基于鱼群算法优化normalized cut的彩色图像分割方法

为了克服传统的谱聚类算法求解normalized cut彩色图像分割时,分割效果差、算法复杂度高的缺点,提出了一种基于鱼群算法优化normalized cut的彩色图像分割方法。先对图像进行模糊C-均值聚类预处理,然后用鱼群优化算法替代谱聚类算法求解Ncut的最小值,最后通过最优个体鱼得到分割结果。实验表明,该方法耗时少,且分割效果好。

On the optimal harvesting of size-structured population dynamics

This work is concerned with a kind of optimal control problem for a size-structured biological population model.Well-posedness of the state system and an adjoint system are proved by means of Banach＇s fixed point theorem.Existence and uniqueness of optimal control are shown by functional analytical approach.Optimality conditions describing the optimal strategy are established via tangent and normal cones technique.The results are of the first ones for this novel structure.

Optimal INR level in elderly and non-elderly patients with atrial fibrillation receiving warfarin: a report from the COOL-AF nationwide registry in Thailand

Background Asian population are at increased risk of bleeding during the warfarin treatment,so the recommended optimal international normalized ratio(INR)level may be lower in Asians than in Westerners.The aim of this prospective multicenter study was to determine the optimal INR level in Thai patients with non-valvular atrial fibrillation(NVAF).Methods Patients with NVAF who were on warfarin for stroke prevention were recruited from 27 hospitals in the nationwide COOL-AF registry in Thailand.We collected demographic data,medical history,risk factors for stroke and bleeding,concomitant disease,electrocardiogram and laboratory data including INR and antithrombotic medications.Outcome measurements included ischemic stroke/transient ischemic attack(TIA)and major bleeding.Optimal INR level was assessed by the calculation of incidence density for six INR ranges(<1.5,1.5–1.99,2–2.49,2.5–2.99,3–3.49,and≥3.5).Results A total of 2,232 patients were included.The mean age of patients was 68.5±10.6 years.The mean follow-up duration was 25.7±10.6 months.There were 63 ischemic stroke/TIA and 112 major bleeding events.The lowest prevalence of ischemic stroke/TIA and major bleeding events occurred within the INR range of 2.0–2.99 for patients<70 years and 1.5–2.99 for patients≥70 years.Conclusions The INR range associated with the lowest risk of ischemic stroke/TIA and bleeding in the Thai population was 2.0–2.99 for patients<70 years and 1.5–2.99 for patients≥70 years.The rates of major bleeding and ischemic stroke/TIA were both higher than the rates reported in Western population.

An Ensemble of Optimal Deep Learning Features for Brain Tumor Classification

Owing to technological developments,Medical image analysis has received considerable attention in the rapid detection and classification of diseases.The brain is an essential organ in humans.Brain tumors cause loss of memory,vision,and name.In 2020,approximately 18,020 deaths occurred due to brain tumors.These cases can be minimized if a brain tumor is diagnosed at a very early stage.Computer vision researchers have introduced several techniques for brain tumor detection and classification.However,owing to many factors,this is still a challenging task.These challenges relate to the tumor size,the shape of a tumor,location of the tumor,selection of important features,among others.In this study,we proposed a framework for multimodal brain tumor classification using an ensemble of optimal deep learning features.In the proposed framework,initially,a database is normalized in the form of high-grade glioma(HGG)and low-grade glioma(LGG)patients and then two pre-trained deep learning models(ResNet50 and Densenet201)are chosen.The deep learning models were modified and trained using transfer learning.Subsequently,the enhanced ant colony optimization algorithm is proposed for best feature selection from both deep models.The selected features are fused using a serial-based approach and classified using a cubic support vector machine.The experimental process was conducted on the BraTs2019 dataset and achieved accuracies of 87.8%and 84.6%for HGG and LGG,respectively.The comparison is performed using several classification methods,and it shows the significance of our proposed technique.

