(202t) Predictive Functional Control On Programmable Logic Controller Using Code Generation | AIChE

(202t) Predictive Functional Control On Programmable Logic Controller Using Code Generation



Processes with large delay and multivariable interactions are often controlled by Model Predictive Controller (MPC). MPC is typically deployed on a computer or server in an industrial control setting. However, many industrial processes are controlled by Programmable Logic Controller(s) (PLC), which have less processing power and memory than standard computers.

Due to the restrictions mentioned earlier, simple control strategies are typically used to control these processes using PLC platform. These simple control strategies include basic Proportional Integral Derivative (PID) control and enhanced PID control, such as cascade and ratio control. Predictive Functional Control (PFC) [1] is a model based control algorithm which is computationally less intensive than standard MPC.  The performance of PFC is similar to the performance of MPC in most of the scenarios of industrial importance. Therefore PFC is a good advanced control option for industrial processes controlled using PLC platform.

Advanced control is developed in high level mathematical programming languages, while PLC is programmed using lower level programming languages. Development of advanced control algorithms using lower level languages is a tedious task, as advanced math libraries, testing and debugging tools needed for algorithm development are not available in low level languages. Code generation is an attractive option to develop advanced control algorithms for PLC platform. There are many practical challenges in deploying advanced control on PLC using code generation because of involvement of multiple programming languages, multiple development environments and third party tools. This paper implements PFC on PLC using code generation approach and examines the practical aspects of this approach for advanced control deployment on PLC. This approach enables development of complex control algorithms on PLC with quick turnaround time compared to Structured Text (ST) language used for developing control logic on PLC.  A case study is included to demonstrate the feasibility of code generation approach to deploy PFC on PLC for waste water treatment plant.

References

  1. J. Richalet and D. O’Donovan, Predictive Functional Control: Principles & Industrial Applications, Springer 2009.

Topics