Method Reduces Energy Consumption Research Articles

Sodium carbonate-enhanced calcium carbide slag activation of sewage sludge for the preparation of green cementitious materials

The utilization of calcium carbide residue (CCR) to activate solid wastes can effectively promote sustainable development. This paper explores adding sodium carbonate (Na2CO3) to CCR activators to improve the early reactivity of alkali-activated materials (AAMs) made from sewage sludge ash (SSA). The research focuses on the effects of Na2CO3 on setting, hardening, mechanical strength, and environmental benefits. The findings indicate that Na2CO3 can shorten the paste’ hardening time and significantly enhance the early mechanical properties. The introduction of Na2CO3 accelerates the consumption of CCR during the early phases and facilitates the formation of C-(N)-A-S-H gels. Meanwhile, it can significantly reduce the porosity and refine the pore structure. When the dosage reaches 8%, the total intruded pore volume of the sample decreases to 0.15 mL/g. This activation method reduces energy consumption, CO2 emissions, and costs by 69%, 50%, and over 90%, respectively, making it an efficient, eco-friendly approach to AAM production.

Research on low-power driving fatigue monitoring method based on spiking neural network.

Fatigue driving is one of the leading causes of traffic accidents, and the rapid and accurate detection of driver fatigue is of paramount importance for enhancing road safety. However, the application of deep learning models in fatigue driving detection has long been constrained by high computational costs and power consumption. To address this issue, this study proposes an approach that combines Self-Organizing Map (SOM) and Spiking Neural Networks (SNN) to develop a low-power model capable of accurately recognizing the driver's mental state. Initially, spatial features are extracted from electroencephalogram (EEG) signals using the SOM network. Subsequently, the extracted weight vectors are encoded and fed into the SNN for fatigue driving classification. The research results demonstrate that the proposed method effectively considers the spatiotemporal characteristics of EEG signals, achieving efficient fatigue detection. Simultaneously, this approach successfully reduces the model's power consumption. When compared to traditional artificial neural networks, our method reduces energy consumption by approximately 12.21-42.59%.

Enhancing Machinability and Sustainability: The Effects of Hybrid MQL+CO2 Cooling on the Drilling of AA7075T6 with TiO2 and C-Reinforced Composites

This study investigates the impact of various auxiliary cooling techniques on machinability, energy consumption, carbon emissions, and economic factors in the drilling process of AA7075T6 aluminium alloy using TiO2 and C-reinforced composites. The study employed various cooling conditions (dry, MQL, CO2, and hybrid MQL+CO2), with different cutting speeds and feed rates, to evaluate their effects on drilling characteristics. The findings indicated that the combined MQL and CO2 cooling notably enhanced the drilling process by reducing cutting forces by 32% and surface roughness by 65% compared to dry cutting. This synergy between lubrication and cooling significantly improves machinability, resulting in higher-quality machining outputs with smoother surfaces and more precise circularity. Energy analysis revealed that the MQL+CO2 method reduces energy consumption to 64% observed under dry conditions, underscoring its efficiency through better heat dissipation and reduced friction. Furthermore, this method demonstrates a significant reduction in carbon emissions, contributing to environmental sustainability. Economically, although initial costs associated with the implementation of cooling systems are higher, they are offset by reduced tool wear and energy costs, making it a viable solution for sustainable manufacturing practices.

An energy-aware cluster-based routing in the Internet of things using particle swarm optimization algorithm and fuzzy clustering

The effectiveness and longevity of IoT infrastructures heavily depend on the limitations posed by communication, multi-hop data transfers, and the inherent difficulties of wireless links. In dealing with these challenges, routing, and data transmission procedures are critical. Among the fundamental concerns are the attainment of energy efficiency and an ideal distribution of loads among sensing devices, given the restricted energy resources at the disposal of IoT devices. To meet these challenges, the present research suggests a novel hybrid energy-aware IoT routing approach that mixes the Particle Swarm Optimization (PSO) algorithm and fuzzy clustering. The approach begins with a fuzzy clustering algorithm to initially group sensor nodes by their geographical location and assign them to clusters determined by a certain probability. The proposed method includes a fitness function considering energy consumption and distance factors. This feature guides the optimization process and aims to balance energy efficiency and data transmission distance. The hierarchical topology uses the advanced PSO algorithm to identify the cluster head nodes. The MATLAB simulator shows that our method outperforms previous approaches. Various metrics have demonstrated significant improvements over DEEC and LEACH. The method reduces energy consumption by 52% and 16%, improves throughput by 112% and 10%, increases packet delivery rates by 83% and 15%, and extends the network lifespan by 48% and 27%, respectively, compared to DEEC and LEACH approaches.

