A rough neural network for material proportioning system

摘要

A rough membership function neural network for raw material proportioning is presented in this paper. Approximation neurons and decision-based decider neurons have been used in the design of the rough neural classification system. Data obtained from simulations are used for neuron implementation and testing. The simulation results show that the outputs are very close to the target values. The network performs good control on the composition of mixed material throughout the test.