Distributed Stochastic Optimization with Compression for Non-Strongly Convex Objectives

We are investigating the distributed optimization problem,where a network of nodes works together to minimize a global objective that is a finite sum of their stored local functions.Since nodes exchange optimization parameters through the wireless network,large-scale training models can create communication bottlenecks,resulting in slower training times.To address this issue,CHOCO-SGD was proposed,which allows compressing information with arbitrary precision without reducing the convergence rate for strongly convex objective functions.Nevertheless,most convex functions are not strongly convex(such as logistic regression or Lasso),which raises the question of whether this algorithm can be applied to non-strongly convex functions.In this paper,we provide the first theoretical analysis of the convergence rate of CHOCO-SGD on non-strongly convex objectives.We derive a sufficient condition,which limits the fidelity of compression,to guarantee convergence.Moreover,our analysis demonstrates that within the fidelity threshold,this algorithm can significantly reduce transmission burden while maintaining the same convergence rate order as its no-compression equivalent.Numerical experiments further validate the theoretical findings by demonstrating that CHOCO-SGD improves communication efficiency and keeps the same convergence rate order simultaneously.And experiments also show that the algorithm fails to converge with low compression fidelity and in time-varying topologies.Overall,our study offers valuable insights into the potential applicability of CHOCO-SGD for non-strongly convex objectives.Additionally,we provide practical guidelines for researchers seeking to utilize this algorithm in real-world scenarios.

Multi-Objective Optimization with Artificial Neural Network Based Robust Paddy Yield Prediction Model

Agriculture plays a vital role in the food production process that occupies nearly one-third of the total surface of the earth.Rice is propagated from the seeds of paddy and it is a stable food almost used byfifty percent of the total world population.The extensive growth of the human population alarms us to ensure food security and the country should take proper food steps to improve the yield of food grains.This paper concentrates on improving the yield of paddy by predicting the factors that influence the growth of paddy with the help of Evolutionary Computation Techniques.Most of the researchers used to relay on historical records of meteorological parameters to predict the yield of paddy.There is a lack in analyzing the day to day impact of meteorological parameters such as direction of wind,relative humidity,Instant Wind Speed in paddy cultivation.The real time meteorological data collected and analysis the impact of weather parameters from the day of paddy sowing to till the last day of paddy harvesting with regular time series.A Robust Optimized Artificial Neural Network(ROANN)Algorithm with Genetic Algorithm(GA)and Multi Objective Particle Swarm Optimization Algorithm(MOPSO)proposed to predict the factors that to be concentrated by farmers to improve the paddy yield in cultivation.A real time paddy data collected from farmers of Tamilnadu and the meteorological parameters were matched with the cropping pattern of the farmers to construct the database.The input parameters were optimized either by using GA or MOPSO optimization algorithms to reconstruct the database.Reconstructed database optimized by using Artificial Neural Network Back Propagation Algorithm.The reason for improving the growth of paddy was identified using the output of the Neural Network.Performance metrics such as Accuracy,Error Rate etc were used to measure the performance of the proposed algorithm.Comparative analysis made between ANN with GA and ANN with MOPSO to identify the recommendations for improving the paddy yield.

B^(2)C^(3)NetF^(2):Breast cancer classification using an end‐to‐end deep learning feature fusion and satin bowerbird optimization controlled Newton Raphson feature selection

