Universidad Politécnica Salesiana

Cabrera D., Guamán A., Zhang S., Cerrada M., Sánchez R., Cevallos J., Long J., Li C.

Reciprocating compression machinery is the primary source of compressed air in the industry. Undiagnosed faults in the machinery’s components produce a high rate of unplanned stoppage of production processes that can even result in catastrophic consequences. Fault diagnosis in reciprocating compressors requires complex and time-consuming feature-extraction processes because typical fault diagnosers cannot deal directly with raw signals. In this paper, we streamline the deep learning and optimization algorithms for effective fault diagnosis on these machines. The proposed approach iteratively trains a group of long short-term memory (LSTM) models from a time-series representation of the vibration signals collected from a compressor. The hyperparameter search is guided by a Bayesian approach bounding the search space in each iteration. Our approach is applied to diagnose failures in intake/discharge valves on double-stage machinery. The fault-recognition accuracy of the best model reaches 93% after statistical selection between a group of candidate models. Additionally, a comparison with classical approaches, state-of-the-art deep learning-based fault-diagnosis approaches, and the LSTM-based model shows a remarkable improvement in performance by using the proposed approach.

Khan S.A., Ponce P., Yu Z., Ponce K.

Economic growth is one of the primary macroeconomic aggregates that governments seek to achieve due to the positive externalities it generates in a nation. Natural resources are one of those factors that can be seen as a blessing or a curse in determining economic growth. Nowadays, it has been shown that countries can base their economic growth on factors other than capital formation and the intensive use of the labour force. Therefore, this research aims to examine the causal link between economic growth and capital formation, the labour force, renewable energy and renewable energy, the technological innovation at a global level and in groups of countries classified according to their level of income, developed and developing countries. Next, second-generation econometric techniques have been used that control the cross-sectional dependence among the countries examined. Then, the long-term coefficients are examined using the Method of Moments Quantile Regression (MMQR). The results find that the regressor variables are positively associated with economic growth, showing heterogeneity in the various quantiles examined. In addition, there is variation in results when examining economic growth by groups of countries, which shows the intensive use of factors. The results obtained reveal several policy implications aimed at sustainable growth in the era of post-COVI-19. • Examine the causal link between economic and natural resource dependence. • Regressor variables are positively associated with economic growth indicator. • Variation found in results when examining economic growth by a group of countries. • The results provide several policy implications for the Post-COVID-19 era.

Pérez-Gosende P., Mula J., Díaz-Madroñero M.

Facility layout planning (FLP) involves a set of design problems related to the arrangement of the elements that shape industrial production systems in a physical space. The fact that they are cons...

Caiza G., Saeteros M., Oñate W., Garcia M.V.

The industrial applications in the cloud do not meet the requirements of low latency and reliability since variables must be continuously monitored. For this reason, industrial internet of things (IIoT) is a challenge for the current infrastructure because it generates a large amount of data making cloud computing reach the edge and become fog computing (FC). FC can be considered as a new component of Industry 4.0, which aims to solve the problem of big data, reduce energy consumption in industrial sensor networks, improve the security, processing and storage real-time data. It is a promising growing paradigm that offers new opportunities and challenges, beside the ones inherited from cloud computing, which requires a new heterogeneous architecture to improve the network capacity for delivering edge services, that is, providing computing resources closer to the end user. The purpose of this research is to show a systematic review of the most recent studies about the architecture, security, latency, and energy consumption that FC presents at industrial level and thus provide an overview of the current characteristics and challenges of this new technology.

Bustamante-Torres M., Romero-Fierro D., Arcentales-Vera B., Pardo S., Bucio E.

In recent years, polymer nanocomposites produced by combining nanofillers and a polymeric matrix are emerging as interesting materials. Polymeric composites have a wide range of applications due to the outstanding and enhanced properties that are obtained thanks to the introduction of nanoparticles. Therefore, understanding the filler-matrix relationship is an important factor in the continued growth of this scientific area and the development of new materials with desired properties and specific applications. Due to their performance in response to a magnetic field magnetic nanocomposites represent an important class of functional nanocomposites. Due to their properties, magnetic nanocomposites have found numerous applications in biomedical applications such as drug delivery, theranostics, etc. This article aims to provide an overview of the filler-polymeric matrix relationship, with a special focus on magnetic nanocomposites and their potential applications in the biomedical field.

Yin A., Yan Y., Zhang Z., Li C., Sánchez R.

The gearbox is one of the most fragile parts of a wind turbine (WT). Fault diagnosis of the WT gearbox is of great importance to reduce operation and maintenance (O&M) costs and improve cost-effectiveness. At present, intelligent fault diagnosis methods based on long short-term memory (LSTM) networks have been widely adopted. As the traditional softmax loss of an LSTM network usually lacks the power of discrimination, this paper proposes a fault diagnosis method for wind turbine gearboxes based on optimized LSTM neural networks with cosine loss (Cos-LSTM). The loss can be converted from Euclid space to angular space by cosine loss, thus eliminating the effect of signal strength and improve the diagnosis accuracy. The energy sequence features and the wavelet energy entropy of the vibration signals are used to evaluate the Cos-LSTM networks. The effectiveness of the proposed method is verified with the fault vibration data collected on a gearbox fault diagnosis experimental platform. In addition, the Cos-LSTM method is also compared with other classic fault diagnosis techniques. The results demonstrate that the Cos-LSTM has better performance for gearbox fault diagnosis.