Brain-computer interfaces inspired spiking neural network model for depression stage identification

BackgroundDepression is a global mental disorder, and traditional diagnostic methods mainly rely on scales and subjective evaluations by doctors, which cannot effectively identify symptoms and even carry the risk of misdiagnosis. Brain-Computer Interfaces inspired deep learning-assisted diagnosis based on physiological signals holds promise for improving traditional methods lacking physiological basis and leads next generation neuro-technologies. However, traditional deep learning methods rely on immense computational power and mostly involve end-to-end network learning. These learning methods also lack physiological interpretability, limiting their clinical application in assisted diagnosis. MethodologyA brain-like learning model for diagnosing depression using electroencephalogram (EEG) is proposed. The study collects EEG data using 128-channel electrodes, producing a 128×128 brain adjacency matrix. Given the assumption of undirected connectivity, the upper half of the 128×128 matrix is chosen in order to minimise the input parameter size, producing 8,128-dimensional data. After eliminating 28 components derived from irrelevant or reference electrodes, a 90×90 matrix is produced, which can be used as an input for a single-channel brain-computer interface image. ResultAt the functional level, a spiking neural network is constructed to classify individuals with depression and healthy individuals, achieving an accuracy exceeding 97.5 %. Comparison with existing methodsCompared to deep convolutional methods, the spiking method reduces energy consumption. ConclusionAt the structural level, complex networks are utilized to establish spatial topology of brain connections and analyse their graph features, identifying potential abnormal brain functional connections in individuals with depression.

Model-free adaptive optimal control for fast and safe start-up of pumped storage hydropower units

Pumped storage unit (PSU) is a special kind of hydropower unit, that undertakes the task of power generation, and performs the function of energy storage. Due to its unique advantages, pumped storage power stations have been extensively developed in China in recent years and play an important role that is difficult to be performed by other energy sources. Although many studies have been carried out on the control of PSUs, most of the control algorithms are based on object models, which results in the control performance depending on the model accuracy to a large extent, limiting the practical application of these algorithms. For this reason, a preliminary application of model-free adaptive control in PSUs is investigated in this paper. Firstly, a refined nonlinear model of PSU is established, secondly, the model-free adaptive controller of PSU and its parameter optimization strategy are designed, and finally, verification is carried out by simulation studies for four different start-up conditions, and a comprehensive comparison is made with control strategies such as proportional-integral-derivative (PID), fractional-order PID, and nonlinear generalized predictive control. The results show that the proposed method has significant improvement in both speed and head control performance compared with other methods, with the regulation time being on average at least 5.53 s faster and the overshoot being on average at least 2.85 % lower. It proves the effectiveness and superiority of the method. In addition, the proposed method reduces energy consumption in the regulation process, and the computational cost meets the real-time control requirements, which has a wide range of application prospects.

Text Semantics-Driven Data Classification Storage Optimization

Data classification storage has emerged as an effective strategy, harnessing the diverse performance attributes of storage devices to orchestrate a harmonious equilibrium between energy consumption, cost considerations, and user accessibility. The traditional strategy of solely relying on access frequency for data classification is no longer suitable for today’s complex storage environment. Diverging from conventional methods, we explore from the perspective of text semantics to address this issue and propose an effective data classification storage method using text semantic similarity to extract seasonal features. First, we adopt a dual-layer strategy based on semantic similarity to extract seasonal features. Second, we put forward a cost-effective data classification storage framework based on text seasonal features. We compare our work with the data classification approach AS-H, which runs at full high performance. In addition, we also compare it with K-ear, which adopts K-means as the classification algorithm. The experimental results show that compared with AS-H and K-ear, our method reduces energy consumption by 9.51–13.35% and operating costs by 13.20–22.17%.

Eco-efficient removal of polymer back sheet fraction and material separation from solar cell waste

This study explores an innovative solution for solar cell waste recycling, addressing current methods' limitations, which often neglect the polymer fraction, and leading to environmental risks. Instead of solely using the traditional thermal treatment, which can release toxic gases from polymer back sheets, this study proposes an alternative approach. The proposed method includes a selective mechanical peeling process supplemented by thermal treatment as the first step to separate the multilayer components of silicon-based PV panels. The experiment used a two-step peeling process: initially isolating the back sheet, then removing non-glass layers from the glass. Subsequent thermal treatments were tested to disintegrate the remaining polymers and separate materials optimally. The proposed method's effectiveness and environmental implications are compared to traditional methods using Raman spectroscopy, FTIR, XRD, XRF, and gas chromatography. Findings indicate the proposed method reduces energy consumption and pollutant emissions while increasing the purity of recovered materials.

