IET Cyber-Physical Systems: Theory and Applications最新文献

In industrial environments, workers should wear workwear for safety considerations. For the same reason, smoking is also prohibited. Due to the supervision of monitoring devices, workers have reduced smoking behaviours and started wearing workwear. To meet the requirements for detecting these behaviours in real-time monitoring videos with high speed and accuracy, the authors proposed an improved YOLOv5 model with the Triplet Attention mechanism. This mechanism strengthens the connection between channel and spatial dimensions, focuses the network on important parts, and improves feature extraction. Compared to the original YOLOv5 model, the addition of the mechanism increases the parameters by only 0.04%. The recall rate of the YOLOv5 model is enhanced while its prediction speed is maintained with only a minimal increase in parameters. Experiment results show that, compared to the original model, the improved YOLOv5 has a recall rate of 78.8%, 91%, and 89.3% for detecting smoking behaviour, not wearing helmets, and inappropriate workwear, respectively.

Within the Industry 4.0 landscape, humans collaborate with cyber and physical elements to form human-cyber-physical systems (HCPS). These environments are increasingly complex and challenging workspaces due to increasing levels of automation and data availability. An effective system design requires suitable frameworks that consider human activities and needs whilst supporting overall system efficacy. Although several reviews of frameworks for technology were identified, none of these focused on the human in the system (moving towards Industry 5). The critical literature review presented provides a summary of HCPS frameworks, maps the considerations for a human in HCPS, and provides insight for future framework and system development. The challenges, recommendations, and areas for further research are discussed.

Online presence is becoming an important part of everyday's life and online communities may represent a significant source of engagement for the elderlies. Nevertheless, many may struggle to be online due to a lack of expertise, and a decentralised architecture may provide a solution by removing intermediaries, such as a webmaster, while not requiring expensive cloud solutions. However, issues concerning accessibility, security, and user experience have to be tackled. The paper focuses mainly on three issues: providing a human-readable domain, moderating content, and creating a reward system based on user reputation. An architecture is proposed based on Ethereum and Swarm. Smart contracts provide an automated set of rules to handle enterprise registration, content creation, and decision-making process, while Swarm serves both as distributed storage and the web host. Besides, in combination with Ethereum Name Service, Swarm provides a secure, distributed, and human-readable point of access to the web interface. The paper also describes an innovative two-token system where one token is meant to be a trustworthy reputation metre and the other is a spendable coin to get rewards. The final result is a fully decentralised, authenticated and moderated platform where users can aggregate and share their content presentations on the Internet.

Electricity substitution is an effective measure for reducing pollutant emissions and promoting the consumption of renewable energy. The current research lacks a quantitative analysis method for the factors affecting regional electricity substitution. This article proposes a decomposition model of the factors affecting electricity substitution based on Logarithmic Mean Divisia Index method to expand the depth and breadth of electricity substitution. Results show that: (1) Macro-level electricity substitution in Guangdong province develops steadily, with 1694.93 × 10⁸ kWh of total electricity substitution quantity during the period 2006–2020. The electricity substitution quantity in production sector accounts for 83.92% of the total, which is much larger than that in household sector. (2) Sustained improvement of labour productivity gives the largest contribution (3592.32 × 10⁸ kWh) to the increase of electricity substitution in production sector during the period 2006–2020. Living standard effect has the largest contribution (452.21 × 10⁸ kWh) to the increase of electricity substitution in household sector during the period 2006–2020. Electrification level effect and population effect have significant impact on electricity substitution development, while industrial structure has little impact on electricity substitution. (3) Energy intensity effect and energy consumption growth effect negatively influence electricity substitution in Guangdong province. The decline of energy intensity in the production sector has driven the decrease of electricity substitution quantity, contributing −2540.36 × 10⁸ kWh to electricity substitution during the period 2006–2020. Policy implications like continuously promoting breakthroughs in electricity substitution technologies, improvement of electrification rate in the sectors with potential for substitution, and improvement the price mechanism for fossil fuels and electricity can promote the deepening development of electricity substitution and the realisation of the carbon neutrality goal.

This article studies a distributed consensus-based estimation problem for discrete time-varying non-linear systems with missing measurements and Denial of Service (DoS) attacks. The probability of missing measurements is independent for each sensor. The communication link between sensor nodes is unreliable and subjected to DoS attacks. To achieve accurate state estimation against missing measurements, a local estimator with compensation mechanism is designed for each sensor node. A stochastic event-triggered mechanism is used to lessen additional information transfer. Based on this, a distributed consensus-based estimator is constructed by continually fusing local neighbours information matrixs and vectors. Moreover, the analysis of the designed estimator boundedness is presented. Finally, the effectiveness of the proposed algorithm is verified by three numerical examples.

