Feature selection based on sparse representation with the measures of classification error rate and complexity of boundary

Deng, Yanli; Jin, Weidong

doi:10.1016/j.ijleo.2015.06.057

Cited by 5 publications

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, it has been widely applied in all kinds of classification and prediction, feature selection and anomaly detection problem [30]. For regression problem, random forest can select feature subset according to mean model fitting error [31]. Since random forest has good feature extraction ability and high precision, this paper adopts random forest to select and extract the key influencing factors of the shield tunneling axis attitude deviation combined recursive feature elimination, the specific process is shown in FIGURE 2.…”

Section: The Key Influencing Factor Analysis About Shield Tunnelmentioning

confidence: 99%

Prediction of Axis Attitude Deviation and Deviation Correction Method Based on Data Driven During Shield Tunneling

et al. 2019

View full text Add to dashboard Cite

Due to the complex shield construction characteristics and the complex effects of geological environment, it is difficult to control the direction of shield tunneling and to determine the reasonable tunneling parameters such as tunneling speed and so on. During the tunneling, shield tunneling machine may rise, shift and snake advance, which are not conducive to control tunnel axis. Aiming at the problem that it is difficult to accurately predict and correct the axis attitude deviation in the shield tunneling process, a prediction of axis attitude deviation and deviation correction method based on data driven during shield tunneling was put forward in this paper. Under certain geological conditions, the relationship between the attitude deviation of the construction axis and tunneling parameters during shield tunneling process with different tunneling mileages is established. Based on the tunneling historical data, the XGBoost prediction model is constructed, and the axis deviation variation is predicted and analyzed with the shield construction parameters. The multi-ring deviation correction parameter calculation model based on the fusion of geometric model and association rules is designed to obtain the internal correlation of the deviation amount, the number of deviation correction rings and the deviation correction parameters of each ring under the maximum deviation of different deviation correction sections, so as to realize the accurate prediction and deviation correction of shield axis deviation in the complex construction process. Under the verification of 155 ring data in a certain subway construction section, the method proposed in this paper has higher prediction accuracy, which is important for improving the safety and quality of shield tunneling. Results from the measured shield construction ring data verify the reasonability of the proposed axis attitude deviation prediction and multi-ring deviation correction method.

show abstract

Section: The Key Influencing Factor Analysis About Shield Tunnelmentioning

confidence: 99%