Physics informed neural network using oscillator model towards sensor applications

In this paper we perform numerical simulations to understand sensor match to instruments of materials, fluid properties, solid-liquid interaction, sensitivity to change in geometry, mass measurement, density, conductivity, ion concentration and transport of polymers. The method is influenced by parameter mass and its transport towards property matching with sensitivity in geometry. The method is called Newton's method unlike energy balance methods because of independent sensor assembly circuitry manufacturing match experiments. We predict the new areas of the geometry available match with experiments. The new areas can be environment in precision sensors. The method can find applications in bio sensors, automobiles, manufacturing, sensors, electronics and chemical sensors.