Valve fault detection and diagnosis based on CMAC neural networks

A self-constructing fuzzy CMAC model and its applications

“…The CMAC model has been successfully applied to various fields, such as robot control [11,29,30], signal processing [19], pattern recognition [9], and diagnosis [13,36]. However, Albus' CMAC model has three major limitations [2,3].…”

BP neural network-based on fault diagnosis of hydraulic servo-valves

“…Presently, there are a lot of methods to improve the training speed of the network, such as improving the error function, adjusting the study rate dynamically and so on. But we must choose other optimization methods in order to overcome the disadvantage of getting into local minimum [3].…”

A novel model-based fault detection method for temperature sensor using fractal correlation dimension

“…This method is validated to evaluate soft sensor faults (biases) for temperature sensors and flow meters in central chilling plant. Other mathematical models including blackbox multivariate polynomial methods, specifically radial basis function and multilayer perceptron, the generic physical component model [11,12], artificial neural network [14], rough set approach [15], transient pattern analysis [16] and others, are used to get deviations for well suited automated FDD in HVAC equipments and systems.…”