Efficient Optimization Research Articles

Modified hybrid fruit fly-based salp swarm optimization strategy for energy efficient optimization in MIMO Wireless Powered Communication Networks

Modified hybrid fruit fly-based salp swarm optimization strategy for energy efficient optimization in MIMO Wireless Powered Communication Networks

YADA - Reference Free Deconvolution of RNA Sequencing Data

Introduction: We present YADA, a cellular content deconvolution algorithm for estimating cell type proportions in heterogeneous cell mixtures based on gene expression data. YADA utilizes curated gene signatures of cell type-specific marker genes, either obtained intrinsically from pure cell type expression matrices or provided by the user. Method: YADA implements an accessible and extensible deconvolution framework uniquely capable of handling marker genes alone as inputs. Adoption barriers are lowered significantly by relying solely on literature-supported cell type-specific signatures rather than full transcriptomic profiles from purified isolates. However, flexible inputs do not necessitate sacrificing rigor - predictions match metrics of current methodologies through an integrated optimization scheme balancing multiple inference algorithms. Efficiency optimizations via compiled runtimes enable rapid execution. Packaging as an importable Python toolkit promotes community enhancement while retaining codebase extensibility. Result: Validation studies demonstrate that YADA matches or exceeds the performance of current deconvolution methods on benchmark datasets. To demonstrate the utility and enable immediate usage, we provide an online Jupyter Notebook implementation coupled with tutorials. Conclusion: YADA provides an accurate, efficient, and extensible Python-based toolkit for cellular deconvolution analysis of heterogeneous gene expression data.

Optimization of desorption parameters using response surface methodology for enhanced recovery of arsenic from spent reclaimable activated carbon: Eco-friendly and sorbent sustainability approach

Desorption and adsorbent regeneration are imperative factors that are required to be taken into account when designing the adsorption system. From the environmental, economic, and practical points of view, regeneration is necessary for evaluating the efficiency and sustainability of synthesized adsorbents. However, no study has investigated the optimization of arsenic species desorption from spent adsorbents and their regeneration ability for reuse as well as safe disposal. This study aims to investigate the desorption ability of arsenic ions adsorbed on hybrid granular activated carbon and the optimization of the independent factors influencing the efficient recovery of arsenic species from the spent activated carbon using central composite design of the response surface methodology. The activated carbon before the sorption process and after the adsorption-desorption of arsenic ions have been characterized using SEM-EDX, FTIR, and TEM. The study found that all the investigated independent desorption variables greatly influence the retrievability of arsenic ions from the spent activated carbon. Using the desirability function for the optimization of the independent factors as a function of desorption efficiency, the optimum experimental conditions were solution pH of 2.00, eluent concentration of 0.10 M, and temperature of 26.63 ℃, which gave maximum arsenic ions recovery efficiency of 91 %. The validation of the quadratic model using laboratory confirmatory experiments gave an optimum arsenic ions desorption efficiency of 97 %. Therefore, the study reveals that the application of the central composite design of the response surface methodology led to the development of an accurate and valid quadratic model, which was utilized in the enhanced optimization of arsenic ions recovery from the spent reclaimable activated carbon. More so, the desorption isotherm and kinetic data of arsenic were well correlated with the Langmuir and the pseudo-second-order models, while the thermodynamics studies indicated that arsenic ions desorption process was feasible, endothermic, and spontaneous.

Design and simulation of a new QCA-based low-power universal gate

Quantum-dot Cellular Automata (QCA) is recognized in electronics for its low power consumption and high-density capabilities, emerging as a potential substitute for CMOS technology. GDI (Gate Diffusion Input) technology is featured as an innovative approach for enhancing power efficiency and spatial optimization in digital circuits. This study introduces an advanced four-input Improved Gate Diffusion Input (IGDI) design specifically for QCA technology as a universal gate. A key feature of the proposed 10-cell block is the absence of cross-wiring, which significantly enhances the circuit’s operational efficiency. Its universal cell nature allows for the carrying out of various logical gates by merely altering input values, without necessitating any structural redesign. The proposed design showcases notable advancements over prior models, including a reduced cell count by 17%, a 29% decrease in total energy usage, and a 44% reduction in average energy loss. This innovative IGDI design efficiently executes 21 combinational and various sequential functions. Simulations in 18 nm technology, accompanied by energy consumption analyses, demonstrate this design’s superior performance compared to existing models in key areas such as multiplexers, comparators, and memory circuits, alongside a significant reduction in cell count.

