A Time-Domain Numerical Simulation for Free Motion Responses of Two Ships Advancing in Head Waves

The constant panel method within the framework of potential flow theory in the time domain is developed for solving the hydrodynamic interactions between two parallel ships with forward speed.When solving problems within a time domain framework,the free water surface needs to simultaneously satisfy both the kinematic and dynamic boundary conditions of the free water surface.This provides conditions for adding artificial damping layers.Using the Runge−Kutta method to solve equations related to time.An upwind differential scheme is used in the present method to deal with the convection terms on the free surface to prevent waves upstream.Through the comparison with the available experimental data and other numerical methods,the present method is proved to have good mesh convergence,and satisfactory results can be obtained.The constant panel method is applied to calculate the hydrodynamic interaction responses of two parallel ships advancing in head waves.Numerical simulations are conducted on the effects of forward speed,different longitudinal and lateral distances on the motion response of two modified Wigley ships in head waves.Then further investigations are conducted on the effects of different ship types on the motion response.

Research on Short-Term Load Forecasting of Distribution Stations Based on the Clustering Improvement Fuzzy Time Series Algorithm

An improved fuzzy time series algorithmbased on clustering is designed in this paper.The algorithm is successfully applied to short-term load forecasting in the distribution stations.Firstly,the K-means clustering method is used to cluster the data,and the midpoint of two adjacent clustering centers is taken as the dividing point of domain division.On this basis,the data is fuzzed to form a fuzzy time series.Secondly,a high-order fuzzy relation with multiple antecedents is established according to the main measurement indexes of power load,which is used to predict the short-term trend change of load in the distribution stations.Matlab/Simulink simulation results show that the load forecasting errors of the typical fuzzy time series on the time scale of one day and one week are[−50,20]and[−50,30],while the load forecasting errors of the improved fuzzy time series on the time scale of one day and one week are[−20,15]and[−20,25].It shows that the fuzzy time series algorithm improved by clustering improves the prediction accuracy and can effectively predict the short-term load trend of distribution stations.

Non-harmonic resonance of viscoelastic structures subjected to time-dependent exponentially decreasing transverse distributed loads

In this paper,the non-harmonic resonance of Bernoulli viscoelastic beams,Kirchhoff viscoelastic plates,Timoshenko viscoelastic beams,and Mindlin viscoelastic plates subjected to time-dependent exponentially decreasing transverse distributed load is investigated for the first time.The constitutive equations are expressed utilizing Boltzmann integral law with a constant bulk modulus.The displacement vector is approximated by employing the separation of variables method.The Laplace transformation is used to transfer equations from the time domain to the Laplace domain and vice versa.The novel point of the proposed method is to express,prove and calculate the critical time in which the displacement will be several times the displacement at time zero.In addition,this new method calculates the maximum deflection at the critical time,explicitly and exactly,without any need to follow the time-displacement curve with a low computational cost.Additionally,the proposed method introduces the critical range of time so that the responses are greater than the responses at time zero.

再生粗骨料混凝土应力-应变关系

再生粗骨料混凝土应力-应变关系是实现其材料到结构力学分析的桥梁纽带,成为再生粗骨料混凝土结构基础理论的基石。介绍了作者团队多年来在再生粗骨料混凝土应力-应变关系方面取得的研究进展:采用模型化再生粗骨料方法,研究了复杂界面过渡区对再生粗骨料混凝土破坏行为的影响,揭示了再生粗骨料混凝土细观损伤本质与演化机理;从静力作用到动力作用,系统地开展了不同工况下再生粗骨料混凝土应力-应变行为试验研究,探明了载荷条件对再生粗骨料混凝土应力与变形的影响规律并建立了相适应的力学与数学模型;进一步考虑再生粗骨料性能时空变异性,发现了再生粗骨料混凝土力学响应的概率分布特征,提出了再生粗骨料混凝土随机损伤本构关系;基于获得的本构模型,完成了再生粗骨料混凝土构件时变可靠度分析和结构动力非线性分析,为再生粗骨料混凝土在实际工程中的安全应用提供了理论支撑;提炼了相关研究结论并对未来研究工作进行了展望。

基于时变权重模型预测的双向DC-DC优化控制方法

双向DC-DC变换器是电动汽车的关键电压变换部件,提高其动态负载下的响应速度与鲁棒性对电动汽车低压供电系统有重要意义。提出一种基于模糊时变权重的MPC(model predictive control)控制方法。基于状态空间平均法建立变换器动态模型。基于MPC原理搭建变换器MPC模型,实现对电压的跟踪控制;在分析权重对动态负载下电压跟踪性能的影响的基础上,设计权重模糊调节器,实现权重的在线调节。基于HM-cSPACE搭建实验平台验证所提方法的准确性。在Matlab/Simulink环境下与PI控制和传统MPC控制进行对比,结果表明:在负载变化时采用时变权重MPC控制的变换器有更好的动态性能和鲁棒性,超调量降低26.1%。该方法对提高双向DC-DC变换器动态负载下的优化控制具有重要意义。

