Modeling and prediction of geothermal reservoir temperature behavior using evolutionary design of neural networks

Porkhial, Soheil; Salehpour, M.; Ashraf, Hafiz Muhammad; Jamali, Ali

doi:10.1016/j.geothermics.2014.07.003

Cited by 45 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In an attempt to address the issue of model bias faced by ANN, group method of data handling (GMDH), automatically synthesize network from the database of inputs and outputs. This process which is called self-organization of input models removes the user from specifying the network architecture in advance and hence removes biases in the model [22][23][24]. Reduced computational time is another advantage of the GMDH method as compared to the standard ANN.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Permeability Using Group Method of Data Handling (GMDH) Neural Network from Well Log Data

et al. 2020

View full text Add to dashboard Cite

Permeability is an important petrophysical parameter that controls the fluid flow within the reservoir. Estimating permeability presents several challenges due to the conventional approach of core analysis or well testing, which are expensive and time-consuming. On the contrary, artificial intelligence has been adopted in recent years in predicting reliable permeability data. Despite its shortcomings of overfitting and low convergence speed, artificial neural network (ANN) has been the widely used artificial intelligent method. Based on this, the present study conducted permeability prediction using the group method of data handling (GMDH) neural network from well log data of the West arm of the East African Rift Valley. Comparative analysis of GMDH permeability model and ANN methods of the back propagation neural network (BPNN) and radial basis function neural network (RBFNN) were further explored. The results of the study showed that the proposed GMDH model outperformed BPNN and RBFNN as it achieved R/root mean square error (RMSE) value of 0.989/0.0241 for training and 0.868/0.204 for predicting, respectively. Sensitivity analysis carried out revealed that shale volume, standard resolution formation density, and thermal neutron porosity were the most influential well log parameters when developing the GMDH permeability model.

show abstract

Section: Introductionmentioning

confidence: 99%

Prediction of Permeability Using Group Method of Data Handling (GMDH) Neural Network from Well Log Data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…Fuzzy control is an approach that allows the definition of intermediate values between clear evaluations such as “true or false,” and it is widely applied in a variety of engineering sciences [ 14 ]. Although the application of artificial intelligence methods for different purposes in geothermal systems has been started in recent years, methods such as artificial neural networks, machine learning can enable optimization in systems by using mostly historical data [ 9 , [18] , [19] , [20] ]. The fuzzy control is a more suitable method for decision mechanisms that will rapidly optimize the instantaneous changes in production and reinjection wells under dynamic geothermal reservoir conditions.…”

Section: Introductionmentioning

confidence: 99%

The importance of long-term well management in geothermal power systems using fuzzy control: A Western Anatolia (Turkey) case study

Haklidir

2020

Energy

View full text Add to dashboard Cite

Effective geothermal power generation depends on two main elements: geothermal reservoir management and maintenance of the power plant. Reservoir management consists of both the fluid production and reinjection of brine to the underground. The management of wells is important to ensure the sustainability of the reservoir. Thus, the flow rate control systems are essential to protect geothermal reservoirs under long-term power production. The second issue is the daily change in electricity prices and the load change process is complex because geothermal well controls are not flexible operations. The well management thus requires control approaches, and fuzzy control can be one effective solution. In this study, a fuzzy control system has been developed to control flow rates of the wells in Kızıldere geothermal field and its performance has been compared with the real data taken from the Kızıldere Power Plant. The results of comparison show that the fuzzy controllers achieved the target energy production in 2 h instead of 5 h, compared to the real data. Based on the real data, the reinjection was only able to stabilize at the end of the fourth hour and the process took only 2 h when using the fuzzy controllers.

show abstract

“…It has been shown that evolutionary methods can be notably used for optimal design of fuzzy models and neural network systems. 22,30 Besides, many efforts have been accomplished for the optimal design of variables in the premise part of TSK-type fuzzy rules by GAs. 31,32 In addition, optimal parts of antecedents and consequents of ANFIS by both GA and SVD methods have been lately exploited.…”

Section: Introductionmentioning

confidence: 99%

Multi-objective optimum design of ANFIS for modelling and prediction of deformation of thin plates subjected to hydrodynamic impact loading

Jamali

Babaei

Nariman-Zadeh

et al. 2016

Proceedings of the Institution of Mechanical Engineers, Part L:

View full text Add to dashboard Cite

Drop hammer impact experiments have been carried out to assess the dynamic plastic response of fully clamped circular and rectangular plates made of aluminum and steel subjected to hydrodynamic impact loading at various energy levels. Also, the effective parameters in forming process are proposed in non-dimensional forms for modeling and prediction of the central deflection of plates using adaptive neuro-fuzzy inference system in conjunction with genetic algorithm and singular value decomposition method. Genetic algorithm is used for optimal scheme of Gaussian membership function’s variables and multi-objective Pareto optimal design of adaptive neuro-fuzzy inference system model. Also, the singular value decomposition method is applied to compute the linear parameters of the adaptive neuro-fuzzy inference system method. The important conflicting objectives of developed adaptive neuro-fuzzy inference system, namely, training error and prediction error, are obtained by dividing date sets into two parts. Hence, various optimal choices of adaptive neuro-fuzzy inference system model are provided which are non-dominated states from each other. Moreover, optimal Pareto front of such model leads to trade-off between the conflicting pair of considered objectives for two series of experiments. The results of this work indicate that multi-objective Pareto optimal design of adaptive neuro-fuzzy inference system predicts central deflection of plates with a good accuracy. In addition, the comparison between the adaptive neuro-fuzzy inference system model and exiting one demonstrates superior performance of the present approach in simulating central deflection of plates.

show abstract

Modeling and prediction of geothermal reservoir temperature behavior using evolutionary design of neural networks

Cited by 45 publications

References 26 publications

Prediction of Permeability Using Group Method of Data Handling (GMDH) Neural Network from Well Log Data

Prediction of Permeability Using Group Method of Data Handling (GMDH) Neural Network from Well Log Data

The importance of long-term well management in geothermal power systems using fuzzy control: A Western Anatolia (Turkey) case study

Multi-objective optimum design of ANFIS for modelling and prediction of deformation of thin plates subjected to hydrodynamic impact loading

Contact Info

Product

Resources

About