Enhancing Efficiency and Energy Optimization: Data-Driven Solutions in Process Industrial Manufacturing

This paper reviews the current state of research in data analytics and machine learning techniques, focusing on their applications in process industrial manufacturing, particularly in control and optimization. Key areas for future research include selection and transfer learning for process monitoring, addressing time-varying characteristics, and enhancing data-driven optimal control with domain-specific knowledge. Additionally, the paper explores reinforcement learning techniques and robust optimization, including distributional robust optimization, for high-level decision-making. Emphasizing the importance of historical knowledge of plants and processes, this paper aims to identify knowledge gaps and pave the way for future research in data-driven strategies for process industries, with a particular emphasis on energy efficiency and optimization.

Deep neural network based time–frequency decomposition for structural seismic responses training with synthetic samples

AbstractTime–frequency decomposition is a powerful tool in assessing the dynamic behaviors of structures. Traditional time–frequency decomposition methods struggle with adaptability, and are limited in handling the structural responses with strong nonlinearity and closely spaced modes. In this study, a cutting‐edge approach based on deep neural network (DNN) is proposed to achieve a precise and adaptive time–frequency decomposition of nonlinear structural responses under seismic excitations. Remarkably, despite being trained on synthetic samples, the proposed method demonstrates outstanding performance in decomposing time–frequency components across various seismic response scenarios. Compared to variational mode decomposition (VMD) and synchroextracting transform (SET), the proposed method exhibited superior precision in time–frequency decomposition and excellent efficiency in parameter optimization. Moreover, the applicability of the proposed method to real‐world complex structures has been verified, which also shows promising generalization capabilities. Future research will aim to enhance the network performance by incorporating additional learning samples with diverse nonlinear characteristics.

A Simple and Efficient Convex Optimization Based Bound-Preserving High Order Accurate Limiter for Cahn–Hilliard–Navier–Stokes System

A Simple and Efficient Convex Optimization Based Bound-Preserving High Order Accurate Limiter for Cahn–Hilliard–Navier–Stokes System

Optimization of Operational Parameters Using Artificial Neural Network and Support Vector Machine for Bio-oil Extracted from Rice Husk.

Bio-oil production from rice husk, an abundant agricultural residue, has gained significant attention as a sustainable and renewable energy source. The current research aims to employ artificial neural network (ANN) and support vector machine (SVM) modeling techniques for the optimization of operating parameters for bio-oil extracted from rice husk ash (RHA) through pyrolysis. ANN and SVM methods are employed to model and optimize the operational conditions, including temperature, heating rate, and feedstock particle size, to enhance the yield and quality of bio-oil. Additionally, ANN modeling is utilized to create a predictive model for bio-oil properties, allowing for the efficient optimization of pyrolysis conditions. This research provides valuable insights into the production and properties of bio-oil from RHA. By harnessing the capabilities of ANN and SVM, this research not only aids in understanding the intricate relationships between process variables and bio-oil properties but also provides a means to systematically enhance the production process. The predictive results obtained from the ANN were found to be good when compared with the SVM. Several models with different numbers of neurons have been trained with different transfer functions. R values for the training, validation, and test phases are around 1.0, i.e., 0.9981, 0.9976, and 0.9978, respectively. The overall R-value was 0.9960 for the proposed network. The findings were considered acceptable, as the overall R-value was close to 1.0. The optimized operational parameters contribute to the efficient conversion of RHA into bio-oil, thereby promoting the use of this sustainable resource for renewable energy production. This approach aligns with the growing emphasis on reducing the environmental impact of traditional fossil fuels and advancing the utilization of alternative and environmentally friendly energy sources.