考虑供需灵活性量化的源-荷-储多时间尺度协同优化

风电、光伏类新能源出力具有随机性和波动性,且随着新能源渗透率的增高,系统灵活性需求也激增。在此背景下,考虑源-荷-储灵活性资源互动,对新能源高渗透系统源-荷-储协同运行问题展开研究。首先,引入向上、向下裕度量化系统灵活性,并利用灵活性供需平衡机理动态评估灵活性需求;其次,提出一种考虑供需灵活性的多时间尺度协同运行方法,构建典型周、典型日、典型时段运行模拟模型,并将碳交易收益和灵活性不足惩罚成本纳入优化目标,考虑调峰需求和爬坡容量对典型时段灵活性的影响,采用多级优化、逐级细化的思路求解模型。算例分析表明,所提方法在保障系统环保性和经济性的基础上,提升了系统各个时间尺度上的灵活性容量,可以促进系统新能源消纳。

基于STE-TCN的中短期电力负荷预测

目的 针对传统电力负荷预测模型对长序列预测精度低的问题,提出一种结合跳级卷积连接与时间编码网络的新型时序卷积神经网络(TCN)模型——STE-TCN模型。方法 首先对TCN模型加入跨周期的膨胀卷积通道(Skip-convolution)提取电力数据周期信息;再进行特征融合得到Skip-TCN网络,使网络抓取周期规律,增加信息利用长度;最后设计日期编码网络(Time encoding network)捕捉生活周期和季节性特征,与Skip-TCN进行特征融合得到STE-TCN模型,实现对电力负荷数据长序列预测。结果 实验表明:在与TCN模型和传统时序网络的对比下,Skip-TCN的预测精度均有提升,在预测长度更长的测试上提升尤为明显。结论 实验结果验证了通过对更长跨度时序关系的捕捉,STE-TCN网络改进方法有效提升了对长序列电力负荷的预测精度。

应用n-LSTM的云平台任务CPU负载预测方法

云平台任务的CPU负载预测有助于云平台资源的优化配置,以改善资源利用率.它是有效管理云资源的重要手段.为提高任务CPU负载预测精度,本文主要做了以下工作:1)利用热度图提取用于进行CPU负载预测的资源使用特征;2)设计并实现了一种基于n-LSTM的云平台任务的CPU负载预测方法DPFE-n-LSTM;3)分别在阿里云平台数据集和Google云平台数据集上进行了实验,结果表明,相对于目前已经提出的CPU负载预测模型BP、LSTM和CNN-LSTM,DPFE-n-LSTM方法具有更好的预测性能.

Optimal guidance strategy for flexible load based on hybrid direct load control and time of use

The time-of-use(TOU)strategy can effectively improve the energy consumption mode of customers,reduce the peak-valley difference of load curve,and optimize the allocation of energy resources.This study presents an Optimal guidance mechanism of the flexible load based on strategies of direct load control and time-of-use.First,this study proposes a period partitioning model,which is based on a moving boundary technique with constraint factors,and the Dunn Validity Index(DVI)is used as the objective to solve the period partitioning.Second,a control strategy for the curtailable flexible load is investigated,and a TOU strategy is utilized for further modifying load curve.Third,a price demand response strategy for adjusting transferable load is proposed in this paper.Finally,through the case study analysis of typical daily flexible load curve,the efficiency and correctness of the proposed method and model are validated and proved.

基于改进注意力机制的时间卷积网络-长短期记忆网络短期电力负荷预测

为充分挖掘蕴含在电力负荷数据中的有效时序信息,提高短期电力负荷预测准确度,本文提出一种基于改进注意力机制的时间卷积网络(TCN)-长短期记忆(LSTM)网络负荷预测模型。首先,将时序数据输入TCN模型中进行时序特征提取;然后,将所提取的时序特征与非时序数据组合,并输入LSTM模型中进行训练;最后,采用贝叶斯优化方法进行超参数寻优以获得TCN-LSTM模型的最优参数,引入通过多层感知器(MLP)改进的注意力机制以减少历史信息丢失并加强重要信息的影响,完成短期负荷预测。通过对比多种深度学习模型的预测效果表明,本文所提模型的短期电力负荷预测准确度更高。

Load-redistribution strategy based on time-varying load against cascading failure of complex network

