An innovative hydraulic turbine governor for predictive maintenance of hydropower plant

摘要

An important part of the enterprises budget is devoted to inspecting, repairing and modifying components, all this being done often too early or too late. Therefore, modern hydropower plants are now active by pursuing transition from conventional scheduled maintenance as too expensive, to predictive maintenance more adapted to detecting and diagnosing just in time the degradations in order to anticipate the failures. The predictive maintenance is mainly based on conditioning monitoring, fault diagnosis and prognosis, which allow for taking the right maintenance decision at the right time and at the right place. These three characteristics of the decision lead, as a consequence, to a combination of the conventional hierarchical model of the operating system with a new distributed one founded on the co-operation of "intelligent" field components embedding data processing, storage and communication capabilities. So, the operational IT architecture which supports the predictive maintenance strategy, has to integrate an interoperability with these types of components. The paper aims at describing the development of an "intelligent" interoperable governor that supports some predictive maintenance functions. It allows for not only performing fault and degradation diagnosis on frequency, gate opening and other signals, but also automatically reconfiguring its operation mode if some faults happen. The prototype of the proposed governor is tested and validated to be then developed into an industrial product.