Systematic Review of Computational Fluid Dynamics Modelling and Simulation Techniques Employed in Vertical Axis Hydrokinetic Turbines

At present, Computational Fluid Dynamics (CFD) is being utilized to explore tidal and hydrokinetic turbine systems, advancing comprehension of turbulent flow phenomena and raising the accuracy of performance predictions for these fluid-driven turbines. Consequently, this utilization of CFD contributes to the optimization of efficiency in these devices. Turbines are subject to variations from the best design point due to the influence of fluctuating flow rates, despite the findings of recent research that have identified the ideal design points. In contrast to conventional literature reviews, systematic analysis offers numerous advantages. The methodological strategy applied in this study entailed cross-referencing the findings obtained from Web of Science (WOS) and Scopus databases to establish the comprehensiveness as well as reliability of researcher data. Improving the review process, elevating the prominence of the field of study, and establishing critical priorities to mitigate research bias are all potential strategies for enhancing the quality of these evaluations. This research divides its findings into three core themes: (1) Performance assessment for practical implications, (2) Numerical analysis for theoretical insights, and (3) Design parameter optimization for engineering relevance. These themes collectively provide a well-rounded examination of the research outcomes across practical, theoretical, and engineering dimensions. Finally, the research findings have the potential to act as a significant point of reference for informing the best design considerations pertaining to vertical axis turbines.

Leachate treatment via electrocoagulation–coal‐based powdered activated carbon process: Efficiencies, mechanisms, kinetics, and costs

AbstractThis study aims to improve COD, NH3‐N, and turbidity removal from Bingöl's leachate using a single‐reactor integrated electrocoagulation (EC)–coal‐based powdered activated carbon (CBPAC) process under various experimental conditions. In the EC‐CBPAC process, three stainless‐steel cathodes and three aluminum electrodes were connected to the negative and positive terminals of the power supply, respectively. The initial concentrations in the leachate were 1044 mg O2/L for COD, 204 mg/L for NH3‐N, and 57 NTU (or 71.25‐mg (NH2)2H2SO4/L) for turbidity, respectively. After a 40‐min EC‐CBPAC process, with a CBPAC dosage of 5 g/L and pH of 5 for COD and turbidity, and 9.5 for NH3‐N, the optimum removal efficiencies for COD, NH3‐N, and turbidity were achieved at 92%, 40%, and 91%, respectively. When the EC process was applied without CBPAC under the same experimental conditions, the removal efficiencies of COD, NH3‐N, and turbidity were 87%, 28%, and 54%, respectively. Before and after the EC‐CBPAC process, the Brunauer–Emmett–Teller (BET) surface area, pore volume, and mean pore diameter of the CBPAC were found to be (888 m2/g, 0.498 cm3/g, and 22.28 Å) and (173 m2/g, 0.18 cm3/g, and 42.8 Å), respectively. The optimum pseudo‐first‐order (PFO) rate constants for COD, turbidity, and NH3‐N were determined to be 3.15 × 10−2, 4.77 × 10−2, and 8.8 × 10−3 min−1, respectively. With the current density increasing from 15 to 25 mA/cm2, energy consumption, unit energy consumption, and total cost increased from 68.7 to 122.4 kWh/m3, 6.948 to 15.226 kWh/kg COD, and 0.85 to 1.838 $/kg COD, respectively.Practitioner points EC‐CBPAC process has greater COD, NH3‐N, and turbidity removal efficiency than EC process. COD and turbidity achieved their optimum disposal efficiencies at 92% and 91%, respectively, at pH 5 The most efficient disposal efficiency for NH3‐N was observed to be 40% at pH 9.5. EC‐CBPAC process increased removal efficiencies for COD, NH3‐N, and turbidity by 20%, 19%, and 38%, respectively, compared with EC alone. The turbidity, NH3‐N, and COD disposal fitted PSO model due to high correlation values (R2 0.94–0.99).