Cascading failure can cause great damage to complex networks, so it is of great significance to improve the network robustness against cascading failure. Many previous existing works on load-redistribution strategies require global information, which is not suitable for large scale networks, and some strategies based on local information assume that the load of a node is always its initial load before the network is attacked, and the load of the failure node is redistributed to its neighbors according to their initial load or initial residual capacity. This paper proposes a new load-redistribution strategy based on local information considering an ever-changing load. It redistributes the loads of the failure node to its nearest neighbors according to their current residual capacity, which makes full use of the residual capacity of the network. Experiments are conducted on two typical networks and two real networks, and the experimental results show that the new load-redistribution strategy can reduce the size of cascading failure efficiently.

Space-time evolution rules of acoustic emission location of unloaded coal sample at different loading rates

By using MTS815 rock mechanics test system,a series of acoustic emission(AE) location experiments were performed under unloading confining pressure,increasing the axial stress.The AE space-time evolution regularities and energy releasing characteristics during deformation and failure process of coal of different loading rates are compared,the influence mechanism of loading rates on the microscopic crack evolution were studied,combining the AE characteristics and the macroscopic failure modes of the specimens,and the precursory characteristics of coal failure were also analyzed quantitatively.The results indicate that as the loading rate is higher,the AE activity and the main fracture will begin earlier.The destruction of coal body is mainly the function of shear strain at lower loading rate and tension strain at higher rate,and will transform from brittleness to ductility at critical velocities.When the deformation of the coal is mainly plasticity,the amplitude of the AE ringing counting rate increases largely and the AE energy curves appear an obvious ''step'',which can be defined as the first failure precursor point.Statics of AE information shows that the strongest AE activity begins when the axial stress level was 92-98%,which can be defined as the other failure precursor point.As the loading rate is smaller,the coal more easily reaches the latter precursor point after the first one,so attention should be aroused to prevent dynamic disaster in coal mining when the AE activity reaches the first precursor point.

基于MDTimeGAN的序列数据生成方法

非侵入式负荷分解是能源管理领域的一个热门研究课题,其在各种工业和商业场景中都得到广泛应用。针对负荷分解数据集中存在的样本不平衡问题,提出一种基于多判别器时间序列生成对抗网络(MDTimeGAN)的序列数据生成方法。通过对原始序列提取时域、频域、时频域以及自相关特征,并在TimeGAN模型基础上采用4种不同的判别器对时间序列的多维度特征进行判别,从而提高对原始数据的判别能力,提升数据质量。在3种公开数据集上进行横向和纵向对比实验,结果表明,与对比模型相比,MDTimeGAN模型生成的数据能够更好地覆盖原始数据的分布,在数据分布方面保持良好的性能,生成数据符合时间序列数据的特点。

基于ConvLSTM-LSTM的短期负荷预测方法

长短时记忆(LSTM)网络和结合卷积神经网络(CNN)的CNN-LSTM预测模型由于其网络模型本身的缺陷,限制了预测精度的提高。针对以上问题,提出了一种结合卷积长短时记忆(ConvL⁃STM)网络的ConvLSTM-LSTM负荷预测模型。利用ConvLSTM网络充分提取时序特征,将提取到的信息输入到LSTM网络中进行进一步的选择性记忆,并输出预测结果。将该模型与CNN-LSTM网络模型、LSTM网络模型、以及门控循环单元(GRU)网络模型进行了对比,结果显示所构建的Con⁃vLSTM-LSTM模型的预测效果均优于对比模型,在精度评价指标平均绝对百分比误差(MAPE)上,分别减小了1.10%、1.54%、1.91%。

Nelson-Siegel模型中的时变载荷因子及其对提高经济形势预测的重要作用

现有文献中Nelson-Siegel模型大多将载荷因子λ提前固定为常数或作为待估静态参数,较少考虑载荷因子λ的时变化。本文基于得分驱动时变参数建模方法,在状态空间模型框架内考虑时变载荷因子λ,构建GAS-λ-DNS模型,并提取相应时变载荷因子。结果显示,时变载荷因子λ_(t)呈现出极强的波动性,同时与经济周期密切相关;将时变载荷因子λ_(t)应用于预测工业产出增速,发现载荷因子相比于传统宏观预测因子具有额外增量信息,引入λ_(t)可以显著提高模型预测精度。本文结论对Nelson-Siegel利率期限结构建模和宏观经济预测因子选取具有参考价值。

