Scale adaptive fitness evaluation‐based particle swarm optimisation for hyperparameter and architecture optimisation in neural networks and deep learning

Research into automatically searching for an optimal neural network(NN)by optimi-sation algorithms is a significant research topic in deep learning and artificial intelligence.However,this is still challenging due to two issues:Both the hyperparameter and ar-chitecture should be optimised and the optimisation process is computationally expen-sive.To tackle these two issues,this paper focusses on solving the hyperparameter and architecture optimization problem for the NN and proposes a novel light‐weight scale‐adaptive fitness evaluation‐based particle swarm optimisation(SAFE‐PSO)approach.Firstly,the SAFE‐PSO algorithm considers the hyperparameters and architectures together in the optimisation problem and therefore can find their optimal combination for the globally best NN.Secondly,the computational cost can be reduced by using multi‐scale accuracy evaluation methods to evaluate candidates.Thirdly,a stagnation‐based switch strategy is proposed to adaptively switch different evaluation methods to better balance the search performance and computational cost.The SAFE‐PSO algorithm is tested on two widely used datasets:The 10‐category(i.e.,CIFAR10)and the 100−cate-gory(i.e.,CIFAR100).The experimental results show that SAFE‐PSO is very effective and efficient,which can not only find a promising NN automatically but also find a better NN than compared algorithms at the same computational cost.

Design of Evolutionary Algorithm Based Energy Efficient Clustering Approach for Vehicular Adhoc Networks

In a vehicular ad hoc network(VANET),a massive quantity of data needs to be transmitted on a large scale in shorter time durations.At the same time,vehicles exhibit high velocity,leading to more vehicle disconnections.Both of these characteristics result in unreliable data communication in VANET.A vehicle clustering algorithm clusters the vehicles in groups employed in VANET to enhance network scalability and connection reliability.Clustering is considered one of the possible solutions for attaining effectual interaction in VANETs.But one such difficulty was reducing the cluster number under increasing transmitting nodes.This article introduces an Evolutionary Hide Objects Game Optimization based Distance Aware Clustering(EHOGO-DAC)Scheme for VANET.The major intention of the EHOGO-DAC technique is to portion the VANET into distinct sets of clusters by grouping vehicles.In addition,the DHOGO-EAC technique is mainly based on the HOGO algorithm,which is stimulated by old games,and the searching agent tries to identify hidden objects in a given space.The DHOGO-EAC technique derives a fitness function for the clustering process,including the total number of clusters and Euclidean distance.The experimental assessment of the DHOGO-EAC technique was carried out under distinct aspects.The comparison outcome stated the enhanced outcomes of the DHOGO-EAC technique compared to recent approaches.

Simultaneous Identification of Thermophysical Properties of Semitransparent Media Using a Hybrid Model Based on Artificial Neural Network and Evolutionary Algorithm

A hybrid identification model based on multilayer artificial neural networks(ANNs) and particle swarm optimization(PSO) algorithm is developed to improve the simultaneous identification efficiency of thermal conductivity and effective absorption coefficient of semitransparent materials.For the direct model,the spherical harmonic method and the finite volume method are used to solve the coupled conduction-radiation heat transfer problem in an absorbing,emitting,and non-scattering 2D axisymmetric gray medium in the background of laser flash method.For the identification part,firstly,the temperature field and the incident radiation field in different positions are chosen as observables.Then,a traditional identification model based on PSO algorithm is established.Finally,multilayer ANNs are built to fit and replace the direct model in the traditional identification model to speed up the identification process.The results show that compared with the traditional identification model,the time cost of the hybrid identification model is reduced by about 1 000 times.Besides,the hybrid identification model remains a high level of accuracy even with measurement errors.

