A New Training Algorithm for Neuro-Fuzzy Networks

Abstract

In this paper a new iterative construction algorithm for local model networks is presented. The algorithm is focussed on building models with sparsely distributed data as they occur in engine optimization processes. The validity function of each local model is fitted to the available data using statistical criteria along with regularisation and thus allowing an arbitrary orientation and extent in the input space. Local models are consecutively placed into those regions of the input space where the model error is still large thus guaranteeing maximal improvement through each new local model. The orientation and extent of each validity function is also adapted to the available training data such that the determination of the local regression parameters is a well posed problem. The regularisation of the model can be controlled in a distinct manner using only two user-defined parameters. Examples from an industrial problems illustrate the efficiency of the proposed algorithm.