Search Results (1,265)

Currently, various pruning strategies including different methods and distribution types are commonly used to reduce the number of FLOPs and parameters in deep learning models. However, their impact on actual energy savings remains insufficiently studied, particularly in resource-constrained settings. To address this, we introduce PruneEnergyAnalyzer, an open-source Python tool designed to evaluate the energy efficiency of pruned models. Starting from the unpruned model, the tool calculates the energy savings achieved by pruned versions provided by the user, and generates comparative visualizations based on previously applied pruning hyperparameters such as method, distribution type (PD), compression ratio (CR), and batch size. These visual outputs enable the identification of the most favorable pruning configurations in terms of FLOPs, parameter count, and energy consumption. As a demonstration, we evaluated the tool with 180 models generated from three architectures, five pruning distributions, three pruning methods, and four batch sizes, using another previous library (e.g. FlexiPrune). This experiment revealed the significant impact of the network architecture on Energy Reduction, the non-linearity between FLOPs savings and energy savings, as well as between parameter reduction and energy efficiency. It also showed that the batch size strongly influences the energy consumption of the pruned model. Therefore, this tool can support researchers in making pruning policy decisions that also take into account the energy efficiency of the pruned model. Full article

An efficient integrated energy system (IES) can enhance the potential of building energy conservation and carbon mitigation. However, imbalances between user-side demand and supply side output present formidable challenges to the operational dispatch of building energy systems. To mitigate heat rejection and improve dispatch optimization, an integrated building energy system incorporating waste heat recovery via an absorption heat pump based on the flow temperature model is adopted. A comprehensive analysis was conducted to investigate the correlation among heat pump operational strategies, thermal comfort, and the dynamic thermal storage capacity of piping network systems. The optimization calculations and comparative analyses were conducted across five cases on typical season days via the CPLEX solver with MATLAB R2018a. The simulation results indicate that the operational modes of absorption heat pump reduced the costs by 4.4–8.5%, while the absorption rate of waste heat increased from 37.02% to 51.46%. Additionally, the utilization ratio of battery and thermal storage units decreased by up to 69.82% at most after considering the pipeline thermal inertia and thermal comfort, thus increasing the system’s energy-saving ability and reducing the pressure of energy storage equipment, ultimately increasing the scheduling flexibility of the integrated building energy system. Full article

This article presents a new method of parametric shaping of buildings with buffer spaces characterized by complex forms and effective thermal operation in the moderate climate of the Central Europe Plane. The parameterization of an elaborated thermal qualitative model of buildings with buffer spaces and its configuration based on computer simulations of thermal operation of many discrete models are the specific features of the method. The model uses various original building shapes and a new parametric artificial neural network (a) to automate the calculations and recording of results and (b) to predict a number of new buildings with buffer spaces characterized by effective thermal operation. The configuration of the parametric quantitative model was carried out based on the simulation results of 343 discrete models defined by means of ten independent variables grouping the properties of the building and buffer space related to their forms, materials and air circulation. The analysis performed for the adopted parameter variability ranges indicates a varied impact of these independent variables on the thermal operation of buildings located in a moderate climate. The infiltration and ventilation and physical properties of the windows and walls are the independent variables that most influence the energy savings utilized by the examined buildings with buffer spaces. The optimal values of these variables allow up to 50–60% of the energy supplied by the HVAC system to be saved. The accuracy and universality of the method will continuously be increased in future research by increasing the types and ranges of independent variables. Full article

