Multicollinearity occurs in a dataset due to correlation between the predictors. Models derived from such data without a check on multicollinearity may lead to erroneous system analysis. This problem can be eliminated by the selection of appropriate predictors from the dataset. Variable reduction methods like B2, B4, VIF, KIF and factor analysis (FA) can be used to overcome this problem. Such methods are useful particularly when used in conjunction with modelling methods that do not automate variable selection, such as artificial neural network (ANN) and fuzzy logic. The literature reveals that the current problem is aptly described in the field of statistics but is paid little attention in the field of machine learning. In this paper, multicollinearity is presented involving the estimation of fat content inside the body. Commonly used statistical methods such as stepwise regression, radial basis function partial least squares, partial robust M-regression, ridge regression and principal component regression are applied to this problem. The machine learning methods FA-ANN and genetic programming are also applied. The results are discussed with the interpretation and comparison of the modelling methods summarised in order to guide users on the proper techniques for tackling the multicollinearity problem.

This paper provides a novel intelligent scheduling strategy for a real-world transportation dynamic scheduling case from an engine workshop of general motor company (GMEW), which is a key production line throughout the manufacturing process. In order to reduce the carbon emission in the scheduling process and make up for ignoring the energy consumption of each part in the scheduling when optimizing the carbon emission of the workshop and the factory. This paper first formulates a fuzzy random chance-constrained programming model of inverse scheduling problem (ISP) with energy consumption. A multi-strategy parallel genetic algorithm based on machine learning (RL-MSPGA) is proposed, which uses machine learning to improve the genetic algorithm. First, the parallel idea is developed to accelerate the process of evolution of genetic algorithm, and the initial population is divided into clusters by $k$ -means clustering algorithm. Second, similar individuals are evenly distributed to different sub-populations to ensure the diversity and uniformity of sub-populations. Third, in the process of evolution, the sub-populations communicate with each other, and extend the excellent individuals to replace the poor ones in other populations, so as to improve the overall quality of the population. Fourth, the self-learning of the crossover probability is realized by the self-learning of the self-sensing environment, which makes the crossover probability adapt to the evolutionary process according to experience. Finally, the real instance is used to validate the different algorithms. It can effectively adjust the completion time and the proportion of energy consumption, thus providing the possibility for the production of energy-saving enterprises. This implies that the suggested model is reasonable and the provided algorithm can effectively solve the inverse shop scheduling problem.

The accurate estimation of state of health (SOH) for lithium-ion batteries is significant to improve the reliability and safety of batteries in operation. However, many existing studies on battery SOH estimation are conducted on the premise of large sizable labeled training data acquisition without considering the time cost and experimental cost. To solve such issues, this article proposes a novel capacity prediction method for SOH estimation based on the battery equivalent circuit model (ECM), deep learning, and transfer learning. First, an actual charge–discharge experiment is carried out, and a simulation of the corresponding cycling process is conducted for virtual data acquisition using the battery equivalent model. Second, a convolutional neural network (CNN)-based feature extraction network is selected by conducting a performance comparison. Then, a capacity estimation model consisting of a feature extraction network, regressor, and feature alignment metric calculation modules is generated. Several transfer learning methods are chosen for feature alignment metric calculation. Finally, a capacity estimation performance comparison is done for the final selection of the feature alignment metric calculation methods. The results illustrate that the capacity prediction model established using virtual data and the generative adversarial network (GAN)-based transfer learning method has ideal prediction performance (with the 0.0941 of the maximum test error in all capacity estimation situation).

Lithium-ion (Li-ion) battery pack is vital for storage of energy produced from different sources and has been extensively used for various applications such as electric vehicles (EVs), watches, cookers, etc. For an efficient real-time monitoring and fault diagnosis of battery operated systems, it is important to have a quantified information on the state-of-health (SoH) of batteries. This paper conducts comprehensive literature studies on advancement, challenges, concerns, and futuristic aspects of models and methods for SoH estimation of batteries. Based on the studies, the methods and models for SoH estimation have been summarized systematically with their advantages and disadvantages in tabular format. The prime emphasis of this review was attributed toward the development of a hybridized method which computes SoH of batteries accurately in real-time and takes self-discharge into its account. At the end, the summary of research findings and the future directions of research such as nondestructive tests (NDT) for real-time estimation of battery SoH, finding residual SoH for the recycled batteries from battery packs, integration of mechanical aspects of battery with temperature, easy assembling–dissembling of battery packs, and hybridization of battery packs with photovoltaic and super capacitor are discussed.