电梯系统垂直振动的频率敏感度分析与鲁棒设计

考虑曳引绳刚度在电梯运行过程中具有变刚度的特性,利用Lagrange方程对电梯系统建立了7自由度的振动微分方程,通过试验设计方法得出随机变量与系统频率响应的数值关系,利用人工神经网络技术获得随机变量与频率响应之间的显性函数关系式。根据激振频率与固有频率之比不超过规定值的关系准则,定义了电梯系统纵向振动的可靠性模式,运用频率可靠性分析方法,提出多频激励情况下的系统的频率可靠性及其敏感度的分析方法。在此基础上,结合可靠性敏感度理论与稳健设计方法,将可靠性敏感度融入到目标优化设计中,建立系统频率可靠性鲁棒设计模型。通过工程算例,将可靠性敏感度理论和稳健设计方法应用到电梯系统设计当中,得到满足鲁棒设计要求的随机参数值。结果表明,该方法具有较高的计算效率和求解精度,可以作为电梯系统稳健设计的理论依据。

基于构件的Web系统前馈神经网络可靠性模型

对基于构件结构复杂度较高的Web系统进行可靠性评估时,基于状态或基于路径的软件可靠性评估模型计算复杂度较高,鲁棒性不足。为此,提出了一种计算复杂度低、鲁棒性强的基于构件的前馈神经网络可靠性模型CBPRM。CBPRM将Web系统中各构件的可靠性作为前馈神经网络输入,并基于构件可靠性敏感度对神经元进行动态优化,Web系统可靠性评估由前馈神经网络输出实现。理论分析和实验结果表明,在基于构件结构复杂度较高的Web系统可靠性评估中,CBPRM的计算复杂度低于对比模型,并可确保可靠性评估精度。

计及有序充电的电动汽车接纳能力量化评估方法

通过研究可靠性指标对电动汽车规模的敏感度,可从新的角度找到定量评估电动汽车接纳能力的方法。首先建立了电动汽车充电负荷模型与有序峰谷电价时段优化策略;其次,提出了最大接纳量、可靠性灵敏度等指标,可定量表征系统承受电动汽车接入能力;最后,结合传统可靠性评估方法,提出了计及有序充电的电动汽车接纳能力量化评估方法,采用重要抽样方法可提高抽样模拟效率。通过IEEE24-RTS算例验证了所提指标及方法的可行性。该量化评估方法和指标可拓展应用于实际电网调控电动汽车接入规模,以保证系统的安全可靠性。

System reliability analysis of seismic pseudo-static stability of rock wedge based on nonlinear Barton–Bandis criterion

Based on the nonlinear Barton–Bandis(B–B)failure criterion,this study considers the system reliability of rock wedge stability under the pseudo-static seismic load.The failure probability(Pf)of the system is calculated based on the Monte−Carlo method when considering parameter correlation and variability.Parameter analysis and sensitivity analysis are carried out to explore the influence of parameters on reliability.The relationships among the failure probability,safety factor(Fs),and variation coefficient are explored,and then stability probability curves of the rock wedge under the pseudo-static seismic load are drawn.The results show that the parameter correlation of the B–B failure criterion has a significant influence on the failure probability,but correlation increases system reliability or decreases system reliability affected by other parameters.Under the pseudo-static seismic action,sliding on both planes is the main failure mode of wedge system.In addition,the parameters with relatively high sensitivity are two angles related to the joint dip.When the coefficient of variation is consistent,the probability of system failure is a function of the safety factor.

Reliability and sensitivity of bogie frame of high-speed train with strength degradation

High-speed bogie frame is a key mechanical component in a train system. The reliability analysis of the bogie is necessary to the safety of high-speed train. Reliability analysis of a bogie frame was considered. The equivalent load method was employed to account for random repeated loads in structural reliability analysis. Degradation of material strength was regarded as a Gamma process. The probabilistic perturbation method was, then, employed for response moment computation. Example of a high-speed train bogie structure under time-variant load was employed for reliability and sensitivity analyses. Monte-Carlo simulation verifies the accuracy and efficiency of the proposed method in time-variant reliability analysis. The analysis results show that the reliability calculation considering the strength degradation and repeated load is closer to the practicality than the method of considering reliability calculation only. Its decreasing velocity is faster than the traditional reliability. The reliability sensitivity value changes over time. The analysis results provide a variation trend of reliability and sensitivity to design and usage of bogie frame.