Start-to-end modeling and transmission efficiency optimization for a cyclotron-based proton therapy beamline

Utilizing first-order beam dynamics models is adequate for studying the beam properties during the conceptual design of a cyclotron-based proton therapy beamline. After finishing lattice design, particle-matter interaction simulations for passive elements (e.g., degrader, collimators, energy slit) are required. The cascade simulation is used for lattice updates in each iteration, which is complicated. In addition, when the models involve particle tracking and particle-matter interaction, their optimization process is time-consuming. Therefore, this study proposes a start-to-end modeling method using Monte Carlo Beam Delivery Simulation (BDSIM) software that considers more realistic factors, such as particle-matter interaction and the realistic vacuum chamber, to precisely evaluate working parameters, along with an efficient optimization method that utilizes multi-objective Bayesian optimization (MOBO) to improve transmission efficiency. Taking the Huazhong University of Science and Technology proton therapy facility (HUST-PTF) as an example, beam loss along the beamline is located, quantified, and subsequently reduced by tuning the quadrupole strengths based on MOBO. The results show that: (i) By considering the particle-matter interaction and the realistic vacuum chamber, the precision in the prediction of the beam properties is improved; (ii) After optimization, the transmission efficiency of the entire beamline is relatively increased by an average of 6.52 % under different energy settings, especially 11.39 % at 70 MeV.

Enhanced weathering for CO2 removal using carbonate-rich crushed returned concrete; a pilot study from SE Ireland

A 1.21-ha pilot study in SE Ireland investigated crushed returned concrete (CRC), applied at a rate of 7.5 tonnes/ha as a soil amendment for carbon dioxide removal (CDR) by enhanced weathering (EW) over a 10-month period. Most, but not all amended sites showed enhanced concentrations of calcium and bicarbonate in shallow soil waters compared with adjacent controls. Soil pH increased rapidly by 0.3 to >1 pH units (5.75 ± 0.25 (n = 8) in controls to 6.5 ± 0.7 (n = 24) at amended sites), with greatest increases in soils that had lowest initial pH (<6.0). Aside from one site where concentrations continued to increase, calcium (1.76–0.39 mM Ca) and bicarbonate (4.06–0.64 mM) values in soil waters declined gradually over the course of the study, consistent with rapid initial dissolution and progressive exhaustion of the fine-grained fraction, as indicated by a shrinking-core model. Most sites exhibited high bicarbonate to cation ratios (1.2–2.2) consistent with CRC dissolution by soil carbonic acid, although two sites adjacent to the field-boundary river that floods periodically showed evidence for strong acid (nitric) dissolution. Enhanced bicarbonate levels in soil waters at amended sites compared with controls, coupled with a downward-flush hydrological model driven by annual effective rainfall to calculate bicarbonate export and a simple stoichiometric correction for nitric acid weathering, indicate gross CDR rates in the range zero to 0.52 tonnes CO2/ha (69 kg CO2/tonne), limited by bicarbonate export rather than weathering rates. Further optimisation of the CDR efficiency of this material should be possible using strategies to avoid or pre-neutralise non-carbonic soil acids and by taking account of sub-aerial carbonation at the post-crushing, pre-amendment stage that arises when the material is crushed in preparation for EW applications. A simplified life-cycle analysis indicates that a discount of about 11% should be applied to account for CO2 emissions from crushing, transport and spreading operations.

Optimization of energy use efficiency and environmental assessment in soybean and peanut farming using the imperialist competitive algorithm

This study proposes using input-output analysis to improve energy use efficiency and assess environmental indicators in soybean and peanut farming. The research findings show that soybean production consumes 43170.20 MJ ha−1 of total energy and yields 40417.28 MJ ha−1 of output energy, while peanut production requires 28677.36 MJ ha−1 of total energy and produces 28677.36 MJ ha−1 of output energy. The LCA results indicate that on-farm emissions are associated with production inputs for both soybean and peanut systems, with diesel fuel and chemical fertilizers being the main contributors. Soybean farming shows higher levels of diesel fuel pollutants due to reduced usage, while peanuts have a higher prevalence of contaminants associated with diesel fuel. Using ICA to calculate optimized energy consumption results in 36672.77 MJ ha−1 for soybean production and 23485.45 MJ ha−1 for peanut production, leading to energy savings of 6075.33 MJ ha−1 and 5191.91 MJ ha−1, respectively. Additionally, the environmental impact category quantities for ICA analysis of soybean and peanut production reveal the resources per ton for soybean production to be 111.25 USD2013 and for peanut production to be 45.43 USD2013 in terms of energy use. The use of ICA demonstrates potential in maximizing energy efficiency and evaluating environmental indicators in soybean and peanut farming. By optimizing resource allocation and decision-making processes, ICA can help farmers reduce energy consumption and minimize environmental impact, ultimately leading to more sustainable and eco-friendly farming practices that benefit both the environment and the agricultural industry. Further research and implementation of ICA in farming practices have the potential to significantly improve energy efficiency and environmental sustainability in the soybean and peanut farming industry.