This study presents a breakthrough in sustainable injection molding by uniquely combining a backpropagation neural network (BPNN) with particle swarm optimization (PSO) to overcome traditional optimization challenges. The BPNN’s exceptional ability to learn complex nonlinear relationships between six key process parameters (including melt temperature and holding pressure) and product quality is amplified by PSO’s intelligent search capability, which efficiently navigates the high-dimensional parameter space. Together, this hybrid approach achieves what neither method could accomplish alone: the BPNN accurately models the intricate process-quality relationships, while PSO rapidly converges on optimal parameter sets that simultaneously meet strict quality targets (66–70 g weight, 3–5 mm thickness) and minimize energy consumption. The significance of this integration is demonstrated through three key outcomes: First, the BPNN-PSO combination reduced optimization time by 40% compared to traditional trial-and-error methods. Second, it achieved remarkable prediction accuracy (RMSE 0.8229 for thickness, 1.5123 for weight) that surpassed standalone BPNN implementations. Third, the method’s efficiency enabled SMEs to achieve CAE-level precision without expensive software, reducing setup costs by approximately 25%. Experimental validation confirmed that the optimized parameters decreased energy use by 28% and material waste by 35% while consistently producing parts within specifications. This research provides manufacturers with a practical, scalable solution that transforms injection molding from an experience-dependent craft to a data-driven science. The BPNN-PSO framework not only delivers superior technical results but does so in a way that is accessible to resource-constrained manufacturers, marking a significant step toward sustainable, intelligent production systems. For SMEs, this framework offers a practical pathway to achieve both economic and environmental sustainability, reducing reliance on resource-intensive CAE tools while cutting production costs by an estimated 22% through waste and energy savings. The study provides a replicable blueprint for implementing data-driven sustainability in injection molding operations without compromising product quality or operational efficiency. Full article

This comprehensive review maps the fast-evolving landscape in which artificial intelligence (AI) and deep-learning (DL) techniques converge with the Internet of Things (IoT) to manage energy, comfort, and sustainability across smart environments. A PRISMA-guided search of four databases retrieved 1358 records; after applying inclusion criteria, 143 peer-reviewed studies published between January 2019 and April 2025 were analyzed. This review shows that AI-driven controllers—especially deep-reinforcement-learning agents—deliver median energy savings of 18–35% for HVAC and other major loads, consistently outperforming rule-based and model-predictive baselines. The evidence further reveals a rapid diversification of methods: graph-neural-network models now capture spatial interdependencies in dense sensor grids, federated-learning pilots address data-privacy constraints, and early integrations of large language models hint at natural-language analytics and control interfaces for heterogeneous IoT devices. Yet large-scale deployment remains hindered by fragmented and proprietary datasets, unresolved privacy and cybersecurity risks associated with continuous IoT telemetry, the growing carbon and compute footprints of ever-larger models, and poor interoperability among legacy equipment and modern edge nodes. The authors of researches therefore converges on several priorities: open, high-fidelity benchmarks that marry multivariate IoT sensor data with standardized metadata and occupant feedback; energy-aware, edge-optimized architectures that lower latency and power draw; privacy-centric learning frameworks that satisfy tightening regulations; hybrid physics-informed and explainable models that shorten commissioning time; and digital-twin platforms enriched by language-model reasoning to translate raw telemetry into actionable insights for facility managers and end users. Addressing these gaps will be pivotal to transforming isolated pilots into ubiquitous, trustworthy, and human-centered IoT ecosystems capable of delivering measurable gains in efficiency, resilience, and occupant wellbeing at scale. Full article

With the increasing integration of renewable energy and smart technologies in residential energy systems, proactive household energy management (HEM) have become critical for reducing costs, enhancing grid stability, and achieving sustainability goals. This study proposes a ultra-short-term forecasting-driven proactive energy consumption optimization strategy that integrates advanced forecasting models with multi-objective scheduling algorithms. By leveraging deep learning techniques like Graph Attention Network (GAT) architectures, the system predicts ultra-short-term household load profiles with high accuracy, addressing the volatility of residential energy use. Then, based on the predicted data, a comprehensive consideration of electricity costs, user comfort, carbon emission pricing, and grid load balance indicators is undertaken. This study proposes an enhanced mixed-integer optimization algorithm to collaboratively optimize multiple objective functions, thereby refining appliance scheduling, energy storage utilization, and grid interaction. Case studies demonstrate that integrating photovoltaic (PV) power generation forecasting and load forecasting models into a home energy management system, and adjusting the original power usage schedule based on predicted PV output and water heater demand, can effectively reduce electricity costs and carbon emissions without compromising user engagement in optimization. This approach helps promote energy-saving and low-carbon electricity consumption habits among users. Full article