Automated Expert Knowledge-Based Deep Reinforcement Learning Warm Start via Decision Tree for Hybrid Electric Vehicle Energy Management

<div>Deep reinforcement learning has been utilized in different areas with significant progress, such as robotics, games, and autonomous vehicles. However, the optimal result from deep reinforcement learning is based on multiple sufficient training processes, which are time-consuming and hard to be applied in real-time vehicle energy management. This study aims to use expert knowledge to warm start the deep reinforcement learning for the energy management of a hybrid electric vehicle, thus reducing the learning time. In this study, expert domain knowledge is directly encoded to a set of rules, which can be represented by a decision tree. The agent can quickly start learning effective policies after initialization by directly transferring the logical rules from the decision tree into neural network weights and biases. The results show that the expert knowledge-based warm start agent has a higher initial learning reward in the training process than the cold start. With more expert knowledge, the warm start shows improved performance in the initial learning stage compared to the warm start method with less expert knowledge. The results indicate that the proposed warm start method requires 76.7% less time to achieve convergence than the cold start. The proposed warm start method is also compared with the conventional rule-based method and equivalent consumption minimization strategy. The proposed warm start method reduces energy consumption by 8.62% and 3.62% compared with the two baseline methods, respectively. The results of this work can facilitate the expert knowledge-based deep reinforcement learning warm start in hybrid electric vehicle energy management problems.</div>

Biomass waste fern leaves as a material for a sustainable method of activated carbon production for CO2 capture

In this work, we report the use of activated carbon synthesized from a sustainable material - fern leaves - as a sorbent for carbon dioxide capture applications. The resource-friendly technology for activated carbon production was applied and described. The activated carbons were prepared by chemical and physical activation and carbonization at the same time at the temperature range of 500–900 °C. This method reduces energy consumption and resources. KOH and CO2 were used as activating agents. The evaluation of the CO2 adsorption ability of the activated carbon was supported by different methods including: elemental analysis using X-ray fluorescence spectroscopy, ash content, surface area and porosity measurements, Raman spectroscopy, X-ray spectroscopy and scanning electron microscopy. Results indicated that the optimum temperature of the synthesis was 700 °C. The highest achieved adsorption of CO2 was equal to 6.77 mmol/g and 3.58 mmol/g at 0 °C and 25 °C, respectively. The activated carbons synthesized from fern leaves showed high CO2 adsorption and selectivity. Moreover, the abundance and low cost of fern leaves make them very promising carbon sources for CO2 sorbents production.

Dynamic Virtual Machine Allocation in Cloud Computing Using Elephant Herd Optimization Scheme

Cloud computing is a computing technology that is expeditiously evolving. Cloud is a type of distributed computing system that provides a scalable computational resource on demand including storage, processing power and applications as a service via Internet. Cloud computing, with the assistance of virtualization, allows for transparent data and service sharing across cloud users, as well as access to thousands of machines in a single event. Virtual machine (VM) allocation is a difficult job in virtualization that is governed as an important aspect of VM migration. This process is performed to discover the optimum way to place VMs on physical machines (PMs) since it has clear implications for resource usage, energy efficiency, and performance of several applications, among other things. Hence an efficient VM placement problem is required. This paper presents a VM allocation technique based on the elephant herd optimization scheme. The proposed method is evaluated using real-time workload traces and the empirical results show that the proposed method reduces energy consumption, and maximizes resource utilization when compared to the existing methods.

Eco-Driving Cruise Control for 4WIMD-EVs Based on Receding Horizon Reinforcement Learning

Aiming to improve the distance per charge of four in-wheel independent motor-drive electric vehicles in intelligent transportation systems, a hierarchical energy management strategy that weighs their computational efficiency and optimization performance is proposed. According to the information of an intelligent transportation system, a method combining reinforcement learning with receding horizon optimization is proposed at the upper level, which solves the cruising velocity for eco-driving in a long predictive horizon based on the online construction of a velocity planning problem. At the lower level, a multi-objective optimal torque allocation method that considers energy saving and safety is proposed, where an analytical solution based on the state feedback control was obtained with the vehicle following the optimal speed of the upper level and tracking the centerline of the target path. The energy management strategy proposed in this study effectively reduces the complexity of the intelligent energy-saving control system of the vehicle and achieves a fast solution to the whole vehicle energy optimization problem, integrating macro-traffic information while considering both power and safety. Finally, an intelligent, connected hardware-in-the-loop (HIL) simulation platform is built to verify the method formulated in this study. The simulation results demonstrate that the proposed method reduces energy consumption by 12.98% compared with the conventional constant-speed cruising strategy. In addition, the computational time is significantly reduced.

