Decentralized Control of Solid Oxide Fuel Cells

Environmental concerns and limited availability of natural resources have made solid oxide fuel cells (SOFC) a promising alternative for large-scale electricity production. A control system is needed for SOFCs to meet operational objectives and ensure safe operation. Previous studies showed that the control of SOFC is challenging due to its slow response time, tight operating constraints and unavailability of some key measurements. In this work, simple yet reliable decentralized proportional-integral-derivative (PID) controllers are systematically designed based on a benchmark nonlinear dynamic model of SOFC [1]. The design procedure involves selection of controlled variables (CVs), input-output pairing selection, and PID controller tuning. For CV selection, a novel method is proposed such that maintaining the CVs at constant setpoints facilitates constraint satisfaction for all key variables. Closed-loop simulations demonstrate that the proposed decentralized PI controllers are able to closely meet the control objectives of SOFC even in the presence of changing operating conditions.