Data compensation method and system of sensor

By evaluating the sensor fatigue status and adjusting the sensitivity parameters in real time, the sensor's response sensitivity reduction caused by long-term work is solved, and the sensor's high accuracy and stability in long-term use is achieved, extending its service life and reducing maintenance costs.

CN119043401AActive Publication Date: 2024-11-29SHANGHAI NENGYU TECH DEV

Patent Information

Application Number
CN202411525983.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-30
Publication Date
2024-11-29
Estimated Expiration
2044-10-30

AI Technical Summary

Technical Problem

The sensor's response sensitivity decreases due to fatigue effects during long-term work, which affects the accuracy and stability of measurement results. The prior art is difficult to effectively solve the performance degradation problem in long-term continuous work.

Method used

By evaluating the working time and historical data of the sensor, adjusting its sensitivity parameters, monitoring the changes in output data in real time, comparing and correcting the data before and after adjustments, and restoring the initial response sensitivity of the sensor, using disturbance modeling and error feedback mechanisms for Adaptive adjustment.

Benefits of technology

It effectively delays the decay of sensor performance, extends service life, improves the continuity and stability of measurement results, ensures high accuracy of the sensor output data in harsh environments, reduces maintenance needs, and maintains the stable performance of the sensor.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119043401A_ABST
    Figure CN119043401A_ABST
Patent Text Reader

Abstract

The invention discloses a data compensation method and system for a sensor, and relates to the technical field of sensors, and the method comprises the steps: S1, evaluating the fatigue state of the sensor based on the working time length and historical data of the sensor, S2, adjusting the sensitivity parameter of the sensor according to the fatigue state, S3, monitoring the output data change of the sensor in real time, and S4, carrying out the adjustment of the sensitivity parameter of the sensor. S4, comparing data differences of the sensors before and after adjustment; S5, performing compensation correction based on the data differences to recover the initial response sensitivity of the sensors; according to the data compensation method and system of the sensor, the problem that the accuracy and the stability of a measurement result are reduced due to the fact that the performance of the sensor is often degraded along with the prolonging of the working time of the sensor, particularly the reduction of the response sensitivity of the sensor in the prior art is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of sensors, and in particular to a data compensation method and system for a sensor. Background Art

[0002] In modern industry and technology, sensors are widely used as key equipment in various scenarios, such as automation control, environmental monitoring, medical equipment, etc. However, as the working time of sensors increases, their performance tends to decline, especially their response sensitivity. This is mainly because sensors may be affected by fatigue effects during long-term continuous operation, resulting in reduced accuracy and stability of measurement results.

[0003] In order to deal with this problem, a variety of data compensation methods have been proposed in the prior art. These methods usually rely on algorithms or hardware means to monitor and correct the output data of the sensor in real time to improve its data accuracy. The advantage of this type of method is that it can automatically adjust the measurement data without stopping the sensor, thereby restoring or maintaining the performance level of the sensor to a certain extent. Although the existing compensation methods can improve the working efficiency of the sensor in the short term, there are still some inevitable problems. Sensors that work continuously for a long time will gradually lose sensitivity due to factors such as material aging and changes in environmental conditions, which will eventually affect the accuracy of their measurement results. Summary of the invention

[0004] The purpose of the present invention is to provide a sensor data compensation method and system to solve the problem that the prior art cannot solve that as the working time of the sensor increases, its performance tends to decline, especially its response sensitivity decreases, resulting in reduced accuracy and stability of the measurement results.

[0005] To achieve the above object, the present invention provides the following technical solution: a data compensation method for a sensor, the method comprising: S1, assessing the fatigue status based on the sensor’s working hours and historical data; S2. adjusting the sensitivity parameter of the sensor according to the fatigue state; S3, real-time monitoring of sensor output data changes; S4, comparing the data difference of the sensor before and after adjustment; S5, performing compensation correction based on the data difference to restore the initial response sensitivity of the sensor; Wherein, the S2 includes modeling the fatigue state of the sensor as a disturbance, and then estimating and compensating it. The specific dynamic formula is: , ; in, represents the estimated value of the output, represents the estimated value of the disturbance term, Indicates the gain value, represents the control input, represents the sensitivity gain, Represents the actual output data of the sensor, Indicates time; S5 includes adjusting the weight of the sensitivity parameter, and the specific formula is: ; in, represents the weight of the sensitivity parameter, represents the learning rate, represents the output error, represents the weight parameter value after one update, Indicates time, Indication error Sensitivity The partial derivative of .

