Self-organization and associative memory: 3rd edition
Self-organization and associative memory: 3rd edition
Introduction to artificial neural systems
Introduction to artificial neural systems
The interpolation capabilities of the binary CMAC
Neural Networks
Real-time computer control (2nd ed.): an introduction
Real-time computer control (2nd ed.): an introduction
Artificial Neural Networks: Theory and Applications
Artificial Neural Networks: Theory and Applications
Neural Networks and Natural Intelligence
Neural Networks and Natural Intelligence
Expert Systems with Applications: An International Journal
A nature inspired Ying-Yang approach for intelligent decision support in bank solvency analysis
Expert Systems with Applications: An International Journal
HebbR2-Taffic: A novel application of neuro-fuzzy network for visual based traffic monitoring system
Expert Systems with Applications: An International Journal
Expert Systems with Applications: An International Journal
A novel associative memory approach to speech enhancement in a vehicular environment
Expert Systems with Applications: An International Journal
GA-TSKfnn: Parameters tuning of fuzzy neural network using genetic algorithms
Expert Systems with Applications: An International Journal
Supervised Pseudo Self-Evolving Cerebellar algorithm for generating fuzzy membership functions
Expert Systems with Applications: An International Journal
Hi-index | 0.01 |
This paper describes a novel application of an associative memory called the Modified Cerebellar Articulation Controller (MCMAC) (Int. J. Artif. Intell. Engng, 10 (1996) 135) in a continuous variable transmission (CVT) control system. It allows the on-line tuning of the associative memory and produces an effective gain-schedule for the automatic selection of the CVT gear ratio. Various control algorithms are investigated to control the CVT gear ratio to maintain the engine speed within a narrow range of efficient operating speed independently of the vehicle velocity. Extensive simulation results are presented to evaluate the control performance of a direct digital PID control algorithm with auto-tuning (Trans. ASME, 64 (1942)) and anti-windup mechanism. In particular, these results are contrasted against the control performance produced using the MCMAC (Int. J. Artif. Intell. Engng, 10 (1996) 135) with momentum, neighborhood learning and Averaged Trapezoidal Output (MCMAC-ATO) as the neural control algorithm for controlling the CVT. Simulation results are presented that show the reduced control fluctuations and improved learning capability of the MCMAC-ATO without incurring greater memory requirement. In particular, MCMAC-ATO is able to learn and control the CVT simultaneously while still maintaining acceptable control performance.