CNC Thermal Compensation Based on Mind Evolutionary Algorithm Optimized BP Neural Network

Thermal deformation error is one of the most important factors affecting the CNCs’ accuracy, so research is conducted on the temperature errors affecting CNCs’ machining accuracy;on the basis of analyzing the unpredictability and pre-maturing of the results of the genetic algorithm, as well as the slow speed of the training speed of the particle algorithm, a kind of Mind Evolutionary Algorithm optimized BP neural network featuring extremely strong global search capacity was proposed;type KVC850MA/2 five-axis CNC of Changzheng Lathe Factory was used as the research subject, and the Mind Evolutionary Algorithm optimized BP neural network algorithm was used for the establishment of the compensation model between temperature changes and the CNCs’ thermal deformation errors, as well as the realization method on hardware. The simulation results indicated that this method featured extremely high practical value.

Evolution Performance of Symbolic Radial Basis Function Neural Network by Using Evolutionary Algorithms

Radial Basis Function Neural Network(RBFNN)ensembles have long suffered from non-efficient training,where incorrect parameter settings can be computationally disastrous.This paper examines different evolutionary algorithms for training the Symbolic Radial Basis Function Neural Network(SRBFNN)through the behavior’s integration of satisfiability programming.Inspired by evolutionary algorithms,which can iteratively find the nearoptimal solution,different Evolutionary Algorithms(EAs)were designed to optimize the producer output weight of the SRBFNN that corresponds to the embedded logic programming 2Satisfiability representation(SRBFNN-2SAT).The SRBFNN’s objective function that corresponds to Satisfiability logic programming can be minimized by different algorithms,including Genetic Algorithm(GA),Evolution Strategy Algorithm(ES),Differential Evolution Algorithm(DE),and Evolutionary Programming Algorithm(EP).Each of these methods is presented in the steps in the flowchart form which can be used for its straightforward implementation in any programming language.With the use of SRBFNN-2SAT,a training method based on these algorithms has been presented,then training has been compared among algorithms,which were applied in Microsoft Visual C++software using multiple metrics of performance,including Mean Absolute Relative Error(MARE),Root Mean Square Error(RMSE),Mean Absolute Percentage Error(MAPE),Mean Bias Error(MBE),Systematic Error(SD),Schwarz Bayesian Criterion(SBC),and Central Process Unit time(CPU time).Based on the results,the EP algorithm achieved a higher training rate and simple structure compared with the rest of the algorithms.It has been confirmed that the EP algorithm is quite effective in training and obtaining the best output weight,accompanied by the slightest iteration error,which minimizes the objective function of SRBFNN-2SAT.

基于改进实数编码遗传算法的神经网络超参数优化

针对神经网络超参数优化效果差、容易陷入次优解和优化效率低的问题,提出一种基于改进实数编码遗传算法(IRCGA)的深度神经网络超参数优化算法——IRCGA-DNN(IRCGA for Deep Neural Network)。首先,采用实数编码方式表示超参数的取值,使超参数的搜索空间更灵活;然后,引入分层比例选择算子增加解集多样性;最后,分别设计了改进的单点交叉和变异算子,以更全面地探索超参数空间,提高优化算法的效率和质量。基于两个仿真数据集,验证IRCGA-DNN的毁伤效果预测性能和收敛效率。实验结果表明,在两个数据集上,与GA-DNN(Genetic Algorithm for Deep Neural Network)相比,所提算法的收敛迭代次数分别减少了8.7%和13.6%,均方误差(MSE)相差不大;与IGA-DNN(Improved GA-DNN)相比,IRCGA-DNN的收敛迭代次数分别减少了22.2%和13.6%。实验结果表明,所提算法收敛速度和预测性能均更优,能有效处理神经网络超参数优化问题。

基于改进鲸鱼算法优化神经网络的GPS高程拟合方法

采取混沌映射和自适应惯性权重结合的策略对标准鲸鱼算法进行改进,从而提高算法的全局寻优能力和收敛速度,并针对BP神经网络的劣势,利用改进鲸鱼算法对BP神经网络进行优化处理。在此基础上建立改进鲸鱼算法优化BP神经网络的GPS高程异常拟合预测模型,并通过两组不同地形特征工程中的GPS数据对模型进行验证。结果表明,利用改进鲸鱼算法优化的BP模型进行GPS高程拟合时可取得更高的精度和稳定性。

