A COMBINED PARTICLE FILTER METHOD FOR PREDICTING STRUCTURAL PERFORMANCE DEGRADATION
GuanXuexue, ChenJianqiao*,, ZhengYaochen*,, ZhangXiaosheng Department of Mechanics, Huazhong University of Science and Technology, Wuhan 430074, China; Hubei Key Laboratory for Engineering Structural Analysis and Safety Assessment, Wuhan 430074, China 中图分类号:O346.2 文献标识码:A
关键词:贝叶斯原理;证据理论;粒子滤波;差分进化自适应Metropolis算法 Abstract Structural performances will degrade with time due to the influence of loading, environmental and material factors. To assess the status of a structure in service, the structural deterioration process is usually described through physical models with uncertain model parameters. Prior distributions of model parameters are often determined by using the data collected from similar structures. To improve the accuracy of the model, Bayesian inference incorporated with available data is often used to update the distribution of the parameters. In this work, an effective Bayesian method PF-DREAM is proposed. In this approach, firstly, the mixing combination rule of the Dempster-shafer theory (DST) is utilized to get the prior distribution. Thereafter, for evaluating the complicated multidimensional integral in the Bayesian inference formula and obtaining the posterior distribution, a differential evolution adaptive Metropolis (DREAM) approach integrated with the particle filter (PF) is developed. As compared with the original PF method, the proposed PF-DREAM method can enhance the sample particles’ diversity and improve the quality of the model. To illustrate the efficiency and accuracy of the proposed method, a lithium-ion battery problem and a fatigue crack propagation problem are presented. Results demonstrated that the proposed method can provide more accurate results in parameters updating as well as response prediction. As more data is incorporated, the model’s variance becomes smaller, and the predicted mean trajectory is more reliable in terms of the actual deteriorate curves. It is pointed out that PF-DREAM method can be applied to high-dimensional problems and implicit function problems with the same algorithm presented in this paper, only accompanying more iteration numbers and greater computational load for obtaining convergent results.
Keywords:Bayesian inference;Dempster-shafer theory;particle filter;differential evolution adaptive Metropolis -->0 PDF (8212KB)元数据多维度评价相关文章收藏文章 本文引用格式导出EndNoteRisBibtex收藏本文--> 关雪雪, 陈建桥, 郑瑶辰, 张晓生. 预测结构性能退化的混合粒子滤波方法[J]. 力学学报, 2018, 50(3): 677-687 https://doi.org/10.6052/0459-1879-18-014 GuanXuexue, ChenJianqiao, ZhengYaochen, ZhangXiaosheng. A COMBINED PARTICLE FILTER METHOD FOR PREDICTING STRUCTURAL PERFORMANCE DEGRADATION[J]. Chinese Journal of Theoretical and Applied Mechanics, 2018, 50(3): 677-687 https://doi.org/10.6052/0459-1879-18-014
本算例,针对4组电池的容量退化,将其中3组数据作为先验数据信息,预测第4组电池容量退化轨迹. 数据来源说明如下: 对象为石墨阳极锂氧化钴阴极额定容量为1.1 Ah的商业锂离子电池,采用美国Arbin BT2000的锂电池实验系统在室温条件下进行循环充放电老化实验,放电电流为1.1 A,其中4组电池B1、B2、B3、B4的容量退化数据如图3中绿色十字符所示,数据来自于马里兰大学先进寿命周期工程中心. 显示原图|下载原图ZIP|生成PPT 图3电池容量双指数退化模型拟合曲线 -->Fig.3Fitting curves of the double exponential degradation model of battery capacities -->
新窗口打开 Table 2 表 2 表 2利用PF-DREAM方法更新的模型参数后验分布统计特征 Table 2Statistical characteristics of posterior distribution of model parameters using PF-DREAM method
新窗口打开 为了形象地描述更新后的模型参数对响应预测结果的准确性,将基于 , 数据点更新后的物理模型预测轨迹均值绘制于图4、图5. 图4为采用PF方法更新后的响应预测,图5为采用本文 方法更新后的响应预测. 显示原图|下载原图ZIP|生成PPT 图4测量数据与响应预测 (PF更新) -->Fig. 4Field data vs. predictive results for battery B4 using the PF updating -->
