Multi-sensor system detection method and device, computer equipment and storage medium

Abstract

The invention relates to a multi-sensor system detection method and device, computer equipment and a storage medium. The multi-sensor system detection method comprises the following steps: obtaining output test signals of multiple sensors; determining a measured value according to the output test signal; determining dynamic uncertainty according to the measured value; determining a support degreeaccording to the measured values of any two sensors and the corresponding dynamic uncertainty; detecting whether the support degree is matched with a preset value; and when the support degree is matched with the preset value, sending a consistency signal to a monitoring system. The dynamic uncertainty corresponding to each sensor is determined by outputting the measured value of the test signal; the dynamic uncertainty and the measured values of any two sensors are taken as conditions for solving the support degree; according to the matching relationship between the support degree and the preset value, whether the two sensors are consistent is judged, and the uncertainty is accurately obtained, so that the probability of misjudging a normally working sensor as a failure sensor is reduced,and the detection accuracy of the failure sensor is improved.

Description

technical field [0001] The invention relates to the technical field of sensor detection, in particular to a multi-sensor system detection method, device, computer equipment and storage medium. Background technique [0002] With the application of multi-sensor systems in traditional measurement systems more and more widely, it is necessary to explore the detection methods of sensor failure in multi-sensor systems and to check the consistency of the output data of each sensor. [0003] However, the traditional multi-sensor system inspection method requires a relatively large amount of calculations to measure the confidence distance between sensors, and it is difficult to achieve real-time processing. Each iteration requires re-comparison and sorting operations, which is required for a network with many sensors. The calculation time is still relatively long. Moreover, when data processing is performed, for example, the consistency judgment is made through the measured value and...

AI Technical Summary

Problems solved by technology

[0003] However, the traditional multi-sensor system inspection method requires a relatively large amount of calculations to measure the confidence distance between sensors, and it is difficult to achieve real-time processing. Each iteration requires re-comparison and sorting operations, which is required for a network with many sensors. The calculation time is still relatively long. Moreover, when data processing is performed, for example, the consistency judgment is made through the measured value and uncertainty of the sensor. When the performance of the sensor may change or the measured signal contains a large randomness, it is very difficult to Uncertainty is difficult to obtain, and it is difficult to calculate accurately, which increases the probability of misjudging a normal sensor as a failed sensor, thereby reducing the detection accuracy of a failed sensor

Images