Prediction of Fault Symptoms in Elevator Knowledge Base Based on Improved LSTM-AE Algorithm

doi:10.3778/j.issn.1002-8331.2111-0134

Abstract

Abstract: To solve the problem of fault symptom prediction in the operation and maintenance knowledge base system, an improved LSTM-AE algorithm for elevator equipment is proposed. Firstly, in view of redundancy problem in the elevator operating parameter sequence, the attribute correlation density ranking（ACDR） method is used to filter the feature vector group. Moreover, aiming at the non-stationary problem of the running speed characteristic sequence, variational mode decomposition（VMD） algorithm is used for noise reduction and smoothing. Finally, a sliding window attention mechanism fused with BILSTM is introduced into the LSTM-AE model to improve the ability of time-series feature extraction, and softmax classifier is used to fuse the reconstruction error of each feature sequence to realize prediction of the elevator fault symptom. The experimental results show that compared with traditional LSTM-AE algorithm, the proposed improved LSTM-AE algorithm has a 13% increase in the normal sample accuracy rate and a 11% reduction in the abnormal sample error rate. It can predict common elevator faults more precisely, and be suitable for constructing a reliable fault symptom prediction model of elevator operation and maintenance knowledge base.

Key words: fault symptom prediction, feature redundancy, variational mode decomposition, noise reduction and smoothing, time-series feature extraction

摘要： 针对运维知识库系统中故障征兆预测问题，提出面向电梯设备的改进LSTM-AE算法。使用属性子集选择（ACDR）方法筛选特征向量组，剔除电梯运行参数中的冗余特征。同时，针对运行速度特征序列的非平稳性问题，使用变分模态分解（VMD）算法作降噪平稳化处理。在LSTM-AE模型中引入融合BILSTM的滑动窗口注意力机制，提高模型的时序特征提取能力，并通过softmax分类器融合各特征序列的重构误差实现电梯故障征兆预测。实验结果表明，相较经典LSTM-AE算法，提出的改进LSTM-AE算法正常样本判准率提高13%，异常样本误判率降低11%，能够对常见电梯故障进行准确预测，适于构建可靠的电梯运维知识库故障征兆预测模型。

关键词: 故障征兆预测, 特征冗余, 变分模态分解, 降噪平稳化, 时序特征提取

SUN Qinggang, WANG Cheng. Prediction of Fault Symptoms in Elevator Knowledge Base Based on Improved LSTM-AE Algorithm[J]. Computer Engineering and Applications, 2023, 59(7): 311-318.

孙庆港, 王呈. 改进LSTM-AE算法的电梯知识库故障征兆预测[J]. 计算机工程与应用, 2023, 59(7): 311-318.

