One of the principal applications of fuzzy logic is in control system design. Fuzzy logic controllers (FLC) can be used to control
systems where the use of conventional control techniques may be Problematic. The tuning of fuzzy controllers has tende
d to rely on human expert knowledge, but where the number of rules and fuzzy sets is large. The Problem of generation desirable fuzzy rule is very important in the development of fuzzy systems. The purpose of this paper is to present a generation method of fuzzy control rules by learning from examples using genetic algorithms (GA). We propose real coded genetic algorithms (RCGA) for learning fuzzy rules, and an iterative process for obtaining set of rules which covers the examples set with a covering value previously defined.
Scientists nowadays increased their interest in artificial intelligence because
of the rapid development in the modern age. This development increased
the complexity of systems in order to consent society rapid needs in getting
systems of better r
eliability and high performance.
Artificial Intelligence solved many difficult and sticky problems.
We are going to define the genetic algorithms(GA) which is one of the
artificial intelligence branch because of its ability to solve many complex
problems in Different scientific aspects either in computer science[8] or
operational research and image processing[7] or social science[9].
In this essay, we used GA to find the maximum value of continuous function
within a limited rang to study the effect of some of the most important GA
parameters on the performance and accuracy of the results. We noticed the
effects of probability of mutation, population size and the number of the
repeated operations on the results accuracy and execution time in choosing
the Roulette Wheel Selection procedure. After that, we compare
between the Roulette Wheel procedure and the Elitism Selection procedure.
This study has reached to that ANN (5-9-1) (five neurons in input
layer_nine neurons in hidden layer _ one neuron in output layer) is the
optimum artificial network that hybrid system has reached to it with
mean squared error equals (1*10^-4) (0.7
m3/sec), where this software
has summed up millions of experiments in one step and in limited time, it
has also given a zero value of a number of network connections, such as
some connections related of relative humidity input because of the lake
of impact this parameter on the runoff when other parameters are
avaliable.
This study recommend to use this technique in forecasting of
evaporation and other climatic elements.
