Hybrid Neural Genetic Algorithm Research Articles

Predictive capability evaluation and mechanism of Ce (III) extraction using solvent extraction with Cyanex 572

Owing to the high toxicity of cerium toward living organisms, it is necessary to remove cerium from aqueous solutions. In this regard, the extraction of cerium (Ce (III)) from nitrate media by Cyanex 572 under different operating conditions was examined in this study. The effect of contact time, pH, extractant concentration, and nitrate ion concentration were investigated to characterize the extraction behavior of cerium and based on these outcomes, an extraction mechanism was suggested. The analysis of infrared spectra of Cyanex 572 before and after the extraction of cerium indicated that cerium extraction was performed via a cation-exchange mechanism. Then, the predictive models based on intelligent techniques [artificial neural network (ANN) and hybrid neural-genetic algorithm (GA-ANN)] were developed to predict the cerium extraction efficiency. The GA-ANN model provided better predictions that resulted higher R2 and lower MSE compared to ANN model for predicting the extraction efficiency of cerium by Cyanex 572. The interactive effects of each process variable on cerium extraction were also investigated systematically. pH was the most influential parameter on cerium extraction, followed by extractant concentration, nitrate ion concentration and contact time. Finally, the separation of cerium from other rare earth elements like La (III), Nd (III), Pr (III), and Y (III) was conducted and observed that the present system provides a better separation of cerium from rare heavy earth than light rare earths.

Selectivity index and separation efficiency prediction in industrial magnetic separation process using a hybrid neural genetic algorithm

It is essential to know the process efficiency in the industrial magnetic separation process under different operating conditions because it is required to control the process parameters to optimize the process efficiency. To our knowledge, there is no information about using artificial intelligence for modeling the magnetic separation process. Hence, finding a robust and more accurate estimation method for predicting the separation efficiency and selectivity index is still necessary. In this regard, a feed-forward neural network was developed to predict the separation efficiency and selectivity index. This model was trained to present a predictive model based on the percentage of iron, iron oxide and sulfur in mill feed and cobber feed, 80% passing size in mill feed and cobber feed and plant capacity. Therefore, this work aims to develop an intelligent technique based on an artificial neural network and a hybrid neural-genetic algorithm for modeling the concentration process. Results indicated that the values of mean square error and coefficient of determination for the testing phase were obtained 0.635 and 0.86 for selectivity index and of 4.646 and 0.84 for separation efficiency, respectively. In order to improve the performance of neural network, genetic algorithm was used to optimize the weights and biases of neural network. The results of modeling with GA-ANN technique indicated that the mean square error and coefficient of determination for the testing phase were achieved by 0.276 and 0.95 for selectivity index and of 1.782 and 0.92 for separation efficiency, respectively. The other statistical criteria for the GA-ANN model were better than those of the ANN model.

The nickel ion removal prediction model from aqueous solutions using a hybrid neural genetic algorithm

Prediction of Ni(II) removal during ion flotation is necessary for increasing the process efficiency by suitable modeling and simulation. In this regard, a new predictive model based on the hybrid neural genetic algorithm (GANN) was developed to predict the Ni(II) ion removal and water removal during the process from aqueous solutions using ion flotation. A multi-layer GANN model was trained to develop a predictive model based on the important effective variables on the Ni(II) ion flotation. The input variables of the model were pH, collector concentration, frother concentration, impeller speed and flotation time, while the removal percentage of Ni(II) ions and water during ion flotation were the outputs. The most effective input variables on Ni(II) removal and water removal were evaluated using the sensitivity analysis. The sensitivity analysis of the model shows that all input variables have a significant impact on the outputs. The results show that the proposed GANN models can be used to predict the Ni(II) removal and water removal during ion flotation.

Recovery prediction of copper oxide ore column leaching by hybrid neural genetic algorithm

The artificial neural network (ANN) and hybrid of artificial neural network and genetic algorithm (GANN) were applied to predict the optimized conditions of column leaching of copper oxide ore with relations of input and output data. The leaching experiments were performed in three columns with the heights of 2, 4 and 6 m and in particle size of <25.4 and <50.8 mm. The effects of different operating parameters such as column height, particle size, acid flow rate and leaching time were studied to optimize the conditions to achieve the maximum recovery of copper using column leaching in pilot scale. It was found that the recovery increased with increasing the acid flow rate and leaching time and decreasing particle size and column height. The efficiency of GANN and ANN algorithms was compared with each other. The results showed that GANN is more efficient than ANN in predicting copper recovery. The proposed model can be used to predict the Cu recovery with a reasonable error.

