An artificial neural network based model to predict spatial soil type distribution using piezocone penetration test data (CPTu)

Abstract

Soil types mapping and the spatial variation of soil classes are essential concerns in both geotechnical and geoenvironmental engineering. Because conventional soil mapping systems are time-consuming and costly, alternative quick and cheap but accurate methods need to be developed. In this paper, a new optimized multi-output generalized feed forward neural network (GFNN) structure using 58 piezocone penetration test points (CPTu) for producing a digital soil types map in the southwest of Sweden is developed. The introduced GFNN architecture is supported by a generalized shunting neuron (GSN) model computing unit to increase the capability of nonlinear boundaries of classified patterns. The comparison conducted between known soil type classification charts, CPTu interpreting procedures, and the outcomes of the GFNN model indicates acceptable accuracy in estimating complex soil types. The results show that the predictability of the GFNN system offers a valuable tool for the purpose of soil type pattern classifications and providing soil profiles.

Soil color analysis based on a RGB camera and an artificial neural network towards smart irrigation: A pilot study

“…To validate the proposed system introduced in this study, a comparison was accompanied with results presented in [ 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 ], where ANN and Convolutional Neural Network (CNN) was adopted to identify the soil types or soil properties as shown in Table 2 . The MSE/root MSE (RMSE), coefficient of determination, ANN inputs related to soil property such as Red, Blue, and Green colors, and ANN structure were considered for comparative purpose, where the values in these parameters were obtained from the computations in previous works and introduced in their results.…”

“… Reference No. of inputs to ANN ANN structure MSE/RMSE (Training) MSE/RMSE (Testing) R 2 [ 43 ]/2018 (ANN) 3 (R, G, B) 3:1:60 1 × 10 −4 1 × 10 −4# 0.99 [ 44 ]/2017 (ANN) 8 (R, G, B, NIR, FC, NDVI, EVI, VHI) 8:14:10 0.05 ---- 0.94 [ 45 ]/2019 (ANN) 3 (R, G, B) 3:1:60 6.55 × 10 −4 5.25 × 10 −4# 0.99 [ 46 ]/2019 (CNN) 3 (R, G, B) Several layers --- 3.27∗ 0.96 [ 47 ]/2020 (CNN) 6 (soil property) Several layers --- 4.8∗ 0.86 [ 48 ]/2019 (CNN) Soil spectral data Several layers ---- 7.55∗ 0.7 [ 49 ]/2017 (ANN) 4 4:8:6:14 0.181∗ 0.163∗ 0.93 [ 50 ]/2020 (ANN) 5 (Color, Gravel, Sand, Silt Clay) 5:1:10 0.041 # 0.045 # 0.99 ANFIS (gebellmf) 3 (R, G, B) No. of mfs (3 3 3) 3.388 × 10 −3 3.378 × 10 −3 0.94 ANFIS (gebellmf) 3 (R, G, B) No.…”

An Automatic Classification Method of Well Testing Plot Based on Convolutional Neural Network (CNN)

“…The overfitting is a common problem in the training process of neural network models and it greatly reduces the generalization ability of neural network models [33]. The main reasons for the overfitting problem are insufficient training samples and a complex structure of networks.…”

lp Norm Smooth Inversion of Magnetic Anomaly Based on Improved Adaptive Differential Evolution

“…DE [11] is a population-based metaheuristic global optimization algorithm. Because of its fast convergence and easy implementation, it has been widely used to solve the optimization of practical problems, such as electromagnetic optimization [12,13], pattern recognition [14], signal processing [15][16][17][18][19][20], engineering application [21][22][23], and other inversion problems [24][25][26][27]. Considering the dependency of control parameters [28], some scholars tried to adjust them by using adaptive or self-adaptive manners.…”