calibration coefficient according to the error value and update the calibration parameters of the electromagnetic sensor;

[0043] S504: Switch the electromagnetic sensor back from the calibration mode to the normal operating mode, and continue to perform measurements using the updated calibration parameters.

[0044] Furthermore, the accuracy optimization method further includes, when using the electromagnetic sensor to monitor data, by deploying multiple electromagnetic sensors, the cloud platform obtains the monitoring data of the multiple electromagnetic sensors through wireless communication. The cloud platform remotely monitors and dynamically calibrates the electromagnetic sensors, and uses a data fusion algorithm to fuse the monitoring data. The cloud platform calculates high-precision measurement data and dynamically calibrates the multiple electromagnetic sensors.

[0045] Furthermore, the cloud platform remotely monitors and dynamically calibrates the electromagnetic sensors, which specifically includes the following steps:

[0046] S601: The electromagnetic sensor uploads measurement data and performance parameters to the cloud platform through wireless communication;

[0047] S602: The cloud platform receives and stores the monitoring data from multiple electromagnetic sensors and establishes a historical record of the performance of the electromagnetic sensors;

[0048] S603: The cloud platform analyzes the monitoring data of the electromagnetic sensors, and monitors the working status and measurement accuracy of each electromagnetic sensor in real time. Among them, when analyzing the monitoring data of the electromagnetic sensors, the average value, variance and fluctuation index of the monitoring data are statistically calculated;

[0049] S604: When it is found that the performance index of an electromagnetic sensor is abnormal, the cloud platform determines whether the electromagnetic sensor needs to be dynamically calibrated;

[0050] S605: For the situation where the electromagnetic sensor cannot be solved by its own dynamic calibration, the cloud platform generates optimized calibration parameters;

[0051] S606: The cloud platform issues a calibration instruction to the corresponding electromagnetic sensor, triggering the electromagnetic sensor to execute the dynamic calibration process;

[0052] S607: After the electromagnetic sensor completes the dynamic calibration, it feeds back the calibration result to the cloud platform, and the cloud platform records the calibration history.

[0053] Furthermore, the specific steps for the cloud platform to calculate high-precision measurement data include the following:

[0054] S611: The cloud platform receives the measurement data from multiple electromagnetic sensors, and performs time synchronization and data alignment;

[0055] S612: Use a data fusion algorithm to perform fusion processing on the measurement data of multiple electromagnetic sensors;

[0056] S613: Output the fused high-precision measurement result as the final measurement output;

[0057] S614: The cloud platform feeds back the final measurement output to each electromagnetic sensor as the basis for optimizing the performance of the electromagnetic sensor.

[0058] The beneficial effects of the present invention are as follows:

[0059] 1. The accuracy optimization method for a single electromagnetic sensor in the present invention starts from multiple aspects such as signal processing, noise suppression, anomaly rejection, parameter optimization, and dynamic calibration, and systematically solves the problems faced by the measurement accuracy and stability of electromagnetic sensors;

[0060] 2. The present invention automatically calculates the calibration coefficient and updates the calibration parameters by regularly switching the working mode of the electromagnetic sensor and comparing the reference signal and the measurement signal, enabling the electromagnetic sensor to achieve real-time dynamic calibration without manual intervention, effectively improving the long-term stability of the measurement accuracy;

[0061] 3. Based on the dynamic calibration function of the electromagnetic sensor itself, the present invention ensures the continuous optimization of the performance of a single sensor. The cloud platform, through remote monitoring and dynamic calibration functions, centrally manages multiple sensors for overall optimization and handling of special situations. The two complement each other, further improving the measurement accuracy of the electromagnetic sensor. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 is a flowchart of the method of the present invention;

[0063] Figure 2 is a flowchart of dynamically adjusting the filter parameters of the present invention;

[0064] Figure 3 is a flowchart of dynamically calibrating the electromagnetic sensor of the present invention;

[0065] Figure 4 is a flowchart of remotely monitoring and dynamically calibrating the electromagnetic sensor of the present invention;

[0066] Figure 5 is a flowchart of the cloud platform of the present invention calculating high-precision measurement data. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0067] The present invention will be further described below with reference to the drawings and embodiments. Embodiment 1

[0068] A method for optimizing the accuracy of an electromagnetic sensor, as shown in the attached Figure 1 figure, includes the following steps:

[0069] S100: Obtain the original measurement signal of the electromagnetic sensor and preprocess the original measurement signal.

