Search

Accurate prediction of faults before they occur is vital because the intricate, uncertain, and intercorrelated natures of industrial processes can lead to multiple component failures or to a complete shutdown of the overall prediction cycle. While the first principle-based fault detection approach demands significant expert knowledge and is component-specific, learning-based approaches offer a plausible alternative because of their learning capability of offline data. Learning-based fault detection and diagnosis still deserve in-depth investigation because current approaches must happen offline, are static, and must be supervised; this makes them hardly applicable for the live scenarios of industrial processes. This chapter proposes a novel approach using an evolving type-2 random vector functional link network, which combines the meta-cognitive learning concept with the random vector functional link theory. The efficacy of evolving type-2 random vector functional link networks was validated with an experimental study on diagnosing different fault conditions of induction motors – namely broken rotor bars, unbalanced voltages, stator winding faults, and eccentricity problems – using a laboratory-scale test rig. Our algorithm was compared with other prominent algorithms and was found to deliver state-of-the-art performance in terms of accuracy, simplicity, and scalability.

usc Refereed/Peer-reviewed Randomized neural network (RNN) is a highly feasible solution in the era of big data because it offers a simple and fast working principle in processing dynamic and evolving data streams. This paper proposes a novel RNN, namely recurrent type-2 random vector functional link network (RT2McRVFLN), which provides a highly scalable solution for data streams in a strictly online and integrated framework. It is built upon the psychologically inspired concept of metacognitive learning, which covers three basic components of human learning: what-to-learn, how-to-learn, and when-to-learn. The what-to-learn selects important samples on the fly with the use of online active learning scenario, which renders our algorithm an online semi-supervised algorithm. The how-to-learn process combines an open structure of evolving concept and a randomized learning algorithm of random vector functional link network (RVFLN). The efficacy of the RT2McRVFLN has been numerically validated through two real-world case studies and comparisons with its counterparts, which arrive at a conclusive finding that our algorithm delivers a tradeoff between accuracy and simplicity.

Learning in neural networks can broadly be divided into two categories, viz., off-line (or batch) learning and online (or incremental) learning. In this paper, a review of a variety of supervised neural networks with online learning capabilities is presented. Specifically, we focus on articles published in main indexed journals in the past 10 years (2003–2013). We examine a number of key neural network architectures, which include feedforward neural networks, recurrent neural networks, fuzzy neural networks, and other related networks. How the online learning methodologies are incorporated into these networks is exemplified, and how they are applied to solving problems in different domains is highlighted. A summary of the review that covers different network architectures and their applications is presented. Refereed/Peer-reviewed