Gas sensor array multi-fault isolation algorithm based on reconstruction contribution

Abstract

The invention discloses a gas sensor array multi-fault isolation algorithm based on reconstruction contribution. The gas sensor array multi-fault isolation algorithm comprises the steps or presettingthe number initial value of fault sensors and the judgment threshold value of the SPE statistical magnitude control limit for a gas sensor array; acquiring a fault signal of the gas sensor array, andconstructing all fault directions and a corresponding amplitude set; for each group of fault directions and corresponding amplitudes, solving reconstructed fault signals and SPE statistics; calculating a contribution rate based on the SPE statistics corresponding to the reconstructed fault signal, and determining a group of fault directions with the maximum contribution rate and the SPE statistics; judging whether the SPE statistic is smaller than a judgment threshold value or not; if not, adding one to the number of the fault sensors, updating the preset initial value of the number of the fault sensors, and performing iterative processing; and if so, outputting the determined current fault direction and the number of the current fault sensors. According to the method, the trailing problemof a traditional contribution graph method is well eliminated, mutual interference among multiple variables is avoided, and the accuracy of multi-fault isolation is greatly improved.

Description

technical field [0001] The invention belongs to the technical field of sensor fault diagnosis and relates to a gas sensor array multi-fault isolation algorithm based on reconstruction contribution. Background technique [0002] The gas sensor array is the source of the machine olfactory system to obtain the data information of the measured target gas, and the reliability of its detection results plays an extremely important role in the comprehensive performance evaluation of the entire olfactory sensing system. Therefore, it is of great significance to realize accurate fault location of the faulty gas sensor. [0003] In the current field of multi-sensor fault isolation, research on related algorithms is limited. The most classic fault isolation method is the contribution graph method based on process variable analysis, but this method has fault tailing effect. Due to the randomness of external environmental stimuli, the faulty sensor is easily interfered by the sensor wit...

AI Technical Summary

Problems solved by technology

However, the traditional fault isolation algorithm based on the contribution graph method has a fault tailing effect, that is, the fault variable affects the contribution of the non-fault variable, which makes the accuracy of multi-fault isolation of the gas sensor array low.

However, although the improved algorithm combined with the idea of ​​reconstruction proposed later improves the accuracy of fault isolation to a certain extent, it still cannot completely eliminate the smearing effect.

The current patents have not been able to effectively locate the faults of multiple gas sensors at the same time, thereby solving the problem of multiple fault isolation of the gas sensor array

Images