Dynamic Self-Adaptive Double Population Particle Swarm Optimization Algorithm Based on Lorenz Equation

In order to improve some shortcomings of the standard particle swarm optimization algorithm, such as premature convergence and slow local search speed, a double population particle swarm optimization algorithm based on Lorenz equation and dynamic self-adaptive strategy is proposed. Chaotic sequences produced by Lorenz equation are used to tune the acceleration coefficients for the balance between exploration and exploitation, the dynamic self-adaptive inertia weight factor is used to accelerate the converging speed, and the double population purposes to enhance convergence accuracy. The experiment was carried out with four multi-objective test functions compared with two classical multi-objective algorithms, non-dominated sorting genetic algorithm and multi-objective particle swarm optimization algorithm. The results show that the proposed algorithm has excellent performance with faster convergence rate and strong ability to jump out of local optimum, could use to solve many optimization problems.

Variational Data Assimilation Method Using Parallel Dual Populations Particle Swarm Optimization Algorithm

In recent years,numerical weather forecasting has been increasingly emphasized.Variational data assimilation furnishes precise initial values for numerical forecasting models,constituting an inherently nonlinear optimization challenge.The enormity of the dataset under consideration gives rise to substantial computational burdens,complex modeling,and high hardware requirements.This paper employs the Dual-Population Particle Swarm Optimization(DPSO)algorithm in variational data assimilation to enhance assimilation accuracy.By harnessing parallel computing principles,the paper introduces the Parallel Dual-Population Particle Swarm Optimization(PDPSO)Algorithm to reduce the algorithm processing time.Simulations were carried out using partial differential equations,and comparisons in terms of time and accuracy were made against DPSO,the Dynamic Weight Particle Swarm Algorithm(PSOCIWAC),and the TimeVarying Double Compression Factor Particle Swarm Algorithm(PSOTVCF).Experimental results indicate that the proposed PDPSO outperforms PSOCIWAC and PSOTVCF in convergence accuracy and is comparable to DPSO.Regarding processing time,PDPSO is 40%faster than PSOCIWAC and PSOTVCF and 70%faster than DPSO.

Multi-objective particle swarm optimization algorithm using Cauchy mutation and improved crowding distance

Purpose-Multi-objective is a complex problem that appears in real life while these objectives are conflicting.The swarm intelligence algorithm is often used to solve such multi-objective problems.Due to its strong search ability and convergence ability,particle swarm optimization algorithm is proposed,and the multi-objective particle swarm optimization algorithm is used to solve multi-objective optimization problems.However,the particles of particle swarm optimization algorithm are easy to fall into local optimization because of their fast convergence.Uneven distribution and poor diversity are the two key drawbacks of the Pareto front of multi-objective particle swarm optimization algorithm.Therefore,this paper aims to propose an improved multi-objective particle swarm optimization algorithm using adaptive Cauchy mutation and improved crowding distance.Design/methodology/approach-In this paper,the proposed algorithm uses adaptive Cauchy mutation and improved crowding distance to perturb the particles in the population in a dynamic way in order to help the particles trapped in the local optimization jump out of it which improves the convergence performance consequently.Findings-In order to solve the problems of uneven distribution and poor diversity in the Pareto front of multi-objective particle swarm optimization algorithm,this paper uses adaptive Cauchy mutation and improved crowding distance to help the particles trapped in the local optimization jump out of the local optimization.Experimental results show that the proposed algorithm has obvious advantages in convergence performance for nine benchmark functions compared with other multi-objective optimization algorithms.Originality/value-In order to help the particles trapped in the local optimization jump out of the local optimization which improves the convergence performance consequently,this paper proposes an improved multi-objective particle swarm optimization algorithm using adaptive Cauchy mutation and improved crowding distance.