Reliability-based Robust Optimization Design of Automobile Components with Non-normal Distribution Parameters

In the reliability designing procedure of the vehicle components, when the distribution styles of the random variables are unknown or non-normal distribution, the result evaluated contains great error or even is wrong if the reliability value R is larger than 1 by using the existent method, in which case the formula is necessary to be revised. This is obviously inconvenient for programming. Combining reliability-based optimization theory, robust designing method and reliability based sensitivity analysis, a new method for reliability robust designing is proposed. Therefore the influence level of the designing parameters’ changing to the reliability of vehicle components can be obtained. The reliability sensitivity with respect to design parameters is viewed as a sub-objective function in the multi-objective optimization problem satisfying reliability constraints. Given the first four moments of basic random variables, a fourth-moment technique and the proposed optimization procedure can obtain reliability-based robust design of automobile components with non-normal distribution parameters accurately and quickly. By using the proposed method, the distribution style of the random parameters is relaxed. Therefore it is much closer to the actual reliability problems. The numerical examples indicate the following: (1) The reliability value obtained by the robust method proposed increases (】0.04%) comparing to the value obtained by the ordinary optimization algorithm; (2) The absolute value of reliability-based sensitivity decreases (】0.01%), and the robustness of the products’ quality is improved accordingly. Utilizing the reliability-based optimization and robust design method in the reliability designing procedure reduces the manufacture cost and provides the theoretical basis for the reliability and robust design of the vehicle components.

OPTIMAL COMBINING PREDICTION FOR BLENDING EFFICIENCY

By means of analysing the mechanism of blending materials,a general blending efficiency model was proposed.Applying this general model to an example 9 a suitable formula of blending efficiency which is more accurate than those in papers[2-3]was obtained.Finally,a high-precision optimal combining prediction formula for calculating blending efficiency was proposed.

Optimal IoT Based Improved Deep Learning Model for Medical Image Classification

Recently medical image classification plays a vital role in medical image retrieval and computer-aided diagnosis system.Despite deep learning has proved to be superior to previous approaches that depend on handcrafted features;it remains difficult to implement because of the high intra-class variance and inter-class similarity generated by the wide range of imaging modalities and clinical diseases.The Internet of Things(IoT)in healthcare systems is quickly becoming a viable alternative for delivering high-quality medical treatment in today’s e-healthcare systems.In recent years,the Internet of Things(IoT)has been identified as one of the most interesting research subjects in the field of health care,notably in the field of medical image processing.For medical picture analysis,researchers used a combination of machine and deep learning techniques as well as artificial intelligence.These newly discovered approaches are employed to determine diseases,which may aid medical specialists in disease diagnosis at an earlier stage,giving precise,reliable,efficient,and timely results,and lowering death rates.Based on this insight,a novel optimal IoT-based improved deep learning model named optimization-driven deep belief neural network(ODBNN)is proposed in this article.In context,primarily image quality enhancement procedures like noise removal and contrast normalization are employed.Then the preprocessed image is subjected to feature extraction techniques in which intensity histogram,an average pixel of RGB channels,first-order statistics,Grey Level Co-Occurrence Matrix,Discrete Wavelet Transform,and Local Binary Pattern measures are extracted.After extracting these sets of features,the May Fly optimization technique is adopted to select the most relevant features.The selected features are fed into the proposed classification algorithm in terms of classifying similar input images into similar classes.The proposed model is evaluated in terms of accuracy,precision,recall,and f-measure.The investigation evident the performance of incorporating optimization techniques for medical image classification is better than conventional techniques.

Covid-19 Forecasting with Deep Learning-based Half-binomial Distribution Cat Swarm Optimization

About 170 nations have been affected by the COvid VIrus Disease-19(COVID-19)epidemic.On governing bodies across the globe,a lot of stress is created by COVID-19 as there is a continuous rise in patient count testing positive,and they feel challenging to tackle this situation.Most researchers concentrate on COVID-19 data analysis using the machine learning paradigm in these situations.In the previous works,Long Short-Term Memory(LSTM)was used to predict future COVID-19 cases.According to LSTM network data,the outbreak is expected tofinish by June 2020.However,there is a chance of an over-fitting problem in LSTM and true positive;it may not produce the required results.The COVID-19 dataset has lower accuracy and a higher error rate in the existing system.The proposed method has been introduced to overcome the above-mentioned issues.For COVID-19 prediction,a Linear Decreasing Inertia Weight-based Cat Swarm Optimization with Half Binomial Distribution based Convolutional Neural Network(LDIWCSO-HBDCNN)approach is presented.In this suggested research study,the COVID-19 predicting dataset is employed as an input,and the min-max normalization approach is employed to normalize it.Optimum features are selected using Linear Decreasing Inertia Weight-based Cat Swarm Optimization(LDIWCSO)algorithm,enhancing the accuracy of classification.The Cat Swarm Optimization(CSO)algorithm’s convergence is enhanced using inertia weight in the LDIWCSO algorithm.It is used to select the essential features using the bestfitness function values.For a specified time across India,death and confirmed cases are predicted using the Half Binomial Distribution based Convolutional Neural Network(HBDCNN)technique based on selected features.As demonstrated by empirical observations,the proposed system produces significant performance in terms of f-measure,recall,precision,and accuracy.