[0006] Preferably, the S1 comprises: The specific formula for predicting the fatigue degree of the sensor is: ; in, Indicates that over time Changing fatigue status, Indicates time, represents the initial workload of the sensor, represents the elastic modulus of the sensor material, represents the rheological coefficient, An exponential expression of change over time.

[0007] Preferably, the specific formula for real-time monitoring of the output data change of the sensor in S3 is: ; in, Indicates the current output data of the sensor, Represents the trend item of the output data, represents the rate of fluctuation in the sensor output, represents the time step, Indicates time, Indicates that the sensor will +1 output data, represents the random fluctuation term.

[0008] Preferably, S4 includes: The KL divergence in information theory is used to compare the data differences before and after adjustment. The specific formula is: ; in, represents the probability distribution of sensor output before adjustment, represents the probability distribution of the adjusted sensor output, represents the difference in probability distribution of output data before and after adjustment, Represents the index of a data point.

[0009] Preferably, the S1 further comprises: Collect the historical data H of the sensor and standardize it to the range of [0,1] to obtain H norm ; The specific formula is: norm =(HH min ) / (H max -H min ), where H represents the historical data of the sensor, H norm represents the standardized historical data, H min Indicates the minimum value of historical data, H max Indicates the maximum value of historical data.

[0010] Calculate the weighted score W based on working hours T =T×k; If W T +H>th, the sensor is judged to be in a fatigue state, where th is a set safety threshold parameter, T represents the actual operating time of the sensor, k represents the fatigue acceleration constant obtained based on the sensor characteristics and experimental data, and W T It represents the weighted score calculated based on the working time of the sensor, and H represents the historical data of the sensor.

[0011] Preferably, S2 further includes: Save the sensor reading before adjusting the sensitivity and record it as ; After adjusting the sensor sensitivity, obtain the sensor reading under the new setting and record it as ; Use the difference formula to calculate the difference before and after the adjustment and check whether it exceeds the allowable offset, specifically: ; in, Indicates the difference between the baseline reading and the adjusted reading. Indicates the sensor reading after adjusting the sensitivity. Indicates the sensor reading before adjusting the sensitivity. Indicates the amount of allowable deviation; If exceeded, measures are taken to increase or decrease the sensor response speed.

[0012] Preferably, S2 further includes: Identify the current state vector S of the sensor, and then calculate the sensitivity gain value based on the current state vector S of the sensor. The specific formula is: G=α·S, where G represents the sensitivity gain value, α represents the empirical coefficient, and S represents the current state vector of the sensor; According to the calculated sensitivity gain G, adjust the sensitivity item in the sensor parameter table. The adjustment formula is: Z = 100 × ΔP / α; Among them, Z represents the quantized value of the state information set, α represents the empirical coefficient, and ΔP represents the change in the current output data of the sensor.

[0013] Preferably, the S5 further includes: The output sample points of the sensor are used to calculate the difference between the new parameters and the ideal state. The specific formula is: ; in, represents the difference function, Indicates the sensor at time The true response, Indicates time, Indicates the sensor at time The expected ideal response; According to the difference function The positive or negative value of determines whether it is necessary to continue adjusting or maintain the existing parameter configuration. The judgment conditions are: ,in, Indicates an operation instruction. If >0, then =1, indicating that the error is positive, restore the original setting. ≤0, the existing parameter configuration is maintained.

[0014] A data compensation system for a sensor, using the data compensation method for the sensor, the system comprising: Fatigue status assessment module, used to assess the fatigue status of sensors based on their working hours and historical data; A sensitivity adjustment module, used for adjusting the sensitivity parameters of the sensor according to the fatigue state; Data monitoring module, used to monitor the output data changes of the sensor in real time; A data difference comparison module is used to compare the data difference of the sensor before and after adjustment; The compensation correction module is used to perform compensation correction based on the data difference to restore the initial response sensitivity of the sensor.