Optimal Operation of Energy Internet Based on User Electricity Anxiety and Chaotic Spatial Variation Particle Swarm Optimization

Ignoring load characteristics and not considering user feeling with regard to the optimal operation of Energy Internet（EI） results in a large error in optimization. Thus, results are not consistent with the actual operating conditions. To solve these problems, this paper proposes an optimization method based on user Electricity Anxiety（EA） and Chaotic Space Variation Particle Swarm Optimization（CSVPSO）. First, the load is divided into critical load, translation load, shiftable load, and temperature load. Then, on the basis of the different load characteristics,the concept of the user EA degree is presented, and the optimization model of the EI is provided. This paper also presents a CSVPSO algorithm to solve the optimization problem because the traditional particle swarm optimization algorithm takes a long time and particles easily fall into the local optimum. In CSVPSO, the particles with lower fitness value are operated by using cross operation, and velocity variation is performed for particles with a speed lower than the setting threshold. The effectiveness of the proposed method is verified by simulation analysis.Simulation results show that the proposed method can be used to optimize the operation of EI on the basis of the full consideration of the load characteristics. Moreover, the optimization algorithm has high accuracy and computational efficiency.

Intelligent Optimization Algorithms to VDA of Models with on/off Parameterizations

Some variational data assimilation （VDA） problems of time- and space-discrete models with on/off parameterizations can be regarded as non-smooth optimization problems. Same as the sub-gradient type method, intelligent optimization algorithms, which are widely used in engineering optimization, can also be adopted in VDA in virtue of their no requirement of cost function＇s gradient （or sub-gradient） and their capability of global convergence. Two typical intelligent optimization algorithms, genetic algorithm （GA） and particle swarm optimization （PSO）, are introduced to VDA of modified Lorenz equations with on-off parameterizations, then two VDA schemes are proposed, that is, GA based VDA （GA-VDA） and PSO based VDA （PSO-VDA）. After revealing the advantage of GA and PSO over conventional adjoint methods in the ability of global searching at the existence of cost function＇s discontinuity induced by on-off switches, sensitivities of GA-VDA and PSO-VDA to population size, observational noise, model error and observational density are detailedly analyzed. It＇s shown that, in the context of modified Lorenz equations, with proper population size, GA-VDA and PSO-VDA can effectively estimate the global optimal solution, while PSO-VDA consumes much less computational time than GA-VDA with the same population size, and requires a much lower population size with nearly the same results, both methods are not very sensitive to observation noise and model error, while PSO-VDA shows a better performance with observational noise than GA-VDA. It is encouraging that both methods are not sensitive to observational density, especially PSO-VDA, using which almost the same perfect assimilation results can be obtained with comparatively sparse observations.

Application of improved PSO to power transmission congestion management optimization model

The parameters of particles were encoded firstly, then the constraint conditions and fitness degree were processed, and the calculation steps of the improved PSO algorithm were presented. Finally, the issues with the adoption of the improved PSO algorithm were solved and the results were analyzed. The results show that it is beneficial to obtaining the optimal solution by increasing the number of particles but that will also increase the operation time. On the aspects of solving continuous differentiable non-linear optimization model with equality and inequality constraints, the optimization result of PSO algorithm is the same as that of the interior point method. Compared with genetic algorithms (GA), PSO algorithm is more effective in the local optimization, and unlike GA, it will not be early maturity. Meanwhile, PSO algorithm is also more effective in the boundary optimization than genetic algorithm.

基于FMCW雷达的人体生命体征信号预测算法

将FMCW雷达检测到的人体生命体征信号,用于预测未来一段时间内人体生命体征信号是否异常,具有明显的应用价值。该方向当前研究主要针对如何进一步降低重构误差、提升生命体征信号的预测精度。为此,本文提出一种自适应变分模态分解-长短期记忆神经网络的生命体征信号预测方法。针对静止状态下的人体,通过雷达采集到的生命体征信号,采用粒子群算法优化变分模态分解VMD的模态分量个数K和惩罚系数α的值,实现自适应选取后用于VMD分解,再将分解后的模态分量进行叠加重构。采用粒子群算法优化长短期记忆网络模型中的网络层数、学习率、正则化系数等3个参数,自适应选取合适的参数组合,将重构后的信号通过优化后的LSTM网络进行预测。实验结果显示本文所提预测方法在10位志愿者的预测结果与原始数据的均方根误差平均值为0.017 188 9,平均绝对误差的平均值为0.007 158,相较于当前其他研究,预测精度上有明显提升。

