A rest-time-based prognostic model for remaining useful life prediction of lithium-ion battery

Neural Computing and Applications - This paper proposes a novel empirical model for the remaining useful life prediction of lithium-ion battery. The proposed model is capable of modeling both the...     

基于深度自编码网络的软件缺陷预测方法

软件缺陷预测是提升软件质量的有效方法,而软件缺陷预测方法的预测效果与数据集自身的特点有着密切的相关性。针对软件缺陷预测中数据集特征信息冗余、维度过大的问题,结合深度学习对数据特征强大的学习能力,提出了一种基于深度自编码网络的软件缺陷预测方法。该方法首先使用一种基于无监督学习的采样方法对6个开源项目数据集进行采样,解决了数据集中类不平衡问题;然后训练出一个深度自编码网络模型。该模型能对数据集进行特征降维,模型的最后使用了三种分类器进行连接,该模型使用降维后的训练集训练分类器,最后用测试集进行预测。实验结果表明,该方法在维数较大、特征信息冗余的数据集上的预测性能要优于基准的软件缺陷预测模型和基于现有的特征提取方法的软件缺陷预测模型,并且适用于不同分类算法。     

锂离子电池剩余寿命预测方法研究综述

电池故障预测和健康管理(Prediction and Health Management, PHM)评价的主要方法是确定电池的健康状态和剩余使用寿命(Remaining Useful Life, RUL),以此保证锂离子电池安全可靠地工作和实现寿命优化。锂电池RUL预测不仅是PHM中的热点问题和挑战问题,其预测方法的准确性也会直接影响电池管理系统(Battery Management System, BMS)的整体性能。介绍了单体电芯测评标准,对影响锂电池循环寿命的主要因素进行详细分析。简述电池日历寿命和循环寿命。概括和总结了近几年锂离子电池剩余寿命预测方法,比较不同方法的优缺点。提出了当前实际应用中预测锂电池RUL仍存在的关键问题并进行探讨。     

一种改进的锂离子电池剩余寿命预测算法

锂离子电池故障往往会使系统性能下降甚至瘫痪,故障部件剩余寿命的精确估计对整个系统的寿命预测和健康管理至关重要。粒子滤波是一种有效的序列信号处理方法,然而应用于锂离子电池剩余寿命预测准确性并不高。根据锂离子电池电学特性,提出一种改进的粒子滤波算法,基于锂离子电池容量退化指数模型,结合训练数据对锂离子电池剩余寿命进行预测。仿真及实验结果表明,改进的粒子滤波算法对锂离子电池剩余寿命预测误差小于5%。     

基于自编码器的评分预测算法

评分预测是推荐系统的一个组成部分,通过一个实数表达对用户的偏好进行预测,在学术界被广泛研究。神经网络具有很强的特征提取能力,能获取数据深层次的特征。使用神经网络中的一种网络即自编码器,通过扩展使其具有处理像评分矩阵这种有缺失数据的矩阵的能力,并通过实验证明其预测结果与当前主流的评分预测算法SVD的性能接近。     

Combining empirical mode decomposition and deep recurrent neural networks for predictive maintenance of lithium-ion battery

Predictive maintenance of lithium-ion batteries has been one of the popular research subjects in recent years. Lithium-ion batteries can be used as the energy supply for industrial equipment, such as automated guided vehicles and battery electric vehicles. Predictive maintenance plays an important role in the application of smart manufacturing. This mechanism can provide different levels of pre-diagnosis for machines or components. Remaining useful life (RUL) prediction is crucial for the implementation of predictive maintenance strategies. RUL refers to the estimated useful life remaining before the machine cannot operate after a certain period of operation. This study develops a hybrid data science model based on empirical mode decomposition (EMD), grey relational analysis (GRA), and deep recurrent neural networks (RNN) for the RUL prediction of lithium-ion batteries. The EMD and GRA methods are first adopted to extract the characteristics of time series data. Then, various deep RNNs, including vanilla RNN, gated recurrent unit, long short-term memory network (LSTM), and bidirectional LSTM, are established to forecast state of health (SOH) and the RUL of lithium-ion batteries. Bayesian optimization is also used to find the best hyperparameters of deep RNNs. Experimental results with the lithium-ion batteries data of NASA Ames Prognostics Data Repository show that the proposed hybrid data science model can accurately predict the SOH and RUL of lithium-ion batteries. The LSTM network has the optimal results. The proposed hybrid data science model with multiple artificial intelligence-based technologies also demonstrates critical digital-technology enablers for digital transformation of smart manufacturing and transportation.     

