Polynomial Identification of Strain Gauge Thermal Output...

Igor Vujović

Ivica Kuzmanić

Zlatan Kulenović

Abstract

Strain gauges are used in different areas, especially in thedesign and development of new technical constructions andmode/testing. Also, strain gauges are inc01porated in the functionalpart of many instruments and devices. They are mostlyused as sensors in transducers designed to measure such mechanicalquantities as forces, moments, pressures, accelerations,etc. They have an important role in shipbuilding and marinetransport in general. In this paper we have suggested andshown an approach to the identification of strain gauge thermaloutput curve on the example of a product available at themarket. Neural network simulated in MatLab has been applied.The neural network has been adapted to simulate a realsystem with 1 o-9 order of magnitude error_ As it is well known,strain gauges measure deformations of 1 o-o order of magnitude.It is obvious that the network eJTOr cannot influence the measurementresults because of its being smaller by three orders ofmagnitude.