Processing Method of Missing Data in Dam Safety Monitoring

Wei Wei,Chongshi Gu,Xiao Fu,Tianwei Zhang

doi:10.1155/2021/9950874

Abstract

A large amount of data obtained by dam safety monitoring provides the basis to evaluate the dam operation state. Due to the interference caused by equipment failure and human error, it is common or even inevitable to suffer the loss of measurement data. Most of the traditional data processing methods for dam monitoring ignore the actual correlation between different measurement points, which brings difficulties to the objective diagnosis of dam safety and even leads to misdiagnosis. Therefore, it is necessary to conduct further study on how to process the missing data in dam safety monitoring. In this study, a data processing method based on partial distance combining fuzzy C-means with long short-term memory (PDS-FCM-LSTM) was proposed to deal with the data missing from dam monitoring. Based on the fuzzy clustering performed for the measurement points of the same category deployed on the dam, the membership degree of each measurement point to cluster center was described by using the fuzzy C-means clustering algorithm based on partial distance (PDS-FCM), so as to determine the clustering results and preprocess the missing data of corresponding measurement points. Then, the bidirectional long short-term memory (LSTM) network was applied to explore the pattern of changes of measurement values under identical clustering conditions, thus processing the data missing from monitoring effectively.

Highlights

As an indicator of the safety and work performance of concrete dam, the original measurement data always attract much attention for its integrity and accuracy [1,2,3]
Proposed by Takagi and Sugeno, the TS nonlinear regression model [17,18,19] is capable of describing nonlinear problems linearly by constructing local linear models and associating these linear models with membership function. This model ignores the correlation between input and output, which leads to the low accuracy in processing sequence information. rough the combination between autocoding and the genetic algorithm, Abdella et al put forward a processing method which takes the missing data as the independent variable for cost function, and the genetic algorithm was applied to optimize the cost function of the missing data to be solved [20, 21]
Based on the aforementioned method, Nelwamondo et al introduced the dynamic programming theory to build multiple self-encoders and selected the optimal model for each incomplete sample to process the missing data [22]. Such methods ignore the correlation between different sample attributes, which causes the accuracy of missing data processing to be low. e newly proposed autoregressive integrated moving average model (ARIMA) hybrid method shows the advantages of both the autoregressive integrated moving average and artificial neural network, which improves its generality

Summary

Introduction

As an indicator of the safety and work performance of concrete dam, the original measurement data always attract much attention for its integrity and accuracy [1,2,3]. Proposed by Takagi and Sugeno, the TS nonlinear regression model [17,18,19] is capable of describing nonlinear problems linearly by constructing local linear models and associating these linear models with membership function This model ignores the correlation between input and output, which leads to the low accuracy in processing sequence information. Based on the aforementioned method, Nelwamondo et al introduced the dynamic programming theory to build multiple self-encoders and selected the optimal model for each incomplete sample to process the missing data [22] Such methods ignore the correlation between different sample attributes, which causes the accuracy of missing data processing to be low. The bidirectional LSTM network is introduced to further process the long-sequence missing data, ensuring the accurate processing of missing data. e basic principle is detailed as follows

Processing Method of Dam Missing Data

Preprocessing Method of Missing Data

Numerical Example

TCN1 TCN2

Conclusion