一种基于PSO-VMD和LSTM的复杂山地风电场观测风速数据质量控制算法

复杂山地风电场普遍存在观测风速数据质量差引起风资源评估误差大、风功率预测精度低的问题。而复杂山地风速呈现较强的间隙性、波动性和非平稳性,导致常规质量控制方法无法有效提高数据质量。针对此,提出一种基于粒子群改进变分模态分解和长短期记忆网络的集成学习算法(PVL),并应用于复杂山地观测风速的质量控制以提高风速数据的质量。以广西某复杂山地风场内5基观测塔2015—2016年逐10 min风速数据为案例进行PVL应用效果检验,并与传统单站及空间回归法、反距离加权法进行对比。应用表明,PVL比传统方法具有更高的寻误率,且在异地形、多风况上具有更强的适应性。

基于双层优化VMD-LSTM的农村超短期电力负荷预测

稳定的供电是农村发展建设的有力保障,而电力负荷水平是建设效果的重要衡量标准,因此建立精确的负荷预测模型可以更准确直观显现电力负荷情况,为供电公司制定决策提供有力支撑。由于LSTM负荷预测模型在数据预测方面存在收敛性差、预测精度不高等问题,为提高模型的预测精度,提出一种基于双层优化VMD-LSTM的超短期电力负荷预测方法。首先提出麻雀算法优化变分模态分解(sparrow variational mode decomposition,SVMD),通过SVMD将原始数据转化为模态分量(intrinsic mode functions,IMF);其次采用改进樽海鞘群算法(association salp swarm algorithm,ASSSA)优化LSTM模型。通过引入4种策略增强标准樽海鞘算法优化能力;最后将各模态分量分别代入到新模型并进行叠加预测。选取辽宁省某市某乡村10kV变压器真实历史负荷数据,以均方根误差(RMSE)、平均绝对误差(MAE)、平均绝对百分比误差(MAPE)、拟合度(R^(2))作为评价指标,并与其他基础预测模型进行对比,结果表明,改进后的算法在计算精度、稳定性方面均优于其他基础预测模型。

PSO寻优VMD的衬板测厚超声回波信号噪声的压制

针对复杂环境下破碎机衬板超声测厚的回波信号噪声难以压制的问题,提出一种基于粒子群算法(Particle Swarm Optimization,PSO)寻优变分模态分解(Variational Mode Decomposition,VMD)的噪声压制方法。首先研究分解尺度[K]和惩罚因子[α]两个初始参数对VMD算法分解效果的影响;然后通过PSO算法对VMD算法中的最佳参数组合进行寻优;再将该组最佳参数组合代入VMD算法中进行信号分解,通过互相关系数法选取本征模态函数(Intrinsic Mode Function,IMF)进行信号重构,得到降噪后的信号。仿真与实测结果表明,PSO寻优VMD的噪声压制方法在提高信噪比、降低均方误差、提高波形相似系数等性能指标方面效果显著,能有效压制衬板超声回波信号中的噪声,提高破碎机衬板测厚超声回波信号的可靠性。

基于多影响因素联合的某抽水蓄能电站主厂房洞室围岩变形预测

为保证地下洞室围岩环境的安全状态,提出了一种以变分模态分解(VMD)方法分解原始数据和粒子群优化算法(PSO)提高预测精度为基础,基于多影响因素联合核极限学习机(KELM)方法的洞室围岩变形预测方法。该方法首先采用VMD方法将监测位移分解为受趋势性因素影响的趋势项位移和受周期性因素影响的周期项位移,去除影响因素的干扰项,其次将演化状态及影响因素作为PSO-KELM的输入数据,预测各影响因素所对应的趋势项或周期项位移,最后叠加两种分项位移,并将多影响因素结合的KELM方法与其他预测方法进行精度比较。对某抽水蓄能电站工程实测数据的验证结果表明,预测结果与原始位移的RRMSE相差仅0.76%,且二者的R为0.986,所提预测方法具有较高的预测精度,可为同类工程的围岩变形预测提供参考。

