In a smart home linked to a smart grid (SG), demand-side management (DSM) has the potential to reduce electricity costs and carbon/chlorofluorocarbon emissions, which are associated with electricity used in today’s modern society. To meet continuously increasing electrical energy demands requested from downstream sectors in an SG, energy management systems (EMS), developed with paradigms of artificial intelligence (AI) across Internet of things (IoT) and conducted in fields of interest, monitor, manage, and analyze industrial, commercial, and residential electrical appliances efficiently in response to demand response (DR) signals as DSM. Usually, a DSM service provided by utilities for consumers in an SG is based on cloud-centered data science analytics. However, such cloud-centered data science analytics service involved for DSM is mostly far away from on-site IoT end devices, such as DR switches/power meters/smart meters, which is usually unacceptable for latency-sensitive user-centric IoT applications in DSM. This implies that, for instance, IoT end devices deployed on-site for latency-sensitive user-centric IoT applications in DSM should be aware of immediately analytical, interpretable, and real-time actionable data insights processed on and identified by IoT end devices at IoT sources. Therefore, this work designs and implements a smart edge analytics-empowered power meter prototype considering advanced AI in DSM for smart homes. The prototype in this work works in a cloud analytics-assisted electrical EMS architecture, which is designed and implemented as edge analytics in the architecture described and developed toward a next-generation smart sensing infrastructure for smart homes. Two different types of AI deployed on-site on the prototype are conducted for DSM and compared in this work. The experimentation reported in this work shows the architecture described with the prototype in this work is feasible and workable.

Internet of Things (IoT) and artificial intelligence (AI) can realize the concept of “smart city.” Video surveillance in smart cities is, usually, based on a centralized framework in which large amounts of real-time media data are transmitted to and processed in the cloud. However, the cloud relies on network connectivity of the Internet that is sometimes limited or unavailable; thus, the centralized framework is not sufficient for real-time processing of media data needed for smart video surveillance. To tackle this problem, edge computing - a technique for accelerating the development of AIoT (AI across IoT) in smart cities - can be conducted. In this paper, a distributed real-time object detection framework based on edge-cloud collaboration for smart video surveillance is proposed. When collaborating with the cloud, edge computing can serve as converged computing through which media data from distributed edge devices of the network are consolidated by AI in the cloud. After AI discovers global knowledge in the cloud, it to be shared at the edge is deployed remotely on distributed edge devices for real-time smart video surveillance. First, the proposed framework and its preliminary implementation are described. Then, the performance evaluation is provided regarding potential benefits, real-time responsiveness and low-throughput media data transmission.

Convolutional Neural Networks (CNNs) based approaches are popular for various image/video related tasks due to their state-of-the-art performance. However, for problems like object detection and segmentation, CNNs still suffer from objects with arbitrary shapes, sizes, occlusions, and varying viewpoints. This problem makes it mostly unsuitable for fire detection and segmentation since flames can have an unpredictable scale and shape. In this paper, we propose a method that detects and segments fire-regions with special considerations of their arbitrary sizes and shapes. Specifically, our approach uses a self-attention mechanism to augment spatial characteristics with temporal features, allowing the network to reduce its reliance on spatial factors like shape or size and take advantage of robust spatial-temporal dependencies. As a whole, our pipeline has two stages: In the first stage, we take out region proposals using Spatial-Temporal features, and in the second stage, we classify whether each region proposal is flame or not. Due to the scarcity of generous fire datasets, we adopt a transfer learning strategy to pre-train our classifier with the ImageNet dataset. Additionally, our Spatial-Temporal Network only requires semi-supervision, where it only needs one ground-truth segmentation mask per frame-sequence input. The experimental results of our proposed method significantly outperform the state-of-the-art fire detection with a 2 ~ 4% relative enhancement in F1-score for large scale fires and a nearly ~ 60% relative improvement for small fires at a very early stage.

