A Novel Scheme for Key Performance Indicator Prediction and Diagnosis With Application to an Industrial Hot Strip Mill

Efficient Use of Hybrid Adaptive Neuro-Fuzzy Inference System Combined with Nonlinear Dimension Reduction Method in Production Processes

“…Automatic fault prediction is an important topic of research in metal industry [1]. Since the beginning of the first industrial revolution, industries are striving to produce fault-free products in least possible amount of time.…”

A Quality-Based Nonlinear Fault Diagnosis Framework Focusing on Industrial Multimode Batch Processes

“…Then, the run-out table cooling section allows the strip to cool to the desired temperature, which allows the steel being of good mechanical property. The detailed descriptions of this process can be found in Peng et al [5] and Ding et al [1]. It can be observed that for the HSMR process, the four key quality variables are thickness, width, flatness and temperature, of which the first three are primarily determined by the finishing mill rolling process (FMRP).…”

“…ITH the rapid development of the scale, degree of automation and integration of industrial processes, challenges have emerged for practitioners with problems of frequently switched operating points, different operating batches and strong nonlinearities [1]. The complexity also makes it even harder to guarantee the product quality, the most core profitable indicator for an industrial process.…”

“…Another handicap for ensuring the product quality is the strong nonlinearity in HSMR, in this case the traditional linear theory will not be sensible. In recent years, to solve such kinds of problem, quality-related fault detection and diagnosis (QFDD) methods have been raised and proved to be useful for quality control [1], [5], [6], [7], [8].…”

PLS-based memory control scheme for enhanced process monitoring

“…If anomalies in industrial systems are not detected promptly, they can affect plant productivity, profitability, and safety [2]. Process monitoring is employed by various process industries [1], [3], [4], [5]. Partial least square (PLS) is among the most widely used multivariate statistical process monitoring method for monitoring multivariate processes [6], [7].…”