基于CEEMD的重大事件对香港住宅价格影响的实证分析

应用互补集成经验模态分解（CEEMD）方法对香港1997―2018年的住宅价格月度数据进行了分解，将经过重构后的数据分成高频序列、低频序列与残差项。将BP多断点检测应用于低频序列，并结合样本时段内的重大事件进行实证分析。结果表明：1997年亚洲金融风暴对房价的影响大于2008年金融危机；外部经济体的救市政策间接地影响香港房价；在经济不景气的大环境下“孙九招”政策没有立即见效；资本投资者入境计划、住房供给调整与按揭贷款调整对房价的影响较为显著；税收调整对房价影响不显著、对交易量影响显著；SARS爆发使住宅价格下降约1%。     

BP neural network recognition algorithm for scour monitoring of subsea pipelines based on active thermometry

Scour monitoring is an important concern for subsea pipeline systems. The active-thermometry-based scour monitoring is based on the difference of heat transfer properties between sediment and sand, recognizes the surrounding media though temperature changing patterns during heating and cooling processes, and hence detects the free spans. Based on the scour monitoring system, a two-layer BP neural network was employed to process the monitoring data and achieved media recognition. The network's inputs were normalized temperature time histories. The network's outputs denoted different media: sediment and water. To validate the method, three experiments were conducted; one was used for training the network and the other two for testing. Also, the effect of noise on the network's performance was studied through simulation. The results demonstrated the feasibility and robustness of the neural network.     

Analysis of an elasto-piezoelectric system of hemivariational inequalities with thermal effects

In this paper we prove the existence and uniqueness of a weak solution for a dynamic electo-viscoelastic problem that describes a contact between a body and a foundation. We assume the body is made from thermoviscoelastic material and consider nonmonotone boundary conditions for the contact. We use recent results from the theory of hemivariational inequalities and the fixed point theory.     

广东省能源需求预测模型构建及实证分析

为了对广东省的能源需求进行准确的预测,首先分析了影响广东省能源需求的各种因素,构建了预测指标体系.在此基础上,针对能源系统非线性等复杂系统特征,结合粒子群算法和BP神经网络的优点,构建了改进的PSO-BP神经网络的预测模型,并通过主成分分析法对指标体系进行数据降维,以降低神经网络的规模和复杂程度.以广东省1985-2013年的能源需求数据进行模拟与仿真,并对2014-2018年的能源需求量进行预测,理论分析和实证研究表明,该方法能够很好的反映广东省能源需求的特征,预测结果较为准确合理.     

基于多级BP神经网络的无线电罗盘多故障诊断研究

【】本文针对过拟合现象提出了基于提前停止法的学习率可变BP算法,运用多级BP神经网络诊断思想,提出了基于多级BP神经网络的多故障诊断方法。文中根据BP神经网络改进算法和网络撕裂法,对具体的机载无线电罗盘测向电路建模仿真,将复杂的无线电罗盘电路分解为3个子网络,并对每个子网络建立合适的故障集,快速准确诊断得出电路中的故障模块。     

基于GA-BP的移动通信设备故障诊断

针对通信设备故障发生随机性强,影响因素多,对应的故障诊断有高度非线性和不确定性的特点,采用BP神经网络算法,优化的GA-BP神经网络算法和POS-BP神经网络算法分别搭建基站设备故障诊断模型,提取设备故障历史数据进行MATLAB仿真,准确预测设备故障类型,帮助提高代维公司调度管理的智能化水平,提高基站设备运维的执行效率。仿真结果表明：本文的BP,GA-BP和POS-BP神经网络算法都能够实现设备故障类别的预测,且GA-BP神经网络算法相比BP和POS-BP神经网络算法对通信设备故障诊断有更好的适应性。     

基于BP神经网络的LED可靠性模型研究

根据LED可靠性与相关参数的映射关系,建立拓扑结构为6-12-1的BP神经网络。以实测白光LED芯片的理想因子、结温、色温漂移等参数为输入量,以寿命为输出量,计算模型精度。研究结果表明,该模型有良好的外推能力及鲁棒性,可在短时间内成功预测LED寿命,神经网络训练结果相关系数为99.8%,检验组误差小于3%。     

MACRO AND MICROSCOPIC PARAMETER MAPPING RELATIONSHIP FOR RED SANDSTONE 1)

颗粒流离散元软件PFC2D在岩土工程中的应用十分广泛,存在的主要问题是如何标定其细观参数,目前大多使用"试凑法",此方法的缺点在于工作量大、效率低等,本文提出以反向传播算法(back propagation,BP)神经网络的方式代替此方法,利用PFC2D 内置的FISH 以及Python 语言对其进行二次开发,使之自动运行并获取40组宏细观参数样本。结果表明：BP神经网络可以快速准确地建立宏细观参数映射关系,误差均保持在0.01 以内,模拟得到的应力--应变曲线与室内试验曲线高度吻合,且无需大量的数据样本便可创建网络,效率较高;另外,经验证本文选用的平直节理模型,可以有效地解决平行粘结模型UCS/TS值偏小问题,确定平直节理模型可以更好地模拟岩石。     

Improved neural networks based on genetic algorithm for pulse recognition

Pulse diagnosis is an important part of Chinese medicine and has played an important role in the development of Chinese medical science. However, the pulse is traditionally determined by cutting it off, which leads to a lack of objective standard pulse identification methods and affects their accuracy and feasibility. This research has studied and discussed the processing and identification of four kinds of pulse: normal pulse, wiry pulse, smooth pulse, and thready pulse. Four frequency-domain characteristics of the pulse wave and six kinds of wavelet scale energy characteristic information were extracted, and a three-layer BP (backprocessing) neural network was established. The LM (Levenberg–Marquard) algorithm and a genetic algorithm were used to improve the BP neural network, to train on and predict experimental samples, and to obtain classification accuracies of 90% and 95% respectively. Moreover, improved BP neural network based on a genetic algorithm has shown highly superior performance in terms of convergence speed and low error rate.