Neural network fusion and inversion model for NDIR sensor measurement

Sławomir Cięszczyk; Paweł Komada

doi:10.1117/12.2220064

17 December 2015 Neural network fusion and inversion model for NDIR sensor measurement

Sławomir Cięszczyk, Paweł Komada

Proceedings Volume 9816, Optical Fibers and Their Applications 2015; 98160R (2015) https://doi.org/10.1117/12.2220064
Event: 16th Conference on Optical Fibers and Their Applications, 2015, Lublin and Naleczow, Poland

Abstract

This article presents the problem of the impact of environmental disturbances on the determination of information from measurements. As an example, NDIR sensor is studied, which can measure industrial or environmental gases of varying temperature. The issue of changes of influence quantities value appears in many industrial measurements. Developing of appropriate algorithms resistant to conditions changes is key problem. In the resulting mathematical model of inverse problem additional input variables appears. Due to the difficulties in the mathematical description of inverse model neural networks have been applied. They do not require initial assumptions about the structure of the created model. They provide correction of sensor non-linearity as well as correction of influence of interfering quantity. The analyzed issue requires additional measurement of disturbing quantity and its connection with measurement of primary quantity. Combining this information with the use of neural networks belongs to the class of sensor fusion algorithm.

Citation Download Citation

Sławomir Cięszczyk and Paweł Komada "Neural network fusion and inversion model for NDIR sensor measurement", Proc. SPIE 9816, Optical Fibers and Their Applications 2015, 98160R (17 December 2015); https://doi.org/10.1117/12.2220064

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available