面向异质变分超图自动编码器的超边链接预测模型

如何采用超边建模网络数据中的多元关联关系,实现潜在超边链接关系的预测具有重要的现实意义。现有方法主要集中于研究具有成对关系的网络数据,然而,直接将现有的链接预测方法用于超图网络中的超边链接预测具有一定的局限性。因此,提出基于异质变分超图自动编码器的超边链接预测模型(heterogeneous variational hypergraph autoencoder,HVGAE)。首先,利用超图卷积实现变分超图自动编码器,将超图网络数据转换成一种低维空间表示;其次,加入节点近邻度函数,最大程度地保留其结构信息,从而构建异质超图网络超边链接预测模型。针对三种不同类型的超图网络进行实验,结果表明相比其他的基准方法,HVGAE模型获得了较好的预测结果,说明其能够较好地解决超图网络中的超边链接预测问题。     

无监督聚类在锂离子电池分类中的应用

单体电池的一致性,决定了电池组的性能,如何选出性能一致的单体电池又一直是电池组研究中的重点所在。本文采集了100个合格锂离子电池的6项性能指标(老化前后电压、容量、内阻、1C放电平台、电芯厚度),运用主成分分析(PCA)、核主成分分析(KPCA)、随机森林(RF)3种无监督聚类方法,对数据结构进行了研究。结果表明,数据指标之间存在复杂的非线性关系,主成分分析和核主成分分析,均未能形成明显聚类,但随机森林数据在低维空间显然形成4类,任意从中选4个电池组成电池组作循环性能仿真测试,结果显示由由该方法挑选出的单体电池具有较好的一致性。     

Learning social representations with deep autoencoder for recommender system

World Wide Web - With the development of online social media, it attracts increasingly attentions to utilize social information for recommender systems. Based on the intuition that users are...     

A deep wavelet sparse autoencoder method for online and automatic electrooculographical artifact removal

Electrooculographical (EOG) artifacts are problematic to electroencephalographical (EEG) signal analysis and degrade performance of brain–computer interfaces. A novel, robust deep wavelet sparse autoencoder (DWSAE) method is presented and validated for fully automated EOG artifact removal. DWSAE takes advantage of wavelet transform and sparse autoencoder to become a universal EOG artifact corrector. After being trained without supervision, the sparse autoencoder performs EOG correction on time–frequency coefficients collected after brain wave signal wavelet decomposition. Corrected coefficients are then used for wavelet reconstruction of uncontaminated EEG signals. DWSAE is compared with five other methods: second-order blind identification, information maximization, joint approximation diagonalization of eigen-matrices, wavelet neural network (WNN) and wavelet thresholding (WT). Experimental results on a visual attention task dataset, a mental state recognition dataset and a semi-simulated contaminated EEG dataset show that DWSAE is capable of suppressing EOG artifacts effectively, while preserving the nature of background EEG signals. The mean square error of signals before and after correction by DWSAE on a semi-simulated contaminated EEG segment of 30 s is the lowest (65.62) when compared to the results produced by WNN and WT. DWSAE addresses limitations posed by these methods in three ways. First, DWSAE can be performed automatically and online in a single channel of EEG data; this has advantages over independent component analysis-based methods. Second, its results are robust and stable in comparison with those of other wavelet-based methods. Third, as an unsupervised learning scheme, DWSAE does not require the off-line training that is necessary for WNN and other supervised learning machine learning-based methods.

     

Stacked autoencoder for operation prediction of coke dry quenching process

Coke dry quenching (CDQ) is widely adopted for waste heat recovery in iron and steel plants. In this work, an economic benefit index was introduced to evaluate the performance of the CDQ system and stacked autoencoder (SAE) based deep neural networks are adopted for CDQ operation prediction. Based on the prediction results, a guidance is provided for online adjustment of the supplementary air flow rate, hence the efficiency and safety of the CDQ system can be improved. The case study on a real plant shows that the proposed method increases the economic efficiency of the CDQ process by 4.39%.     

A context-augmented deep learning approach for worker trajectory prediction on unstructured and dynamic construction sites

Predicting workers’ trajectories on unstructured and dynamic construction sites is critical to workplace safety yet remains challenging. Existing prediction methods mainly rely on entity movement information but have not fully exploited the contextual information. This study proposes a context-augmented Long Short-Term Memory (LSTM) method, which integrates both individual movement and workplace contextual information (i.e., movements of neighboring entities, working group information, and potential destination information) into an LSTM network with an encoder-decoder architecture, to predict a sequence of target positions from a sequence of observations. The proposed context-augmented method is validated using construction videos and the prediction accuracy achieved is 8.51 pixels in terms of final displacement error (FDE), with an observation time of 3 s and prediction time of 5 s—5.4% smaller than using the position-based method. Compared to conventional one-step-ahead predictions, the proposed sequence-to-sequence method predicts trajectories over multiple steps to avoid error accumulation and effectively reduces the FDE by 70%. In addition, qualitative analysis is conducted to provide insights to select appropriate prediction methods given different construction scenarios. It was found that the context-aware model leads to better performance comparing to the position-based method when workers are conducting collaborative activities.     

