Search

This study explores the efficacy of Long Short-Term Memory (LSTM) networks in predicting recycling rates and enhancing resource allocation in waste management systems. It addresses the limitations of traditional statistical models and machine learning algorithms that struggle with sequential data and temporal dependencies. The methodology comprised collecting extensive datasets from public repositories, configuring the LSTM network architecture, training the model with historical data, and testing various activation functions and hyperparameters. The model’s performance was rigorously compared to traditional models and alternative machine learning algorithms using metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and R-squared (R2). The findings demonstrate that the LSTM model significantly outperformed traditional approaches, achieving an MAE of 3.5%, an RMSE of 2.8%, and an R2 of 0.92. These results underscore the superior capability of LSTM networks to capture complex temporal patterns in recycling data, offering substantial improvements in predictive accuracy and reliability. Consequently, the study highlights the potential of LSTM networks to revolutionize waste management strategies, contributing to more effective and sustainable practices.

This research explores the enhancement of electrochemical performance in magnesium batteries by optimising magnesium alloy anodes, explicitly focusing on Mg-Al and Mg-Ag alloys. The study’s objective was to determine the impact of alloy composition on anode voltage stability and overall battery efficiency, particularly under extended cycling conditions. The research assessed the anodes’ voltage behaviour and internal resistance across magnesium bis(trifluoromethanesulfonyl)imide (Mg(TFSI)2) electrolyte formulations using a systematic setup involving cyclic voltammetry on the anode and electrochemical impedance spectroscopy. The Mg-Al alloy demonstrated superior performance, with minimal voltage drop and lower resistance increase than the Mg-Ag alloy. The results showed that the Mg-Al alloy maintained over 85% energy efficiency after 100 cycles, significantly outperforming the Mg-Ag alloy, which exhibited increased degradation and efficiency reduction to approximately 80%. These findings confirm that incorporating aluminium into magnesium anodes stabilises the anode voltage and enhances the overall battery efficiency by mitigating degradation mechanisms. Consequently, the Mg-Al alloy is identified as an up-and-coming candidate for use in advanced battery technologies, offering energy density and cycle life improvements. This study lays the groundwork for future research to refine magnesium alloy compositions further to boost battery performance.

This research investigates the application of machine learning models to optimise renewable energy systems and contribute to achieving Net Zero emissions targets. The primary objective is to evaluate how machine learning can improve energy forecasting, grid management, and storage optimisation, thereby enhancing the reliability and efficiency of renewable energy sources. The methodology involved the application of various machine learning models, including Long Short-Term Memory (LSTM), Random Forest, Support Vector Machines (SVMs), and ARIMA, to predict energy generation and demand patterns. These models were evaluated using metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). Key findings include a 15% improvement in grid efficiency after optimisation and a 10–20% increase in battery storage efficiency. Random Forest achieved the lowest MAE, reducing prediction error by approximately 8.5%. The study quantified CO2 emission reductions by energy source, with wind power accounting for a 15,000-ton annual reduction, followed by hydropower and solar reducing emissions by 10,000 and 7500 tons, respectively. The research concludes that machine learning can significantly enhance renewable energy system performance, with measurable reductions in errors and emissions. These improvements could help close the “ambition gap” by 20%, supporting global efforts to meet the 1.5 °C Paris Agreement targets.

This research aims to enhance the surface quality, mechanical properties, and biocompatibility of PEEK (polyether–ether–ketone) biomimetic dental implants through laser polishing. The objective is to improve osseointegration and implant durability by reducing surface roughness, increasing hydrophilicity, and enhancing mechanical strength. The methodology involved fabricating PEEK implants via FDM and applying laser polishing. The significant findings showed a 66.7% reduction in surface roughness, Ra reduced from 2.4 µm to 0.8 µm, and a 25.3% improvement in hydrophilicity, water contact angle decreased from 87° to 65°. Mechanical tests revealed a 6.3% increase in tensile strength (96 MPa to 102 MPa) and a 50% improvement in fatigue resistance (100,000 to 150,000 cycles). The strength analysis result showed a 10% increase in stiffness storage modulus from 1400 MPa to 1500 MPa. Error analysis showed a standard deviation of ±3% across all tests. In conclusion, laser polishing significantly improves the surface, mechanical, and biological performance of PEEK implants, making it a promising approach for advancing biomimetic dental implant technology.