Bio-Inspired Sleep Control for Improving the Energy Efficiency of a MEC System

The energy consumption of a multi-access edge computing (MEC) system must be reduced to save operational costs. Determining a set of active MEC servers (MECSs) that can minimize the energy consumption of the MEC system while satisfying the service delay requirements of the tasks is an NP-complete problem. To solve this problem, we take a bio-inspired approach. We note that the sleep control problem of the MECS differentiates the operational mode among neighboring MECSs. Therefore, by mimicking the cell differentiation process in a biological system, we designed a distributed sleep control method. Each MECS periodically gathers the utilization and delta levels of the neighboring MECSs. Subsequently, by using the gathered information and the Delta–Notch inter-cell signaling model, a MECS autonomously decides whether to sleep. We evaluated the performance of our method through extensive simulations. Compared with a conventional method, the proposed method reduces energy consumption in a MEC system by more than 13% while providing a comparable service delay. In addition, our method reduces the variations in the service delay by more than 35%.

Real-Time Virtual Machine Scheduling in Industry IoT Network: A Reinforcement Learning Method

The widespread adoption of Industrial Internet of Things (IIoT)-based applications has driven the emergence and development of cloud-related computing paradigms with the ability to seamlessly leverage cloud resources. Heterogeneous resources, mobility factors in IoT, and dynamic behavior make it challenging for the corresponding virtual machine (VM) scheduling problem to address the processing effectiveness of application requests in these kinds of cloud environments. Based on reinforcement learning theory, this article proposes an online VM scheduling scheme (OSEC) for joint energy consumption and cost optimization that divides the scheduling process into two parts: VM allocation and VM migration. First, all the VMs and the physical machines (PMs) are regarded as a set of states and actions in the cloud environment, and the Q-learning feedback is used to achieve the iterative computation of Q-values to obtain the optimal parallel allocation sequence for multiple VMs. Then, VMs are migrated among the active PMs according to a grouping policy and the best-fit principle to achieve dynamic consolidation of the resources in the data center. Finally, experimental results show that compared with state-of-the-art algorithms under different conditions, the proposed method reduces energy consumption by approximately 18.25%, VM execution costs by approximately 21.34%, and service level agreement (SLA) violations by approximately 90.51%.

Research on ATO Control Method for Urban Rail Based on Deep Reinforcement Learning

Aiming at the problems of punctuality, parking accuracy and energy saving of urban rail train operation, an intelligent control method for automatic train operation (ATO) based on deep Q network (DQN) is proposed. The train dynamics model is established under the condition of satisfying the safety principle and various constraints of automatic driving of urban rail train. Considering the transformation rules and sequences of working conditions between train stations, the agent in the DQN algorithm is used as the train controller to adjust the train automatic driving strategy in real time according to the train operating state and operating environment, and optimizes the generation of the train automatic driving curve. Taking the Beijing Yizhuang Subway line as an example, the simulation test results show that the DQN urban rail train control method reduces energy consumption by 12.32% compared with the traditional train PID control method, and improves the running punctuality and parking accuracy; at the same time, the DQN train automatically driving control method can adjust the train running state in real time and dynamically, and has good adaptability and robustness to the change of train running environment parameters.

V-Stability Based Control for Energy-Saving Towards Long Range Sailing

Ensuring an adequate supply of energy is a challenge in the development of autonomous sailboats for long-range sailing capability, as the continuous high-frequency execution of control commands leads to high energy consumption. However, as the control of sailboat is much more complex than the other USVs, especially in a complicated and windy environment, simply reducing control frequency will lead to large path tracking error, and even unsuccessful navigation. Based on the V-stability controller of sailboats, this letter proposes an energy-saving (E-saving) control scheme for a sailboat, to enable the energy planning of actuators. While keeping the sailboat traveling within the acceptable range along a desired path, we develop an approach to analyze and reach a satisfying tradeoff between path tracking error and consumed energy. The E-saving method is examined in simulated environments and field experiments using an autonomous sailboat (OceanVoy). The results show that E-saving method reduces energy consumption by approximately 11% compared to that of the previous V-stability controller, indicating that our method can elongate sailing distance.

