Implementation of a hybrid controller for ventilation control using soft computing

Many industrial facilities utilize pressure control gradients to prevent migration of hazardous species from containment areas to occupied zones, often using proportional-integral-derivative (PID) control systems. When operators rebalance the facility, variation from the desired gradients can occur and the operating conditions can change enough that the PID parameters are no longer adequate to maintain a stable system. As the goal of the ventilation control system is to optimize the pressure gradients and associated flows for the facility. Linear quadratic tracking (LQT) is a method that provides a time-based approach to guiding facility interactions. However, LQT methods are susceptible to modeling and measurement errors, and therefore the additional use of soft computing methods is proposed for implementation to account for these errors and nonlinearities.