The green conversion of proton exchange membrane fuel cells (PEMFCs) has received particular attention in both stationary and transportation applications. However, the poor durability of PEMFC represents a major problem that hampers its commercial application since dynamic operating conditions, including physical deterioration, have a serious impact on the cell performance. Under these circumstances, prognosis and health management (PHM) plays an important role in prolonging durability and preventing damage propagation via the accurate planning of a condition-based maintenance (CBM) schedule. In this specific topic, health deterioration modeling with deep learning (DL) is the widely studied representation learning tool due to its adaptation ability to rapid changes in data complexity and drift. In this context, the present paper proposes an investigation of further deeper representations by exposing DL models themselves to recurrent expansion with multiple repeats. Such a recurrent expansion of DL (REDL) allows new, more meaningful representations to be explored by repeatedly using generated feature maps and responses to create new robust models. The proposed REDL, which is designed to be an adaptive learning algorithm, is tested on a PEMFC deterioration dataset and compared to its deep learning baseline version under time series analysis. Using multiple numeric and visual metrics, the results support the REDL learning scheme by showing promising performances.

Graph theory is a vast field that constitutes a very important body of knowledge. Indeed, this book is just an introduction aiming at clarifying some essential points in this vital field: basic notions, some basic algorithms that are used to solve some classical and famous problems like path finding, tree finding, flow finding, ...etc. Finally, graph theory can be summarized by what Napoleon said: "A little drawing is better than a big speech".

Federated learning (FL) is a data-privacy-preserving, decentralized process that allows local edge devices of smart infrastructures to train a collaborative model independently while keeping data localized. FL algorithms, encompassing a well-structured average of the training parameters (e.g., the weights and biases resulting from training-based stochastic gradient descent variants), are subject to many challenges, namely expensive communication, systems heterogeneity, statistical heterogeneity, and privacy concerns. In this context, our paper targets the four aforementioned challenges while focusing on reducing communication and computational costs by involving recursive least squares (RLS) training rules. Accordingly, to the best of our knowledge, this is the first time that the RLS algorithm is modified to completely accommodate non-independent and identically distributed data (non-IID) for federated transfer learning (FTL). Furthermore, this paper also introduces a newly generated dataset capable of emulating such real conditions and of making data investigation available on ordinary commercial computers with quad-core microprocessors and less need for higher computing hardware. Applications of FTL-RLS on the generated data under different levels of complexity closely related to different levels of cardinality lead to a variety of conclusions supporting its performance for future uses.

The clean energy conversion characteristics of proton exchange membrane fuel cells (PEMFCs) have given rise to many applications, particularly in transportation. Unfortunately, the commercial application of PEMFCs is hampered by the early deterioration and low durability of the cells. In this case, accurate real-time condition monitoring plays an important role in extending the lifespan of PEMFCs through accurate planning of maintenance tasks. Accordingly, among the widely used modeling tools such as model-driven and data-driven, machine learning has received much attention and has been extensively studied in the literature. Small-scale machine learning (SML) and Deep Learning (DL) are subcategories of machine learning that have been exploited so far. In this context and since SML usually contains non-expansive approximators, this study was dedicated to improving its feature representations for better predictions. Therefore, a recurrent expansion experiment was conducted for several rounds to investigate a linear regression model under time series prognosis of PEMFCs. The results revealed that the prediction performance of SML tools under stationary conditions could be further improved.

Proton Exchange Membrane Fuel Cells (PEMFCs) are a key challenger for the world’s future clean and renewable energy solution. Yet, fuel cells are susceptible to operating conditions and hydrogen impurities, leading to performance loss over time in service. Hence, performance degradation prediction is gaining attention recently for fuel cell system reliability. In this work, we present a knowledge transfer approach for online voltage drop prediction. A dual-path convolution neural network is proposed to extract linearity and non-linearity from historical data and performs multi-steps ahead prediction with uncertainty quantification. Online voltage prediction is then evaluated with and without knowledge transfer using two different PEMFC datasets. Results indicate that our proposed approach with transfer knowledge can predict the voltage drop accurately with a small uncertainty range compared to the conventional approach.

In modern Smart Grids (SGs) ruled by advanced computing and networking technologies, condition monitoring relies on secure cyberphysical connectivity. Due to this connection, a portion of transported data, containing confidential information, must be protected as it is vulnerable and subject to several cyber threats. SG cyberspace adversaries attempt to gain access through networking platforms to commit several criminal activities such as disrupting or malicious manipulation of whole electricity delivery process including generation, distribution, and even customer services such as billing, leading to serious damage, including financial losses and loss of reputation. Therefore, human awareness training and software technologies are necessary precautions to ensure the reliability of data traffic and power transmission. By exploring the available literature, it is undeniable that Machine Learning (ML) has become the latest in the timeline and one of the leading artificial intelligence technologies capable of detecting, identifying, and responding by mitigating adversary attacks in SGs. In this context, the main objective of this paper is to review different ML tools used in recent years for cyberattacks analysis in SGs. It also provides important guidelines on ML model selection as a global solution when building an attack predictive model. A detailed classification is therefore developed with respect to data security triad, i.e., Confidentiality, Integrity, and Availability (CIA) within different types of cyber threats, systems, and datasets. Furthermore, this review highlights the various encountered challenges, drawbacks, and possible solutions as future prospects for ML cybersecurity applications in SGs.