In recent years, carbon dioxide (CO₂) emissions have increased at the fastest rates ever recorded. This is a trend that contradicts global efforts to stabilise greenhouse gas (GHG) concentrations and prevent long-term climate change. Over 90% of global transport relies on oil-based fuels. The continued use of diesel and petrol raises concerns related to oil costs, supply security, GHG emissions, and the release of air pollutants and volatile organic compounds. This study explored electric vehicle (EV) charging networks by assessing environmental impacts through GHG and petroleum savings, developing dynamic pricing strategies, and forecasting infrastructure needs. A substantial dataset of over 259,000 EV charging records from Palo Alto, California, was statistically analysed. Machine learning models were applied to generate insights that support sustainable and economically viable electric transport planning for policymakers, urban planners, and other stakeholders. Findings indicate that GHG and gasoline savings are directly proportional to energy consumed, with conversion rates of 0.42 kg CO₂ and 0.125 gallons per kilowatt-hour (kWh), respectively. Additionally, dynamic pricing strategies such as a 20% discount on underutilised days and a 15% surcharge during peak hours are proposed to optimise charging behaviour and improve station efficiency. Full article

This paper proposes a speed-adaptive autonomous driving path-tracking framework based on the soft actor–critic (SAC) and pure pursuit (PP) methods, named the SACPP controller. The framework first analyzes the obstacles around the vehicle and plans an obstacle-free reference path with the minimum curvature using the hybrid A* algorithm. Next, based on the generated reference path, the current state of the vehicle, and the vehicle motor energy efficiency diagram, the optimal speed is calculated in real time, and the vehicle dynamics preview point at the future moment—specifically, the look-ahead distance—is predicted. This process relies on the learning of the SAC network structure. Finally, PP is used to generate the front wheel angle control value by combining the current speed and the predicted preview point. In the second layer, we carefully designed the evaluation function in the tracking process based on the uncertainties and performance requirements that may occur during vehicle driving. This design ensures that the autonomous vehicle can not only quickly and accurately track the path, but also effectively avoid surrounding obstacles, while keeping the motor running in the high-efficiency range, thereby reducing energy loss. In addition, since the entire framework uses a lightweight network structure and a geometry-based method to generate the front wheel angle, the computational load is significantly reduced, and computing resources are saved. The actual running results on the i7 CPU show that the control cycle of the control framework exceeds 100 Hz. Full article

Numerous protocols have emerged to address the energy depletion problem in Wireless Sensor Networks (WSNs). Among these protocols, the Dynamic Multi-Hop Routing (DMR) protocol adopts a dynamic technique for routing data across the network. The use of the DMR protocol has shown promising results in reducing energy consumption, prolonging the network lifetime, and increasing throughput. To improve the performance of WSNs, this paper proposes the Enhanced Dynamic Multi-Hop Routing (EDMR) protocol as a modification of the DMR protocol. The EDMR protocol introduces an effective sleeping mechanism that selectively deactivates clusters that do not generate significantly updated data for a specific duration. This mechanism reduces redundant transmissions, thereby saving energy and prolonging the network lifetime. The EDMR protocol incorporates static and dynamic approaches to support two major categories of applications: monitoring and event-driven applications. The proposed protocol is evaluated against the DMR protocol, the Enhanced Dynamic Multi-Hop Technique (EMDHT-LEACH) protocol, and the Low-Energy Adaptive Clustering Hierarchy (LEACH) protocol. The simulation results demonstrate that the EDMR protocol mitigates energy depletion, extends the network lifetime, increases stability, and improves network throughput toward the Base Station (BS), while reducing packet redundancy compared with the other protocols. Full article

