Sistema de control en red inañámbrico (WNCS) mediante visión artificial y herramienta de mesajería Telegram, para sistemas de nivel

Abstract

The present work talks over the process to implement the LQG wireless control to a level plant, the plant consists of a filling tank and a supply tank that emulates the supply and consumption of fuel inside the filling tank using the Raspberry PI card. 3, which performs level monitoring using machine vision, processes the images in real time to effectively control the plant process. In addition, the wireless actuator drive was developed using the Arduino ESP8266 module that controls the servo motor coupled to a half-turn mechanical valve, which controls the flow of fuel out of the filling tank. For the execution of the project, it was divided into three fundamental parts, the first being the obtaining of the mathematical model that describes the behavior of the plant using Matlab ident, the second is the implementation of the LQG (Linear Quadratic Gaussian) control algorithm to the plant, and the third is monitoring remote by means of the Telegram tool that works with the server of the same application, that is, it can work from any smart device with access to the internet and in turn to the mobile application. The control elements interact through the Node Red graphical programming software, which uses the MQTT protocol for the communication of so-called clients with the server (broker) through the concept of publication and subscription to a certain topic.