CIVILICA We Respect the Science
(ناشر تخصصی کنفرانسهای کشور / شماره مجوز انتشارات از وزارت فرهنگ و ارشاد اسلامی: ۸۹۷۱)

Predicting Rotor Outlet Flow Field of Centrifugal Turbomachinery Using Fuzzy Time Series

عنوان مقاله: Predicting Rotor Outlet Flow Field of Centrifugal Turbomachinery Using Fuzzy Time Series
شناسه ملی مقاله: EMECCONF03_058
منتشر شده در سومین کنفرانس بین المللی مهندسی برق،مهندسی مکانیک، کامپیوتر و علوم مهندسی در سال 1398
مشخصات نویسندگان مقاله:

Seif Dalilsafaei - Mechanical Engineering Department, Amirkabir University of Technology, Tehran ۱۵۹۱۶۳۴۳۱۱, Iran
Nader Montazerin - Mechanical Engineering Department, Amirkabir University of Technology, Tehran ۱۵۹۱۶۳۴۳۱۱, Iran.
Mohammad Hossein Fazel Zarandi - Industrial Engineering Department, Amirkabir University of Technology, Tehran ۱۵۹۱۶۳۴۳۱۱, Iran

خلاصه مقاله:
The purpose of the present study is to create new data of the output flow fields of centrifugal Turbomachinery rotors using a fuzzy time series algorithm. The experimental method to study the flow field contains the use of a particle image velocity system by which the flow field information is recorded at different spatial locations and at a specific time step. Data acquisition using advanced measurement systems including the laser measurement systems requires the establishment of the necessary precautions, periodic calibrations, and experienced operators, in addition to high time and financial costs, which make the necessity of generating and evaluating further information based on the experimental data extracted by the prediction algorithms addressed in this study.Since the data of the particle image measurement system have been extracted at constant time steps in the experimental, it is possible to implement time series algorithms to predict new data. In this research, the selected fuzzy time series model is used based on clustering, which is separately analyzed for each point of the flow field surface based on the system s own information. This has led to the accuracy increase in the prediction of flow field surfaces, which also indicates the quality of the obtained results. Furthermore, in order to present the results of the present study, the velocity contour representation has been used which makes it possible to investigate and compare the experimental and experimental results, in addition to provide an accurate understanding of the predicted flow field data.

کلمات کلیدی:
Turbulent Flow Field, Prediction, Fuzzy Time Series, Centrifugal Turbomachinery, Particle Image Velocimetry

صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/933250/