一种基于智能算法的GNSS高程拟合方法

广义回归神经网络(GRNN)是一种新型的前馈神经网络模型,具有训练次数少、耗时短、非线性参数的预报能力较强等优点。但GRNN唯一的调节参数SPREAD不能自动获取限制其进一步的应用。针对该缺陷,本文采用果蝇优化算法(FOA)与GRNN相结合构建FOAGRNN模型对GRNN进行优化,自动获取调节参数的值。为了检验FOAGRNN模型的GNSS高程拟合精度,进行了实验分析。实验结果证明了FOAGRNN模型的GNSS高程拟合精度可达6mm。为进一步检验FOAGRNN模型的优越性,采用与平面拟合模型、二次曲面拟合模型进行对比。实验结果表示FOAGRNN模型的拟合精度要优于平面拟合模型和二次曲面拟合模型,证明了FOAGRNN模型在数据样本较少的情况下,其GNSS高程拟合精度仍然可以达到较高精度。

学习路径规划方法

旨在通过对学习路径规划研究现状的分析,为未来学习路径规划领域相关研究的发展提供借鉴。具体而言,首先介绍学习路径规划的定义以及学习路径规划方法中常用参数;然后按照使用算法的类别的不同,对学习路径规划算法进行详细分类,总结各类学习路径规划方法的优缺点;接着,对学习路径规划方法使用的数据集与评估方法进行介绍;最后,总结学习路径规划方法面临的挑战并对其未来发展趋势进行预测。

露天爆破飞石距离智能预测研究

为了在露天爆破中更准确地预测出飞石的抛掷距离,研究引入多科得分的概念,利用多科得分思维进化算法(Multidisciplinary Score Mind Evolutionary Algorithm,MSMEA)对BP神经网络(Back-Propagation Neural Network)进行优化并建立模型来预测飞石距离。通过分析隐含层神经元个数、种群规模、子种群规模、优胜及临时子种群个数建立了64个多科得分思维进化算法优化BP神经网络模型(Back-Propagation Neural Network Optimized by Multidisciplinary Score Mind Evolutionary Algorithm,MSMEA BP),并选取了其中最优的MSMEA BP模型。为了验证预测模型的有效性,分别用MSMEA BP模型、思维进化算法优化BP神经网络模型(Back-Propagation Neural Network Optimized by Mind Evolutionary Algorithm,MEA BP)和BP神经网络模型对10组爆破飞石距离进行预测。结果显示,MSMEA BP模型得到的预测结果与真实值之间的平均相对误差、决定系数、均方根误差、均方根百分比误差分别达到3.67%、0.9808、7.3571、1.33%,依次优于MEA BP模型和BP神经网络模型,表明在相同训练条件下,采用多科得分思维进化算法对BP神经网络模型进行优化,可以克服BP神经网络易陷入局部最优解的问题,进而显著提高模型的预测精度。该方法为预测爆破飞石距离提供了一个新思路。

基于RNN信息累积的动态多目标优化算法

动态多目标优化问题广泛存在于现实生活中,在环境发生变化后,进化算法需具备快速收敛、快速追踪帕累托最优前沿和维持多样性的能力。对于环境变化程度严重且变化频繁的情况,传统的预测方法无法有效获得帕累托最优前沿解。针对该问题,提出了一种基于循环神经网络(Recurrent Neural Networks,RNN)信息累积的动态多目标优化算法(IA-RNN)。首先,提出了一种基于RNN信息累积的非线性预测方法,利用RNN递归进行信息累积,提高了历史信息利用率,增强了预测的能力。其次,设计了一种基于个体的线性预测方法,利用参数矩阵对个体线性变化进行预测。线性预测与RNN非线性预测协同进化,可快速追踪帕累托最优前沿。最后,设计了一种基于最小二乘法的参数修正策略,通过当前环境的近似帕累托最优前沿解反向指导参数修正,减小了误差堆积的影响。将IA-RNN与5种代表性动态多目标优化算法在14个DF基准测试问题上进行比较。实验证明,IA-RNN算法的收敛性和多样性更优。