[0070] S200: Use an adaptive noise suppression algorithm to dynamically adjust the parameters of the adaptive filter to suppress environmental noise.

[0071] S300: Establish an abnormal data recognition model by collecting historical data, and monitor and eliminate abnormal data in real time through the abnormal data recognition model.

[0072] S400: Analyze the electromagnetic sensor data and adjust the working parameters of the sensor.

[0073] S500: Use the built-in reference signal source to periodically perform dynamic calibration on the electromagnetic sensor and update the calibration coefficient.

[0074] In this embodiment, the original measurement signal of the electromagnetic sensor is obtained, the original measurement signal is preprocessed, and at the same time, the external electromagnetic environment parameters of the electromagnetic sensor are obtained, including the electromagnetic field intensity, frequency, etc.

[0075] In this embodiment, an adaptive noise suppression algorithm is used to dynamically adjust the parameters of the adaptive filter to suppress environmental noise.

[0076] Specifically, as shown in the appendix Figure 2 The specific steps of using the adaptive noise suppression algorithm to dynamically adjust the filter parameters to suppress environmental noise are as follows:

[0077] S201: Collect real-time environmental noise data in the working environment of the electromagnetic sensor in real time, and perform spectral analysis on the noise signal using the fast Fourier transform (FFT) to obtain the frequency characteristics of the noise;

[0078] S202: According to the noise frequency characteristics, use the least mean square (LMS) algorithm to adjust the parameters of the adaptive filter, including the cut-off frequency and order, etc.;

[0079] S203: Compare the output signal of the electromagnetic sensor processed by the adaptive filter with the original signal, and calculate the residual noise level;

[0080] S204: When the residual noise level exceeds the preset threshold, repeat S201 until the residual noise is effectively suppressed.

[0081] In this embodiment, by collecting historical data, an abnormal data recognition model is established, and through the abnormal data recognition model, abnormal data is monitored and eliminated in real time.

[0082] Specifically, the abnormal data recognition model is constructed using a random forest. The specific construction steps are as follows:

[0083] S301: Collect the monitoring data of the electromagnetic sensor in the normal working state as the training set ; In the normal samples, mark some monitoring data under abnormal conditions manually or simulate them as abnormal samples;

[0084] S302: Perform Bootstrap sampling on the training set data to obtain sample sets ;

[0085] S303: Train a decision tree classifier for each sample set using the random forest algorithm ;

[0086] When constructing the decision tree, for each node, randomly select features from all the features; for the selected features, calculate the Gini index, and select the optimal feature for classification;

[0087] S304: Integrate the decision tree classifier into a random forest model, i.e., the abnormal data recognition model;

[0088] S305: For newly input electromagnetic sensor data , use the abnormal data recognition model for anomaly monitoring:

[0089] Calculate the output in each decision tree classifier ;

[0090] Calculate the average output value:

[0091] ;

[0092] where, is the predicted value of the abnormal data recognition model, is the number of training set samples, is the predicted value output by the

[0093] If is less than the preset threshold, then determine as abnormal data, otherwise as normal data.

[0094] When constructing the abnormal data recognition model, by performing Bootstrap sampling and randomly selecting features to construct multiple decision trees, the single decision tree can avoid the defect of overfitting. Then, by aggregating multiple decision trees to form a random forest, the generalization ability and robustness of the abnormal data recognition model are improved.

[0095] In this embodiment, analyze the monitoring data of the electromagnetic sensor and adjust the working parameters of the electromagnetic sensor.

[0096] Analyze the electromagnetic sensor data to adjust the working parameters of the sensor. By calculating the statistical features of the electromagnetic sensor, including the mean and variance, etc., and using the statistical features as the optimization objective, use the parameter adjustment model to adjust the working parameters of the sensor.