Modeling and optimization of polymer enhanced ultrafiltration using hybrid neural-genetic algorithm based evolutionary approach

A stochastic genetic algorithm (GA) based strategy along with artificial neural network (ANN) was applied to optimize the retention of reactive red 120 (RR 120) dye from its aqueous solutions by way of polymer (polyethyleneimine (PEI)) enhanced ultrafiltration (PEUF). The optimal feed forward back propagation ANN (4-10-1) model network, trained initially through Levenberg–Marquardt (LM) algorithm, was suitably manoeuvred by the GA approach to predict the membrane performance index (PFI) response, evaluated as the product of dye rejection and permeation flux, for a randomly generated population of chromosomes. Each chromosome was constituted by four principal genes, namely, cross-flow rate, transmembrane pressure, polymer to dye ratio, and pH. The local exploitation capacity of the canonical GA was enhanced further by combining hill-climbing (HC) local search with the optimization levels of standard GA. The near-optimal and economically feasible factor levels were predicted by the hybrid ANN-GA-HC strategy, keeping PFI maximization and the constrained PEUF process dynamics in perspective; the optimal process factor settings experimentally yielded a pragmatic PFI of 143.8L/m2h, corresponding to high (99.9%) dye rejection, and a satisfactory permeation flux (144L/m2h).

A novel Grouping Genetic Algorithm–Extreme Learning Machine approach for global solar radiation prediction from numerical weather models inputs

This paper presents a novel scheme for global solar radiation prediction, based on a hybrid neural-genetic algorithm. Specifically a grouping genetic algorithm (GGA) and an Extreme Learning Machine algorithm (ELM) have been merged in a single algorithm, in such a way that the GGA solves the optimal selection of features, and the ELM carries out the prediction. The proposed scheme is also novel because it uses as input of the system the output of a numerical weather meso-scale model (WRF), i.e., atmospherical variables predicted by the WRF at different nodes. We consider then different problems associated with this general algorithmic framework: first, we evaluate the capacity of the GGA–ELM for carrying out a statistical downscaling of the WRF to a given point of interest (where a measure of solar radiation is available), i.e., we only take into account predictive variables from the WRF and the objective variable at the same time tag. In a second evaluation approach, we try to predict the solar radiation at the point of interest at different time tags t+x, using predictive variables from the WRF. Finally, we tackle the complete prediction problem by including previous values of measured solar radiation in the prediction. The proposed algorithm and its efficiency for selecting the best set of features from the WRF are analyzed in this paper, and we also describe different operators and dynamics for the GGA. Finally, we evaluate the performance of the system with these different characteristics in a real problem of solar radiation prediction at Toledo’s radiometric observatory (Spain), where the proposed system has shown an excellent performance in all the subproblems considered, in terms of different error metrics.

Aspherical lens design using hybrid neural-genetic algorithm of contact lenses.

The design of complex contact lenses involves numerous uncertain variables. How to help an optical designer to first design the optimal contact lens to reduce discomfort when wearing a pair of glasses is an essential design concern. This study examined the impact of aberrations on contact lenses to optimize a contact lens design for myopic and astigmatic eyes. In general, two aspherical surfaces can be assembled in an optical system to reduce the overall volume size. However, this design reduces the spherical aberration (SA) values at wide contact radii. The proposed optimization algorithm with optical design can be corrected to improve the SA value and, thus, reduce coma aberration (TCO) values and enhance the modulation transfer function (MTF). This means integrating a modified genetic algorithm (GA) with a neural network (NN) to optimize multiple-quality characteristics, namely the SA, TCO, and MTF, of contact lenses. When the proposed optional weight NN-GA is implemented, the weight values of the fitness function can be varied to adjust system performance. The method simplifies the selection of parameters in the optimization of optical systems. Compared with the traditional CODE V built-in optimal scheme, the proposed scheme is more flexible and intuitive to improve SA, TCO, and MTF values by 50.03%, 45.78%, and 24.7%, respectively.

