Fabrication and Failure Prediction of Carbon-alum solid composite electrolyte based humidity sensor using ANN

Abstract

Using low-cost materials, carbon, and alum, a new solid composite electrolyte system was fabricated and characterized using various techniques. Complex impedance spectroscopy was used to measure its ionic conductivity. A maximum conductivity of 3.4×10−5 S/cm was obtained when alum was doped with 40% carbon. An Arrhenius behavior was reported when the temperature dependence of conductivity was analyzed. Scanning electron microscopy was used to study the surface morphology of the synthesized electrolyte. Fourier transform infrared and X-ray diffraction results confirmed the formation of composite material. The synthesized solid composite electrolyte exhibited excellent humidity sensing behavior in the relative humidity range of 15%–65%. Various humidity characteristics of the sample were measured such as hysteresis loop, recovery, and response time of the sensor. An expert system was modeled using artificial intelligence techniques and failure of the sensor was predicted with 97.2% accuracy using artificial neural networks.

Keywords

• artificial neural networks
• composite material
• humidity sensor
• ftir
• sem
• xrd