پيش بيني رواناب حوزه هاي بالادست با استفاده از مدل ANFIS براي بهره برداري بهينه از مخازن سدها

چكيده انگليسي :

Upstream Basin River Inflow Forecasting by Using ANFIS Model in order to Optimizing Dam Reservoirs Operation ohammad Nazifkar M m nazifkar@cv iut ac ir 1 May 2011 Department of Civil Engineering Isfahan University of Technology Isfahan 84156 83111 IranDegree M Sc Language FarsiDr Keyvan Asghari kasghari@cc iut ac ir AbstractHydrological processes as complex systems and their modeling procedures either physically based ornumerically are difficult tasks due to their inter relational effects of parameters and variables One of theimportant of these processes related to the management of water resources is the derivation of reservoiroperation rules in or to determine the optimal reservoir releases as the main sources of supply water Theaccuracy and reliability of the water supply determination is based on good estimate and reliable forecastingof reservoir inflow and the duration of projection In addition the inherent uncertainty in runoff predictionmakes the modeling procedures very sensitive to both data and type of models This research focuses on ause of a newly developed integrated model namely Adaptive Neuro Fuzzy Inference System ANFIS toinvestigate the potentials of neuro fuzzy systems in modeling runoff time series into Kardeh reservoir inKhorasan province and to access its performance relative to artificial neural networks ANNs This methodhas demonstrated very satisfactory performances in applying to different hydrological process analysis However when the method encounters large number of parameters including input and output variables itproduces many rules and faces curse of dimensionality during computational steps Employing differentfuzzy clustering preprocessing procedures classifies the input data into different independent groups andselects the most effective set of input data in order to reduce the dimension of dataset and enhance theefficiency of algorithm by formation of fewer numbers of fuzzy laws Subtractive Clusrering SUBCLUST and Fuzzy C Means FCM are two methods of fuzzy clustering employed in this research In Fuzzy C Means number of clusters is determined by the user and system gets radius of the clusters but in SubtractiveClusrering cluster radius is determined by the user and system brings the number of clusters obtained Inaccordance with the main objective of this research two daily and monthly models have been developedusing Adaptive Neuro Fuzzy Inference System with a preprocessing step of clustering the input dataset daily models for short term decisions such as reducing flood losses and monthly models for medium termdecisions such as operation of reservoirs Generally the results showed that the Adaptive Neuro FuzzyInference System forecasted inflow series preserve more characteristics of actual inflow data in comparisonwith artificial neural networks Also Fuzzy C Means has demonstrated a relatively better result comparedwith Subtractive Clusrering algorithm as far as the classification of the data was concerned The last phaseof this research was to utilize the forecasted inflow for determining the operational rules of the Kardehreservoir Optimal operational rule curve based on the result of Harmony Search algorithm were derived andcompared with traditional simulated rule curves The results indicated more uniform monthly supply ofwater demand during a year relative to previously simulated outcomes Key WordsRunoff Modeling Adaptive Neuro Fuzzy Inference System Clustering Optimization

نظيف كار، محمد

پيش بيني رواناب حوزه هاي بالادست با استفاده از مدل ANFIS براي بهره برداري بهينه از مخازن سدها

مدل سازي بارش , خوشه بندي فازي , بهينه سازي