Determination of dew point pressure in gas condensate reservoirs based on a hybrid neural genetic algorithm

Knowing dew point pressure considers as one of the preliminary requirements in retrograde gas condensate reservoir simulations. When the pressure declines below the dew point pressure, the condensate dropouts form, which could lead to a substantial decrease in gas relative permeability and well deliverability. Different methods such as equation of states, empirical correlations and experimental procedures have been proposed to determine the dew point pressure. However, due to their convergence problem, being expensive and time consuming, great efforts have been taken to develop an alternative method. In this study, a new method based on artificial neural network has been developed and optimized by genetic algorithm as an evolutionary technique. A data set consists of 308 sample collected from different sources and literature including one of Iranian gas-condensate field is used. Reservoir temperature, mole percentage of gas components and heavy fractions properties were considered as input parameters to this model. The performance of the proposed model was compared with some of the common correlations and Peng–Robinson equation of state. The results confirmed the accuracy and capability of this model in determination of dew point pressure based on 2.46%, 3.66%, 95.91%, 0.02% and 24.39% as average absolute deviation, root mean square error, correlation of determination, minimum and maximum percentage error; respectively. The sensitivity analysis is also performed on variables to determine the impact and importance of each parameter on prediction of dew point pressure. The results show that plus fraction properties and C3–C4 fraction have the greatest positive and negative impacts on estimation of dew point pressure; respectively.

A Hybrid Neural Network and Genetic Algorithm Based Model for Short Term Load Forecast

Aim of this research is to develop a hybrid prediction model based on Artificial Neural Network (ANN) and Genetic Algorithm (GA) that integrates the benefits of both techniques to increase the electrical load forecast accuracy. Precise Short Term Load Forecast (STLF) is of critical importance for the secure and reliable operation of power systems. ANNs are largely implemented in this domain due to their nonlinear mapping nature. The ANN architecture optimization, the initial weight values of the neurons, selection of training algorithm and critical analysis and selection of the most appropriate input parameters are some important consideration for STLF. Levenberg-Marquardt (LM) algorithm for the training of the neural network is implemented in the first stage. The second stage is based on a hybrid model which combines the ANN and GA.

Developing a Formula Based on a Hybrid Neural Genetic Algorithm for the Prediction of Minimum Miscibility Pressure

Minimum miscibility pressure is a fundamental parameter in miscible injection projects. Experimental determination of minimum miscibility pressure is very costly and time-consuming; therefore, attempts have been made to utilize artificial neural networks for determination of minimum miscibility pressure. Despite the wide range of applications and flexibility of artificial neural networks, design and structural optimization of neural networks is still strongly dependent upon the designer's experience. To mitigate this problem, this article presents a new approach based on a hybrid neural genetic algorithm to determine the minimum miscibility pressure for pure and impure carbon dioxide injections. Then, equations for minimum miscibility pressure prediction by using the optimize weights of network have been generated. With the formulas obtained, the user may use such results without running the artificial neural network software. The new model yielded the accurate prediction of the experimental slim-tube carbon dioxide minimum miscibility pressure with the lowest relative mean squared error and average absolute errors among all tested carbon dioxide minimum miscibility pressure correlations.

Recovery and grade prediction of pilot plant flotation column concentrate by a hybrid neural genetic algorithm

Today flotation column has become an acceptable means of froth flotation for a fairly broad range of applications, in particular the cleaning of sulfides. Even after having been used for several years in mineral processing plants, the full potential of the flotation column process is still not fully exploited. There is no prediction of process performance for the complete use of available control capabilities. The on-line estimation of grade usually requires a significant amount of work in maintenance and calibration of on-stream analyzers, in order to maintain good accuracy and high availability. These difficulties and the high cost of investment and maintenance of these devices have encouraged the approach of prediction of metal grade and recovery. In this paper, a new approach has been proposed for metallurgical performance prediction in flotation columns using Artificial Neural Network (ANN). Despite of the wide range of applications and flexibility of NNs, there is still no general framework or procedure through which the appropriate network for a specific task can be designed. Design and structural optimization of NNs is still strongly dependent upon the designer’s experience. To mitigate this problem, a new method for the auto-design of NNs was used, based on Genetic Algorithm (GA). The new proposed method was evaluated by a case study in pilot plant flotation column at Sarcheshmeh copper plant. The chemical reagents dosage, froth height, air, wash water flow rates, gas holdup, Cu grade in the rougher feed, flotation column feed, column tail and final concentrate streams were used to the simulation by GANN. In this work, multi-layer NNs with Back Propagation (BP) algorithm with 8-17-10-2 and 8-13-6-2 arrangements have been applied to predict the Cu and Mo grades and recoveries, respectively. The correlation coefficient (R) values for the testing sets for Cu and Mo grades were 0.93, 0.94 and for their recoveries were 0.93, 0.92, respectively. The results discussed in this paper indicate that the proposed model can be used to predict the Cu and Mo grades and recoveries with a reasonable error.