含电动汽车的主动配电网多目标分层优化调度

为了协调电动汽车车主和主动配电网2个不同主体之间的利益关系,针对电动汽车接入后主动配电网的优化调度问题,文中提出一种考虑电动汽车充电综合满意度和主动配电网运行效益的多目标分层优化方法。上层模型注重最大化电动汽车车主的充电利益,采用归一化法向约束法求解电动汽车的最优充放电计划,并将其输入下层优化模型。下层模型旨在最大化主动配电网的运行效益,根据电动汽车的充放电计划调整可控分布式电源的输出功率,采用二阶锥松弛转换法和带权极小模理想点法求解该非线性多目标问题。仿真结果表明,所提含电动汽车的主动配电网多目标分层优化方法能够在促使电动汽车充电综合满意度超过0.9的同时,减少有功网损约94.12%、运行成本约30.90%,实现电动汽车车主和主动配电网的双赢。

引入相量算子和流向算子的天鹰优化算法

针对天鹰优化算法搜索效率不足,容易陷入局部最优的缺点,提出多策略改进天鹰优化算法(MIAO).引入广义正态分布优化算法(GNDO),将该算法得出的结果与天鹰优化算法第1阶段得出的结果进行比较,筛选出这2种优化算法下的最优值.该操作扩大了搜索空间,提高了解的质量.引入相量算子,将第2阶段变为自适应的非参数优化,提高算法的高维优化能力.针对天鹰优化算法在迭代后期存在种群多样性降低、局部开发能力不足的问题,在天鹰算法的第3阶段引入流向算子,使信息可以在每个个体间相互传递,提高种群信息的利用率,增强天鹰优化算法的开发性能.通过对16个测试函数寻优对比分析以及Wilcoxon秩和检验可知,MIAO的寻优能力和收敛速度都有较大的提升.为了验证MIAO算法的实用性和可行性,采用所提算法求解减速器设计问题,通过实际工程优化问题的实验对比分析可知,MIAO算法在处理现实优化问题上具有一定的优越性.

新文科背景下高等院校毕业论文存在问题及优化路径探讨——以师范类地方本科院校汉语言文学专业为例

新文科建设理念的提出为汉语言文学专业毕业论文在交叉学科选题、人工智能AI辅助、个性化指导等方面的质量提升提供技术实现路径。提升高校师范类汉语言文学专业毕业论文质量需要建设综合性的制度保障路径,实现学校、二级学院、基层教学组织和指导教师等各个层面协同发力,推动教学大纲、专业课程与毕业论文课程需求的有机衔接。高校教师要发挥指导教师的学术专长,引导学生结合学情、个体特点打造出最有价值的选题。

RCMNAAPE在旋转机械故障诊断中的应用

针对精细复合多尺度排列熵(RCMPE)无法充分提取旋转机械振动信号中的故障信息,从而导致旋转机械故障识别准确率不稳定这一缺陷,提出了一种基于精细复合多尺度归一化幅值感知排列熵(RCMNAAPE)、拉普拉斯分数(LS)和灰狼算法优化支持向量机(GWO-SVM)的旋转机械故障诊断方法。首先,利用幅值感知排列熵替换了RCMPE中的排列熵,提出了RCMNAAPE,并将其用于提取旋转机械振动信号的故障特征生成特征样本;随后,采用了LS从原始的高维故障特征向量中筛选出较少的能够更准确描述故障状态的特征,构造敏感特征样本;最后,将低维的故障特征向量输入由灰狼算法优化的支持向量机中进行了训练和测试,完成了旋转机械样本的故障识别和分类,利用滚动轴承和齿轮箱故障数据集将RCMNAAPE-LS-GWO-SVM与其他故障诊断方法进行了对比分析,并开展了评估。研究结果表明:基于RCMNAAPE-LS-GWO-SVM的故障诊断方法能够有效识别旋转机械的各类故障,其识别准确率高于其他对比的故障诊断方法,其中滚动轴承故障的识别准确率达到99.33%,齿轮箱故障的识别准确率达到98.67%。虽然,该方法的特征提取效率不佳,平均特征提取时间分别为153.02 s和163.98 s,仅优于精细复合多尺度模糊熵(RCMFE),但其综合性能更加优异。