密度峰值聚类在塔机损伤诊断中的应用研究

建立塔机有限元模型,获取塔机完好状态和各损伤工况的各采集点的动态位移。提出了两种模型建立方法,基于悬臂梁的双输入单输出模型和基于时域数据的动态双输入单输出模型,对基于时域数据的双输入单输出模型首先利用最小二乘法计算参数初值,进一步利用粒子群优化方法进行参数优化,提高了模型精度。以完好工况的塔机数据为基础,建立基于悬臂梁的双输入单输出模型和基于时域数据的双输入单输出模型,计算参数,建立损伤识别模型,用待检状态的位移数值拟合模型,用两种模型计算出的残差方差做损伤因子,利用密度峰值聚类方法对损伤因子进行分析,实现了对塔机的损伤判定和损伤位置的确定。这种基于密度峰值聚类的诊断方法可对塔机微小损伤进行智能诊断和位置确定,该方法只需要塔机完好状态的数据和待检状态的数据即可自动诊断,解决了塔机损伤识别中损伤数据难以获取,因而无法实现智能训练和诊断的问题。

改进粒子群算法求解生产计划与柔性作业车间调度集成问题

为解决机加工企业制定的生产计划与车间调度方案不兼容的问题,建立以最小化最大完工时间、最小化加工成本为目标函数的生产计划与柔性作业车间调度集成模型。提出一种改进粒子群算法(IPSO)作为全局优化算法。在传统粒子群优化算法(PSO)的基础上,引入遗传算子交叉的方式改进群体进化,同时设计随机性的边界变异,提高种群多样性,避免局部最优,学习因子及惯性权重采用幂函数动态变化,增强其搜索能力,更快收敛。最后通过生产实例,验证了IPSO在解决生产计划与车间调度集成问题上的可行性。同时将PSO、灰狼优化算法(GWO)和遗传算法(GA)作为对比算法,在15个Brandimarte基本算例上开展实验,得到的结果均优于其他算法,证明了IPSO求解柔性作业车间调度问题时的有效性和优越性。

一种基于改进VMD-PSO-CNN-LSTM的短期电价预测方法

为了提升电价预测的准确性和预测模型的稳定性,提出一种基于改进VMD-PSO-CNN-LSTM的短期电价预测方法。首先,通过研究变分模态分解(variational mode decomposition,VMD)与电价影响因素的相关影响程度,并引入最大信息系数(MIC)构建VMD参数优化模型;然后,利用卷积神经网络(convolutional neural networks,CNN)与长短期记忆(long short-term memory,LSTM)神经网络对VMD分解得到的各模态分量进行预测。同时,根据深度可分离卷积结合电价时间规律,在CNN卷积部分构建多尺度的卷积特征提取结构,并利用粒子群优化算法优化包括CNN卷积层数量、CNN卷积神经元数量、LSTM隐藏层数量、LSTM记忆时间以及全连接层数等在内的参数,从而实现模型预测准确性和稳定性的提升。最后,对澳洲电力市场日前电价进行分析预测并与对照算法对比,结果表明该文算法具有更高的精度和更好的稳定性。

基于改进PSO-VMD-MCKD的滚动轴承故障诊断

针对滚动轴承信号在强噪声背景下故障特征提取困难的问题,提出一种变分模态分解(Variational Modal Decomposition,VMD)和最大相关峭度解卷积(Maximum Correlated Kurtosis Deconvolution,MCKD)相结合的故障诊断方法。首先基于VMD方法选取故障信号的最优模态分量,然后采用MCKD算法增强最优分量信号中的冲击成分,最后通过包络谱分析提取滚动轴承的故障频率。利用粒子群优化算法(Particle Swarm Optimization,PSO)对VMD算法中的参数α和K以及MCKD算法中的参数L和M进行寻优,并对PSO算法中惯性因子和学习因子的更新方法加以改进,以提高参数寻优过程的收敛速度。仿真分析和试验结果表明,所提出的诊断方法可以有效提取被强噪声淹没的滚动轴承故障特征。