[0015] It can be seen from the above technical solution that the present invention has the following beneficial effects: The data compensation method and system of the sensor evaluate the fatigue state of the sensor based on the working time and historical data of the sensor, adjust the sensitivity parameters of the sensor according to the fatigue state, monitor the output data changes of the sensor in real time, compare the data difference of the sensor before and after the adjustment, and perform compensation correction based on the data difference to restore the initial response sensitivity of the sensor, thereby ensuring the continuity and stability of the measurement result, being able to effectively delay the performance degradation of the sensor caused by long-term work, extending the service life of the sensor, reducing the need for frequent replacement and maintenance, and significantly improving the accuracy and reliability of the sensor output data. Even under harsh environmental conditions, the high accuracy of the measurement data can be maintained, the stable performance of the sensor can be guaranteed in various complex and dynamic working environments, and the reliability of the sensor in long-term work can be guaranteed. A closed-loop control feedback mechanism is used to ensure that the performance of the sensor is always maintained in the best state through continuous real-time monitoring and adjustment, thereby solving the problem that the prior art cannot solve that as the working time of the sensor is prolonged, its performance often deteriorates, especially its response sensitivity decreases, resulting in reduced accuracy and stability of the measurement result. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a flow chart of the method of the present invention; Figure 2 This is a schematic diagram of module connection of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] like Figure 1 As shown, a data compensation method for a sensor comprises the following steps: S1, assessing the fatigue status based on the sensor’s working hours and historical data; S2. adjusting the sensitivity parameter of the sensor according to the fatigue state; S3, real-time monitoring of sensor output data changes; S4, comparing the data difference of the sensor before and after adjustment; S5. Perform compensation correction based on the data difference to restore the initial response sensitivity of the sensor.

[0019] The working principle of this method is based on the fatigue phenomenon of sensors in long-term work. By analyzing the historical data and working hours of the sensor, its fatigue degree is calculated, and the sensitivity parameters are adjusted accordingly. After the sensitivity is adjusted, the effectiveness of the adjustment is determined through real-time monitoring and data comparison, and finally the initial sensitivity of the sensor is restored through the data compensation model. This method ensures that the sensor can maintain stable detection accuracy after long-term use. By automatically evaluating and adjusting the fatigue state of the sensor, it ensures that the sensor can maintain high-precision output for a long time. In response to the attenuation of the sensor sensitivity, its response characteristics are adjusted and compensated in time, so that the performance of the sensor after long-term work is close to the initial state, extending the service life of the sensor, reducing the replacement frequency, and reducing maintenance costs. By monitoring the output changes of the sensor in real time, performance anomalies can be quickly discovered and corrected, improving the reliability of the system.

[0020] Step S1 includes: The specific formula for predicting the fatigue degree of the sensor is: ; in, Indicates that over time Changing fatigue status, Indicates time, represents the initial workload of the sensor, represents the elastic modulus of the sensor material, represents the rheological coefficient, An exponential expression of change over time.

[0021] During long-term operation, the sensor will experience fatigue due to stress and environmental factors. According to the above formula, the fatigue state of the sensor The calculation is divided into two parts: the first part / Representing the stress response of the sensor material itself, part 2 It reflects the rheological effect of the sensor material under the action of external force as time goes by. By calculating these two items, the fatigue degree of the sensor can be accurately evaluated. This implementation method can accurately quantify the fatigue state of the sensor through a mathematical model, which is more accurate and scientific than simple empirical estimation. This method can not only improve the accuracy of fatigue assessment, but also provide reliable reference data for subsequent sensitivity adjustment and data compensation, ensuring the long-term stable operation of the sensor.

[0022] Step S2 includes modeling the fatigue state of the sensor as a disturbance, and then estimating and compensating it. The specific dynamic formula is: , ; in, represents the estimated value of the output, represents the estimated value of the disturbance term, Indicates the gain value, represents the control input, represents the sensitivity gain, Represents the actual output data of the sensor, Indicates time; This method models the fatigue state of the sensor as a disturbance and uses a dynamic estimation method to adjust the output value of the sensor. First, the difference between the sensor output and the actual output is calculated through the disturbance model, and the sensor output is corrected in real time according to the estimated value. Secondly, through the adaptive adjustment mechanism of the sensitivity parameters, the system can automatically optimize the parameters so that the sensor maintains an accurate sensitivity response during operation. Through real-time dynamic disturbance estimation and compensation, this method can effectively reduce the fluctuation of the sensor output and ensure the stability and accuracy of the output data. This method adopts a sensitivity weight adjustment strategy based on error feedback, so that the sensitivity parameters of the sensor can be adaptively optimized according to the actual situation, ensuring that the high-precision response of the sensor can be maintained under various conditions. Through the weight optimization of the sensitivity parameters, the output error of the sensor can be greatly reduced, thereby improving the accuracy and reliability of its measurement. The weight update mechanism based on error feedback can ensure that the sensitivity parameters are continuously optimized, thereby minimizing the output error and improving the compensation effect.