Currently,the improvement in AI is mainly related to deep learning techniques that are employed for the classification,identification,and quantification of patterns in clinical images.The deep learning models show more remarkable performance than the traditional methods for medical image processing tasks,such as skin cancer,colorectal cancer,brain tumour,cardiac disease,Breast cancer(BrC),and a few more.The manual diagnosis of medical issues always requires an expert and is also expensive.Therefore,developing some computer diagnosis techniques based on deep learning is essential.Breast cancer is the most frequently diagnosed cancer in females with a rapidly growing percentage.It is estimated that patients with BrC will rise to 70%in the next 20 years.If diagnosed at a later stage,the survival rate of patients with BrC is shallow.Hence,early detection is essential,increasing the survival rate to 50%.A new framework for BrC classification is presented that utilises deep learning and feature optimization.The significant steps of the presented framework include(i)hybrid contrast enhancement of acquired images,(ii)data augmentation to facilitate better learning of the Convolutional Neural Network(CNN)model,(iii)a pre‐trained ResNet‐101 model is utilised and modified according to selected dataset classes,(iv)deep transfer learning based model training for feature extraction,(v)the fusion of features using the proposed highly corrected function‐controlled canonical correlation analysis approach,and(vi)optimal feature selection using the modified Satin Bowerbird Optimization controlled Newton Raphson algorithm that finally classified using 10 machine learning classifiers.The experiments of the proposed framework have been carried out using the most critical and publicly available dataset,such as CBISDDSM,and obtained the best accuracy of 94.5%along with improved computation time.The comparison depicts that the presented method surpasses the current state‐ofthe‐art approaches.

Hybrid Optimization Based PID Controller Design for Unstable System

PID controllers play an important function in determining tuning para-meters in any process sector to deliver optimal and resilient performance for non-linear,stable and unstable processes.The effectiveness of the presented hybrid metaheuristic algorithms for a class of time-delayed unstable systems is described in this study when applicable to the problems of PID controller and Smith PID controller.The Direct Multi Search(DMS)algorithm is utilised in this research to combine the local search ability of global heuristic algorithms to tune a PID controller for a time-delayed unstable process model.A Metaheuristics Algorithm such as,SA(Simulated Annealing),MBBO(Modified Biogeography Based Opti-mization),BBO(Biogeography Based Optimization),PBIL(Population Based Incremental Learning),ES(Evolution Strategy),StudGA(Stud Genetic Algo-rithms),PSO(Particle Swarm Optimization),StudGA(Stud Genetic Algorithms),ES(Evolution Strategy),PSO(Particle Swarm Optimization)and ACO(Ant Col-ony Optimization)are used to tune the PID controller and Smith predictor design.The effectiveness of the suggested algorithms DMS-SA,DMS-BBO,DMS-MBBO,DMS-PBIL,DMS-StudGA,DMS-ES,DMS-ACO,and DMS-PSO for a class of dead-time structures employing PID controller and Smith predictor design controllers is illustrated using unit step set point response.When compared to other optimizations,the suggested hybrid metaheuristics approach improves the time response analysis when extended to the problem of smith predictor and PID controller designed tuning.

Machine Learning Assisted Design of Natural Rubber Composites with Multi⁃Performance Optimization

Multi⁃performance optimization of tread rubber composites is a key issue of great concern in automotive industry.Traditional experimental design approach via“trial and error”or intuition is ineffective due to mutual inhibition among multiple properties.A“Uniform design⁃Machine learning”strategy for performance prediction and multi⁃performance optimization of tread rubber composites was proposed.The wear resistance,rolling resistance,tensile strength and wet skid resistance were simultaneously optimized.A series of feasible optimization designs were screened via statistical analysis and machine learning analysis,and were experimentally prepared.The verification experiments demonstrate that the optimization design via machine learning analysis meets the optimization requirements of all target performance,especially for Akron abrasion and 60℃tanδ(about 21%and 9%lower than the design targets,respectively)due to the inhibition of mechanical degradation and good dispersion of fillers.

Optimization of CNC Turning Machining Parameters Based on Bp-DWMOPSO Algorithm

Cutting parameters have a significant impact on the machining effect.In order to reduce the machining time and improve the machining quality,this paper proposes an optimization algorithm based on Bp neural networkImproved Multi-Objective Particle Swarm(Bp-DWMOPSO).Firstly,this paper analyzes the existing problems in the traditional multi-objective particle swarm algorithm.Secondly,the Bp neural network model and the dynamic weight multi-objective particle swarm algorithm model are established.Finally,the Bp-DWMOPSO algorithm is designed based on the established models.In order to verify the effectiveness of the algorithm,this paper obtains the required data through equal probability orthogonal experiments on a typical Computer Numerical Control(CNC)turning machining case and uses the Bp-DWMOPSO algorithm for optimization.The experimental results show that the Cutting speed is 69.4 mm/min,the Feed speed is 0.05 mm/r,and the Depth of cut is 0.5 mm.The results show that the Bp-DWMOPSO algorithm can find the cutting parameters with a higher material removal rate and lower spindle load while ensuring the machining quality.This method provides a new idea for the optimization of turning machining parameters.