This research investigates using Polyether ether ketone (PEEK) in dental prosthetics, focusing on enhancing the mechanical properties, adhesion capabilities, and environmental sustainability through AI-driven data analysis and advanced surface treatments. The objectives include improving PEEK’s adhesion to dental types of cement, assessing its biocompatibility, and evaluating its environmental impact compared to traditional materials. The methodologies employed involve surface treatments such as plasma treatment and chemical etching, mechanical testing under ASTM standards, biocompatibility assessments, and lifecycle analysis. AI models predict and optimize mechanical properties based on extensive data. Significant findings indicate that surface-treated PEEK exhibits superior adhesion properties, maintaining robust mechanical integrity with no cytotoxic effects and supporting its use in direct contact with human tissues. Lifecycle analysis suggests PEEK offers a reduced environmental footprint due to lower energy-intensive production processes and recyclability. AI-driven analysis further enhances the material’s performance prediction and optimization, ensuring better clinical outcomes. The study concludes that with improved surface treatments and AI optimization, PEEK is a promising alternative to conventional dental materials, combining enhanced performance with environmental sustainability, paving the way for broader acceptance in dental applications.

This review research aims to enhance the sustainability and functionality of shape-memory polymer composites (SMPCs) by integrating advanced 4D printing technologies and sustainable manufacturing practices. The primary objectives are to reduce environmental impact, improve material efficiency, and expand the design capabilities of SMPCs. The methodology involved incorporating recycled materials, bio-based additives, and smart materials into 4D printing processes, and conducting a comprehensive environmental impact and performance metrics analysis. Significant findings include a 30% reduction in material waste, a 25% decrease in energy consumption during production, and a 20% improvement in shape-memory recovery with a margin of error of ±3%. Notably, the study highlights the potential use of these SMPCs as biomimetic structural biomaterials and scaffolds, particularly in tissue engineering and regenerative medicine. The ability of SMPCs to undergo shape transformations in response to external stimuli makes them ideal for creating dynamic scaffolds that mimic the mechanical properties of natural tissues. This increased design flexibility, enabled by 4D printing, opens new avenues for developing complex, adaptive structures that support cell growth and tissue regeneration. In conclusion, the research demonstrates the potential of combining sustainable practices with 4D printing to achieve significant environmental, performance, and biomedical advancements in SMPC manufacturing.

This research investigates the environmental sustainability and biomedical applications of shape memory polymers (SMPs), focusing on their integration into 4D printing technologies. The objectives include comparing the carbon footprint, embodied energy, and water consumption of SMPs with traditional materials such as metals and conventional polymers and evaluating their potential in medical implants, drug delivery systems, and tissue engineering. The methodology involves a comprehensive literature review and AI-driven data analysis to provide robust, scalable insights into the environmental and functional performance of SMPs. Thermomechanical modeling, phase transformation kinetics, and heat transfer analyses are employed to understand the behavior of SMPs under various conditions. Significant findings reveal that SMPs exhibit considerably lower environmental impacts than traditional materials, reducing greenhouse gas emissions by approximately 40%, water consumption by 30%, and embodied energy by 25%. These polymers also demonstrate superior functionality and adaptability in biomedical applications due to their ability to change shape in response to external stimuli. The study concludes that SMPs are promising sustainable alternatives for biomedical applications, offering enhanced patient outcomes and reduced environmental footprints. Integrating SMPs into 4D printing technologies is poised to revolutionize healthcare manufacturing processes and product life cycles, promoting sustainable and efficient medical practices.

This research aims to optimize piezoelectric implants for orthopedic applications, enhancing energy harvesting efficiency and mechanical integrity. Our objectives include comparing piezoelectric materials (PZT, PVDF, and BaTiO3) and employing advanced theoretical modeling, finite element analysis (FEA), and validation to identify optimal configurations. Methodologically, this study integrates machine learning and AI-driven techniques to refine design parameters and predict performance outcomes. Significant findings have revealed that PZT demonstrated the highest sensitivity (2 V/mm), achieving a maximum power output of 4.10 Watts, surpassing traditional solutions by over 100%. The optimization process ensured uniform stress distribution, reducing mechanical failure risk, with predictive models showing high accuracy (R-squared value of 97.77%). Error analysis indicated minimal discrepancies, with an average error margin of less than 2%. The conclusions highlight the significant potential of optimized piezoelectric implants in developing durable, efficient, and patient-friendly orthopedic solutions, setting a new standard in intelligent medical device innovation and contributing to enhanced patient care and improved clinical outcomes.