The purpose of this paper is to propose an improvement strategy based on multi-criteria decision making approaches, including fuzzy analytic hierarchy process (AHP), preference ranking organisation method for enrichment evaluation II (PROMETHEE) and vi\v sekriterijumsko kompromisno rangiranje (VIKOR) for the objective of simplifying and organising the improvement process in complex manufacturing processes. Firstly, the proposed strategy started with the selection of decision makers’, such as company leaders, to determine performance indicators. Then fuzzy AHP is used to quantify the weight of each defined indicators. Finally, the weights carried out from fuzzy AHP approach are used as input in VIKOR and PROMETHE II to rank the operations according to their improvement priority. The results obtained from each outranking method are compared and the best method is determined.

This paper proposes a new approach for remaining useful life prediction that combines a bond graph, the Gaussian Mixture Model and similarity techniques to allow the use of both physical knowledge and the data available. The proposed method is based on the identification of relevant variables that carry information on degradation. To this end, the causal properties of the bond graph (BG) are first used to identify the relevant sensors through the fault observability. Then, a second stage of analysis based on statistical metrics is performed to reduce the number of sensors to only the ones carrying useful information for failure prognosis, thus, optimizing the data to be used in the prognosis phase. To generate data in the different system state, a simulator based on the developed BG is used. A Gaussian Mixture Model is then applied on the generated data for fault diagnosis and clustering. The Remaining Useful Life is estimated using a similarity technique. An application on a mechatronic system is considered for highlighting the effectiveness of the proposed approach.

The grasshopper optimization algorithm is one of the recently population-based optimization techniques inspired by the behaviours of grasshoppers in nature. It is an efficient optimization algorithm and since demonstrates excellent performance in solving continuous problems, but cannot resolve directly binary optimization problems. Many optimization problems have been modelled as binary problems since their decision variables varied in binary space such as feature selection in data classification. The main goal of feature selection is to find a small size subset of feature from a sizeable original set of features that optimize the classification accuracy. In this paper, a new binary variant of the grasshopper optimization algorithm is proposed and used for the feature subset selection problem. This proposed new binary grasshopper optimization algorithm is tested and compared to five well-known swarm-based algorithms used in feature selection problem. All these algorithms are implemented and experimented assessed on twenty data sets with various sizes. The results demonstrated that the proposed approach could outperform the other tested methods.

In the last recent years, a large community of researchers and industrial practitioners has been attracted by combining different prognostics models as such strategy results in boosted accuracy and robust performance compared to the exploitation of single models. The present work is devoted to the investigation of three new fusion schemes for the remaining useful life forecast. These integrated frameworks are based on aggregating a set of Gaussian process regression models thanks to the Induced Ordered Weighted Averaging Operators. The combination procedure is built upon three proposed analytical weighting schemes including exponential, logarithmic and inverse functions. In addition, the uncertainty aspect is supported in this work, where the proposed functions are used to weighted average the variances released from competitive Gaussian process regression models. The training data are transformed into gradient values, which are adopted as new training data instead of the original observations. A lithium-ion battery data set is used as a benchmark to prove the efficiency of the proposed weighting schemes. The obtained results are promising and may provide some guidelines for future advances in performing robust fusion options to accurately estimate the remaining useful life.

In this study, we investigate a production planning problem in hybrid manufacturing remanufacturing production system. The objective is the determine the best mix between the manufacturing of new products, and the remanufacturing of recovered products, based on economic and environmental considerations. It consists to determine the best manufacturing and remanufacturing plans to minimising the total economic cost (start-up and production costs of new and remanufactured products, storage costs of new and returned products and disposal costs) and the carbon emissions (new products, remanufactured products and disposed products). The hybrid system consists of a set of machines used to produce new products and remanufactured products of different grades (qualities). We assume that remanufacturing is more environmentally efficient, because it allows to reduce the disposal of used products. A multi-objective mathematical model is developed, and a non dominated sorting genetic algorithm (NSGA-II) based approach is proposed. Numerical experience is presented to study the impact of carbon emissions generated by new, remanufactured and disposed products, over a production horizon of several periods.

In this study, we investigate a production planning problem in hybrid manufacturing remanufacturing production system. The objective is the determine the best mix between the manufacturing of new products, and the remanufacturing of recovered products, based on economic and environmental considerations. It consists to determine the best manufacturing and remanufacturing plans to minimising the total economic cost (start-up and production costs of new and remanufactured products, storage costs of new and returned products and disposal costs) and the carbon emissions (new products, remanufactured products and disposed products). The hybrid system consists of a set of machines used to produce new products and remanufactured products of different grades (qualities). We assume that remanufacturing is more environmentally efficient, because it allows to reduce the disposal of used products. A multi-objective mathematical model is developed, and a non dominated sorting genetic algorithm (NSGA-II) based approach is proposed. Numerical experience is presented to study the impact of carbon emissions generated by new, remanufactured and disposed products, over a production horizon of several periods.

Cooperation in distribution network has attracted the interest of researchers. In this study we analyse an inventory problem in distribution network, where we propose a cooperative platform that allow the members of the network to share and use local inventory of other members to meet their local demand. We develop a MIP models representing the traditional network and the network with the cooperative platform. Then we solve it using LINGO solver. We found that the proposed approach has reduced the total cost of the network and reduce the overstock and stock-out situation, which lead to improve the quality of service.