An Energy-Efficient Multi-swarm Optimization in Wireless Sensor Networks

Wireless Sensor Networks are a group of sensors with inadequate power sources that are installed in a particular region to gather information from the surroundings.Designing energy-efficient data gathering methods in large-scale Wireless Sensor Networks(WSN)is one of the most difficult areas of study.As every sensor node has afinite amount of energy.Battery power is the most significant source in the WSN.Clustering is a well-known technique for enhan-cing the power feature in WSN.In the proposed method multi-Swarm optimiza-tion based on a Genetic Algorithm and Adaptive Hierarchical clustering-based routing protocol are used for enhancing the network’s lifespan and routing opti-mization.By using distributed data transmission modification,an adaptive hier-archical clustering-based routing algorithm for power consumption is presented to ensure continuous coverage of the entire area.To begin,a hierarchical cluster-ing-based routing protocol is presented in terms of balancing node energy con-sumption.The Multi-Swarm optimization(MSO)based Genetic Algorithms are proposed to select an efficient Cluster Head(CH).It also improves the network’s longevity and optimizes the routing.As a result of the study’sfindings,the pro-posed MSO-Genetic Algorithm with Hill climbing(GAHC)is effective,as it increases the number of clusters created,average energy expended,lifespan com-putation reduces average packet loss,and end-to-end delay.

Enhanced Water Quality Control Based on Predictive Optimization for Smart Fish Farming

The requirement for high-quality seafood is a global challenge in today’s world due to climate change and natural resource limitations.Internet of Things(IoT)based Modern fish farming systems can significantly optimize seafood production by minimizing resource utilization and improving healthy fish production.This objective requires intensive monitoring,prediction,and control by optimizing leading factors that impact fish growth,including temperature,the potential of hydrogen(pH),water level,and feeding rate.This paper proposes the IoT based predictive optimization approach for efficient control and energy utilization in smart fish farming.The proposed fish farm control mechanism has a predictive optimization to deal with water quality control and efficient energy consumption problems.Fish farm indoor and outdoor values are applied to predict the water quality parameters,whereas a novel objective function is proposed to achieve an optimal fish growth environment based on predicted parameters.Fuzzy logic control is utilized to calculate control parameters for IoT actuators based on predictive optimal water quality parameters by minimizing energy consumption.To evaluate the efficiency of the proposed system,the overall approach has been deployed to the fish tank as a case study,and a number of experiments have been carried out.The results show that the predictive optimization module allowed the water quality parameters to be maintained at the optimal level with nearly 30%of energy efficiency at the maximum actuator control rate compared with other control levels.

Optimization Model: Resource Distribution for Risk Factors of Type 2 Diabetes Prevention

Type 2 Diabetes, a lifestyle disease, can be prevented/delayed by adopting a healthy lifestyle. Awareness of the same amongst the citizens can be one of the best ways to initiate a decline in the positive census of the disease. We use this paper to illustrate an optimization model where the budget can be distributed based on the census data of the risk factors involved. It uses a non-linear programming model and can easily be modified into a linear one. The alternative options and constraints too, are mentioned in the paper. The results show that the mid-western states need more share of the allocation based on risk factors. The model distributes the percentage of the budget allocated to different states based on a fixed risk factor constraint.