基于自适应变分模态分解的组合模型风电功率预测

风电机组出力的高波动与随机性,影响电力系统安全稳定运行与风电预测精度,针对此提出结合风电功率波动特性研究的风电功率预测方法。首先从时间与机组规模尺度分析风电功率波动特性,并指导选取合适的风电数据用于风电功率预测;然后建立基于最小二乘支持向量机的风电机组短期功率预测模型,采用自适应变分模态分解实现风电数据分频,并采用改进粒子群优化最小二乘支持向量机模型中影响回归预测的模型参数。实验结果表明,预测模型自适应性较强,通过预测误差评价指标,可证明预测方法的有效性。

基于VMD-LSTM-IPSO-GRU的电力负荷预测

为了挖掘电力负荷数据中的潜藏信息,提高短期负荷预测的精度,针对电力负荷强非线性、非平稳性等特点,提出一种基于变分模态分解(variational mode decomposition,VMD)、长短时记忆神经网络(long-term and short-term memory network,LSTM)、改进的粒子群算法(improve particle swarm optimization,IPSO)和门控循环单元(gated recurrent unit neural network,GRU)的混合预测模型。首先,使用相关性分析确定输入因素,再将负荷数据运用VMD算法结合样本熵分解为一系列本征模态分量(intrinsic mode fuction,IMF)和残差量,进而合理地确定分解层数和惩罚因子;其次,根据过零率将这些量划分为低频和高频,低频分量使用LSTM网络,高频分量利用IPSO-GRU网络分别进行预测;最后,将预测结果重构得到电力负荷的最终结果。仿真结果表明:相对于其他模型,所提混合模型可有效的提取模态特征,具有更高的预测精度。

基于VMD-PSO-LSSVM的降雨量预测研究

降雨事件具有高度随机性,为了对复杂的降雨量进行科学有效地预测,提出VMD-PSO-LSSVM的降雨量预测模型。首先利用VMD方法分解原始降雨量序列;然后运用粒子群算法优化最小二乘支持向量机的关键参数,使用精准构建的预测模型对一系列子序列进行预测;最后合成所有的预测子序列,获得最终预测结果。仿真结果表明,VMD-PSO-LSSVM模型预测结果误差更小,准确度更高,可以成为有效的降雨量预测工具,为农业和水利部门制定水资源管理决策提供参考,降低旱涝灾害的风险。

基于熵权法的旋转机械故障诊断研究

在大数据背景下,基于机器学习的旋转机械故障诊断研究发展迅速。对于非线性非平稳的振动信号,提出了基于熵权法优化PSO-SVM故障诊断模型,强化特征提取能力。本文数据来自CWRU数据集和抽水蓄能电站实测数据,首先,分别对每个数据集的四种故障滑动采样、平滑降噪等预处理,其次,对故障样本VMD分解,利用样本熵、能量熵、模糊熵、功率谱熵构建特征向量,采用熵权法选取权值最大的特征向量作为EWM-PSO-SVM模型输入,得到诊断结果,同时与其他方法进行对比证实方法有效性与准确性。

基于混合粒子群优化的交直流混联配电网扩展规划算法

在交直流混联配电网建设过程中,主要采用传统的粒子群算法求解配电网扩展规划方案,容易陷入局部最优,使得最终生成方案的规划成本较高。因此,提出基于混合粒子群优化的交直流混联配电网扩展规划算法。针对配电公司和分布式电源运营商两个配电网运营主体,设计配电公司为主、运营商为辅的博弈关系图,并构建以系统经济性最优、DG利用率最大为目标的双层配电网扩展规划模型。考虑交直流混联配电网运营要求,设置设备投资约束、网络连通性约束和元件运行特性约束条件。结合粒子群算法和模拟退火算法组成混合粒子群优化算法,应用对双层规划模型进行迭代求解,输出最优交直流混联配电网扩展规划方案。实验结果表明:所提算法应用后得出的配电网扩展方案年规划成本仅为135.07万元,满足了配电网扩展经济性要求。