12–17 Jun 2016
University of Ottawa
America/Toronto timezone
Welcome to the 2016 CAP Congress! / Bienvenue au congrès de l'ACP 2016!

SELECTION OF OPTIMAL REGRESSION MODELS FOR INSTRUMENT CALIBRATION

16 Jun 2016, 09:45
15m
SITE J0106 (University of Ottawa)

SITE J0106

University of Ottawa

SITE Building, 800 King Edward Ave, Ottawa, ON
Oral (Non-Student) / orale (non-étudiant) Instrumentation and Measurement Physics / Physique des instruments et mesures (DIMP-DPIM) R1-6 General Instrumentation I (DIMP) / Physique générale des instruments I (DPIM)

Speaker

Nimal De Silva (University of Ottawa)

Description

Instrument calibration is an integral part of many physical and chemical measurement systems. Commonly, instrument response is plotted against a series of known values (standards) of the measurand to establish a calibration line by linear regression. For example, the relationship is established between emitted light intensity versus concentration in Emission Spectrometry, or ion counts against concentration in quantitative Mass Spectrometry. In the simplest and ideal situation, signal intensity is assumed to be directly proportional to the concentration. In reality, however, significant errors can originate from uncertainty of the measurement, as well as non-linear dependence of the response within the chosen dynamic range. In this presentation, we will discuss about the selection of optimal calibration models and evaluate their relative effectiveness using simulated and practical examples.

Author

Nimal De Silva (University of Ottawa)

Co-authors

Jack Cornett (University of Ottawa) Nicolas Guérin (CNL) Ping Zhang (University of Ottawa) Smita Mohanty (University of Ottawa)

Presentation materials

There are no materials yet.