[0023] The specific formula for real-time monitoring of the output data change of the sensor in step S3 is: ; in, Indicates the current output data of the sensor, Represents the trend item of the output data, represents the rate of fluctuation in the sensor output, represents the time step, Indicates time, Indicates that the sensor will +1 output data, represents the random fluctuation term; The formula for sensor output data changing over time It describes the changing trend of the sensor output within the time step, including the trend term and fluctuation term of the current data. By calculating the formula in real time, the system can capture the dynamic changes of the sensor output data and use these data for real-time monitoring. This method is applicable to various sensor types, especially scenarios where the output is greatly affected by the external environment. By real-time monitoring of the trend and fluctuation of the sensor output, the system can quickly respond to the performance fluctuation of the sensor and provide efficient compensation measures. By detecting the fluctuation rate of the output data, possible abnormal conditions of the sensor can be discovered in time, and measures can be taken in advance to avoid further deterioration of the sensor performance.

[0024] Step S4 includes: The KL divergence in information theory is used to compare the data differences before and after adjustment. The specific formula is: ; in, represents the probability distribution of sensor output before adjustment, represents the probability distribution of the adjusted sensor output, represents the difference in probability distribution of output data before and after adjustment, Represents the index of a data point.

[0025] KL divergence is an indicator used to measure the difference between two probability distributions. In this embodiment, by calculating the difference in probability distribution of the output data before and after the sensor is adjusted, the system can quantitatively evaluate the effect of data compensation. Specifically, the smaller the KL divergence, the smaller the difference between the output data of the sensor after adjustment and the output data before adjustment, and the better the compensation effect. KL divergence provides an accurate mathematical tool for measuring the changes in the output data of the sensor before and after adjustment, which can intuitively reflect the effect of compensation. This method can sensitively capture the subtle distribution differences of sensor data and ensure that the compensated data is restored to the normal working state of the sensor as much as possible. By continuously adjusting the compensation parameters to minimize the KL divergence, the system can continuously optimize the compensation effect and ensure that the sensor can maintain high-precision output under different working conditions.

[0026] Step S5 includes adjusting the weight of the sensitivity parameter, and the specific formula is: ; in, represents the weight of the sensitivity parameter, represents the learning rate, represents the output error, represents the weight parameter value after one update, Indicates time, Indication error Sensitivity The partial derivative of In this method, the weight of the sensitivity parameter is adaptively adjusted according to the error of the sensor output. Specifically, the error E is calculated by the difference between the actual output and the expected output of the sensor, and the partial derivative measures the impact of the sensitivity parameter on the error. By using the gradient descent method, the sensitivity weight is adjusted at each iteration to reduce the error. This method can ensure that the sensitivity parameters can be continuously optimized and ultimately achieve compensation for sensor performance. By introducing the learning rate, the system can dynamically adjust the sensitivity weight according to the actual error of the sensor, ensuring that the sensitivity parameters are continuously optimized and reducing errors. The error-based adaptive adjustment strategy can improve the accuracy of sensor data compensation, making the sensor output data more stable and reliable. Through the dynamic adjustment of the sensitivity weight, the system can adapt to the performance changes of the sensor under different environmental conditions and ensure that the sensor sensitivity remains in the best state.

[0027] Step S1 also includes: Collect the historical data H of the sensor and standardize it to the range of [0,1] to obtain H norm ; The specific formula is: norm =(HH min ) / (H max -H min ), where H represents the historical data of the sensor, H norm represents the standardized historical data, H min Indicates the minimum value of historical data, H max Indicates the maximum value of historical data; Calculate the weighted score W based on working hours T =T×k; If W T +H>th, the sensor is judged to be in a fatigue state, where th is a set safety threshold parameter, T represents the actual operating time of the sensor, k represents the fatigue acceleration constant obtained based on the sensor characteristics and experimental data, and W T represents the weighted score calculated based on the sensor working time, and H represents the historical data of the sensor; This method quantifies the performance changes of sensors by collecting and standardizing their historical data. After the historical data is standardized, the influence of the data range can be eliminated and unified to the range of [0,1], which is convenient for comparison with other calculation results. The working time is combined with the fatigue acceleration constant to calculate the weighted score, which is used to evaluate the workload of the sensor. tBy accumulating with historical data and comparing with the threshold, it can be determined whether the sensor has entered a fatigue state. Once the weighted score exceeds the threshold, the system will trigger a compensation or adjustment mechanism to correct the sensor performance. By standardizing the historical data, data of different scales can be unified, enhancing the accuracy of data comparison, and facilitating application in a variety of scenarios. Based on a comprehensive evaluation of working hours and historical data, the fatigue state of the sensor can be predicted in advance to prevent a sudden drop in performance. By setting a reasonable threshold, the system can automatically start the data compensation and sensitivity adjustment mechanism before the sensor enters a fatigue state, thereby improving the reliability and service life of the sensor. By combining working hours and historical data, the system can dynamically evaluate the working status of the sensor, provide a more flexible monitoring solution, and adapt to complex working environments.

