Non-linear PI control inspired by biological control systems

A non-linear modification to PI control has been developed which is appropriate for plants requiring exact set-point matching and disturbance attenuation in the presence of infrequent step changes in load disturbances or set-point. This novel control algorithm, labeled PII (proportional with intermittent integral), is motivated by a model of a signal transduction pathway active in mammalian blood pressure regulation. It is possible that set-point regulation and disturbance rejection are achieved through independent control loops in mammals. This control paradigm implements, somewhat independently, two control loops to achieve the goals of disturbance attenuation and set-point matching. By developing a disturbance rejection controller with no consideration of set-point matching, it is possible to design for greater disturbance attenuation. As a first step, the disturbance attenuator is a simple proportional controller, and set-point matching will be achieved by intermittently invoking an integral controller. The mechanisms observed in a neuronal signal transduction model are used to tie these controllers together. Superior performance of this approach as compared to traditional PI control has been shown on models of level control loops and cyclopentenol production in a CSTR reactor. The cyclopentenol process contains a maximum steady-state production level. PI control cannot be used at this peak operating point as the sign change in plant gain results in an unstable closed-loop system. Application of this new approach to this problem results in stable exact setpoint matching for achievable set-points.