FORECASTING BY USING DIFFERENTIAL EVOLUTION BASED CLUSTERING
Latafat Abbas Gardashova, Hashim A. Alimammadzade · 2015
In this paper it is analyzed a new approach to forecasting based on fuzzy neural network and differential evolution based clustering. The proposed method is applied to forecasting of kerosene production in the oil refinery enterprise. Experimental results have demonstrated efficiency of the proposed method and its advantages as compared to the existing classical methods. Keywords: Fuzzy Inference Neural Network, diferential evolution, fuzzy clustering I.INTRODUCTION In the area of oil refinery plant control and management a number of problems should be settled by using a data obtained from the solution of the forecasting tasks. The uncertainty and incompleteness of the initial data as well as complexity and weakness of the conventional methods of forecasting complicate proper solving of the forecasting problems.The forecasting problems for the oil refinery plant are characterized by informal background description, subjectivity of the states estimations. Some of forecasting tasks are complicated by the nonstationarity and nongraduality of the time-dependent fluctuations.It is necessary to choose adequate means for solving considered problems which would be able to carry out predictably hour-to-hour,day-to-day, week-to-week, month-to-month or year-to- year fluctuations of the technological and economic indexes in such complex environment. In 1965 after Lotfi Zadeh's(1) fuzzy theory fuzzy logic new era began in the development of sciences including forecasting science. Since the middle of 70s by using this notion new forecasting methods were created and applied in practice. Several works have been written about the theoretical and practical problems of forecasting in Azerbaijan as well(2-7). There is no universal forecasting method for economical indicators. There are many forecasting methods because of the diversity of forecasting conditions. We can classify the economical forecasting into two groups: qualitative and quantitative. Quantitative methods are consist of traditional statistical methods, artificial neural network and other modern methods. Sometimes we may have to use new type of time series like fuzzy time series, cost of this kind of time series is linguistic. Thus it is required to use soft computing methods based on application of fuzzy time series and fuzzy logic. Fuzzy Neural Network is effective for these purposes and fuzzy numbers, identity function and fuzzy operations are used. The basic idea of FNN is that results are acquired according to the fuzzy logic apparatus. Fuzzy Neural Network is used to find the parameters of identity functions. This system may use the information that was known beforehand, learn them , may even get new numbers, forecasting time series, etc. FNN is used to forecast different social economical problems, as well. FNN has been applied to forecast the regional electrical charges and consuming of electric power in Turkey(8). The solution of forecasting with the help of FNN has resulted better than that of ANN. But FNN has also some shortcomings. Fuzzy Neural Network is not a dynamic network , it does not have a memory. But these shortcomings may be solved with help of Recurrent Fuzzy Neural Network. In Recurrent Fuzzy Neural Network neurons of certain layer may get signals from itself, both from the outside and from other neurons in the same layer with itself. Thus unlike non-recurrent network , recurrent networks have memories and it enables them to remember the information about the situation in the given period (9-10). It must be noted that Recurrent Fuzzy Neural Networks are very effective. The methods that have been used are characterized with the small calculations complexity and feature of learning from the experiments. Fuzzy clustering is applied to improve the results of FRNN forecasting. Qualitative methods of forecasting is used in case of impossibility of considering several factors because of insufficiencies and complexity of forecasting objects. In this case the application of expert assessment is applied in forecasting. In (7) is offered method for short-term forecasting which combines quantitative method( such as processing fuzzy time series using recurrent neural networks with DE -based learning) and qualitative method(such as the modified Fuzzy Delphi).