RBF neural network modeling method based on feature clustering

Abstract

The invention relates to an RBF neural network modeling method based on feature clustering, which belongs to the field of automatic control, information technology and advanced manufacture. The invention particularly relates to an RBF neural network modeling method based on feature extraction function clustering, which can solve the modeling problem that data can be scattered. The method is characterized by comprising the following steps: defining a feature extraction function based on existing mechanism knowledge, determining an RBF network center in a clustering algorithm based on the feature extraction function, and determining a weight value from the hidden layer to the output layer of the RBF network in a least square method. The invention also provides a clustering algorithm based on the feature extraction function, which is not used for directly clustering data, but is used for clustering data with scattering features through introduction of the feature extraction function based on the mechanism knowledge. The obtained clustering center is used as the RBF network center, and the weight value from the hidden layer to the output layer of the RBF network can be obtained with a linear interpolation method. The invention can effectively solve the modeling problem that the data has scattering features, and can achieve high modeling accuracy.

Description

A Modeling Method of RBF Neural Network Based on Feature Clustering technical field The invention belongs to the fields of automatic control, information technology and advanced manufacturing. In particular, it relates to an RBF neural network modeling method aiming at the modeling problem of data with scattered features. Background technique In many modeling environments oriented to detection, control and optimization of real industrial processes, the data required for modeling often have scattered characteristics. How to determine the center of the RBF network is a key problem when using the RBF network to solve the modeling problem with scattered data. At present, the common method to determine the center of RBF network is the clustering method, but most of the proposed clustering methods (such as the fuzzy C-means clustering method) directly use the sample data for clustering, so the effectiveness of the above method is largely depends on the distribution of the samp...

AI Technical Summary

Problems solved by technology

However, when the data is highly dispersed, the clustering effect of the above clustering method is poor, resulting in inaccurate determination of the center of the RBF network, which ultimately affects the training speed and accuracy of the RBF network

Therefore, it is difficult to effectively solve the modeling problem of data with scattered characteristics by using RBF network based on conventional clustering methods

Images