[0097] Specifically, the parameter adjustment model is constructed based on the gradient descent algorithm, and the specific calculation formula is:

[0098] ;

[0099] is used to represent the mean square error between the measured value and the predicted value of the electromagnetic sensor, reflecting the prediction error of the parameter adjustment model under the current parameter , specifically: ​

[0100] ;

[0101] is the adaptive weight coefficient, which changes dynamically with the increase of the number of iterations and is used to balance the weights of the loss function and the regularization term specifically:

[0102] ;

[0103] is the L2 regularization term, which is used to constrain the change range of the parameter to avoid overfitting caused by too large parameter changes, specifically:

[0104] ;

[0105] Among them, is the optimization objective function. By solving , we get when it is the smallest ; is the value of the electromagnetic sensor parameter to be updated; is the number of samples of the electromagnetic sensor measurement data; is the th sample standard value; is the predicted value of the parameter adjustment model for the th sample; is the th sample predicted value; is the number of iterations of the parameter adjustment model, which is used to adjust the speed of weight change; is the initial working parameter of the electromagnetic sensor.

[0106] The electromagnetic sensor adjusts its own working parameters according to the parameters optimized by the parameter adjustment model, so as to improve the measurement accuracy. Among the parameters include key parameters such as the gain coefficient and bias voltage of the electromagnetic sensor. These key parameters are configured by setting the corresponding register values in the electromagnetic sensor chip. Applying the key parameters in the parameters to the electromagnetic sensor can improve the quality of the output data of the electromagnetic sensor.

[0107] In this embodiment, an internal reference signal source is used to dynamically calibrate the electromagnetic sensor regularly and update the calibration coefficient.

[0108] Specifically, there is an internal reference signal source and a programmatically controlled switching circuit in the electromagnetic sensor;

[0109] The reference signal source dynamically adjusts the generated calibration signal according to the working state of the electromagnetic sensor and environmental changes, etc., to generate a stable reference signal; by comparing the reference signal with the measurement signal of the sensor itself, the calibration coefficient can be accurately calculated to improve the calibration accuracy.

[0110] The switching circuit is used to automatically switch the working mode of the electromagnetic sensor, switching from the normal working mode to the reference signal source or switching the reference signal source to the normal working mode; by periodically starting the switching circuit, the working mode of the electromagnetic sensor is switched, so that it is connected to the built-in reference signal source for calibration, realizing the dynamic real-time calibration of the electromagnetic sensor, making the calibration of the electromagnetic sensor not rely on manual intervention, and improving the degree of automation.

[0111] Specifically, as shown in the appendix Figure 3 The dynamic calibration of the electromagnetic sensor by the built-in signal source at regular intervals specifically includes the following steps:

[0112] S501: Periodically switch the working mode of the electromagnetic sensor, connect the electromagnetic sensor to the built-in reference signal source, and the electromagnetic sensor is converted from the normal working mode to the calibration mode;

[0113] S502: On the premise that the output signal of the electromagnetic sensor and the output signal of the reference signal source are synchronously collected, that is, ensuring the time synchronization of the signals, compare the difference between the output signal of the electromagnetic sensor and the output signal of the reference signal source, calculate the error value between the two. The error value can be calculated using absolute error, relative error, and root mean square error. By calculating the error value, it can be judged whether the measurement accuracy of the electromagnetic sensor meets the requirements; if the output signal types of the electromagnetic sensor and the reference signal source are different, the signals are converted into digital signals to make them comparable in the same domain;

[0114] S503: Calculate the calibration coefficient according to the error value to adjust the measurement result of the electromagnetic sensor, make the measurement result closer to the standard value of the reference signal source, apply the calculated calibration coefficient to the electromagnetic sensor, and update its working parameters to improve the measurement accuracy of the electromagnetic sensor;

[0115] S504: Switch the working mode of the electromagnetic sensor from the calibration mode back to the normal working mode, and continue to measure using the updated calibration parameters.