Federated Learning via Plurality Vote.

Federated learning allows collaborative clients to solve a machine-learning problem while preserving data privacy. Recent studies have tackled various challenges in federated learning, but the joint optimization of communication overhead, learning reliability, and deployment efficiency is still an open problem. To this end, we propose a new scheme named federated learning via plurality vote (FedVote). In each communication round of FedVote, clients transmit binary or ternary weights to the server with low communication overhead. The model parameters are aggregated via weighted voting to enhance the resilience against Byzantine attacks. When deployed for inference, the model with binary or ternary weights is resource-friendly to edge devices. Our results demonstrate that the proposed method can reduce quantization error and converges faster compared to the methods directly quantizing the model updates.

Liquid–solid two-phase flow and separation behavior in a novel cyclone separator

In response to the prevalent issue of sand presence in liquid, particularly prominent in petroleum engineering, a novel cyclone separator has been meticulously engineered for fine-particle separation. Experiments and numerical simulation methods have provided a profound understanding of the flow-field characteristics and separation efficacy of this device. The internal architecture of the swirling flow inside the separator features a distinctive central vortex core, complemented by a turbulent secondary vortex formation in the lower section of the underflow. As the axial height increases, the secondary vortex gradually dissipates. An analysis of pressure and velocity distribution within the cyclone separator confirms the establishment of a stable cyclone field in the built-in cyclone and a tendency for the flow field within the tank to exhibit uniformity with increasing height. These flow-field characteristics show that the cyclone separator has a good separation effect on fine-rust particle impurities. Furthermore, the separation efficiency of the novel cyclone separator demonstrates a positive correlation with increasing particle size. Of the parameters studied, variation of the inlet velocity is the best method for obtaining optimum separation efficiency for a cyclone desander with a fixed particle size. Specifically, when the inlet velocity reaches 3 m/s, the desander attains an impressive separation efficiency of up to 70%.

Coffee bean graded based on deep net models

Coffee is a widely consumed beverage, and sorting coffee beans is a critical process that ensures high-quality graded coffee products. Coffee beans were graded into nine grades in robusta types. To automate the grading process, a deep learning-based approach was developed using a large dataset of high-resolution images and data augmentation techniques. In contrast to previous studies focusing on robusta type graded into six coffee bean grads, our research extends this framework by employing robusta type into nine grades with an outperformed accuracy. The proposed work uses four deep learning models, namely residual network 34(Resnet34), inception version 3 (Inception v3), efficient network bayesian optimization (EfficientNet-B0), and visual geometry group-16(VGG-16), where trained and evaluated for coffee bean classification into nine grades. The EfficientNet-B0 model exhibited outperformed accuracy, achieving 100% in distinguishing good and bad coffee beans, even in challenging lighting and background conditions.

Efficiency and mass optimization for space nuclear power systems with closed Brayton cycle loops

Space nuclear power systems should be strictly limited in mass and size. Lack of detailed designs, mass estimation in the general design phase usually relies on simple and empirical models that leave too much design margin from the limits. Accurate mass estimation of space nuclear power systems is key to minimize the design margin and maximize the system performance in the general design phase. In this paper, new mass estimation models are proposed for the main components of space nuclear power systems, including nuclear reactor, shield, turbine, compressor, heat exchanger, and thermal radiator. The mass estimation models are more accurate by incorporating principles of nuclear physics, thermal hydraulics, and structural mechanics to the component designs. Combined with thermodynamic models and genetic algorithm, the MEGREZ code is developed to optimize the system performance by setting efficiency and mass as dual objectives. The code verification results indicate that the deviations between and the Jupiter Icy Moons Orbiter project parameters and the code calculated values are only 0.4% for system efficiency and 2.1% for system mass. The optimization for the Jupiter Icy Moons Orbiter project shows that the optimal molecular weight of helium-xenon mixture is the minimum value for single-stage turbomachinery limits. Additionally, the impacts of potential technological degradation, such as increased heat exchanger thermal resistances, higher frictional factors, and decreased turbomachinery efficiency, on system performances are discussed. The results show that turbomachinery efficiency presents the strongest influence on the system-specific mass; a 0.04 reduction in the polytropic efficiency of the compressor or turbine results in more than a 10% increase in system-specific mass.