A high-order accurate numerical algorithm for three-dimensional transport prediction

The numerical solution of the three-dimensional pollutant transport equation is obtained with the method of fractional steps; advection is solved by the method of moments and diffusion by cubic splines. Topography and variable mesh spacing are accounted for with coordinate transformations. First estimate wind fields are obtained by interpolation to grid points surrounding specific data locations. Mass consistency is ensured by readjusting the three-dimensional wind field with a Sasaki variational technique. Numerical results agree with results obtained from analytical Gaussian plume relations for ideal conditions. The numerical model is used to simulate the transport of tritium released from the Savannah River Plant on 2 May, 1974. Predicted ground level air concentration 56 km from the release point is within 38% of the experimentally measured value.     

基于SOC的锂离子电池组均衡控制

不一致性问题极大地降低了锂离子电池组的整体性能,均衡控制是目前能有效改善电池组间不一致性的唯一办法。在分析了目前主流均衡设计方案的基础上,针对Buck-Boost均衡电路,提出了以锂电池荷电状态（SOC）为均衡对象的均衡控制策略。同时,设计了一种新式的基于双模型自适应扩展卡尔曼滤器的SOC估算方法。实验结果表明,该均衡控制策略改善了电池组间的不一致性,提高了容量利用率。     

Dynamic energy model of a lithium-ion battery

As an electrochemical component, a lithium-ion battery is clearly a multi-disciplinary system. The choice was made to model it via Bond Graph formalism. Although this tool has been developed since the 1970s, the novelty is its application to lithium-ion batteries, which turns the modeling presented here into an original energy approach. The main objective is to develop and validate a lithium-ion battery model that could be implemented in a global system for energy monitoring. However, nearly every phenomenon occurring in the battery is taken into account for a possible ageing or thermal study. In the first part, the energy modeling approach is described. In the second part, the lithium-ion battery operation is explained. In the third part, the Bond Graph model is proposed. At last, experimental validations are presented.     

DRA-ODM: a faster and more accurate deep recurrent attention dynamic model for object detection

World Wide Web - The traditional object detection model based on convolutional neural network contains a large amount of parameters, so it has poor performance when applied to high-real-time and...     

A new supervised multi-head self-attention autoencoder for health indicator construction and similarity-based machinery RUL prediction

Remaining useful life (RUL) prediction plays a significant role in the prognostic and health management (PHM) of rotating machineries. A good health indicator (HI) can ensure the accuracy and reliability of RUL prediction. However, numerous existing deep learning-based HI construction approaches rely heavily on the prior knowledge, and they are difficult to capture the key information in the process of machinery degradation from raw signals, thereby affecting the performance of RUL prediction. To tackle the aforementioned problem, a new supervised multi-head self-attention autoencoder (SMSAE) is proposed for extracting the HI that effectively reflects the degraded state of rotating machinery. By embedding the multi-head self-attention (MS) module into autoencoder and imposing the constraint of power function-type labels on the hidden variable, SMSAE can directly extract the HIs from raw vibration signals. As the current HI evaluation indexes don’t consider the global monotonicity and variation law of HI, two improved monotonicity and robustness indexes are designed for the better evaluation of HI. With the proposed HI, a two-stage residual life prediction framework based on similarity is developed. Extensive experiments have been performed on an actual wind turbine gearbox bearing dataset and a well-known open commercial modular aero-propulsion system simulation (C-MAPSS) dataset. The comparative results verify that the constructed SMSAE HI has better comprehensive performance than the typical HIs, and the proposed prediction method is competitive with the state-of-the-art methods.     

Stacked autoencoder based deep random vector functional link neural network for classification

Extreme learning machine (ELM), which can be viewed as a variant of Random Vector Functional Link (RVFL) network without the input–output direct connections, has been extensively used to create multi-layer (deep) neural networks. Such networks employ randomization based autoencoders (AE) for unsupervised feature extraction followed by an ELM classifier for final decision making. Each randomization based AE acts as an independent feature extractor and a deep network is obtained by stacking several such AEs. Inspired by the better performance of RVFL over ELM, in this paper, we propose several deep RVFL variants by utilizing the framework of stacked autoencoders. Specifically, we introduce direct connections (feature reuse) from preceding layers to the fore layers of the network as in the original RVFL network. Such connections help to regularize the randomization and also reduce the model complexity. Furthermore, we also introduce denoising criterion, recovering clean inputs from their corrupted versions, in the autoencoders to achieve better higher level representations than the ordinary autoencoders. Extensive experiments on several classification datasets show that our proposed deep networks achieve overall better and faster generalization than the other relevant state-of-the-art deep neural networks.     

出租车目的地预测的深度学习方法

出租车目的地预测可以掌握出租车的流动方向,便于出租车调度。已有的预测方法多仅利用轨迹序列的原始特征作为预测模型的输入,忽略了原始特征背后的时空数据,造成轨迹时空信息缺失。针对以上问题,提出出租车目的地预测的深度学习方法DLDP。首先采用滑动窗口,基于速度、转角利用统计量计算得到轨迹的高层特征。其次,自动编码器将高层特征转换为固定长度的潜在空间表示,得到轨迹的深度特征。最后,将深度特征和原始特征相结合,一同作为LSTM的输入进行预测。实验表明,DLDP比传统RNN预测模型的准确率提高了9%,平均距离误差减少了1 km。