Optimization of DRASTIC method by supervised committee machine artificial intelligence to assess groundwater vulnerability for Maragheh–Bonab plain aquifer, Iran

Fijani, Elham; Nadiri, Ata Allah; Moghaddam, Asghar Asghari; Tsai, Frank T.-C.; Dixon, Barnali M.

doi:10.1016/j.jhydrol.2013.08.038

Cited by 147 publications

(47 citation statements)

References 48 publications

(23 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The training/testing data split can have a significant impact on the results of the models. The cross-validation technique (Chang et al 2013;Fijani et al 2013) was used in this study to divide the data sets. Based on this approach, 93 data points were divided in two sets; training and testing.…”

Section: Data Collection and Preparationmentioning

confidence: 99%

Combining the advantages of neural networks using the concept of committee machine in the groundwater salinity prediction

Barzegar

Moghaddam

2016

Model. Earth Syst. Environ.

Self Cite

View full text Add to dashboard Cite

Salinity is one of the main factors in groundwater quality monitoring. The main objectives of this study are to investigate and compare the accuracy of three different neural computing techniques, multi-layer perceptron neural network (MLP), radial basis function neural network (RBFNN), and generalized regression neural network (GRNN), in prediction of groundwater salinity of the Tabriz plain confined aquifer, expressed by electrical conductivity [EC (lS/cm)], and to employ an integrated method to combine the advantages of neural network models utilizing the concept of committee machine. To develop the models, 93 data records of groundwater samples were collected from East Azarbaijan regional water company. The data set including Ca 2? , Mg 2? , Na ? , SO 4 2-and Cl -concentrations as the inputs and salinity [EC (lS/ cm)] as an output were divided into two subsets; training and testing based on cross validation approach. After training and testing of the models, the performance of the models were evaluated using root mean square errors (RMSE), determination coefficient (R 2 ) and mean absolute error (MAE). The performance criteria of the constructed neural network models showed that RBFNN model has the best performance in predicting salinity. The committee neural network (CNN) combined the results of salinity predicted from MLP, RBFNN and GRNN, each of them has a weight factor showing its contribution in overall prediction. The optimal weights were derived by a genetic algorithm (GA). The results of salinity prediction derived from CNN showed that the CNN performs better than any one of the individual ANNs acting alone for predicting groundwater salinity.

show abstract

Section: Data Collection and Preparationmentioning

confidence: 99%

Combining the advantages of neural networks using the concept of committee machine in the groundwater salinity prediction

Barzegar

Moghaddam

2016

Model. Earth Syst. Environ.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The cross-validation technique (Chang et al 2013;Fijani et al 2013) is used in this research to divide the data sets. Based on this approach, 243 data points are divided in two sets; training and validation.…”

Section: Data Preparationmentioning

confidence: 99%

“…The same clusters of input and outputs and rules are used for NF construction. Hybrid algorithm which is the combination of the least-squares method and the back propagation gradient descent method is applied to optimize and adjust the Gaussian membership function parameters and coefficients of the output linear equations (ZounematKermani and Teshnehlab 2008;Fijani et al 2013). The number of fuzzy rules is 3.…”

Section: Neuro Fuzzy (Nf)mentioning

confidence: 99%

“…Incorporating the fuzzy rules and neural network with DRASTIC variables has shown improved vulnerability prediction for pesticides (Rahman 2008). Fijani et al (2013) used Supervised Committee Machine Artificial Intelligence to optimize DRASTIC in Maragheh-Bonab plain aquifer. They concluded that combination use of artificial intelligent models, as the Supervised Committee Machine Artificial Intelligence model, can potentially provide better assessment of the groundwater vulnerability by combining the advantages of individual artificial intelligent models.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A supervised committee machine artificial intelligent for improving DRASTIC method to assess groundwater contamination risk: a case study from Tabriz plain aquifer, Iran

Barzegar

Moghaddam

Baghban

2015

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

Vulnerability maps are designed to show areas of greatest potential for groundwater contamination on the basis of hydrogeological conditions and human impacts. The objective of this research is (1) to assess the groundwater vulnerability using DRASTIC method and (2) to improve the DRASTIC method for evaluation of groundwater contamination risk using AI methods, such as ANN, SFL, MFL, NF and SCMAI approaches. This optimization method is illustrated using a case study. For this purpose, DRASTIC model is developed using seven parameters. For validating the contamination risk assessment, a total of 243 groundwater samples were collected from different aquifer types of the study area to analyze NO À 3 concentration. To develop AI and CMAI models, 243 data points are divided in two sets; training and validation based on cross validation approach. The calculated vulnerability indices from the DRASTIC method are corrected by the NO À 3 data used in the training step. The input data of the AI models include seven parameters of DRASTIC method. However, the output is the corrected vulnerability index using NO À 3 concentration data from the study area, which is called groundwater contamination risk. In other words, there is some target value (known output) which is estimated by some formula from DRASTIC vulnerability and NO À 3 concentration values. After model training, the AI models are verified by the second NO À 3 concentration dataset. The results revealed that NF and SFL produced acceptable performance while ANN and MFL had poor prediction. A supervised committee machine artificial intelligent (SCMAI), which combines the results of individual AI models using a supervised artificial neural network, was developed for better prediction of vulnerability. The performance of SCMAI was also compared to those of the simple averaging and weighted averaging committee machine intelligent (CMI) methods. As a result, the SCMAI model produced reliable estimates of groundwater contamination risk.