Energy-Efficient Clustering Using Optimization with Locust Game Theory

Wireless sensor networks(WSNs)are made up of several sensors located in a specific area and powered by a finite amount of energy to gather environmental data.WSNs use sensor nodes(SNs)to collect and transmit data.However,the power supplied by the sensor network is restricted.Thus,SNs must store energy as often as to extend the lifespan of the network.In the proposed study,effective clustering and longer network lifetimes are achieved using mul-ti-swarm optimization(MSO)and game theory based on locust search(LS-II).In this research,MSO is used to improve the optimum routing,while the LS-II approach is employed to specify the number of cluster heads(CHs)and select the best ones.After the CHs are identified,the other sensor components are allo-cated to the closest CHs to them.A game theory-based energy-efficient clustering approach is applied to WSNs.Here each SN is considered a player in the game.The SN can implement beneficial methods for itself depending on the length of the idle listening time in the active phase and then determine to choose whether or not to rest.The proposed multi-swarm with energy-efficient game theory on locust search(MSGE-LS)efficiently selects CHs,minimizes energy consumption,and improves the lifetime of networks.The findings of this study indicate that the proposed MSGE-LS is an effective method because its result proves that it increases the number of clusters,average energy consumption,lifespan extension,reduction in average packet loss,and end-to-end delay.

协调公平与效率的“四水四定”研究Ⅰ:方法与模型

针对水资源刚性约束落地缺抓手的问题,按照以水定发展、协调定规模的理念,提出“四水四定”的分析思路和控制指标,厘清各类发展目标与水资源利用的协调关系。以公平为社会目标,以效益为经济目标,提出“尊重现状、保障刚需、效率驱动、区域均衡”的协调优化准则,建立目标函数和约束条件,构建“四水四定”多目标协调模型,以满足保障刚性需求基础上的水量高效流转,确定发展规模和水量分配方案。模型选取人均GDP最大实现效率驱动,使有限水量产生更大效益;选取区域人均GDP差异最小、发展匹配度最高实现发展公平,缩小区域经济差异;通过约束条件反映水资源约束、刚性需求保障、水源用户配置关系、经济社会发展合理范围等要求;设置罚函数协调约束条件不能满足时的目标值调整。通过归一化方法将多目标转化为单目标,基于现状设置初始条件,采用线性规划等方法求解,得出协调产业发展和水量优化分配的整体结果,通过结果评价反馈调整,形成可服务于经济社会发展和水资源管控决策的方案。

过盈配合工况下收缩管吸能特性分析及结构参数优化

轨道车辆收缩管在轴向冲击载荷及过盈配合下的吸能特性是吸能结构研究中的重要课题,其吸能过程依靠收缩管与锥形衬套之间的摩擦做工及收缩管的塑性变形实现,过盈配合通过影响收缩管与衬套间的接触力提高收缩管的能量耗散。为提高收缩管的吸能特性,研究过盈配合对吸能能力的影响,采用试验与有限元模型相结合的方法,基于多目标优化算法获取包括过盈量在内的收缩管最优结构参数。进行空心轴与轴套过盈静压试验,验证接触力理论公式的正确性;利用台车冲击试验研究收缩管碰撞过程中的吸能特性,构建有限元模型进行验证;推导过盈配合条件下收缩管接触表面压力的理论模型;使用全因子和Hammersley设计方法对过盈配合工况下收缩管结构参数进行研究,包括厚度(T)、过盈量(I_(ntf))、锥角长边(α_(x))和收缩比(R_(atio))。基于此,利用移动最小二乘法(MLSM)构建峰值力(PCF)、比吸能(SEA)和平均力(MCF)的近似模型。主效应分析表明,收缩比对PCF、SEA和MCF的影响最为显著。以获取最小PCF、最大SEA以及MCF最接近设计要求650 kN为优化目标,进行全局响应面(GRSM)优化,获得最优结构配置,与初始设计相比,SEA提高28.54%,PCF降低12.37%。代理模型最优解得出的MCF与有限元模型误差为2.01%,证明构建的代理模型具有较高的精度,可用于优化计算。优化后的结构有效提高了收缩管的吸能能力,降低了碰撞的初始峰值力。

Advancing COVID-19 Diagnosis with CNNs: An Empirical Study of Learning Rates and Optimization Strategies