[0116] By periodically switching the working mode of the electromagnetic sensor and comparing the reference signal and the measurement signal, automatically calculating the calibration coefficient and updating the calibration parameters, the electromagnetic sensor can realize the real-time dynamic calibration of the electromagnetic sensor without manual intervention, effectively improving the long-term stability of the measurement accuracy. Embodiment 2

[0117] In this embodiment, the accuracy optimization method further includes, when using electromagnetic sensors to monitor data, deploying multiple electromagnetic sensors, and the cloud platform uses wireless communication to obtain the monitoring data of the multiple electromagnetic sensors, remotely monitors and dynamically calibrates the electromagnetic sensors, and uses a data fusion algorithm to fuse the monitoring data, calculates high-precision measurement data, and dynamically calibrates the multiple electromagnetic sensors. Among them, the data fusion algorithm can adopt methods such as weighted average and Kalman filtering.

[0118] When deploying multiple electromagnetic sensors, the multiple electromagnetic sensors are optimized according to the geometric shape and size of the measured target, including array layouts such as circular and rectangular, and at the same time, the relative positions and spacings between the electromagnetic sensors in the array are optimized according to the characteristics of the measured target, including being evenly distributed around the target or densely arranged along a specific direction. At the same time, in addition to electromagnetic sensors, other types of auxiliary sensors can also be deployed, such as optical sensors and acoustic sensors, etc. Through the data fusion algorithm, the measurement data of different types of sensors are fused to further improve the overall measurement accuracy.

[0119] Specifically, as shown in the appendix Figure 4 The cloud platform remotely monitors and dynamically calibrates the electromagnetic sensors specifically includes the following steps:

[0120] S601: The electromagnetic sensors regularly upload the measurement data and performance parameters to the cloud platform through wireless communication.

[0121] S602: The cloud platform receives and stores the data from multiple electromagnetic sensors, and establishes a historical record of the performance of the electromagnetic sensors.

[0122] S603: The cloud platform analyzes the sensor data, and monitors the working status and measurement accuracy of each sensor in real time. Among them, the analysis of the sensor data uses indicators such as the mean, variance, and fluctuation index of the statistical sensor data.

[0123] S604: When the cloud platform finds that the performance index of a certain electromagnetic sensor is abnormal, the cloud platform triggers an alarm and records the abnormal situation. At the same time, the cloud platform judges whether dynamic calibration is required according to the analysis result of the electromagnetic sensor.

[0124] S605: For the situation where the electromagnetic sensor cannot solve the problem through its own dynamic calibration, the cloud platform calculates more optimized calibration parameters or strategies according to the analysis result of the sensor data, and generates customized calibration instructions.

[0125] S606: The cloud platform sends the calibration instructions to the corresponding electromagnetic sensor through wireless communication, triggering the electromagnetic sensor to execute the dynamic calibration process.

[0126] S607: The calibration process of the electromagnetic sensor is monitored in real time by the cloud platform. After the electromagnetic sensor completes dynamic calibration, the electromagnetic sensor feeds back the calibration results to the cloud platform, and the cloud platform records the calibration history of the electromagnetic sensor and the changes in calibration parameters. The cloud platform also provides a user interface to display the calibration status and historical data of the electromagnetic sensor for users to view.

[0127] The cloud platform discovers overall abnormal situations by centrally monitoring the performance status of multiple sensors. Based on the analysis results of a large amount of electromagnetic sensor data, the cloud platform gives more optimized calibration strategies or parameters; for some special situations that cannot be solved by dynamic calibration of certain electromagnetic sensors, the cloud platform can issue customized calibration instructions and record the global calibration history, providing a basis for the performance management and optimization of electromagnetic sensors.

[0128] Specifically, as shown in the appendix Figure 5 The cloud platform calculates high-precision measurement data, which specifically includes the following steps:

[0129] S611: The cloud platform receives measurement data from multiple electromagnetic sensors for the same target and performs preprocessing such as time synchronization and data alignment on these data;

[0130] S612: Use data fusion algorithms to fuse multiple sensor data;

[0131] S613: Output the fused high-precision measurement results as the final measurement output;

[0132] S614: The cloud platform feeds back the fusion result, that is, the final measurement output, to all electromagnetic sensors as the basis for performance optimization.

