Study And Applications On The Method For Integrated Monitoring Of Hydro Turbine Generator Sets Based On Team Intelligence

Hydroelectricity is the main clean energy at present in China. Under the background of the worldwide increasing attention to energy saving and emission reduction, It's a prominent issue that how to improve the utilization of Hydro Turbine Generator Sets (HTGS) as much as possible when the save and stable operation of HTGS can be guaranteed. Since the 1990s, condition based maintenance aimed at improving the reliability of HTGS and economic benefits has become the focus of research. The technique of condition monitoring used to master the health condition of HTGS has made considerate development and progress, and a variety of condition monitoring systems of HTGS have been put into industrial application. These systems are basically special monitoring devices or collective monitoring systems, which play a role in ensuring the save and stable operation of HTGS and improving the facility utilization.However, these systems neither achieve overall monitoring generally, nor can make all special monitoring devices be integrated into the same platform, and thus information sharing and behavior collaborating (such as synchronous sampling, synchronous storage, correlation analysis) among special monitoring devices are difficult to be realized. Additionally, HTGS is a close-coupling, complex, and large system consisted of hydraulic, mechanical, electrical, control and auxiliary equipment, whose health condition can be affected by any one part. The current monitoring methods cannot solve organic fusion issue of state data of HTGS that has the characteristics of multi source, multi time scope, and isomerization, so the synthetic analysis and diagnosis of HTGS cannot be achieved.For this reason, the method for integrated monitoring of HTGS based on Team Intelligence (TI) is proposed. The basic objective of this method is to make all special monitoring devices be integrated into the same platform via real-time fieldbus and then form an organic integrated monitoring team of HTGS; the correlation and coordination mechanisms of these special monitoring devices are established; they jointly realize organic fusion of state data of HTGS that has the characteristics of multi source, multi time scope, and isomerization, and as a result it's convenient for the synthetic analysis and diagnosis of HTGS.By summarizing the distinctive characteristics of teams, the concept of TI is drawn forth and its definition is given, and the formal descriptions of team behaviors possibly existed in a team are conducted. The research of team model is carried out systematically, including organizational structures, task descriptions, role relations, organizational norms and human relations, etc.. The unity of goals, behaviors and tasks of the team is achieved through task descriptions, and the collaborative behaviors can be contributed to the role relations under constraint of organizational norms in a team. The model of Team Intelligence Player (TIP) with three-layer vertical structure based on the message mechanism is established, and the frameworks of message processing layer, behavior control layer, and core functional layer included in the model of TIP are separately designed.According to TI theory, the framework of the integrated monitoring team of HTGS is established, which is composed of control unit TIP, mechanical unit TIP, electrical unit TIP, synthetic unit TIP and other TIPs, and core functions of these TIPs are discussed separately. The basic operating mechanisms of the integrated monitoring team of HTGS are elaborated, including divisions of work, coordination and cooperation, constrains and conflicts, etc.. Organic fusion of state data of HTGS is achieved by key techniques such as time fusion, operating condition fusion and abnormal event fusion, etc..Surrounded by the problems of organization, storage and management of mass state data of HTGS, optimizing organization strategy, selectively intelligent storage strategy, and validity confirmation and management of state data of HTGS are separately studied.Based on integrated monitoring of HTGS, the synthetic analysis approach to HTGS is discussed. Making the conventional threshold analysis combine with operating conditions and abnormal events of HTGS, the method for real-time fault prewarnings based on the dynamic threshold is studied, whose application examples in the fault prewarning of the transformer are introduced. On basis of the overall and strong relevance state data provided by integrated monitoring of HTGS, the field case of the synthetic analysis for the stability of HTGS is presented. Aimed at common failures or faults of the Oil System of the Governor (OSG), the performance indices that can well reflect them are defined separately, and the calculation methods are given. Based on state data and performance indices, the approaches to the performance analysis and failure detection of OSG are introduced. In the end, the application examples such as the real-time monitoring, daily analysis, oil leakage, and fault prewarning of OSG are given.It has been proved that the integrated monitoring of HTGS based on TI can provide a good platform for carrying out the synthetic analysis and diagnosis of facilities, which plays extremely important roles in improving the accuracy, reliability, and credibility of the analysis and diagnosis results.