Study Of Sensors Modeling Methods In PSpice Based On Neural Network

The sensors as a kind of components for acquiring and transforming information have become an indispensable important part of all detection equipments. The sensors are also widely applied in automatic detection, automatic control and other areas.Circuit simulation is to implement the electric circuit functional simulation and characteristic analysis by expressing the electronic components and circuit modules with mathematical models, while supported by numerical analysis and graphical simulation show method. Circuit simulation may exactly reflect the electric circuit characteristic, and it can also assist optimizing the design of the circuit structure conveniently, quickly and economically. This will have very great significance to shorten the development period of electronic products, reduce the cost, enhance the overall performance and promote market competence.However, there are no generic sensors modeling methods for non-electronic devices in the existing circuit simulation tools. So the simulation for the circuit with sensors can not be made in the existing simulation tools.To address this issue, a sensor modeling method for PSpice based on neural network is proposed in this article. Firstly, a multi-layer feedforward neural network is used to approximate the characteristics of a sensor. Secondly, the achieved structure of the neural network is described in the PSPICE language to form a subcircuit. Finally, the subcircuit is used as the sensor model when the changes of a non-electric quantity imposed on the sensor in real applications is replaced with that of an electric quantity in PSPICE simulation. The efficiency of this method is proved e by experiments.