基于DLT-Kmedoids算法的用电负荷聚类分析

针对高校用电负荷中传统聚类算法直接应用于时间序列聚类效果准确性较低的问题,提出一种融合DTW距离、LB_Keogh距离以及时间窗口的DLT-Kmedoids算法,以提高聚类算法应用于时间序列的准确性以及算法效率。DLT-Kmedoids算法使用DTW计算时序数据之间的距离取代传统的欧氏距离度量方式,提高了相似性度量算法精度,同时也提高了聚类算法的准确性和复杂度,引入LB_Keogh距离在计算DTW距离之前过滤掉大部分不可能是最优匹配序列的序列,对于剩下的序列再使用DTW逐个比较,进一步降低算法的复杂度。最后结合高校建筑用电负荷时间序列数据进行分析,通过与主流聚类算法进行比较,表明该算法对于高校用电负荷数据的聚类任务,能够更准确地识别相似的负荷模式,并以更高的效率进行聚类分析。

黏土-砂混合物三轴压缩超声波特性研究

为研究核废料缓冲回填材料的超声波特征,本文利用岩石三轴压缩伺服仪和超声波测试系统,开展侧限压缩和常规三轴压缩2种应力路径和2种超固结比状态OCR=0.83和2.9的压实黏土-砂混合物的三轴压缩试验和超声波测试。结果表明:侧限压缩试验的纵、横波波速(V_(p)、V_(s))曲线表现出与轴压-轴向应变曲线相似的3阶段特征,但V_(p)曲线具有滞后效应;常规三轴压缩试验的V_(p)增加,V_(s)降低,V_(p)和V_(s)变化与围压有关。动剪切模量和动泊松比(G_(d)和μ_(d))的曲线分别与V_(s)和V_(p)的曲线相似,围压主要影响动剪切模量的大小;V_(p)和V_(s)与有效平均压力(p)有关,V_(p)和V_(s)与p的关系可用二次函数较好描述;横波信号的主频逐渐增大,主振幅先减小后增大,主频变得不易分辨,围压越大,横波信号衰减越强;Ⅰ和Ⅱ区随偏压加载而逐渐融合,横波分裂效应减弱。研究结果可为核废料缓冲回填材料的配比设计提供参考。

基于时域BLT方程的带绝缘线缆束场-线耦合模型

为研究外场辐照下复杂线缆系统(多导体、非线性负载、带绝缘层)终端电压/电流时域响应特性规律,提出了复杂线缆束在入射场激励下的终端响应建模方法。研究在时域内重塑多导体BLT方程,在兼顾计算效率、易于向线缆网络扩展的基础上,解决了线缆束端接非线性负载的终端时域响应求解问题;并通过引入介质层系数,修正了BLT方程源矢量项,解决了常规BLT方程忽视绝缘层对入射场畸变效应而导致的误差问题。最后,将计算结果同时域有限差分法和有限积分法计算结果进行对比,验证了模型在处理复杂线缆场线耦合问题上的准确性。

Effects of drive current rise-time and initial load density distribution on Z-pinch characteristics

A two-dimensional, three-temperature radiation magneto-hydrodynamics model is applied to the investigation of evolutional trends in x-ray radiation power, energy, peak plasma temperature and density as functions of drive current rise-time and initial load density distribution by using the typical experimental parameters of tungsten wire-array Z- pinch on the Qiangguang-I generator. The numerical results show that as the drive current rise-time is shortened, x-ray radiation peak power, energy, peak plasma density and peak ion temperature increase approximately linearly, but among them the x-ray radiation peak power increases more quickly. As the initial plasma density distribution in the radial direction becomes gradually flattened, the peak radiation power and the peak ion-temperature almost exponentially increase, while the radiation energy and the peak plasma density change only a little. The main effect of shortening drive current rise-time is to enhance compression of plasma, and the effect of flattening initial load density distribution in the radial direction is to raise the plasma temperature. Both of the approaches elevate the x-ray peak radiation power

基于N-BEATS的能源互联网短期负荷预测

短期负荷预测在能源互联网的规划中既占重要组成部分,又是能源系统可靠高效运行的基础。在能源互联网中能源的短期负荷预测精度问题是人们重点关注问题。N-BEATS的深度神经结构未使用时序特别组成成分,仅使用一种基于后向和前向残差链路以及非常深的全连接层堆栈的深度神经架构。该结构具有可解释性、适用于广泛的目标域、并且训练速度快等优点。实验使用N-BEATS模型对历史负荷数据进行训练,然后对未来负荷进行短期负荷预测,取得了较高的预测精度。测得平均绝对百分比误差(eMAPE)为1.26%,平均绝对误差(eMAE)为84.238 kW,决定系数(R^(2))为0.9955,实验结果表明采用该方法的预测精度高于传统的预测方法,如在eMAPE方面相比TCN降低了0.61%。