Data in intelligent approach for estimation of disc cutter life using hybrid metaheuristic algorithm

This data in brief presents the monitoring data measured during shield tunnelling of Guangzhou–Shenzhen intercity railway project. The monitoring data includes shield operational parameters, geological conditions, and geometry at the site. The presented data were arbitrarily split into two subsets including the training and testing datasets. The field observations are compared to the forecasting values of the disc cutter life assessed using a hybrid metaheuristic algorithm proposed for “Prediction of disc cutter life during shield tunnelling with artificial intelligent via incorporation of genetic algorithm into GMDH-type neural network” [1]. The presented data can provide a guidance for cutter exchange in shield tunnelling.

other scholars who f

us on designing an
modelling the disc cutter for practical tunnelling applications.Using this data, researchers can assess the behavior of disc cutter during tunnelling especially in rock-soil strata.• Helpful insights can be gleaned from this data.According to this data, other data can be done that would lead to suitable survey studies.


Data Description

The database in this article includes the shield operational parameters, geological conditions, and soil geometry.The operational parameters are initially extracted directly from a built-in data acquisition system in the tunnel boring machine.In this paper, geo

gical as-built ma
s and geological engineering and geotechnical reports from boreholes ( Fig. 1 ) and surface outcrops were considered as sources of geotechnical information for the database [2] .Table 1 lists the data source of the utilized parameters in this paper.The observed data includes operational parameters and geological conditions such as thrust force (TF), cutter rotation speed (RPM), penetration rate (PR), screw rate (SC), grouting pressure (GP), soil pressure (SP), and specific energy (SE), quartz content (Q c ), excavation depth (H), and disc cutter life (H f ).The overburden layers of the tunnel were backfill with a thickness about 6 m, clay soil with a thickness of 3.1-8.25 m, silt clay soil with a thickness reaches to 8.5 m.

The Guangzhou-Shenzhen intercity railway project is one of the greatest infrastructure projects in recent years (see Fig. 2 ) [3] .This project is located on the coast of the Pearl River Delta of Guangdong, China.The overall length of the project is about 116 km long and includes tunn ls of a total length of 22 km.The construction project connects Guangzhou North Station


Table 1

Descriptive statistics for average values of the generated database.

No. H f (m 3 /cutter) TF (kN) RPM (rev/min) SR (rev/min) PR (mm/rev) GP (kPa) SP (kPa) SE (kWh/m 3 ) Qc (%) H (m)   and Bao'an International Airport, Shenzhen.The tunnel section is located in the zone of airport terminal 3, between Bao'an Airport North Station and Bao'an Airport Station.To construct the tunnel, an earth pressure balance shield machine is used [4][5][6] .The cutterhead shield machine is 8.85 m in diameter, and the trailing shield is 8.50 m in diameter, thereby leading to an over-cut annulus of 35.0 mm.The specifications of the earth pressure balance shield machine are listed in the original publication [1] .The main geological formation that encountered during tunnelling are silt clay and weathered rock.Table 2 lists the statistical analyses of the proposed model using different settings.


Experimental Design, Materials and Methods

To predict the disc cutter life, an intelligent model-a

ybrid approach based on integrating group m
thod of data handling (GMDH) with genetic algorithm (GA) was adopted [1] .This model is different from both traditional numerical models [7] and other artificial intelligence models, e.g. the gated recurrent unit (GRU) [6] , evolutionary neural network [8][9][10][11][12] , and long short-term memory (LSTM) [13] .The observed data in this article can be used to identify the applicability of the proposed model in which the collected data are divided into training set (22 variable) and testing set (10 variable).Table 2 lists the statistical results of the proposed GMDH-GA model.Regarding the application of the proposed GMDH-GA, different architectures were tested ( Table 2 ), and an increase in the correlation coefficient (R 2 ) and a decrease in the root mean square error (RMSE) was observed when the number of population of individuals and number of generations were increased.The best estimates were obtained using a network with double hidden GMDH layers.Hence, using the observation of the model number (E) in Table 2 , the disc cutter life can be predicted in an appropriate manner.


Declaration of Competing Interest

The authors declare that they have no known competing financial intere

s or personal relationships that c
uld have appeared to influence the work reported in this paper.

Fig. 1 .
1
Fig. 1.An example of the geological borehole extracted from construction site; (a) Lot ET-3; (b