关键词: 超超临界发电机组, 分散控制系统, 可靠性

Abstract: The steady operation of distributed control systems (DCS) is highly dependent on the system reliabilities. In order to enhance the reliabilities of the DCS system, this paper proposes three principles, i.e., preferential prevention, weakening impact and online maintenance, for the implementation of generalized reliability technologies based on the technical characteristics and control requirements of large thermal power generation units. According to these principles, the realization methods, principles and specific measures are discussed accordingly in the aspects of network communication, hardware performance, system power supply, early-warning function, software optimization and engineering design. To be specifically, the duplicate communication packets through duplex network can improve the fault tolerance of the system, while the redundant power supply plan can improve the power supply reliability design and the standardized algorithm can normalize the software utilization and realize safe actions during faults. In addition, the DCS system designed with these technologies has been successfully used in a 1 000-MW class ultra-supercritical thermal power unit. It has passed the testing under extreme conditions and gone through long-term operation, thus can be popularized broadly.

Key words: ultra-supercritical thermal power generation unit, distributed control system (DCS), reliability