AUTHOR=Hao Xiaoguang , Jin Fei , Wang Bin , Zhang Qinghao , Wu Chuang , Sun Hao 

TITLE=Research on the flow characteristics identification of steam turbine valve based on FCM-LSSVM

JOURNAL=Frontiers in Smart Grids

VOLUME=Volume 2 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/smart-grids/articles/10.3389/frsgr.2023.1129541

DOI=10.3389/frsgr.2023.1129541

ISSN=2813-4311

ABSTRACT=The flow characteristics of turbine inlet valve often deviate from the design value with the impact of aging, deformation and system retrofit of flow path, thus influencing the control precision of unit load and operation stability. In this paper, the fuzzy c-means (FCM) clustering and least-squares support vector machine (LSSVM) are utilized to achieve the accurate identification of the flow characteristics of steam turbine valve. First, FCM clustering is proposed to classify the historical operating data of the power plant and extensive range of variable operating conditions are obtained. Then, in each condition cluster, a relationship model between the turbine input and output variables is constructed by using LSSVM. Finally, we obtain the FCM-LSSVM model to predict actual steam inlet flow by considering valve position command, speed, and real power generation as model inputs. In addition, considering a 330 MW turbine unit as an application instance, the valve flow characteristics of the turbine is verified by using the proposed FCM-LSSVM model. Results show that the model can obtain accurate valve flow characteristics without carrying tests on the turbine, thereby effectively avoiding the impact of characteristic testing on the safe operation of the unit.