基于NNC法和DMC算法的CCHP型微电网两阶段调度

冷热电联产(combined cooling, heating and power,CCHP)系统与微电网的结合有利于促进消纳可再生能源,为了提升CCHP型微电网的经济性、环保性和稳定性,提出了两阶段优化调度模型。离线优化阶段基于需求侧响应策略,建立了基于归一化法向约束法的多目标规划模型,并用熵权-TOPSIS法筛选最优结果。在线优化阶段建立了基于动态矩阵控制算法的有限时域优化模型,对离线优化结果进行跟踪优化和反馈校正,以降低不确定性因素的影响。最后,设计对比方案进行分析,验证了所提优化模型的有效性。

基于机器学习的变刚度纤维增强复合材料最小化结构柔顺性优化设计

纤维增强复合材料层合板的变刚度优化设计,通过逐点优化纤维铺角的可设计性,从而匹配结构中应力状态的空间变化,更高效地发挥纤维增强复合材料层合板在强度与刚度性能上的方向性,为设计师提供了更广阔的设计空间与设计灵活度.然而,基于梯度类算法的传统复合材料变刚度优化设计,因其设计变量众多,不可避免地在结构分析与灵敏度分析中面临大规模计算的挑战.同时,结构在概念设计阶段存在载荷工况随机性问题,如何在初始概念设计阶段,针对随机载荷工况制定高效的设计方案具有重要工程价值.近年来,随着人工智能与高性能计算的快速发展,基于传统优化获得的数据集构建端到端的机器学习模型,为实现实时的复合材料变刚度优化提供了可能.文章采用反向传播(back propagation,BP)神经网络算法,建立了基于机器学习的纤维增强复合材料变刚度优化设计方法.首先,基于正态分布纤维优化(normal distribution fiber optimization,NDFO)插值格式,构建以最小化结构柔顺度为目标的复合材料变刚度优化设计模型,考虑载荷大小与方向的随机性,获得神经网络模型训练所需的样本集数据.其次,以最小均方误差(means square error,MSE)为目标函数,采用BP神经网络模型对样本数据集进行训练.最后,建立基于皮尔逊相关系数(Pearson correlation coefficient)、均方误差的模型评价体系,对生成的神经网络模型进行评价.数值算例讨论了含圆孔MBB梁与悬臂C型梁变刚度优化设计,详细阐述了基于机器学习的复合材料变刚度优化方法的实施过程,系统地对比了所提出的基于机器学习的复合材料变刚度优化与传统基于NDFO插值格式复合材料变刚度优化设计结果在纤维铺角轨迹和目标函数的差异,验证了本方法的有效性.

含小水电配电网重要负荷评估及其水光储优化配置方法

为保证含小水电配电网重要负荷的电能质量安全与稳定,发挥水光储系统互补作用,提出一种负荷重要程度评估方法和含小水电配电网重要负荷的水光储优化配置方法。首先,从用户角度出发,基于用电效益指标与用电体验指标2个维度建立负荷重要程度评估体系,采用G1熵权法量化用户节点的重要程度,确定重要负荷节点并将其作为光储接入备选节点,降低光储优化配置求解维度;其次,建立以水光储全寿命周期成本最小、电压偏差平方之和最小为目标的优化配置模型,引入规格化平面约束法求解两目标水光储优化配置问题,得到均匀的Pareto最优解集;然后,计算各解集综合偏移程度函数,得到最优折中解,作为优化配置最佳参考方案;最后,以改进的IEEE 33节点配电网为算例,证明所提的优化配置方法可以改善重要负荷的电压偏差,同时提高投资者的收益。