[0028] Step S2 also includes: Save the sensor reading before adjusting the sensitivity and record it as ; After adjusting the sensor sensitivity, obtain the sensor reading under the new setting and record it as ; Use the difference formula to calculate the difference before and after the adjustment and check whether it exceeds the allowable offset, specifically: ; in, Indicates the difference between the baseline reading and the adjusted reading. Indicates the sensor reading after adjusting the sensitivity. Indicates the sensor reading before adjusting the sensitivity. Indicates the amount of allowable deviation; If exceeded, the sensor response speed is enhanced or weakened; This method determines whether the sensitivity adjustment is effective by saving the sensor readings before and after the sensitivity adjustment and calculating the difference between the two. If the difference exceeds the preset allowable deviation, it indicates that the sensitivity adjustment of the sensor is not within a reasonable range, which may affect the accuracy of the measurement. Therefore, the system will further adjust the sensor's response speed according to the size of the difference to ensure that it operates within the ideal working range. By analyzing the difference between the readings before and after the adjustment, the effect of the sensor sensitivity adjustment can be accurately judged, and appropriate measures can be taken according to the out-of-tolerance situation to avoid excessive performance deviation. If the difference exceeds the allowable range, the system can automatically increase or weaken the sensor's response speed to ensure that the dynamic characteristics of the sensor remain within the expected range and improve the stability of the system. By calculating the difference before and after the adjustment, the system can monitor the adjustment effect in real time to ensure that the compensation adjustment of the sensor is within the normal range, and there will be no over-correction or under-correction problems. The judgment method based on the reading difference makes the sensor sensitivity adjustment process more intelligent, and can be optimized and adjusted in real time according to the actual situation to improve the measurement accuracy.

[0029] Step S2 also includes: Identify the current state vector S of the sensor, and then calculate the sensitivity gain value based on the current state vector S of the sensor. The specific formula is: G=α·S, where G represents the sensitivity gain value, α represents the empirical coefficient, and S represents the current state vector of the sensor; According to the calculated sensitivity gain G, adjust the sensitivity item in the sensor parameter table. The adjustment formula is: Z = 100 × ΔP / α; Among them, Z represents the quantized value of the state information set, α represents the empirical coefficient, and ΔP represents the change in the current output data of the sensor; The method first identifies the current state vector of the sensor and then uses the empirical coefficient to calculate the sensitivity gain. The state vector represents the multi-dimensional state information of the sensor, including temperature, pressure, time and other related parameters. The calculation of the sensitivity gain helps to dynamically adjust the sensitivity of the sensor to adapt it to the current working conditions. Then, the state information is quantified by the formula Z=100×ΔP / α, so as to further adjust the sensitivity parameters of the sensor. This dynamic adjustment strategy based on sensitivity gain can ensure that the sensor maintains high-precision output under different environmental conditions. Based on the real-time state vector of the sensor, the method dynamically adjusts the sensitivity parameters by calculating the sensitivity gain to ensure that the sensor can adapt to environmental changes. By quantifying the change in the sensor output data, the system can accurately evaluate the state of the sensor and make precise compensation, which improves the accuracy of the compensation effect. The calculation and adjustment of the sensitivity gain provides a scientific basis for sensitivity control and ensures the stability and accuracy of the sensor after long-term use. By analyzing the state vector, the method can adapt to various working conditions, automatically adjust the sensor sensitivity, and enhance the adaptability and flexibility of the system.

