In this project, a control system based on an artificial intelligence technique is developed and compared with a conventional controller in its implementation on the multi-tank plant of the Barión research nursery. This intelligent controller is a dynamically focused adaptive fuzzy PID, which is developed and simulated in Matlab's Simulink with the mathematical model of the plant and later coded in the Codesys platform for its implementation. The fuzzy controller coding is done in such a way that the number of calculations required to generate the control signal is optimized in each sampling cycle. In addition, tools are used that allow the correct operation of the controller when it is implemented in a physical system. For the design and implementation of the intelligent controller, the advantages of the conventional PID controller are taken, such as the topology of a PI-D controller, filtering of the reference signal, derivative filter and Anti Wind-Up system in the integral part. A fuzzy PD controller is used, which is tuned by a fuzzy adaptation system with dynamic focus and a proposal for an intelligent integration system is made. The physical implementation of the controller is done with a Soft-PLC prototype based on Raspberry Pi and Psoc, which is debugged and enabled. The developed controller is mounted on the prototype and compared with a conventional PID controller on the plant's cascade tank system, it is evaluated by means of performance and variability terms such as maximum impulse, stabilization time and error variance.