Fault detection model for a variable speed heat pump

The impact of the COVID pandemic has resulted in many people cultivating a remote working culture and increasing building energy use. A reduction in the energy use of heating, ventilation, and air-conditioning (HVAC) systems is necessary for decreasing the energy use in buildings. The refrigerant charge of a heat pump greatly affects its energy use. However, refrigerant leakage causes a significant increase in the energy use of HVAC systems. The development of refrigerant charge fault detection models is, therefore, important to prevent unwarranted energy consumption and CO2 emissions in heat pumps. This paper examines refrigerant charge faults and their effect on a variable speed heat pump and the most accurate method between a multiple linear regression and multilayer perceptron model to use in detecting the refrigerant charge fault using the discharge temperature of the compressor, outdoor entering water temperature and compressor speed as inputs, and refrigerant charge as the output. The COP of the heat pump decreased when it was not operating at the optimum refrigerant charge, while an increase in compressor speed compensated for the degradation in the capacity during refrigerant leakage. Furthermore, the multilayer perception was found to have a higher prediction accuracy of the refrigerant charge fault with a mean square error of ± 3.7%, while the multiple linear regression model had a mean square error of ± 4.5%. The study also found that the multilayer perception model requires 7 neurons in the hidden layer to make viable predictions on any subsequent test sets fed into it under similar experimental conditions and parameters of the heat pump used in this study.

Teaching Geology in Higher Education Institutions under COVID-19 Conditions

Teaching geology under COVID-19 pandemic conditions led to teaching limitations for educators and learning difficulties for students. The lockdown obstructed face-to-face teaching, laboratory work, and fieldtrips. To minimize the impact of this situation, new distance learning teaching methods and tools were developed. The current study presents the results of an empirical study, where distance learning teaching tools were constructed and used to teach geology to university students. A mineralogical mobile phone application was used to replace laboratory mineral identification and a flow chart to replace laboratory rock identification. Additionally, exercises on faults and maps were developed to fill the gap that was created as field work was impossible. A university course on geology was designed on the basis of the constructed distance learning teaching tools, and more than 100 students from the Department of Civil Engineering attended the course. The results show that the proposed tools helped the students to considerably understand scientific information on geology and supported the learning outcomes. Thus, it is suggested that the teaching tools, constructed for the purposes of the study, could be used in conditions when distance learning is required, or even under typical learning conditions after laboratories, as well as before or after fieldtrips, for better learning outcomes.

Intelligent OpenFlow Switch for SDN Networks Based on COVID-19’s High Network Traffic using Heuristic GA-Fuzzification Control Approach

The COVID-19 outbreak has impacted network operators and data centers in terms of congestion and high traffic that lead to outages and significant pressure on the network. The overhead traffic is generated from web, voice calls, and Internet activity. In this paper, we are investigating data center congestion control for Software Defined Networks (SDN) network data centers. A Software-Defined (SDN) data center is an emerging networking paradigm that simplifies the network architecture by decentralizing plane functionality into a single with centralized decision capabilities. Along with the SDN paradigm, there is a crucial part that is responsible for forwarding packet called OpenFlow switching engine. In a typical SDN environment, the rules are initiated by the SDN controller and pushed to the OpenFlow switches. The traditional OpenFlow switch has no forwarding decision and depends on the incoming policies from the controller’s southbound interface. Additionally, the flow of traffic is initiated from different sources that are assigned to a specific route. However, this significant flow of traffic due to COVID-19 can lead to congestion and degradation of network performance in terms of delay and interruption. To be precise, a single OpenFlow switch could receive a capacity of traffic that floods its forwarding table and lead to link flaps and outages. In order to optimize the OpenFlow switch with regards to how much traffic it can host and to adjust routing capabilities for dynamic changes in the network, we propose an optimized OpenFlow congestion control and fault prediction framework for inbound traffic to overcome the inefficient route planning in the network. The proposed developed optimization algorithm is based on Genetic Evolutionary Algorithm criteria and adds intelligence to the OpenFlow switch by the adoption of Fuzzy Logic prediction capabilities. The experimental evaluation shows that the proposed optimization method adds significant intelligence and optimization to OpenFlow operation. The testbed was implemented experimentally using Raspberry Pi (RPI)cluster with customized SDN and OpenFlow deployment. The probability of the best fitness was 14.11% for Gen 999. The proposed approach adds intelligence and prediction into the OpenFlow switch to overcome the unstable flows of traffic and to predict faults to enhance the traffic capacity levels and manage flows into an entirely uninterrupted production environment. © 2022 IEEE.

Predictive Maintenance 4.0 for Chilled Water System at Commercial Buildings: A Systematic Literature Review

Predictive maintenance plays an important role in managing commercial buildings. This article provides a systematic review of the literature on predictive maintenance applications of chilled water systems that are in line with Industry 4.0/Quality 4.0. The review is based on answering two research questions about understanding the mechanism of identifying the system’s faults during its operation and exploring the methods that were used to predict these faults. The research gaps are explained in this article and are related to three parts, which are faults description and handling, data collection and frequency, and the coverage of the proposed maintenance programs. This article suggests performing a mixed method study to try to fill in the aforementioned gaps.