Ammonia, as a toxic and corrosive gas, is widely present in industrial emissions, agricultural activities, and disease biomarkers. Detecting ammonia is of vital importance to environmental safety and human health. Sensors based on MXene have become an effective means for detecting ammonia gas due to their unique hierarchical structure, adjustable surface chemical properties, and excellent electrical conductivity. This study reviews the latest progress in the use of MXene and its composites for the low-temperature detection of ammonia gas. The strategies for designing MXene composites, including heterojunction engineering, surface functionalization, and active sites, are introduced, and their roles in improving sensing performance are clarified. These methods have significantly improved the ability to detect ammonia, offering high selectivity, rapid responses, and ultra-low detection limits within the low-temperature range. Successful applications in fields such as industrial safety, food quality monitoring, medical diagnosis, and agricultural management have demonstrated the multi-functionality of this technology in complex scenarios. The challenges related to the material’s oxidation resistance, humidity interference, and cross-sensitivity are also discussed. This study aims to briefly describe the reasonable design based on MXene sensors, aiming to achieve real-time and energy-saving environmental and health monitoring networks in the future. Full article

The Industrial Internet of Things (IIoT) offers transformative potential for supply chain management by enabling automation, real-time monitoring, and predictive analytics. However, fragmented standardization, interoperability challenges, and cybersecurity risks hinder its sustainable adoption. This study aims to develop and validate an ISO-based framework to optimize IIoT networks for sustainable supply chain operations. A quantitative time-series research design was employed, analyzing 150 observations from 10–15 industrial firms over five years. Analytical methods included ARIMA, structural equation modeling (SEM), and XGBoost for predictive evaluation. The findings indicate a 6.2% increase in system uptime, a 4.7% reduction in operational costs, a 2.8% decrease in lead times, and a 55–60% decline in security incidents following ISO standard implementation. Interoperability improved by 40–50%, and integration cost savings ranged from 35–40%, contributing to a 25% boost in overall operational efficiency. These results underscore the critical role of ISO frameworks such as ISO/IEC 30141 and ISO 50001 in enhancing connectivity, energy efficiency, and network security across IIoT-enabled supply chains. While standardization significantly improves key performance indicators, the persistence of lead time variability suggests the need for additional optimization strategies. This study offers a structured and scalable methodology for ISO-based IIoT integration, delivering both theoretical advancement and practical relevance. By aligning with internationally recognized sustainability standards, it provides policymakers, practitioners, and industry leaders with an evidence-based framework to accelerate digital transformation, enhance operational efficiency, and support resilient, sustainable supply chain development in the context of Industry 4.0. Full article

This study combines machine vision technology and deep learning models to rapidly assess the activity of anaerobic ammonium oxidation (Anammox) granular sludge. As a highly efficient nitrogen removal technology for wastewater treatment, the Anammox process has been widely applied globally due to its energy-saving and environmentally friendly features. However, existing sludge activity monitoring methods are inefficient, costly, and difficult to implement in real-time. In this study, we collected and enhanced 1000 images of Anammox granular sludge, extracted color features, and used machine learning and deep learning training methods such as XGBoost and the ResNet50d neural network to construct multiple models of sludge image color and sludge denitrification efficiency. The experimental results show that the ResNet50d-based neural network model performed the best, with a coefficient of determination (R²) of 0.984 and a mean squared error (MSE) of 523.38, significantly better than traditional machine learning models (with R² up to 0.952). Additionally, the experiment demonstrated that under a nitrogen load of 2.22 kg-N/(m³·d), the specific activity of Anammox granular sludge reached its highest value of 470.1 mg-N/(g-VSS·d), with further increases in nitrogen load inhibiting sludge activity. This research provides an efficient and cost-effective solution for online monitoring of the Anammox process and has the potential to drive the digital transformation of the wastewater treatment industry. Full article