A non-linear large scale stochastic optimisation model for enhancing the oil production and the recovery factor of the offshore oil reservoirs is proposed. The model aims at minimising the miss-match between mathematical model and the actual dynamic behaviour of the reservoir and the exploitation time, while maximising the oil production and the recovery factor. The model involves the three dimension (3D) oil reservoirs equipped with a few vertical injection and production wells. The limited number of wells is one of the major features of the common oil reservoirs in the middle-east region. The proposed model consists of the primarily mathematical model of the 3D reservoir, a model update algorithm and a large scale constrained non-linear optimisation algorithm. The input to this model is the daily production rate of the oil, natural gas and water produced from the oil reservoir and the output is the optimal injection rate to be injected to the injection wells in order to maximise the oil production and the recovery factor. In order to evaluate the performance of this model, the authors apply this model on part of one of the Iran's offshore oil reservoirs and study the performance improvement due to the proposed model and compare its performance with the performance of the available Improved Oil Recovery (IOR) technique. It is illustrated that the proposed model can increase the oil production from the reservoir up to 47.96% and reduce the exploitation period up to 66.66% compared with those of the available technique.

In order to achieve more efficient and optimised resource scheduling, this research carried out a multi-objective task resource allocation method for low-voltage station edge computing based on hierarchical Bayesian adaptive sparsity. Based on hierarchical Bayesian adaptive sparsity, the multi-objective task resource allocation technical framework for edge computing in low-voltage stations is established, which is composed of end pipe edge cloud; After collecting real-time operation data of power distribution equipment, substation terminals, transmission terminals, etc. in the architecture end, it is transmitted to the data middle platform and service middle platform of the Internet of Things management platform in the cloud through the edge Internet of Things agent; Set and solve the constraint conditions, and build a multi type flexible load hierarchical optimal allocation model; The abnormal area topology identification sub module of multi-objective task resource of low-voltage station area edge computing is used to identify the abnormal area topology of the current low-voltage station area; Taking it as input, the multi-objective task resources of edge computing are allocated, and the multi-objective task resources allocation method of edge computing in low pressure platform area is realized under the differential evolution algorithm. The experimental results show that the proposed method has good convergence effect, strong distribution ability, relatively gentle increase in energy consumption, and the calculated results are basically consistent with the actual values, with good effectiveness.

With the emergence of powerful and low-energy Internet of Things devices, deep learning computing is increasingly applied to resource-constrained edge devices. However, the mismatch between hardware devices with low computing capacity and the increasing complexity of Deep Neural Network models, as well as the growing real-time requirements, bring challenges to the design and deployment of deep learning models. For example, autonomous driving technologies rely on real-time object detection of the environment, which cannot tolerate the extra latency of sending data to the cloud, processing and then sending the results back to edge devices. Many studies aim to find innovative ways to reduce the size of deep learning models, the number of Floating-point Operations per Second, and the time overhead of inference. Neural Architecture Search (NAS) makes it possible to automatically generate efficient neural network models. The authors summarise the existing NAS methods on resource-constrained devices and categorise them according to single-objective or multi-objective optimisation. We review the search space, the search algorithm and the constraints of NAS on hardware devices. We also explore the challenges and open problems of hardware NAS.

A Cyber-Physical System (CPS) is a spatiotemporal multi-dimensional and heterogeneous hybrid autonomous system composed of deep integration of information resources and physical systems. With the development of digitisation and digitalisation, a large number of data acquisition equipment, computing equipment, and electrical equipment are interconnected between the power grid and the information communication network. The power grid has thus been restructured as a mature and highly complex CPS. In order to promote the development of power grid CPS technologies and provide a reference for relevant researchers in the field, the origin and concept of CPS and features in power grid CPS are introduced firstly. Then the key technologies of power grid CPS simulation are discussed and further analysed from three perspectives, including modelling theory, simulation methods, and system-level simulation. On this basis, the application of CPS simulation technology in future power grids has been prospected.

False Data Injection (FDI) attacks can be injected into the communication links of microgrids and result in significant damages. An approach to detect false data injection (FDI) attacks is introduced and a method to maintain the average voltages of buses within an acceptable range is suggested. To this aim, the impacts of FDI are first studied and a detection mechanism is extracted based on the attack effects. Then, multiple local integral feedback is employed to maintain the voltages within a permitted interval. Next, the procedure to return the values to their initial conditions (following the removal of the attack) is addressed. Finally, the proposed methods are simulated by multiple attack scenarios and validated using experimental tests.