[0133] Based on the dynamic calibration function of the electromagnetic sensor itself, the performance of a single sensor is continuously optimized. The cloud platform centrally manages multiple sensors through remote monitoring and dynamic calibration functions for overall optimization and special situation handling. The two complement each other to further improve the measurement accuracy of electromagnetic sensors.

[0134] The above-described embodiments only represent the preferred embodiments of the present invention, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present invention, several deformations, improvements, and substitutions can be made, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the appended claims.

Claims

1. A method for optimizing the accuracy of an electromagnetic sensor, characterized in that, It includes the following steps: S100: Obtain the original measurement signal of the electromagnetic sensor and preprocess the original measurement signal; S200: Use an adaptive noise suppression algorithm to dynamically adjust the parameters of the adaptive filter and suppress environmental noise; S300: Establish an abnormal data identification model by collecting historical measurement data, and perform real-time monitoring and elimination of abnormal data through the abnormal data identification model; S400: Analyze the data of the electromagnetic sensor and adjust the working parameters of the electromagnetic sensor; S5 S500: Use the built-in reference signal source to periodically perform dynamic calibration on the electromagnetic sensor and update the calibration coefficient; Analyze the data measured by the electromagnetic sensor to adjust the working parameters of the electromagnetic sensor, calculate the statistical characteristics of the measurement data, including the mean value and variance, and use the parameter adjustment model to adjust the working parameters of the electromagnetic sensor with the statistical characteristics as the optimization target.

2. The precision optimization method for an electromagnetic sensor according to claim 1, wherein The specific steps for using the adaptive noise suppression algorithm to dynamically adjust the parameters of the adaptive filter to suppress environmental noise are: S201: Real-time collect the real-time environmental noise data in the working environment of the electromagnetic sensor, and use spectrum analysis to obtain the frequency characteristics of the noise; S202: According to the frequency characteristics of the noise, use the least mean square (LMS) algorithm to adjust the parameters of the adaptive filter, including the cut-off frequency and the order; S203: Compare the output signal of the electromagnetic sensor processed by the adaptive filter with the original signal, and calculate the residual noise level; S204: When the residual noise level exceeds the preset threshold, repeat S201 until the residual noise is effectively suppressed.

3. A method for optimizing the accuracy of an electromagnetic sensor according to claim 2, characterized in that, The abnormal data identification model is constructed using a random forest. The specific construction steps are: S301: Collect the data of the electromagnetic sensor in the normal working state as the training set D; S302: Bootstrap sample the data in the training set D to obtain N sample sets D = {D1, D2, …, D N}; S303: For each sample set D i train a decision tree classifier h using the random forest algorithm i ; S304: Integrate the decision tree classifier h i into a random forest model, i.e., an abnormal data recognition model; S305: For the newly input electromagnetic sensor data x, use the abnormal data identification model for abnormal monitoring: Calculate the output h of x in each decision tree classifier h i in i (x); Calculate the average output value: where H(x) is the predicted value of the abnormal data recognition model, N is the number of samples in the training set D, and h i (x) is the predicted value output by the i-th decision tree classifier; If H(x) is less than the preset threshold, then determine that x is abnormal data, otherwise it is normal data.

4. A method for optimizing the accuracy of an electromagnetic sensor according to claim 3, characterized in that, The parameter adjustment model is constructed based on the gradient descent algorithm. The specific calculation formula is: L new (θ) = w k L(θ) + λreg(θ); L(θ) is the mean square error loss function, which is used to represent the difference between the measured value and the true value of the electromagnetic sensor. Specifically: w k is the adaptive weight coefficient, which is used to balance the weights of the loss function L(θ) and the regularization term rge(θ), specifically: rge(θ) is the L2 regularization term, which is used to constrain the change range of the parameter θ to avoid overfitting caused by excessive parameter changes. Specifically: Among them, L new (θ) is the optimization objective function. By solving L new (θ), the value of θ when L new (θ) is minimized is obtained; θ is the value of the electromagnetic sensor parameter to be updated; N is the number of samples of the electromagnetic sensor measurement data; y i is the standard value of the i-th sample; f(x i , θ) is the predicted value of the parameter adjustment model for the i-th sample; x i is the predicted value of the i-th sample; k is the number of iterations of the parameter adjustment model, and α is used to adjust the speed of weight change; θ0 is the initial working parameter of the electromagnetic sensor.

