A Recursive Algorithm Based on Fuzzy Neural Network for Target Number

To overcome this deficiency, a compound fuzzy neural network named IMJ was introduced which combines the process’s rule-based reasoning and function approximation functions. This paper presents the concept of target numbers, and introduces a number of recursive algorithm target issues. Suppose given positive integer n is A (1), A (2), ..., A (n-1), A (n), the need to find the corresponding number of targeting, guaranteed to make the sum of the different elements and the combination of values in the vicinity of the target number (may be equal or the absolute value of the difference between the minimum case) for the most number of combinations. Such compound fuzzy neural network emphasizes the use of prior knowledge and studying that makes it possible to effectively improve the approximation and generalization of complex nonlinear systems in large segments range. We made a study of the most number of combinations that can be considered a target for loading problems in the best height leveling. The common fuzzy neural network does not introduce the object-oriented prior knowledge from a practical sense, such as introducing knowledge of operating and controlling experience as well as process field, resulting in poor using of “fuzzy” advantages.