Pneumatic servo valve models based on artificial neural networks

Abstract

This paper presents a study where static models for the flow stage of three-port pneumatic servo valves are obtained using artificial neural networks. A new approach to build the models is introduced in detail. This approach considers two distinct models: for simulation purposes, the air mass flow is determined for a given working pressure and command input; for control purposes, the command input is determined given the working pressure and the desired air mass flow. This approach enables the use of air mass flow as the synthesized control action, thus rendering the overall pneumatic system affine. Therefore, control techniques requiring this condition on the system model can be directly used. The use of this approach in two different industrial pneumatic servo valves is presented in detail. A comparison between the different characteristics of each servo valve is also provided.