5. A method for optimizing the accuracy of an electromagnetic sensor according to claim 4, characterized in that A reference signal source and a programmatically controllable switching circuit are built into the electromagnetic sensor; The reference signal source is used to generate a stable reference signal; The switching circuit is used to automatically switch the working mode of the electromagnetic sensor, switching the normal working mode to the reference signal source or switching the reference signal source to the normal working mode.

6. The accuracy optimization method for an electromagnetic sensor according to claim 5, characterized in that, Performing dynamic calibration on the electromagnetic sensor periodically through the built-in reference signal source specifically includes the following steps: S501: Periodically switch the working mode of the electromagnetic sensor, connect the electromagnetic sensor to the built-in reference signal source, and the electromagnetic sensor is converted from the normal working mode to the calibration mode; S502: On the premise of synchronously collecting the output signal of the electromagnetic sensor and the output signal of the reference signal source, compare the output signal of the electromagnetic sensor and the output signal of the reference signal source within the same time, and calculate the error value between the two. The error value can be calculated using absolute error, relative error, and root mean square error; S503: Calculate the calibration coefficient based on the error value, and update the calibration parameters of the electromagnetic sensor; S504: Switch the electromagnetic sensor from the calibration mode back to the normal working mode, and continue to perform measurements using the updated calibration parameters.

7. The precision optimization method for an electromagnetic sensor according to claim 6, characterized in that, The accuracy optimization method further includes when monitoring data using the electromagnetic sensor, by deploying multiple electromagnetic sensors, the cloud platform obtains the monitoring data of multiple electromagnetic sensors through wireless communication. The cloud platform remotely monitors and dynamically calibrates the electromagnetic sensors, and uses a data fusion algorithm to fuse the monitoring data. The cloud platform calculates high-precision measurement data and dynamically calibrates multiple electromagnetic sensors.

8. A method for optimizing the accuracy of an electromagnetic sensor according to claim 7, characterized in that The cloud platform remotely monitors and dynamically calibrates the electromagnetic sensor specifically includes the following steps: S601: The electromagnetic sensor uploads measurement data and performance parameters to the cloud platform through wireless communication; S602: The cloud platform receives and stores the monitoring data from multiple electromagnetic sensors, and establishes a historical record of the performance of the electromagnetic sensors; S603: The cloud platform analyzes the monitoring data of the electromagnetic sensor, and real-time monitors the working state and measurement accuracy of each electromagnetic sensor. Among them, when analyzing the monitoring data of the electromagnetic sensor, the average value, variance, and fluctuation index of the monitoring data are statistically calculated; S604: When it is found that the performance index of an electromagnetic sensor is abnormal, the cloud platform determines whether the electromagnetic sensor needs to be dynamically calibrated; S605: For the situation where the electromagnetic sensor cannot be solved by its own dynamic calibration, the cloud platform generates optimized calibration parameters; S606: The cloud platform sends a calibration instruction to the corresponding electromagnetic sensor, triggering the electromagnetic sensor to execute the dynamic calibration process; S607: After the electromagnetic sensor completes the dynamic calibration, it feeds back the calibration result to the cloud platform, and the cloud platform records the calibration history.

9. The accuracy optimization method for an electromagnetic sensor according to claim 8, wherein The cloud platform calculates high-precision measurement data specifically includes the following steps: S611: The cloud platform receives the measurement data from multiple electromagnetic sensors, and performs time synchronization and data alignment; S612: Use a data fusion algorithm to fuse the measurement data of multiple electromagnetic sensors; S613: Output the fused high-precision measurement result as the final measurement output; S614: The cloud platform feeds back the final measurement output to each electromagnetic sensor as the basis for optimizing the performance of the electromagnetic sensor.

Images