A Novel Hybrid Machine Learning Algorithm for Limited and Big Data Modeling With Application in Industry 4.0

Abstract

To meet the challenges of manufacturing smart products, the manufacturing plants have been radically changed to become smart factories underpinned by industry 4.0 technologies. The transformation is assisted by employment of machine learning techniques that can deal with modelling both big or limited data. This manuscript reviews these concepts and present a case study that demonstrates the use of a novel intelligent hybrid algorithms for Industry 4.0 applications with limited data. In particular, an intelligent algorithm is proposed for robust data modelling of nonlinear systems based on input-output data. In our approach, a novel hybrid data-driven combining the Group-Method of Data-Handling and Singular-Value Decomposition is adapted to find an offline deterministic model combined with Pareto multi-objective optimization to overcome the overfitting issue. An Unscented-Kalman-Filter is also incorporated to update the coefficient of the deterministic model and increase its robustness against data uncertainties. The effectiveness of the proposed method is examined on a set of real industrial measurements. INDEX TERMS Industry 4.0; Big data modelling; Limited data modelling, Multi-objective optimization. NOMENCLATURE CNN

A novel interval type-2 fuzzy Kalman filtering and tracking of experimental data

“… 2020 ; Haessig and Friedland 1998 ), non-modeled dynamics (Khayyam et al. 2020 ), high approximation and prediction errors (Tang et al. 2020 ).…”

“…Among the main aspects characterizing the complexity of problems in science and engineering are the nonlinearities (Bendat 1998;Schoukens and Ljung 2019), uncertainties (Martynyuk et al 2019), noisy and non-stationary environment (Hendricks et al 2008;Moss and McClintock 1989), temporal variability (Tomás-Rodríguez and Banks 2010), among others. Computational modeling approaches which consider these complexities in their formulations do have better performance for facing to conditions of stability and convergence (Bonyadi and Michalewicz 2016;Boutayeb et al 1997), polarized parametric estimation (Chan et al 2020;Haessig and Friedland 1998), non-modeled dynamics (Khayyam et al 2020), high approximation and prediction errors (Tang et al 2020). In data analysis, an increasingly concern from researchers is related to the presence of several types of uncertainties such as inaccuracy and incompleteness of data and information, parametric and structural uncertainties, propagation and accumulation of uncertainties, and unknown initial conditions, that must be taken into account by modeling approaches in order to ensure accurate models for real-world problems (Wang and Zhao 2013).…”

A Mini-Review and Perspective on Current Best Practice and Emerging Industry 4.0 Methods for Risk Reduction in Advanced Composites Manufacturing

“…As Rudin et al 2019 claim, any model that automatically "learns" from a dataset and is not manually constructed, has the potential to be intractable at any point in its domain, by virtue of the model builder implicitly not being hands-on in that process [36]. This perspective may be particularly relevant for use cases with limited data available, due to there being less embedded statistical information, further relying on the performance of increasingly complex and high-fidelity models to compensate [37]. Recent developments seeking to retain the advantages of black-box ML, while increasing model confidence, include the strategic drafting of federal policy towards fostering scientific machine learning (SciML) in which scientific principles can be more tractably encoded into such models Figure 3.…”

An improved adaptive neuro fuzzy inference system model using conjoined metaheuristic algorithms for electrical conductivity prediction

“…The literature reviews, highly emphasized the implication of hybrid machine learning models for diverse environmental engineering problems 48 – 50 . It has been approved as those newly developed versions are the trustworthy computer aid models for solving highly stochastic and non-linear historical big data 51 , 52 .…”