Intelligent flow and level sensors for a filtering system in a water treatment plant

Abstract

Industrial sensors play a fundamental role in distributed control systems, and their increase in number and quantity of parameters involved in industrial processes, makes it necessary for these elements to evolve towards intelligent entities, which in addition to fulfilling more complex tasks, have a certain autonomy, which allows reducing the load on the control equipment, increasing the overall performance of the system. This research work proposes an intelligent sensor architecture, based on the BDI methodology, aimed at controlling the filtering system in a water treatment plant, for level and flow variables. The system design was projected using UML tools, and for the modeling of the smart sensor, Petri nets were used, which allowed the simulation, validation and verification of the system. The Modbus protocol for communications was adopted, and for the implementation the ATmega2560 microcontroller was used, which allowed to test experimentally the behavior of the intelligent sensor; response times and the values of the measured variables. The proposed Architecture allowed to develop an adaptable and flexible intelligent sensor in the industrial environment, with adequate performance and behavior, generating values and response times, in many cases better than other works found in the scientific literature.