References

[1] ABDEL N，MOHAMED，KARAR M.Accurate photovoltaic power forecasting models using deep LSTM-RNN[J].Neural Computing and Applications，2019，31（7）：2727-2740.
[2] GU S，KELLY B，XIU D.Autoencoder asset pricing models[J].Journal of Econometrics，2021，222（1）：429-450.
[3] DAI A M，QUOC V.Semi-supervised sequence learning[C]//Advances in Neural Information Processing Systems，2015：3079-3087.
[4] HOU B，YANG J Y.LSTM-based auto-encoder model for ECG arrhythmias classification[J].IEEE Transactions on Instrumentation and Measurement，2019，69（4）：1232-1240.
[5] YU W，KIM Y，MECHEFSKE C.An improved similarity-based prognostic algorithm for RUL estimation using an RNN autoencoder scheme[J].Reliability Engineering & System Safety，2020，199：106926.
[6] 滕建，滕飞，李天瑞.基于3D卷积和 LSTM 编码解码的出行需求预测[J].计算机科学，2021，48（12）：195-203.
TENG J，TENG F，LI T R.Travel demand prediction based on 3D convolution and LSTM encoding and decoding[J].Computer Science，2021，48（12）：195-203.
[7] SUN Y Q，LI J，LIU J，et al.Using causal discovery for feature selection in multivariate numerical time series[J].Machine Learning，2015，101（1）：377-395.
[8] BACCIU D.Unsupervised feature selection for sensor time-series in pervasive computing applications[J].Neural Computing and Applications，2016，27（5）：1077-1091.
[9] 张坤华，丁立新，万润泽.基于相关性密度的多变量时间序列属性选择[J].计算机应用与软件，2017，34（12）：273-277.
ZHANG K H，DING L X，WAN R Z.Attribute selection of multivariate time series based on correlation density[J].Computer Applications and Software，2017，34（12）：273-277.
[10] YANG H，CHEN Y P.Hybrid deep learning and empirical mode decomposition model for time series applications[J].Expert Systems with Applications，2019，120：128-138.
[11] ANSELL P J，MULLENERS K.Multiscale vortex characteristics of dynamic stall from empirical mode decomposition[J].AIAA Journal，2020，58（2）：600-617.
[12] CHENG Y，WANG Z，CHEN B，et al.An improved complementary ensemble empirical mode decomposition with adaptive noise and its application to rolling element bearing fault diagnosis[J].ISA Transactions，2019，91：218-234.
[13] WITT B，ROHE D P.Digital image correlation as an experimental modal analysis capability[J].Experimental Techniques，2021，45（3）：273-286.
[14] ZHANG Y G，CHEN G，PAN G，et al.A novel hybrid model based on VMD-WT and PCA-BP-RBF neural network for short-term wind speed forecasting[J].Energy Conversion and Management，2019，195：180-197.
[15] HUANG Q Y，XIE L F，YIN G F，et al.Acoustic signal analysis for detecting defects inside an arc magnet using a combination of variational mode decomposition and beetle antennae search[J].ISA Transactions，2020，102：347-364.
[16] KHARE S K，BAJAJ V.An evolutionary optimized variational mode decomposition for emotion recognition[J].IEEE Sensors Journal，2020，21（2）：2035-2042.
[17] WANG Y，CHEN Q，HONG T，et al.Review of smart meter data analytics：applications，methodologies，and challenges[J].IEEE Transactions on Smart Grid，2018，10（3）：3125-3148.
[18] CHEN X，YANG Y，CUI Z，et al.Vibration fault diagnosis of wind turbines based on variational mode decomposition and energy entropy[J].Energy，2019，174：1100-1109.
[19] LONG D，NIU C，ZHOU H L，et al.Application of VMD algorithm in time-frequency analysis of seismic data[J].Progress in Geophysics，2020，35（1）：166-173.
[20] SAGHEER A，KOTB M.Unsupervised pre-training of a deep LSTM-based stacked autoencoder for multivariate time series forecasting problems[J].Scientific Reports，2019，9（1）：1-16.
[21] ALHNAITY B.An autoencoder wavelet based deep neural network with attention mechanism for multi-step prediction of plant growth[J].Information Sciences，2021，560：35-50.
[22] XU B，LIU J，HOU X，et al.Attention by selection：a deep selective attention approach to breast cancer classification[J].IEEE Transactions on Medical Imaging，2019，39（7）：1930-1941.
[23] ZHAO Z，BAO Z，ZHANG Z，et al.Automatic assessment of depression from speech via a hierarchical attention transfer network and attention autoencoders[J].IEEE Journal of Selected Topics in Signal Processing，2019，14（2）：423-434.
[24] 陆莉霞，邹俊忠，郭玉成，等.多模态融合的膝关节损伤预测[J].计算机工程与应用，2021，57（9）：225-232.
LU L X，ZOU J Z，GUO Y C，et al.Prediction of knee joint injury based on multimodal fusion[J].Computer Engineering and Applications，2021，57（9）：225-232.
[25] 陈彬，张荣梅，张琦.DCFM：基于深度学习的混合推荐模型[J].计算机工程与应用，2021，57（3）：150-155.
CHEN B，ZHANG R M，ZHANG Q.DCFM：hybrid recommendation model based on deep learning[J].Computer Engineering and Applications，2021，57（3）：150-155.