Fault diagnosis in manufacturing systems represents one of the most critical challenges dealing with condition-based monitoring in the recent era of smart manufacturing. In the current Industry 4.0 framework, maintenance strategies based on traditional data-driven fault diagnosis schemes require enhanced capabilities to be applied over modern production systems. In fact, the integration of multiple mechanical components, the consideration of multiple operating conditions, and the appearance of combined fault patterns due to eventual multi-fault scenarios lead to complex electromechanical systems requiring advanced monitoring strategies. In this regard, data fusion schemes supported with advanced deep learning technology represent a promising approach towards a big data paradigm using cloud-based software services. However, the deep learning models’ structure and hyper-parameters selection represent the main limitation when applied. Thus, in this paper, a novel deep-learning-based methodology for fault diagnosis in electromechanical systems is presented. The main benefits of the proposed methodology are the easiness of application and high adaptability to available data. The methodology is supported by an unsupervised stacked auto-encoders and a supervised discriminant analysis.

Heating, ventilation and air-conditioning (HVAC) systems are the major energy consumers among buildings’ equipment. Reliable fault detection and diagnosis schemes can effectively reduce their energy consumption and maintenance costs. In this respect, data-driven approaches have shown impressive results, but their accuracy depends on the availability of representative data to train the models, which is not common in real applications. For this reason, transfer learning is attracting growing attention since it tackles the problem by leveraging the knowledge between datasets, increasing the representativeness of fault scenarios. However, to date, research on transfer learning for heating, ventilation and air-conditioning has mostly been focused on learning algorithmic, overlooking the importance of a proper domain similarity analysis over the available data. Thus, this study proposes the design of a transfer learning approach based on a specific data selection methodology to tackle dissimilarity issues. The procedure is supported by neural network models and the analysis of eventual prediction uncertainties resulting from the assessment of the target application samples. To verify the proposed methodology, it is applied to a semi-supervised transfer learning case study composed of two publicly available air-handling unit datasets containing some fault scenarios. Results emphasize the potential of the proposed domain dissimilarity analysis reaching a classification accuracy of 92% under a transfer learning framework, an increase of 37% in comparison to classical approaches.

Smart energy factories are crucial for the development of upcoming energy markets in which emissions, energy use and network congestions are to be decreased. The virtual power plant (VPP) can be implemented in an industrial site with the aim of minimizing costs, emissions and total energy usage. A VPP considers the future situation forecasting and the situation of all energy assets, including renewable energy generation units and energy storage systems, to optimize the total cost of the plant, considering the possibility to trade with the energy market. For a VPP to be constructed, a proper communication system is essential. The energy management system (EMS) enables the monitoring, management and control of the different energy devices and permits the transference of the decisions made by the VPP to the different energy assets. VPP concept is explained together with the methods used for forecasting the future situation and the energy flow inside the facility. To reach its benefits, the optimization of the VPP is assessed. After that, the communication technologies that enable the VPP implementation are also introduced, and the advantages/disadvantages regarding their deployment are stated. With the tools introduced, the VPP can face the challenges of energy markets efficiently.

In the ongoing energy transition, small and medium-sized industrial companies are making energy equipment investments due to the obsolescence of their current equipment as well as social, political and market pressures. These firms typically choose investments with low risk exposure based on a combination of criteria that are not always quantifiable. However, published studies on energy investment to date have not been suitable for industrial SMEs because they do not assess the value of the investment over time, ignore the qualitative aspects of decision-making, and do not consider uncertainties. To fill this gap in the literature, this paper proposes a methodology that considers both quantitative and qualitative parameters and risks over time through an extended two-stage risk-informed approach. The proposed methodology includes fuzzy and statistical techniques for evaluating both qualitative and quantitative parameters, as well as their uncertainties, at the time of decision-making and over the investment lifetime. Fuzzy logic is used in the first stage of the optimisation process to measure qualitative parameters and their uncertainty, while quantitative parameters are expressed using probability density functions to account for their uncertainty and measure the quantitative risk assumed by the investor. This methodology is applied to a case study involving a real industrial SME, and the results show that considering both quantitative and qualitative parameters and uncertainties in the optimisation process leads to a more balanced consideration of economic, environmental and social criteria and reduces the variability of the outcome compared to economic-only approaches that do not account for risks. Specifically, the case study shows that considering these parameters and uncertainties resulted in a 15.7% reduction in the size of the cogeneration system due to its environmental and social impacts, and 4.2% reduction in the variability of the economic result.