Multi-model approach for wind resource assessment

This study presents a multi-model approach for wind resource assessment of a wind farm affected by external wakes. The Weather Research and Forecasting model (WRF), a mesoscale model, is employed to simulate external wind farm wakes, while the Farm Optimization and eXtended yield Evaluation Software (FOXES), an engineering model, is used to simulate the wind farm of interest. This hybrid approach addresses the limitation of both models, mainly the lack of layout effects in mesoscale models and the poor representation of cluster wakes in engineering models. A case study, focusing on the Kaskasi wind farm in the Heligoland cluster, shows that the WRF model predicts larger wake losses compared to FOXES, with the multi-model approach yielding intermediate results. Systematic differences are found as a function of wind speed and seasonality, while the models behave differently as a function of turbulence intensity. The external wake effect was clearly identified for one wind direction sector in WRF and the multi-model approach, while FOXES failed to represent this. The proposed methodology does not only enhance classic resource assessment, but also facilitates efficient layout optimization using cluster waked inflow and allows for wind farm control studies, contributing to both planning and operational phases of wind farm management.

Research on low-carbon optimization methods for regional energy interconnection systems considering multi-energy coupling characteristics

Abstract With the increasing application of multi-energy cogeneration, distributed renewable power generation, and multi-energy storage in various multi-energy systems, their efficient collaborative optimization has become the key to further improving emissions. This article utilizes energy hubs to integrate energy equipment with multi-energy coupling and multi-source supply characteristics and constructs a regional interconnected energy system that considers multi-energy coupling characteristics. A multi-objective optimization model for the above-mentioned energy system is established, including economic, low-carbon, and safe aspects such as operating costs, process emissions, and operational safety, and the raccoon optimization algorithm is applied to solve the model. The experimental results show that compared with classical scheduling algorithms, the proposed method can effectively reduce carbon emissions at a certain operating cost.

Enhancing performance in solar air channels: A numerical analysis of turbulent flow and heat transfer with novel shaped baffles

Solar air heaters are emerging as a promising solution for sustainable heating, as they harness abundant solar energy to provide heat for a variety of applications. However, their optimum efficiency remains a major challenge. Deflectors have significant potential to improve the performance of solar air heaters by redirecting airflow and optimizing heat transfer, making them more efficient and more suitable for large-scale adoption. This article examines the performance of a new baffle design aimed at improving heat transfer in the channel by increasing the exchange surface between fluid and solid surface. It consists of a “≠” -shaped baffle mounted on the top and bottom walls of the channel. An in-depth analysis of the spacing between these baffles was carried out to assess their impact on the thermal and aerodynamic properties of the solar air channel. Computational Fluid Dynamics (CFD) was used to perform the calculations, using the Finite Volume Method (FVM) in conjunction with the SIMPLE algorithm. All studies are carried out using the CFD code Fluent. Crucial steps in this research include studying the influence of these new baffles at different Reynolds numbers, ranging from 15000 to 35000, as well as varying the separation distance between the two baffles, chosen periodically (DBaffle/2, 3DBaffle/4, DBaffle, 5DBaffle/4 and 3DBaffle/2). Some of the numerical results were validated with the available experimental data and showed satisfactory agreement. The study identified an optimal case for the new baffles, by reducing the distance between the baffles, an improvement in system performance was observed, by increasing the flow strength and widening the recycling cells, which has an impact on dynamic behavior and heat transfer. The presence of this new baffle design, with minimal separation distance between them, demonstrated a significant thermal improvement, with a thermal improvement factor increasing by 44% compared to the simple channel without baffles, thus highlighting the effectiveness of the new solar air channel with “≠” -shaped baffles.