耦合调峰与通航需求的梯级水电站群短期多目标优化调度的MILP方法

电网调峰任务与河道通航需求间的矛盾是水电航运梯级调度时所面临的突出问题,区间回水的顶托作用增大了该问题的复杂性。建立考虑回水影响的梯级水电站群短期多目标优化调度的混合整数线性规划模型,模型以剩余负荷平均距与下游尾水位平均距最小为目标,在将非线性约束通过函数聚合后,利用六面体栅格化技术与第二类特殊有序集约束方法实现该约束的线性化。利用法线边界交叉方法对模型进行求解。算例结果表明,所提方法可以充分计及回水顶托的影响,兼顾调峰与通航需求,高效求解多目标调度问题并获得较理想的结果。

高铁接触网系统可靠性评估与维修计划决策

研究目的:针对高速铁路接触网系统四种维修方式,本文以接触网系统可靠性最高和维修费用最低为目标,建立基于维修方式组合的预防维修系统多目标优化模型,提出一种改进多目标粒子群优化算法对该模型进行求解。首先采用PWLMC混沌映射生成初始化种群,增强种群多样性,然后采取非线性递减的惯性权重调整策略,提高算法寻优精度和收敛速度。在种群进化过程中,对每一代粒子使用归一化越界处理方法,防止算法陷入早熟。研究结论:(1)提出的改进多目标粒子群优化算法与现有算法相比,Pareto最优解前端的分布范围更广且种群多样性及获得的最优解精度均有提高;(2)在维修次数取13的情况下,改进算法获得的维修方案及该方案下各部件的动态可靠性能均达到较理想效果;(3)改进多目标粒子群优化算法可在满足实际安全运行的要求下得到最优决策方案,有效减低接触网系统维修费用的同时提高其可靠性,为高速铁路接触网系统可靠性评估与维修计划决策提供有效方法。

基于绿视率和NDVI的城市街道景观分析与优化研究

街道景观空间对市民健康和城市风貌具有重要影响。既往研究中常以归一化植被指数(NDVI)和绿视率(GVI)来分别代表二维和三维的绿色指标,但对二者的指标相关性研究甚少。采用基于深度学习的图像语义分割方法分析百度街景计算代表性街道的GVI,利用GF-1卫星数据计算NDVI,比较分析城市街道的GVI和NDVI指标特征及相关性。结果表明,1)中山市中心城区各代表街道GVI指标参差不齐,从8.06%到36.00%,其中石岐街道兴中道GVI最高;2)各街道观测点的NDVI均值随着缓冲区尺度的增加也随之呈现出不同变化,NDVI均值具有强烈的尺度敏感性;3)50 m GVI和DNVI均值的皮尔逊相关系数最高,达到0.832。在此基础上分析街道景观存在的不足并给出优化建议,为城市街景评估、空间优化、景观提升提供参考。

基于VPP碳流计算的多目标多时间尺度优化调度

为研究虚拟电厂低碳经济运行策略,以虚拟电厂能流和碳流模型为基础,执行了多目标多时间尺度优化调度。首先,基于碳流走向建立虚拟电厂低碳经济运行的非线性多目标规划模型,并采用归一化法向约束法进行求解。然后,建立了基于动态矩阵控制算法的有限时域优化模型,对日前结果进行跟踪优化和反馈校正,以降低不确定性因素的影响。最后,仿真对比实验验证了所提调度策略能在保证碳流计算精度超过95%的前提下实现虚拟电厂低碳经济运行。

基于BayesianOpt-XGBoost的煤电机组碳排放因子预测

以贝叶斯参数优化的XGBoost算法为基础,基于机组特征和煤炭特性建立BayesianOpt-XGBoost预测模型,其发电、供热碳排放因子预测的相关系数R^(2)分别为0.91和0.87,绝对误差百分比为2.51%和2.91%.进一步,通过特征标准化方法减少对煤炭特性的依赖,模型预测R2分别为0.79和0.77,绝对误差百分比为3.94%和2.75%,精度仍可得到保障.基于该模型分析全国各省区煤电机组碳排放因子并与公布数据进行比较,证明了该模型的有效性.对机组预测结果的分析表明对现存的低容量机组进行改造、对新建造电机组采用大容量高参数可以减少碳排放强度.