The rapid spread of the novel Coronavirus (COVID-19) has emphasized the necessity for advanced diagnostic tools to enhance the detection and management of the virus. This study investigates the effectiveness of Convolutional Neural Networks (CNNs) in the diagnosis of COVID-19 from chest X-ray and CT images, focusing on the impact of varying learning rates and optimization strategies. Despite the abundance of chest X-ray datasets from various institutions, the lack of a dedicated COVID-19 dataset for computational analysis presents a significant challenge. Our work introduces an empirical analysis across four distinct learning rate policies—Cyclic, Step Based, Time-Based, and Epoch Based—each tested with four different optimizers: Adam, Adagrad, RMSprop, and Stochastic Gradient Descent (SGD). The performance of these configurations was evaluated in terms of training and validation accuracy over 100 epochs. Our results demonstrate significant differences in model performance, with the Cyclic learning rate policy combined with SGD optimizer achieving the highest validation accuracy of 83.33%. This study contributes to the existing body of knowledge by outlining effective CNN configurations for COVID-19 image dataset analysis, offering insights into the optimization of machine learning models for the diagnosis of infectious diseases. Our findings underscore the potential of CNNs in supplementing traditional PCR tests, providing a computational approach to identify patterns in chest X-rays and CT scans indicative of COVID-19, thereby aiding in the swift and accurate diagnosis of the virus.

地铁列车刨削式防爬吸能结构冲击力平稳行为参数优化

为提升地铁列车在碰撞过程的平稳性和耐撞性,提出一种刨削式吸能防爬器。该吸能结构因其抗偏载能力强,结构紧凑等优点被广泛应用于地铁列车的端部吸能结构中。以某型地铁列车的刨削式防爬器为研究对象,使用有限元仿真的方式建立该吸能结构的对称简化模型,并结合台车冲击的试验数据验证有限元模型的准确性;通过数值仿真的方式,将刨刀前角(α)、刨刀后角(β)、刨刀宽度(w)和刨削深度(d)作为输入变量,以总吸能量(E_(A))和吸能阶段的平均刨削力(M_(PF))为耐撞性指标,以样本方差(S^(2))作为考核参照,通过全因子法对输入变量进行参数化分析;在此基础上,利用径向基函数构建代理模型,并采用多目标遗传算法(MOGA)对该吸能结构的主要设计参数进行优化设计,以确定结构参数的最佳的配置方案。研究结果表明:在该型刨削式防爬吸能结构的主要设计尺寸参数中,β对E_(A)、M_(PF)以及S^(2)的影响均不明显;α对MPF、E_(A)呈负相关,同时与S^(2)呈正相关;w和d对E_(A)与M_(PF)呈正相关、与S^(2)呈负相关;通过优化设计得到的吸能结构的最优的参数设计方案中:d为3.94 mm;α为7.78°,w为34 mm;最优设计方案的E_(A)提高了5.91%,M_(PF)提高了5.89%,S^(2)降低了45.4%,耐撞性与平稳性得到了提高。研究结果可为地铁列车刨削式防爬吸能结构的设计提供工程参考。

计及IGBT结温约束的光伏高渗透配电网无功电压优化控制策略

光伏电源参与配电网无功电压调节是提升光伏高渗透配电网运行经济性和可靠性的有效手段,但光伏电源提供无功支撑会使得光伏电源IGBT最大结温升高、结温波动加剧,进而影响光伏电源和配电网的安全稳定运行。为此,该文提出一种计及IGBT结温约束的光伏高渗透配电网无功电压优化控制策略。首先,利用CatBoost算法计算IGBT结温,提高了IGBT结温计算效率,避免了传统结温算法对IGBT热模型参数的依赖;然后,建立考虑IGBT结温约束的有源配电网多目标无功优化模型,利用二分法求解IGBT结温约束下的光伏电源最大输出功率,实现了IGBT结温约束向二阶锥约束的转换;最后,利用IEEE33节点典型配电系统验证了所提策略在光伏高渗透配电网无功电压优化、光伏电源运行可靠性提升方面的有效性,并提出了综合考虑配电网网损、光伏电源可靠性的光伏电源IGBT结温限值整定原则。

Bi-Objective Optimization: A Pareto Method with Analytical Solutions

Multiple objectives to be optimized simultaneously are prevalent in real-life problems. This paper develops a new Pareto Method for bi-objective optimization which yields analytical solutions. The Pareto optimal front is obtained in closed-form, enabling the derivation of various solutions in a convenient and efficient way. The advantage of analytical solution is the possibility of deriving accurate, exact and well-understood solutions, which is especially useful for policy analysis. An extension of the method to include multiple objectives is provided with the objectives being classified into two types. Such an extension expands the applicability of the developed techniques.