Ortiz L., González J.W., Gutierrez L.B., Llanes-Santiago O.

Microgrids (MG) treat local energy supply issues effectively and from a point of view of the distribution grid, may be a power supply or virtual load. Despite holding a myriad of benefits, MGs also bear a set of challenges, including a higher fault rate. Currently, many articles focus on control techniques; however, little has been written about the techniques of control, hierarchical control, and fault-tolerant control (FTC) applied to MGs, which is the motive of this bibliographic revision on control systems. A brief comparison of the different approaches in the field of present-day research is carried out primarily addressing hierarchical control and fault tolerance. The objective of this investigation is to attract the interest of researchers to the field of control and fault tolerance applied to MGs, such as: modeling, testbed, benchmark systems, control and hierarchical control strategies, fault diagnosis and FTC.

Yu Z., Khan S.A., Ponce P., Zia-ul-haq H.M., Ponce K.

The increasing amount of waste generation and its adverse environmental effects have become a crucial challenge for the global authorities. Under ecological modernization theory and the circular economy model, resource recovery from waste streams emerges as a practical solution depending on the efficient waste management system. The present research explores potential macroeconomic determinants of waste recovery in the Organization of Economic Co-operation and Development countries, covering data from 1995 to 2019. A set of econometric techniques are employed to control the cross-sectional dependence among different variables. Next, the Method of Moments Quantile Regression is applied to analyze the long-term equilibrium relationship of waste recovery with environmental technology, renewable energy consumption, economic growth, globalization, and industrialization. Empirical results reveal the significant influence of targeted variables on waste recovery. Specifically, this study contributes by reporting the positive roles of environmental technology and renewable energy consumption in enhancing waste-to-resource recovery performance. Our findings provide several policy implications for the authorities to promote waste management and recovery towards achieving sustainability . • Environmental technology contributes to waste recovery. • There is heterogeneity in the waste recovery's conditional distribution. • The location-scale effect supports the waste recovery's heterogeneity. • Second generation tests are used to control the cross-sectional dependence. • When compared to other methods, Moments Quantile Regression yields better results.

Barragán-Escandón A., Olmedo Ruiz J.M., Curillo Tigre J.D., Zalamea-León E.F.

This work evaluates the biogas production potential of the Ceibales landfill for feeding a power plant in the southern region of Ecuador. Biogas production is estimated through mathematical models that consider energy generation and technologies available to supply electricity plants. Characteristic landfill data are accounted for to analyze and develop these mathematical models. Once the generation capability of each source is identified, a decision can be made on the most suitable electricity generation technology. A local model (Ecuadorian model) is applied to calculate the potential of biogas and is compared with other models commonly used for evaluating this type of project. This type of renewable energy is attractive because it produces electricity from waste; however, it is not an attractive option unless its application is encouraged, as hydro has been encouraged through the investment of taxpayer resources. Technologies require a boost to become profitable, and even more so if they compete with traditional technologies.

Serrano-Guerrero X., Briceño-León M., Clairand J., Escrivá-Escrivá G.

• The proposed interval prediction methodology limits the predictions' uncertainty. • The prediction interval is straightforward to apply and can be used for any consumer type. • The error is considerably lower than other forecasting methods, especially on holidays. • The prediction interval can detect anomalies and inefficiencies in electricity consumption. Demand prediction has been playing an increasingly important role for electricity management, and is fundamental to the corresponding decision-making. Due to the high variability of the increasing electrical load, and of the new renewable energy technologies, power systems are facing technical challenges. Thus, short-term forecasting has crucial utility for generating dispatching commands, managing the spot market, and detecting anomalies. The techniques associated with machine learning are those currently preferred by researchers for making predictions. However, there are concerns regarding limiting the uncertainty of the obtained results. In this work, a statistical methodology with a simple implementation is presented for obtaining a prediction interval with a time horizon of seven days (15-min time steps), thereby limiting the uncertainty. The methodology is based on pattern recognition and inferential statistics. The predictions made differ from those from a classical approach which predicts point values ​​by trying to minimize the error. In this study, 96 intervals of absorbed active power are predicted for each day, one for every 15 min, along with a previously defined probability associated with the real values ​​being within each obtained interval. To validate the effectiveness of the predictions, the results are compared with those from techniques with the best recent results, such as artificial neural network (ANN) long short-term memory (LSTM) models. A case study in Ecuador is analyzed, resulting in a prediction interval coverage probability (PICP) of 81.1% and prediction interval normalized average width (PINAW) of 10.13%, with a confidence interval of 80%.