سال انتشار: ۱۳۹۰

محل انتشار: اولین کنفرانس بین المللی و سومین کنفرانس ملی سد و نیروگاههای برق آبی

تعداد صفحات: ۶

نویسنده(ها):

Hamideh Kazemi – Department of Civil and Environmental Engineering ,Iran University of Science and Technology
Amir pouya Peidayesh – Department of Civil Engineering and Center of Excellence for Enviro-Hydroinformatics Research, Iran
Abbas Afshar – University of Science and Technology

چکیده:

Calibration is one of the most important operations in water quality modeling. This process has been done by different methods like trial and error method, optimizationmethod. As these method have its own problem, nowadays hybrid algorithm are noteworthy. They can surmount the other algorithm’s shortcoming. In this paper, oneof these algorithms (ANN-PSO) was used to reduce run time of one of water quality model (CE-QUAL-W2). In this process optimization algorithm (particle swarmoptimization) produces calibration parameters for simulating model. As this process isso time consuming, traiend neural network, as surrogate of CE-QUAL estimatesimulation’s behavior. Needed data was driven from Karkheh eservoir in Iran,simulation period was 2 months. Estimator (ANN) embedded in optimization algorithm (PSO) has obviously reduced run time by estimating simulation model’sbehavior while the answers have reliable quality.