多/单目标优化转换下的作战任务重分配

协同作战任务分配技术是近年来军事领域的研究热点之一,以往的研究一般将任务分配划分为预分配和动态分配2个阶段并独立进行研究,但是这种做法忽略了2个阶段之间的内在联系。针对这一问题,以多目标优化下的任务预分配和单目标优化下的任务动态分配为切入点,提出一种多/单目标优化转换思想。在动态任务分配阶段,通过决策者在预分配阶段的选择获取其主观偏好,基于获取的主观偏好将多目标优化转化为单目标优化后,使用合同网协议完成单目标任务重分配。仿真结果证明了所提出的多/单目标优化转换思想的正确性及其在动态任务分配问题中的适用性。

Multi-Objective Cold Chain Path Optimization Based on Customer Satisfaction

To improve customer satisfaction of cold chain logistics of fresh agricultural goods enterprises and reduce the comprehensive distribution cost composed of fixed cost, transportation cost, cargo damage cost, refrigeration cost, and time penalty cost, a multi-objective path optimization model of fresh agricultural products distribution considering client satisfaction is constructed. The model is solved using an enhanced Elitist Non-dominated Sorting Genetic Algorithm (NSGA-II), and differential evolution is incorporated to the evolution operator. The algorithm produced by the revised algorithm produces a better Pareto optimum solution set, efficiently balances the relationship between customer pleasure and cost, and serves as a reference for the long-term growth of organizations. .

层级引导的增强型多目标萤火虫算法

针对多目标萤火虫算法在求解过程中易产生振荡和聚集现象,导致开发能力较弱、求解精度不佳的问题,提出一种层级引导的增强型多目标萤火虫算法(hierarchical guided enhanced multi-objective firefly algorithm,HGEMOFA)。构建层级引导模型,利用非支配排序获得不同层级个体,用优势层个体引导劣势层个体进化,明确引导方向,解决了进化过程中出现的振荡,减少了聚集现象的出现,增强了算法收敛性;引入莱维飞行扰动最优层个体,增强算法的全局搜索能力;每代进化完成后,对当前种群采用变异机制,增强算法的局部开发能力;把变异后的种群和前一代种群合并进行环境选择,筛选出和前一代种群规模相同的子代,避免优势解丢失。实验结果表明:HGEMOFA能有效增强解的收敛性和多样性。

武器装备轻量化结构正向设计方法及应用

为满足武器装备轻量化的迫切需求,提升装备结构优化设计水平,提出了一种代理模型辅助多目标优化的结构正向设计方法。按照从概念设计到参数优化设计的流程,通过优化信息交互和代理模型重构来实现结构多目标优化设计。以轻型车载高炮弹箱架优化设计为例,首先,根据拓扑优化获得的最佳传力路径,建立几何模型和有限元分析模型。其次,通过灵敏度分析缩减设计变量,建立以一阶模态频率和结构强度为约束、最小化质量和最大化刚度为目标的优化模型。然后,使用差分进化算法求解基于Kriging模型的多目标优化问题;在优化过程中,通过比较中间优化结果调整优化模型,并采用基于期望改进的填充样本策略重构代理模型。最后,兼顾结构轻量化和安全,筛选44组优化解中的3组优势方案与初始设计进行比较,结果显示:在满足弹箱架结构性能的前提下,重量分别可减轻31.2%、24.7%和21.8%;所提结构正向设计方法能够明显提升轻量化效果,也可为相关结构优化设计提供参考。