基于特征辨识和变分自编码器网络的工商业空调负荷辨识

空调负荷功率的准确计算是实现其需求侧管理的关键,为此,提出基于负荷曲线特征辨识和变分自编码器网络的工商业用户空调负荷辨识方法。针对用户的连续日负荷曲线,提出基于局部加权线性拟合和快速动态时间规整的负荷曲线形态相似度度量方法,以实现对负荷曲线形态特征的度量。提出基于点排序的聚类结构辨识算法的日负荷序列特征辨识方法,以实现对负荷曲线的分类。针对同一特征类型下的用户日负荷序列,提出基于变分自编码器网络的空调负荷辨识算法,以实现空调负荷功率的准确计算。以浙江某市的加工制造业和商业写字楼宇用户负荷数据验证本文所提方法的有效性。算例仿真结果表明,所提方法可以在无需电表高频采样数据、无须预先获取用户的用电设备信息和用电行为信息的条件下准确辨识用户空调负荷功率,为量化空调负荷参与需求响应的可调潜力提供了基础。

基于SLM-RBF的配电网分布式光伏集群智能划分策略

分布式电源大规模分散接入给配电网的优化调度带来计算上的维数灾难,需要对分布式电源进行集群以降低调控难度,因此合理的集群划分十分重要.同时,配电网实时量测数据不全造成分布式电源进行实时集群划分难度大、时间效率低,因此提出一种智能局部移动(SLM)算法与径向基神经网络相结合的分布式电源集群智能划分策略.首先,选取有功和无功功率调节范围以及有功和无功功率-电压的灵敏度作为集群划分的指标,构造相似度矩阵并基于SLM形成分布式电源的集群划分方案库.然后,离线建立电压拟合模型,拟合可实时观测节点的功率与电压之间的关系;同时,离线建立电压-划分结果模型,在线通过电压得到实时划分结果,创新性地解决了潮流模型缺失时无法进行集群划分的问题,提高了集群划分的实时性.最后,在MATLAB平台通过仿真计算验证了算法的合理性和优越性.

基于多源异构气象数据融合的线路状态识别算法

针对传统输电线路状态判断方法存在费时费力且准确率偏低的问题,文中在多源异构气象数据融合分析的基础上提出了一种输电线路状态判别算法。该算法由气象数据分析模块和输电线路故障判断模块组成,对于LSTM网络中存在超参数选择困难的问题,利用WOA算法进行自动优化,以实现对气象数据的特征提取。而输电线路故障判断模块使用航拍线路图像作为训练集,通过引入GAN网络完成训练,并采用AE网络进行改进,从而提升原网络的性能。在实验测试中,所提算法在数据处理层面的MAE仅为0.1159,其余多个测试指标也均优于对比算法,同时还可根据训练得到的多源数据特征来对当前线路状态进行准确预测,表明了该算法具有良好的综合性能与工程实践能力。

Forecasting solar power generation using evolutionary mating algorithm-deep neural networks

This paper proposes an integration of recent metaheuristic algorithm namely Evolutionary Mating Algorithm (EMA) in optimizing the weights and biases of deep neural networks (DNN) for forecasting the solar power generation. The study employs a Feed Forward Neural Network (FFNN) to forecast AC power output using real solar power plant measurements spanning a 34-day period, recorded at 15-minute intervals. The intricate nonlinear relationship between solar irradiation, ambient temperature, and module temperature is captured for accurate prediction. Additionally, the paper conducts a comprehensive comparison with established algorithms, including Differential Evolution (DE-DNN), Barnacles Mating Optimizer (BMO-DNN), Particle Swarm Optimization (PSO-DNN), Harmony Search Algorithm (HSA-DNN), DNN with Adaptive Moment Estimation optimizer (ADAM) and Nonlinear AutoRegressive with eXogenous inputs (NARX). The experimental results distinctly highlight the exceptional performance of EMA-DNN by attaining the lowest Root Mean Squared Error (RMSE) during testing. This contribution not only advances solar power forecasting methodologies but also underscores the potential of merging metaheuristic algorithms with contemporary neural networks for improved accuracy and reliability.

