Sustainable Computing-Informatics & Systems最新文献

With the rapid development of the digital economy and intelligent industry, the energy consumption of data centers (DCs) has increased significantly. Various optimization methods are proposed to improve the energy efficiency of servers in DCs. However, existing solutions usually adopt model-based heuristics and best practices to select operations, which are not universally applicable. Moreover, existing works primarily focus on the optimization methods for individual components, with a lack of work on the joint optimization of multiple components. Therefore, we propose a multi-agent reinforcement learning-based method, named MRDF, combining DVFS and dynamic fan control to achieve a trade-off between power consumption and performance while satisfying thermal constraints. MRDF is model-free and learns by continuously interacting with the real server without prior knowledge. To enhance the stability of MRDF in dynamic environments, we design a data-driven baseline comparison method to evaluate the actual contribution of a single agent to the global reward. In addition, an improved Q-learning is proposed to deal with the large state and action space of the multi-core server. We implement MRDF on a Huawei Taishan 200 server and verify the effectiveness by running benchmarks. Experimental results show that the proposed method improves energy efficiency by an average of 3.9% compared to the best baseline solution, while flexibly adapting to different thermal constraints.

This study presents a novel Hunger Game Search and Niching Chimp Optimization Algorithms (HGS-NChOA) for optimizing grid-connected desalination plants powered by renewable energy. The primary innovation of this study is the significant advantages provided by the HGS-NChOA method, particularly in terms of reducing the cost of freshwater production and mitigating greenhouse gas emissions. Key outcomes reveal the HGS-NChOA’s superiority in reducing freshwater production costs and greenhouse gas emissions. Notably, the desalination unit capacity decreased from 10.4 m³ to 8.5 m³ , with a cost reduction of 0.223 $/m³ in the PV-battery storage-wind turbine system. Experimental results show a 59% and 49% decrease in computation time for the PV-battery and PV-hydrogen systems, respectively. Sensitivity analysis highlights the significant impact of solar irradiation on investment costs. Overall, HGS-NChOA demonstrates enhanced efficiency and economic viability in managing grid-connected, renewable energy-powered desalination facilities. Sensitivity analysis showed that solar radiation has a more significant impact on investment costs compared to wind speed, with hourly solar radiation fluctuations affecting water production costs by 17.09% to 19.56%. Additionally, the study indicates that integrating a diesel generator into the system can further reduce costs and greenhouse gas emissions, proving HGS-NChOA’s versatility in optimizing hybrid energy systems. Statistical analysis using metrics like Inverted Generational Distance (IGD) and Maximum Spread (MaxS) demonstrated the proposed method’s superior convergence and diversity compared to well-known multi-objective algorithms like MOPSO, MOEA/D, and MOGWO-PSO.

This paper examines the energy efficiency of a non-cooperative and cooperative On-body Wireless Body Area Network (WBAN) and link reliability in a cooperative WBAN using IEEE 802.15.6 based CM3A channel model. The proposed energy optimization framework is based on Whale Optimization Algorithm (WOA) and Particle Swarm Optimization (PSO) techniques. For the optimized code rate, obtained results illustrate that WOA gives 44.7%. and 43.4% better results for the non-cooperative and cooperative cases respectively than the PSO. The link reliability has been investigated by observing the effect of the fading parameter (m) on the outage probability over the Nakagami m fading channel. The obtained results reveal how the fading parameter (m) affected the likelihood of an outage.

Due to the growth of communication networks, energy consumption in information and communication technology industries is increasing dramatically. Among these industries, data centers are operating with a large number of processors and other components, which, due to heavy processing and mass data transmission, in addition to consuming high electrical power, also cause high thermal losses. Since water cooling systems are used for cooling different parts of these centers as well as for cooling the electric power generation unit, these centers are among the major consumers of water and electricity resources. In this article, while examining the important factors affecting water consumption in data centers, useful methods are suggested to reduce the consumption of electrical energy and water. Optimizing energy consumption in data centers is possible in three parts: routing, servicing and use of cooling equipment. For all three parts, improvement methods are suggested in this article. For this purpose, an optimization problem is designed and an algorithm is presented to solve it. In the proposed solution, an energy-smart method based on SDN technology is used for routing, virtual machines equipped with the possibility of reducing power consumption in no-load conditions are used for servicing, and two types of water-cooled and air-cooled systems are used for cooling equipment. is used. The simulation results show that depending on the cooling system, the proposed method reduces water consumption between 24% and 32% compared to the case where the proposed solution is not used.

Mobile Edge Computing (MEC) is deployed closer to User Equipment (UE) and has strong computing power. Not only it relieves the load pressure on the central cloud, but also effectively reduces the transmission delay caused by offloading computation from devices because it is closer to users. Therefore, we study edge computing task offloading based on edge–cloud collaboration scenarios to meet the requirement of low delay and high energy efficiency. In order to improve the convergence accuracy and system energy efficiency, we proposed a hybrid bio-inspired algorithm, the HS-HHO algorithm, which combines the Slime Mode Algorithm (SMA) and the optimized Harris Hawks Optimizer (HHO). For different types of tasks, we design a task clustering scheme based on K-medoids clustering for edge cloud scenarios, which clusters tasks into computation-intensive, data-intensive, and integrated, and is used to optimize the offloading objectives of each type of tasks. Experimental results demonstrate that our proposed HS-HHO algorithm takes into account the time delay while effectively reducing energy consumption and making full use of the computational resources. The HS-HHO algorithm improves the total energy efficiency of the system by about 22% compared with the SMA, HHO, and AO algorithm strategies.