show abstract

“…Furthermore, the vulnerability of the intrinsic aquifer was evaluated using a modified DRASTIC model, and the groundwater value was evaluated based on its quality and aquifer storage [9]. A supervised committee machine with artificial intelligence (SCMAI) model was introduced to improve the DRASTIC method for the groundwater vulnerability assessment of the Maragheh-Bonab plain aquifer in Iran [10]. Finally, an intrinsic vulnerability assessment was verified using a total organic carbon (TOC) concentration in groundwater, which was a novel approach for the validation of groundwater vulnerability methods on a regional scale [11].…”

Section: Introductionmentioning

confidence: 99%

A Vulnerability Evaluation of the Phreatic Water in the Plain Area of the Junggar Basin, Xinjiang Based on the VDEAL Model

Jia

Zhou

Zhou³

et al. 2014

Sustainability

View full text Add to dashboard Cite

A VDEAL (V is the lithology of the vadose zone, D is the groundwater depth, E is the degree of groundwater exploitation, A is the aquifer characteristics and L is the land use pattern.) model, which is suitable for a vulnerability evaluation of the groundwater in arid inland areas, and that is based on the GOD (G is the groundwater status, O is overburden feature and D is groundwater depth) method and DRASTIC (D is the depth of water-table, R is the net recharge, A is the aquifer media, S is the soil media, T is the topography, I is the impact of the vadose and C is the conductivity of the aquifer.) model is proposed in this paper. Five indicators were selected by reference to the DRAV (D is the depth of water-table, R is the net recharge, A is the aquifer media and V is the impact of the vadose.) and VLDA (V is the lithology of the vadose zone , L is the land use pattern, D is the groundwater depth and A is the aquifer characteristics and.) models, namely, the OPEN ACCESS Sustainability 2014, 6 8605 lithology of the vadose zone (V), the groundwater depth (D), the degree of groundwater exploitation (E), the aquifer characteristics (A) and the land use pattern (L). According to monitoring data from 2003 and 2011, the variations of phreatic water quality in the plain area of the Junggar Basin were divided into three types: the water quality may have deteriorated, be unchanged or improved. Four groups of indicator weights were configured to calculate the vulnerability index using the VDEAL model. The changes of phreatic water quality were then compared against the vulnerability index. The normalized weights of V, D, E, A, and L were respectively 0.15, 0.25, 0.10, 0.10, and 0.40; this is according to the principle that the sampling sites of deteriorated water quality are generally distributed in a high-vulnerability region, and the sites of unchanged and improved water quality are distributed in middle vulnerability, low vulnerability and invulnerable regions. The evaluation results of phreatic water vulnerability in the plain area of the Junggar Basin based on the VDEAL model are as follows. The regions with vulnerability indexes of 2.0-4.0, 4.0-6.0, 6.0-8.0, and >8.0, respectively account for 2.2%, 61.0%, 35.9%, and 0.9% of the region. The regions with a higher vulnerability are mainly distributed in the farmlands and the sand and gravel regions with a phreatic water depth of <3 m. Moreover, the regions with a lower vulnerability are generally located in the non-irrigation regions with a sandy loam or silty fine sand and a phreatic water depth of >6 m. The phreatic water in these regions is deficient with regard to the infiltration of irrigation water and the recharge from precipitation.

show abstract

Optimization of DRASTIC method by supervised committee machine artificial intelligence to assess groundwater vulnerability for Maragheh–Bonab plain aquifer, Iran

Cited by 147 publications

References 48 publications

Combining the advantages of neural networks using the concept of committee machine in the groundwater salinity prediction

Combining the advantages of neural networks using the concept of committee machine in the groundwater salinity prediction

A supervised committee machine artificial intelligent for improving DRASTIC method to assess groundwater contamination risk: a case study from Tabriz plain aquifer, Iran

A Vulnerability Evaluation of the Phreatic Water in the Plain Area of the Junggar Basin, Xinjiang Based on the VDEAL Model

Contact Info

Product

Resources

About