基于脉冲耦合神经网络的红外图像分割算法研究

红外图像噪声多、结构复杂,传统的分割方法难以对红外图像实现完整的分割。脉冲耦合神经网络(PCNN)参数众多,而人工调整网络参数耗时耗力。为了避免人工调整网络参数、实现对红外图像较完整的分割,将粒子群优化(PSO)算法和改进的PCNN相结合,提出了一种新的红外图像分割算法。首先,对红外图像进行预处理。其次,为了减少网络参数,对PCNN进行了化简。然后,将预处理图像输入网络进行循环迭代,利用PSO算法计算每次迭代的分割结果的适应函数值,从而确定群体和个体最佳参数。最后,通过种群最佳参数得到分割结果。为了验证算法的分割性能,试验使用了不同的自然风景图像和红外图像。试验结果表明,所提算法可以对红外和自然风景图像实现较完整的分割;无论是分割效果还是定性分析,所提算法都要优于传统算法。

基于似然BP算法的区域供电能效差异变化趋势预测方法

采用目前方法预测区域供电能效差异变化趋势时,存在误差率高、预测值与实际值差距大及预测模型复杂程度高的问题。提出基于似然BP算法的区域供电能效差异变化趋势预测方法。利用RAMP重构具有稀疏噪声的电能数据,并将处理后的无干扰电能原始数据投入到神经网络中,利用似然BP算法对数据样本进行编码、选择、交叉和变异,最终构建预测模型,并运用预测模型完成区域供电能效差异变化趋势的预测。实验结果表明,所提方法误差率低、预测值与实际值差距小、预测模型复杂程度低。

基于灰狼算法优化BP神经网络的GNSS高程拟合

为了提高GNSS高程拟合精度,提出了一种基于灰狼算法优化BP神经网络的GNSS高程拟合方法。采用GWO算法对BP神经网络的初始阈值和权值进行优化,建立了基于GWO-BP神经网络的GNSS高程拟合模型。采用工程实例进行仿真分析,并与其他GNSS拟合方法进行对比,结果表明,所提GNSS高程拟合方法的拟合残差更小,稳定性更好,有利于提高GNSS高程拟合精度,为GNSS高程拟合提供了一种新方法。

基于优化IGNN的飞机爬升段性能燃效估计

针对飞机爬升过程中性能燃效变化受多种不确定性因素影响,呈现出复杂非线性和随机性特征,提出基于优化嵌入型灰色神经(Inlaid Grey Neural Network,IGNN)的爬升段性能燃效估计方法。该方法利用灰色理论弱化原始数据随机性,结合BP神经网络非线性拟合能力强的特点,构建基于IGNN的爬升段性能燃效估计模型;利用思维进化算法(Mind Evolutionary Algorithm,MEA)优化IGNN的初始权值和阈值,解决随机初始化网络权值和阈值对模型精度的不利影响。实验结果表明,该模型估计精度和稳定性更高,可有效准确估计飞机爬升段性能燃效。

基于卷积神经网络的复杂网络鲁棒性优化

近年来,复杂网络的鲁棒性优化问题引起了人们广泛关注.目前,基于单一测度的鲁棒性优化已经取得很大成就,但复杂网络暴露在外会同时受到多种破坏,从而导致网络系统崩溃无法正常运行.因此构建能同时抵抗节点和连边攻击下的网络结构尤为重要,但在优化过程中复杂网络多目标能控性鲁棒性计算非常耗时.为了解决该问题,文中提出了一种基于卷积神经网络(CNN)的多目标复杂网络能控性鲁棒性优化进化算法.该方法在3种合成网络和两种真实网络上的实验结果表明,在优化过程中使用CNN来评估网络的能控性鲁棒性不仅能降低优化时间,也能辅助进化算法找到抗击能力更强的网络结构.