Towards an Accurate Faults Detection Approach in Internet of Medical Things Using Advanced Machine Learning Techniques.

Remotely monitoring people's healthcare is still among the most important research topics for researchers from both industry and academia. In addition, with the Wireless Body Networks (WBANs) emergence, it becomes possible to supervise patients through an implanted set of body sensors that can communicate through wireless interfaces. These body sensors are characterized by their tiny sizes, and limited resources (power, computing, and communication capabilities), which makes these devices prone to have faults and sensible to be damaged. Thus, it is necessary to establish an efficient system to detect any fault or anomalies when receiving sensed data. In this paper, we propose a novel, optimized, and hybrid solution between machine learning and statistical techniques, for detecting faults in WBANs that do not affect the devices' resources and functionality. Experimental results illustrate that our approach can detect unwanted measurement faults with a high detection accuracy ratio that exceeds the 99.62%, and a low mean absolute error of 0.61%, clearly outperforming the existing state-of-art solutions.

Strategy Based on Two Stages for IR Thermographic Inspections of Photovoltaic Plants

Aerial infrared (IR) thermography has been implemented in recent years, proving to be a powerful and versatile technique for performing maintenance at photovoltaic (PV) plants. Its application speed and reliability using unmanned aerial vehicles (UAVs) or drones make it extremely interesting at large PV plants, due to the associated savings in time and costs. Ground-level thermographic inspection is slower and more costly to apply, although it does provide higher optical resolution, due to being conducted closer to the PV modules being inspected. Both techniques used in combination can improve the diagnosis. An IR thermography inspection strategy is proposed for PV plants based on two stages. The first stage of the inspection is aerial, enabling thermal faults to be detected and located quickly and reliably. The second stage of the inspection is done on the ground and applied only to the most relevant incidents revealed in the first stage. This inspection strategy was applied to a 100 kW PV plant, with an improved diagnosis verified via this procedure, as the ground-level inspection detects one-off thermal incidents from objects creating shade and from solar reflections. For PV modules with open circuits or open substrings, the use of one technique or another is immaterial.

Cloud Computing and IoT Based Intelligent Monitoring System for Photovoltaic Plants Using Machine Learning Techniques

This paper proposes an Intelligent Monitoring System (IMS) for Photovoltaic (PV) systems using affordable and cost-efficient hardware and also lightweight software that is capable of being easily implemented in different locations and having the capability to be installed in different types of PV power plants. IMS uses the Internet of Things (IoT) platform for handling data as well as Interoperability and Communication among the devices and components in the IMS. Moreover, IMS includes a personal cloud server for computing and storing the acquired data of PV systems. The IMS also consists of a web monitor system via some open-source and lightweight software that displays the information to multiple users. The IMS uses deep ensemble models for fault detection and power prediction in PV systems. A remarkable ability of the IMS is the prediction of the output power of the PV system to increase energy yield and identify malfunctions in PV plants. To this end, a long short-term memory (LSTM) ensemble neural network is developed to predict the output power of PV systems under different environmental conditions. On the other hand, the IMS uses machine learning-based models to detect numerous faults in PV systems. The fault diagnostic of IMS is based on the following stages. Firstly, major features are elicited through an analysis of Current–Voltage (I–V) characteristic curve under different faulty and normal events. Second, an ensemble learning model including Naive Bayes (NB), K-Nearest Neighbors (KNN), and Support Vector Machine (SVM) is used for detecting and classifying fault events. To enhance the performance in the process of fault detection, a feature selection algorithm is also applied. A PV system has been designed and implemented for testing and validating the IMS under real conditions. IMS is an interoperable, scalable, and replicable solution for holistic monitoring of PV plant from data acquisition, storing, pre-and post-processing to malfunction and failure diagnosis, performance and energy yield assessment, and output power prediction.

Virtual field trip to the Esla Nappe (Cantabrian Zone, NW Spain): delivering traditional geological mapping skills remotely using real data

The restrictions implemented to contain the spread of the COVID-19 pandemic during 2020 and 2021 have forced university-level educators from around the world to seek alternatives to the residential physical field trips that constitute a fundamental pillar of Geoscience programmes. The field-mapping course for second-year Geology BSc students from Cardiff University was replaced with a virtual mapping course set in the same area as previous years, the Esla Nappe (Cantabrian Zone, NW Spain). The course was designed with the aim of providing the students with the same methodology employed in physical mapping, including such skills as gathering discrete data at stops located along five daily itineraries. Data included bedding attitude, outcrop descriptions with a certain degree of ambiguity, photographs and/or sketches, panoramic photos, and fossil images. Data were provided to the students through georeferenced KMZ files in Google Earth. Students were asked to keep a field notebook, define lithological units of mappable scale, identify large structures such as thrust faults and folds with the aid of age estimations from fossils, construct a geological map on a hard-copy topographic map, draw a stratigraphic column and cross sections, and plot the data in a stereonet to perform structural analysis. The exercise allowed for successful training of diverse geological field skills. In light of the assessment of reports and student surveys, a series of improvements for the future is considered. Though incapable of replacing a physical field course, the virtual exercise could be used in preparation for the residential field trip.