This study presents an optimization approach for sizing photovoltaic (PV) and battery energy storage systems (BESSs) within a DC microgrid, aiming to enhance cost-effectiveness, energy reliability, and environmental sustainability. PV generation is modeled based on environmental parameters such as solar irradiance and ambient temperature, while battery charging and discharging operations are managed according to real-time demand. A simulation framework is developed in MATLAB 2021b to analyze PV output, battery state of charge (SOC), and grid energy exchange. For demand-side management, the Long Short-Term Memory (LSTM) deep learning model is employed to forecast future load profiles using historical consumption data. Moreover, a Multi-Layer Perceptron (MLP) neural network is designed for comparison purposes. The dynamic load prediction, provided by LSTM in particular, improves system responsiveness and efficiency compared to MLP. Simulation results indicate that optimal sizing of PV and storage units significantly reduces energy costs and dependency on the main grid for both forecasting methods; however, the LSTM-based approach consistently achieves higher annual savings, self-sufficiency, and Net Present Value (NPV) than the MLP-based approach. The proposed method supports the design of more resilient and sustainable DC microgrids through data-driven forecasting and system-level optimization, with LSTM-based forecasting offering the greatest benefits. Full article

The present study implements novel innovative geostatistical imaging using precision agriculture (PA) under sugarbeet field conditions. Two driplines layout designs (d.l.d.) and soil water content (SWC)–irrigation treatments (A: d.l.d. = 1.00 m driplines spacing × 0.50 m emitters inline spacing; B: d.l.d. = 1.50 m driplines spacing × 0.50 m emitters inline spacing) were applied, with two subfactors of clay loam and clay soils (laboratory soil analysis) for modeling (evaluation of seven models) TDR multi-sensor network measurements. Different sensor calibration methods [method 1(M1) = according to factory; method 2 (M2) = according to Hook and Livingston] were applied for the geospatial two-dimensional (2D) imaging of accurate GIS maps of rootzone soil moisture profiles, soil apparent dielectric Ka profiles, and granular and hydraulic parameters profiles, in multiple soil layers (0–75 cm depth). The modeling results revealed that the best-fitted geostatistical model for soil apparent dielectric Ka was the Gaussian model, while spherical and exponential models were identified to be the most appropriate for kriging modelling, and spatial and temporal imaging was used for accurate profile SWC ${\theta v}_{TDR}$(m³·m⁻³) M1 and M2 maps using TDR sensors. The resulting PA profile map images depict the spatio-temporal soil water and apparent dielectric Ka variability at very high resolutions on a centimeter scale. The best geostatistical validation measures for the PA profile SWC ${\theta v}_{TDR}$ maps obtained were MPE = −0.00248 (m³·m⁻³), RMSE = 0.0395 (m³·m⁻³), MSPE = −0.0288, RMSSE = 2.5424, ASE = 0.0433, Nash–Sutcliffe model efficiency NSE = 0.6229, and MSDR = 0.9937. Based on the results, we recommend d.l.d. A and sensor calibration method 2 for the geospatial 2D imaging of PA GIS maps because these were found to be more accurate, with the lowest statistical and geostatistical errors, and the best validation measures for accurate profile SWC imaging were obtained for clay loam over clay soils. Visualizing sensors’ soil moisture results via geostatistical maps of rootzone profiles have practical implications that assist farmers and scientists in making informed, better and timely environmental irrigation engineering decisions, to save irrigation water, increase water use efficiency and crop production, optimize energy, reduce crop costs, and manage water resources sustainably. Full article

This paper presents methods for modelling and optimising the industrial limestone grinding process carried out using a real limestone plant. Two key process evaluation indicators were developed: specific electric energy consumption and an extended indicator that also includes gas usage. Using process data collected from the SCADA system and results from industrial factorial experiments, regression artificial neural network models were developed, with controllable process parameters used as inputs. In the next phase, black-box optimisation was performed using Bayesian and genetic algorithms to identify optimal mill operating settings. The results demonstrate significant improvements in energy efficiency, with energy savings reaching up to 48% in selected cases. The proposed methodology can be effectively applied to enhance energy performance in similar industrial grinding processes. Full article