[0030] Step S5 also includes: The output sample points of the sensor are used to calculate the difference between the new parameters and the ideal state. The specific formula is: ; in, represents the difference function, Indicates the sensor at time The true response, Indicates time, Indicates the sensor at time The expected ideal response; According to the difference function The positive or negative value of determines whether it is necessary to continue adjusting or maintain the existing parameter configuration. The judgment conditions are: ,in, Indicates an operation instruction. If >0, then =1, indicating that the error is positive, restore the original setting. ≤0, keep the existing parameter configuration; In this method, the system calculates the difference between the actual response of the sensor and the ideal response. , to determine whether the actual output of the sensor is consistent with the expected. When the difference is positive, it means that the response of the sensor deviates from the ideal state. At this time, the system will decide whether to restore to the original parameter setting based on the judgment conditions; when When ≤0, it indicates that the sensor is in normal or better than the ideal state, so the existing parameter configuration can be maintained. Through the difference function, the system can detect the output deviation of the sensor in real time to ensure that timely measures can be taken when the output deviates. The system can automatically determine whether the parameters need to be adjusted based on the positive and negative values, reduce unnecessary adjustment operations, and improve the efficiency of system operation. By judging the conditions, when the sensor is in a normal state, the system will not perform unnecessary operations to ensure the stable operation of the system. The system can quickly return to the original set value to ensure that the sensor can automatically make compensation adjustments when errors occur to ensure the accuracy of the output.

[0031] The present invention also provides a sensor data compensation system, which adopts the sensor data compensation method, and the system comprises: Fatigue status assessment module, used to assess the fatigue status of sensors based on their working hours and historical data; A sensitivity adjustment module, used for adjusting the sensitivity parameters of the sensor according to the fatigue state; Data monitoring module, used to monitor the output data changes of the sensor in real time; A data difference comparison module is used to compare the data difference of the sensor before and after adjustment; The compensation correction module is used to perform compensation correction based on the data difference to restore the initial response sensitivity of the sensor.

[0032] Based on the actual working conditions of the sensor, the system dynamically evaluates its fatigue state and makes compensation corrections in real time through sensitivity adjustment and data monitoring. The data difference comparison module compares the data before and after adjustment, calculates the difference value, and compensates the output of the sensor in combination with the compensation correction module, thereby ensuring that the sensor maintains its accuracy and stability during its service life. Through the collaborative work of the fatigue state evaluation module and the sensitivity adjustment module, the system can automatically detect the fatigue state of the sensor and dynamically adjust the sensitivity, reduce human intervention, and enhance the automation capability of the system. The data monitoring module and the difference comparison module ensure real-time monitoring of the output data of the sensor, and provide a reliable reference for compensation correction through scientific data comparison methods, and finally achieve high-precision data compensation. Through the division of labor and cooperation of multiple functional modules, the system realizes a complete compensation process, covering the entire process from fatigue evaluation, sensitivity adjustment, data monitoring to compensation correction, ensuring the stable operation of the sensor. Through timely adjustment and compensation correction, the system can delay the performance degradation of the sensor, extend its service life, reduce the frequency of sensor replacement, and reduce maintenance costs.

[0033] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A sensor data compensation method, characterized in that: The method comprises: S1, assessing the fatigue status based on the sensor’s working hours and historical data; S2. adjusting the sensitivity parameter of the sensor according to the fatigue state; S3, real-time monitoring of sensor output data changes; S4, comparing the data difference of the sensor before and after adjustment; S5, performing compensation correction based on the data difference to restore the initial response sensitivity of the sensor; Wherein, the S2 includes modeling the fatigue state of the sensor as a disturbance, and then estimating and compensating it. The specific dynamic formula is: , ; in, represents the estimated value of the output, represents the estimated value of the disturbance term, Indicates the gain value, represents the control input, represents the sensitivity gain, Represents the actual output data of the sensor, Indicates time; S5 includes adjusting the weight of the sensitivity parameter, and the specific formula is: ; in, represents the weight of the sensitivity parameter, represents the learning rate, represents the output error, represents the weight parameter value after one update, Indicates time, Indication error Sensitivity The partial derivative of .

2. The sensor data compensation method according to claim 1, characterized in that: The S1 includes: The specific formula for predicting the fatigue level of the sensor is: ; in, Indicates that over time Changing fatigue status, Indicates time, represents the initial workload of the sensor, represents the elastic modulus of the sensor material, represents the rheological coefficient, An exponential expression of change over time.

3. The sensor data compensation method according to claim 1, characterized in that: The specific formula for the real-time monitoring of the output data change of the sensor by S3 is: ; in, Indicates the current output data of the sensor, Represents the trend item of the output data, represents the rate of fluctuation in the sensor output, represents the time step, Indicates time, Indicates that the sensor will +1 output data, represents the random fluctuation term.