Deep Learning–Based Monitoring Sustainable Decision Support System for Energy Building to Smart Cities with Remote Sensing Techniques

In modern society, energy conservation is an important consideration for sustainability. The availability of energy-efficient infrastructures and utilities depend on the sustainability of smart cities. The big streaming data generated and collected by smart building devices and systems contain useful information that needs to be used to make timely action and better decisions. The ultimate objective of these procedures is to enhance the city's sustainability and livability. The replacement of decades-old infrastructures, such as underground wiring, steam pipes, transportation tunnels, and high-speed Internet installation, is already a major problem for major urban regions. There are still certain regions in big cities where broadband wireless service is not available. The decision support system is recently acquiring increasing attention in the smart city context. In this article, a deep learning–based sustainable decision support system (DLSDSS) has been proposed for energy building in smart cities. This study proposes the integration of the Internet of Things into smart buildings for energy management, utilizing deep learning methods for sensor information decision making. Building a socially advanced environment aims to enhance city services and urban administration for residents in smart cities using remote sensing techniques. The proposed deep learning methods classify buildings based on energy efficiency. Data gathered from the sensor network to plan smart cities' development include a deep learning algorithm's structural assembly of data. The deep learning algorithm provides decision makers with a model for the big data stream. The numerical results show that the proposed method reduces energy consumption and enhances sensor data accuracy by 97.67% with better decision making in planning smart infrastructures and services. The experimental outcome of the DLSDSS enhances accuracy (97.67%), time complexity (98.7%), data distribution rate (97.1%), energy consumption rate (98.2%), load shedding ratio (95.8%), and energy efficiency (95.4%).

A two-stage stochastic optimization model for integrated tram timetable and speed control with uncertain dwell times

Modern trams usually own passive transit signal priority (TSP) to avoid interruption from traffic signals along the route. The key to TSP depends on the stick to the recommended travel time between intersections strictly. However, the effectiveness of the TSP can be weakened by dwell time fluctuation due to uncertain passenger demand at the stations. This paper proposes a two-stage stochastic optimization model for timetable and tram control to improve the TSP reliability considering uncertain dwell times. The first stage of the model focuses on designing timetable alternatives, and the second stage evaluates the timetables through expected travel time and energy consumption under different dwell time disturbance scenarios. The Brute force algorithm is developed to attain the optimal tram control, while the non-dominated sorting genetic algorithm II (NSGA-II) and GUROBI solver are both adopted to optimize the timetables. A case study of Nanjing Tram Line 1 in China is performed to demonstrate the effectiveness of the proposed approach. The results show that compared to the existing method, the proposed method reduces energy consumption by 16.0% and the number of stops at intersections decreases by 73.7% with the same travel time.

Lighting control of LEDs considering ambient sound variations to meet the requirements of smart cities

This study provides an innovative solution by including sound levels in the energy management of light-emitting diodes (LEDs). At the same time, this method increases the accuracy of individuals' decisions regarding the environmental parameters of light and sound. The introduced system determines the environment's activity level using sound sensors. By applying this strategy, the amount of light and the energy consumption of LED light sources is restricted under conditions where no activity occurs in the environment or the level of the performed activity is limited. The results of system simulation and implementation demonstrate the effectiveness of this technique. According to the findings, this method reduces energy consumption by more than 40% in the stated scenario when environmental activity is reduced. This approach is recommended as an innovative solution for the energy management of LED light sources to satisfy users' requirements in smart cities.

Clustering Method of Mobile Cloud Computing According to Technical Characteristics of Cloudlets

The rapid increase in the number of mobile phones and IoT devices connected to the network reduces the bandwidth of the Internet communication channel, and as a result, delays occur in the delivery of data processed in remote clouds. Edge computing systems (cloudlet, fog computing, etc.) are used to eliminate resource shortages, energy consumption, and communication channel delays in mobile devices. Edge computing systems place processing devices (computers) close to users. Cloudlet-based mobile cloud computing is widely used to reduce delays in communication channels and energy consumption in mobile devices. Selection of the most suitable cloudlet allowing users to run applications fast in cloud is still a considerable problem. This paper proposes a strategy for the selection of high-performance cloudlets providing fast solutions, considering the complexity of application (file type). It offers a method for cloudlet selection out of large number of cloudlets with different technical capabilities providing faster processing of user application. The timing of user applications in cloudlets with different technical capabilities (operating frequency, number of cores, volume of RAM, etc.) also varies. The proposed method provides faster solution for the user application. User applications are grouped by type of application, and a set of cloudlets are clustered by the number of groups. Clustering is performed first by the parameters corresponding to the operating frequency of the cloudlets, then by the number of cores and the volume of RAM. The proposed method reduces energy consumption of mobile devices by providing faster processing of applications. Thus, the proposed strategy provides an energy consumption reduction on mobile devices, faster processing of results and decrease of network delays.