Electronic nose technology capable of realizing innate identification and acquired training and application thereof

Abstract

The invention belongs to the field of smell analysis and detection, and particularly discloses an electronic nose technology capable of realizing innate identification and acquired training, and an application thereof. The electronic nose technology comprises the following steps of: determining apertures of gas filtering modules in gas sensing arrays according to dynamic molecular dimensions of identification gases in a to-be-detected gas, and then introducing the to-be-detected gas into the gas sensing array; entering a tutor-free mode, respectively carrying out tutor-free mode identificationon signals acquired by means of each gas sensing array, and judging the concentrations of the identification gases corresponding to different molecular dimensions; and entering a tutor mode, taking amode identification mapping relationship between signals established by acquired training and sample characteristics as a database, and inputting all signals acquired by means of the gas sensing arrays to obtain the characteristics of the to-be-detected gas. The electronic nose technology separates gas molecules of different dimensions in a gradient mode based on the molecular sieve effect, so that a gas response signal is subjected to spectral expansion on the molecular size, multi-component gas detection in a complex atmosphere is achieved, and the selectivity of an electronic nose technology is improved.

Description

technical field [0001] The invention belongs to the field of odor analysis and detection, and more specifically relates to an electronic nose technology capable of innate recognition and acquired training and its application. Background technique [0002] Electronic nose technology is an odor recognition technology that distinguishes the difference between odors through gas sensing array + pattern recognition. This technology is achieved by testing samples and pattern training, and then pattern recognition, and this process is also called building a database. The establishment of the database is realized in the application after the electronic nose instrument is manufactured, so it is the recognition ability acquired through acquired training. [0003] In the odor detection application of the electronic nose, it is necessary to identify many characteristics of the sample. Some characteristics are not identified substances, but are composed of multiple different types of gas...

AI Technical Summary

Problems solved by technology

However, the acquired training process has the following defects: the process is cumbersome, the more the number of recognition samples is, the more acquired training is required, and the workload is huge; the sensor drift will increase the recognition error, and to a certain extent, the acquired training database needs to be rebuilt. Establishment; the acquired training databases of different sensors are not portable, resulting in the need to re-establish the database for sensor upgrades and replacements

Images