4. The sensor data compensation method according to claim 1, characterized in that: The S4 includes: The KL divergence in information theory is used to compare the data differences before and after adjustment. The specific formula is: ; in, represents the probability distribution of sensor output before adjustment, represents the probability distribution of the adjusted sensor output, represents the difference in probability distribution of output data before and after adjustment, Represents the index of a data point.

5. The sensor data compensation method according to claim 1, characterized in that: The S1 further comprises: Collect the historical data H of the sensor and standardize it to the range of [0,1] to obtain H norm ; The specific formula is: norm =(HH min ) / (H max -H min ), where H represents the historical data of the sensor, H norm represents the standardized historical data, H min Indicates the minimum value of historical data, H max Indicates the maximum value of historical data; Calculate the weighted score W based on working hours T =T×k; If W T +H>th, the sensor is judged to be in a fatigue state, where th is a set safety threshold parameter, T represents the actual operating time of the sensor, k represents the fatigue acceleration constant obtained based on the sensor characteristics and experimental data, and W T It represents the weighted score calculated based on the working time of the sensor, and H represents the historical data of the sensor.

6. The sensor data compensation method according to claim 1, characterized in that: The S2 further includes: Save the sensor reading before adjusting the sensitivity and record it as ; After adjusting the sensor sensitivity, obtain the sensor reading under the new setting and record it as ; Use the difference formula to calculate the difference before and after the adjustment and check whether it exceeds the allowable offset, specifically: ; in, Indicates the difference between the baseline reading and the adjusted reading. Indicates the sensor reading after adjusting the sensitivity. Indicates the sensor reading before adjusting the sensitivity. Indicates the amount of allowable deviation; If exceeded, measures are taken to increase or decrease the sensor response speed.

7. The sensor data compensation method according to claim 1, characterized in that: The S2 further includes: Identify the current state vector S of the sensor, and then calculate the sensitivity gain value based on the current state vector S of the sensor. The specific formula is: G=α·S, where G represents the sensitivity gain value, α represents the empirical coefficient, and S represents the current state vector of the sensor; According to the calculated sensitivity gain G, adjust the sensitivity item in the sensor parameter table. The adjustment formula is: Z = 100 × ΔP / α; Among them, Z represents the quantized value of the state information set, α represents the empirical coefficient, and ΔP represents the change in the current output data of the sensor.

8. The sensor data compensation method according to claim 1, characterized in that: The S5 further includes: The output sample points of the sensor are used to calculate the difference between the new parameters and the ideal state. The specific formula is: ; in, represents the difference function, Indicates the sensor at time The true response, Indicates time, Indicates the sensor at time The expected ideal response; According to the difference function The positive or negative value of determines whether it is necessary to continue adjusting or maintain the existing parameter configuration. The judgment conditions are: ,in, Indicates an operation instruction. If >0, then =1, indicating that the error is positive, restore the original setting. ≤0, the existing parameter configuration is maintained.

9. A sensor data compensation system, using the sensor data compensation method according to any one of claims 1 to 8, characterized in that: The system comprises: Fatigue status assessment module, used to assess the fatigue status of sensors based on their working hours and historical data; A sensitivity adjustment module, used for adjusting the sensitivity parameters of the sensor according to the fatigue state; Data monitoring module, used to monitor the output data changes of the sensor in real time; A data difference comparison module is used to compare the data difference of the sensor before and after adjustment; The compensation correction module is used to perform compensation correction based on the data difference to restore the initial response sensitivity of the sensor.

Citation Information

Patent Citations

  • Steel bridge finite element model correction method based on non-uniform temperature response monitoring value

    CN105956218A

  • Sensor initial calibration method

    CN113252096A

  • Ultrasonic sensor sensitivity compensation curve construction method based on BP neural network

    CN115098824A

  • Control method and system of electro-hydraulic servo dynamic fatigue testing machine

    CN118090484A

  • Compensation method for pressure measurement

    CN118395087A

Cited By

  • Software test data management system and method based on cloud platform

    CN120276997A

  • A software testing data management system and method based on cloud platform

    CN120276997B

  • Fluid sample loading dynamic calibration method and system based on sensor

    CN120594861A

  • A sensor-based dynamic calibration method and system for fluid loading

    CN120594861B