The advancement of artificial intelligence and autonomous control has resulted in the widespread use of unmanned aerial vehicles (UAVs) in a variety of large-scale practical applications like target tracking, disaster surveillance, and traffic monitoring. Heterogeneous UAVs outperform homogeneous UAVs in terms of energy consumption and performance. The use of several unmanned aerial vehicles (UAVs) inside broad cooperative search systems, including numerous separate locations, provides the difficulty of sophisticated path planning. The computational complexity of NP-hard problems makes coverage path planning a difficult challenge to solve. This difficulty stems from the need to establish the most effective paths for unmanned aerial vehicles (UAVs) to thoroughly explore selected areas of interest. In this paper, Improved Green Anaconda Optimization Algorithm-based Coverage Path Planning Mechanism is proposed for handling the problem of coverage path planning in UAVs. It specifically adopted an improved Green Anaconda Optimization System (IGAOS) to determines possible and potential paths for the UAVs to fully cover the complete regions of interest in an efficient manner. Initially, the regions and models of UAVs are established using linear programming for identifying the best-to-point flight path for each UAV. It is proposed for minimizing the tasks’ time consumption in the system of cooperative search through the exploration of optimal solution depending on the inspiration derived from the hunting and mating strategy of green anacondas. Experiments on deviation ratio, task completion time, and execution time with this IGAOS revealed its advantages over prior PPSOESSA, HFACPP, ACSCPP, and GAGPSCPP approaches.

In Vehicular ad hoc network (VANETs), Quality of Service (QoS)- aware protocols helps in handling the necessitated demand of delay sensitive applications for facilitating intelligent transportation. The fundamental challenge of VANETs lies in the process of establishing vehicle to infrastructure and vehicle-to-vehicle communication that are prone to link failure. Bio-inspired algorithms are identified to provide reliable solutions for securing the links of VANETs. In this paper, Komodo Mlipir Algorithm-based Optimal Route Determination Mechanism (KMAORDM) is proposed for achieving best optimal route that guarantees enhanced QoS. This QoS aware routing protocol utilizes the exceptional potentiality of Komodo Mlipir Algorithm attributed in terms of exploration and exploitation for the objective of evaluating and memorizing the impactful factors that helps in exchanging the necessitated messages to its neighbours. It adopts Mlipir as a best mode of modelling the behaviours of vehicular nodes during the process of data routing in VANETs. It uses a cache strategy which discovered reliable traversed routing paths such that the pre-cached route is used for data transmission. This cache strategy paves the way for preventing the route identification and maintenance process with minimized routing overhead. It also identifies the fittest vehicular node from its one hop distance as the successive forwarder in the absence of pre-cached route for addressing link failure during reliable packet transmission. The simulation results of this proposed KMAORDM approach with different vehicular nodes confirmed minimized communication overhead of 19.32 %, reduced network latency of 18.64 %, maximized throughput of 18.54 % better than benchmarked approaches.

In a demand response scenario, coordinating multiple data centers with an electricity load aggregator provides opportunities to minimize electricity cost and absorb the volatility in the grid that is caused by renewable generation. To enable optimal coordination, this paper introduces a joint data center and aggregator optimization framework that minimizes the cost of data centers while they participate in demand response programs regulated by a load aggregator. The proposed framework, DCAopt, solves three integrated optimization problems: optimizing the quality-of-service of jobs in each data center, coordinating workload sharing among multiple data centers, and assigning (electricity) prices that incentivize demand response. Instead of relying on simplified relations between a data center’s overall utilization rate and the average job delay, DCAopt applies queueing theory and job scheduling simulation techniques to model data centers with heterogeneous workloads, where different workload properties can be measured using data from actual servers. DCAopt solves the aforementioned joint optimization problems via gradient descent. Through evaluation using fine-grained simulations, we demonstrate that our framework finds better solutions to the data-center-aggregator optimization problems. With DCAopt, the energy costs of data centers can be reduced by 5% on average, with a corresponding reduction of a social cost assessed by the aggregator amounting to more than 30% in most cases. In addition, power usage reduction at the data centers is 6% higher compared to data-center-centric power use optimization.

Due to infrastructure developments, wide disturbances have occurred in the power system. There is a need for intelligent monitoring systems across wide power networks for the stability and security of systems. A significant challenge in a comprehensive power monitoring system is identifying the noises in electrical measurements and oscillatory errors. In this research, the disturbances in the power system are monitored using principal component analysis with a Support vector machine and Extreme Learning Machine (ELM) for analyzing the monitored data. In this work, PCA has been used to reduce the curse of dimensionality of the original data. Then, SVM was used to select the relevant and essential features from the disturbance signals. These selected features are fed as input into the Extreme learning machine to classify the power quality events. This machine learning advantage is that it can analyze many wide-area variables in real time and reduce the masking effect of the oscillatory trends and noise on disturbances. Compared to the existing feature selection and classification of PQ disturbance data, the proposed model secured an improved accuracy of 99.16%, and the comparison results prove the model's effectiveness.

Silt is the leading cause of the erosion of the turbine's underwater components during hydropower generation. This erosion subsequently decreases the machine's efficiency. The present study aims to develop statistical correlations for predicting the efficiency of a hydropower plant based on the Kaplan turbine. Historical data from a Kaplan turbine-based hydropower plant was employed to create the model. Curve fitting, multilinear regression (MLR), and artificial neural network (ANN) techniques were used to develop models for predicting the machine's efficiency. The results show that the ANN method is better at predicting the machine's efficiency than the MLR and curve fitting methods. It got an R2-value of 0.99966, a MAPE of 0.0239%, and an RMSPE of 0.1785%. Equipment manufacturers, plant owners, and researchers can use the established correlation to evaluate the machine's condition in real-time. Additionally, it offers utility in formulating effective operations and maintenance (O&M) strategies.