Forest fires are among the most critical natural tragedies threatening forest lands and resources. The accurate and early detection of forest fires is essential to reduce losses and improve firefighting. Conventional firefighting techniques, based on ground inspection and limited by the field-of-view, lead to insufficient monitoring capabilities for large areas. Recently, due to their excellent flexibility and ability to cover large regions, unmanned aerial vehicles (UAVs) have been used to combat forest fire incidents. An essential step for an autonomous system that monitors fire situations is first to locate the fire in a video. State-of-the-art forest-fire segmentation methods based on vision transformers (ViTs) and convolutional neural networks (CNNs) use a single aerial image. Nevertheless, fire has an inconsistent scale and form, and small fires from long-distance cameras lack salient features, so accurate fire segmentation from a single image has been challenging. In addition, the techniques based on CNNs treat all image pixels equally and overlook global information, limiting their performance, while ViT-based methods suffer from high computational overhead. To address these issues, we proposed a spatiotemporal architecture called FFS-UNet, which exploited temporal information for forest-fire segmentation by combining a transformer into a modified lightweight UNet model. First, we extracted a keyframe and two reference frames using three different encoder paths in parallel to obtain shallow features and perform feature fusion. Then, we used a transformer to perform deep temporal-feature extraction, which enhanced the feature learning of the fire pixels and made the feature extraction more robust. Finally, we combined the shallow features of the keyframe for de-convolution in the decoder path via skip-connections to segment the fire. We evaluated empirical outcomes on the UAV-collected video and Corsican Fire datasets. The proposed FFS-UNet demonstrated enhanced performance with fewer parameters by achieving an F1-score of 95.1% and an IoU of 86.8% on the UAV-collected video, and an F1-score of 91.4% and an IoU of 84.8% on the Corsican Fire dataset, which were higher than previous forest fire techniques. Therefore, the suggested FFS-UNet model effectively resolved fire-monitoring issues with UAVs.

Electricity is a vital resource for various human activities, supporting customers’ lifestyles in today’s modern technologically driven society. Effective demand-side management (DSM) can alleviate ever-increasing electricity demands that arise from customers in downstream sectors of a smart grid. Compared with the traditional means of energy management systems, non-intrusive appliance load monitoring (NIALM) monitors relevant electrical appliances in a non-intrusive manner. Fog (edge) computing addresses the need to capture, process and analyze data generated and gathered by Internet of Things (IoT) end devices, and is an advanced IoT paradigm for applications in which resources, such as computing capability, of a central data center acted as cloud computing are placed at the edge of the network. The literature leaves NIALM developed over fog-cloud computing and conducted as part of a home energy management system (HEMS). In this study, a Smart HEMS prototype based on Tridium’s Niagara Framework^® has been established over fog (edge)-cloud computing, where NIALM as an IoT application in energy management has also been investigated in the framework. The SHEMS prototype established over fog-cloud computing in this study utilizes an artificial neural network-based NIALM approach to non-intrusively monitor relevant electrical appliances without an intrusive deployment of plug-load power meters (smart plugs), where a two-stage NIALM approach is completed. The core entity of the SHEMS prototype is based on a compact, cognitive, embedded IoT controller that connects IoT end devices, such as sensors and meters, and serves as a gateway in a smart house/smart building for residential DSM. As demonstrated and reported in this study, the established SHEMS prototype using the investigated two-stage NIALM approach is feasible and usable.

Current methodologies in electrical equipment fault diagnosis are not based on an edge-cloud computing architecture; however, the potential applicability of this architecture for realizing real-time autonomous AI-based fault diagnosing is vast. Collaborative cloud and edge computing brings data computation and storage closer to Internet of Things (IoT) data sources; specifically, data are processed at the periphery of the network (the Internet). In this article, an intelligent fault diagnosis methodology based on infrared thermal imaging (infrared thermography (IRT) images) and deep learning (DL) in an edge-cloud collaborative computing architecture is developed for smart electrical equipment fault diagnosis. In the developed methodology, YOLOv8 is used to perform object detection and classification of objects in captured IRT images of electrical components for on-site and on-line condition monitoring. This algorithm is implemented and run on an nVIDIA ^® Jetson Nano™ developer kit to serve as an autonomous edge AI device in the architecture. The Jetson Nano™ kit is integrated with an FLIR Lepton ^® 3.5 thermal imaging module to capture accurate, calibrated and noncontact temperature data in every pixel of each IRT image. The developed architecture and its preliminary implementation are described and tested. Additionally, the collaborativeness and performance evaluation are shown. As reported, via the proposed methodology, a performance improvement of ~20% in the mAP@.5 in object detection and classification is achieved for an addressed task as an illustrative paradigm of the developed architecture. Additionally, a reduction of ~83% of the bandwidth of the backbone network is gained. Ultimately, the developed architecture may be applied to IRT-based intelligent fault diagnosis in other power equipment in the area of power equipment diagnosis.