显示原图|下载原图ZIP|生成PPT 图5测量数据与响应预测 (PF-DREAM更新) -->Fig. 5Field data vs. predictive results for battery B4 using the PF-DREAM updating -->
(LiuHaibo, JiangChao, ZhengJing, et al. A system reliability analysis method for structures with probability and interval mixed uncertainty ., 2017, 49(2): 456-466 (in Chinese)) [本文引用: 1]
[4]
GeR, ChenJ, WeiJ. Reliability-based design of composites under the mixed uncertainties and the optimization algorithm ., 2008, 21(1): 19-27 [本文引用: 1]
(LiShuaibing, YangRui, LuoXisheng, et al. Instability study of an electrified coaxial jet in a coflowing gas stream ., 2017, 49(5): 997-1007 (in Chinese)) [本文引用: 1]
(ZhouChunxiao,WangRuiqiong, NieZhaokun, et al. Probabilistic modelling of dynamic response of underwater vehicle structure via maximum entropy method ., 2018, 50(1): 114-123 (in Chinese)) [本文引用: 1]
[7]
HuangX, ChenJ. Time-dependent reliability model of deteriorating structures based on stochastic processes and Bayesian inference methods ., 2015, 141(3): 04014123 [本文引用: 1]
[8]
GuanX, HeJ, JhaR, et al. An efficient analytical Bayesian method for reliability and system response updating based on Laplace and inverse first-order reliability computations ., 2012, 97(1): 1-13 [本文引用: 2]
[9]
HeW, WilliardN, OstermanM, et al. Prognostics of lithium-ion batteries based on Dempster-Shafer theory and the Bayesian Monte Carlo method ., 2011, 196(23): 10314-10321 [本文引用: 3]
[10]
RebbaR, MahadevanS. Computational methods for model reliability assessment ., 2008, 93(8): 1197-1207 [本文引用: 1]
[11]
LukeT, JosephBK. Accurate Approximations for posterior moments and marginal densities ., 1986, 81(393): 82-86 [本文引用: 1]
PittMK, SilvaRDS, GiordaniP, et al. On some properties of Markov chain Monte Carlo simulation methods based on the particle filter ., 2012, 171(2): 134-151 [本文引用: 2]
ZioE, PeloniG. Particle filtering prognostic estimation of the remaining useful life of nonlinear components ., 2011, 96(3): 403-409 [本文引用: 1]
[24]
BiH, MaJ, WangF. An improved particle filter algorithm based on Ensemble Kalman filter and Markov Chain Monte Carlo method ., 2015, 8(2): 447-459 [本文引用: 1]
[25]
VrugtJA, BraakCJFT, ClarkMP, et al. Treatment of input uncertainty in hydrologic modeling: Doing hydrology backward with Markov chain Monte Carlo simulation ., 2008, 44(12): 5121-5127 [本文引用: 1]
[26]
VrugtJA, BraakCJFT, DiksCGH, et al. Accelerating Markov chain Monte Carlo simulation by differential evolution with self-adaptive randomized subspace sampling ., 2009, 10(3): 273-290 [本文引用: 1]
[27]
LaloyE, VrugtJA. High-dimensional posterior exploration of hydrologic models using multiple‐try DREAM(ZS) and high-performance computing ., 2012, 50(3): 182-205 [本文引用: 1]
[28]
Vrugt JA. Markov chain Monte Carlo simulation using the DREAM software package: Theory, concepts, and MATLAB implementation ., 2016, 75: 273-316 [本文引用: 1]
[29]
WentworthMT, SmithRC, WilliamsB. Bayesian model calibration and uncertainty quantification for an HIV model using adaptive Metropolis algorithms ., 2017, 26(2): 233-256 [本文引用: 1]
[30]
PostH, VrugtJA, FoxA, et al. Estimation of community land model parameters for an improved assessment of net carbon fluxes at European sites ., 2017, 122(3): 661-689 [本文引用: 1]
[31]
LiangS, JiaH, XuC, et al. A Bayesian approach for evaluation of the effect of water quality model parameter uncertainty on TMDLs: A case study of Miyun reservoir . , 2016, s560-561: 44-54 [本文引用: 1]
[32]
WuWF, NiCC. A study of stochastic fatigue crack growth modeling through experimental data ., 2003, 18(2): 107-118 [本文引用: 1]
[33]
WuWF, NiCC. Statistical aspects of some fatigue crack growth data ., 2007, 74(18): 2952-2963 [本文引用: 1]