Permeability prediction based on reservoir zonation by a hybrid neural genetic algorithm in one of the Iranian heterogeneous oil reservoirs

Permeability is the most important parameter for precise reservoir description and modeling. Despite the advances and modification in different methods for permeability evaluation such as well testing and well logging, the most exact method is core analysis, which is expensive and time consuming. Because of the well logging data availability in most drilled wells, attempts have been made to utilize artificial neural networks for identification of the relationship, which may exist between the logging data and core permeability. In this study, a new approach based on hybrid neural genetic algorithm has been designed to predict permeability from the well logging data in one of the Iranian heterogeneous oil reservoirs. This approach is based on reservoir zonation according to geology characteristics and sorting the data in the same manner. The predicted permeability was compared to core permeability and it shows that permeability prediction based on designing separate networks for each zone is more accurately than designing single network for all of zones.

Optimization of Predicted Cell Voltage & Caustic Current Efficiency in a Chlor-Alkali Membrane Cell with Application of Genetic Algorithm

This paper presents the development of hybrid neural-genetic algorithm model for the prediction of cell voltage and caustic current efficiency (CCE) versus various operating parameters in a lab scale chlor-alkali membrane cell. Each of six process parameters including anolyte pH (2-5), operating temperature (25-90 oC), electrolyte velocity (1.3-5.9 cm/s), brine concentration (200-300 g/L), current density (1-4 kA/m2), and run time (up to 150 min) were thoroughly studied at four levels for low caustic concentrations (5 g/L). According to the obtained results, the predicted cell voltages and current efficiencies using GA modeling were found to be very close to the measured values with an average deviation of about 2.82% & 1.77% for test validation data, respectively. It was also found that the developed model is not only capable to predict the voltage and current efficiency but also to reflect the impacts of the other process parameters on the same functions.

A hybrid model for isomorphism identification in mechanism design based on intelligent manufacturing

Isomorphism discernment of graphs is an important and complicate problem. The problem is vital for graph theory based kinematic structures enumeration. To solve the problem, a Genetic Algorithm (GA) model and a Hopfield Neural Networks (HNNs) model are developed respectively, and some operators are improved to prevent premature convergence. By a comparative study, the advantages and limitations of the two approaches for graph isomorphism problem are discussed. Based on above, a hybrid Neural-Genetic algorithm is proposed. Numerical experiments demonstrate the performance of the hybrid algorithm is more successful compared with the approach applying GA or HNN simply.

Minimum miscibility pressure prediction based on a hybrid neural genetic algorithm

Abstract Enhanced oil recovery (EOR) processes are unavoidable fact, which will be applied in oil upstream industry. It seems the miscible gas injection into oil reservoirs be one of the most effective methods in EOR approaches. A fundamental factor in the design of gas injection project is the minimum miscibility pressure (MMP), whereas local displacement efficiency from gas injection is very much dependent on the MMP. From an experimental point of view, slim tube displacements, and rising bubble apparatus (RBA) tests normally determine the MMP. Because such experiments are very costly and time-consuming, searching for quick and vigorous mathematical determination of gas–oil MMP is usually requested. Artificial neural networks (ANN) have been proved to be an effective alternative for forecasting purposes because of the pattern-matching ability. However, there is no specific recommendation on suitable design of network for different structures and generally, the parameters are selected by trial and error, which confines the approach context dependent. In this study, a hybrid neural genetic algorithm (GA-ANN) is proposed with the purpose of automate the design of neural network for dissimilar type of structures. The neural network is trained considering the reservoir temperature, reservoir fluid composition, and injected gas composition as input parameters and the MMP as desired parameter. Consequently, neural genetic model is compared with results obtained using other conventional models to make evaluation among different techniques. The results show that the neural genetic model can be applied effectively and afford high accuracy and dependability for MMP forecasting.

Optimal Selection of Machining Conditions in the Electrojet Drilling Process Using Hybrid NN-DF-GA Approach

This article presents a hybrid neural network, desirability function, and genetic algorithm (NN-DF-GA) approach for optimal selection of the input process parameters for optimizing the multiresponse parameters of the electrojet drilling (EJD) process. EJD is a promising nontraditional machining technique that is used for machining microholes (<1 mm in diameter) in difficult-to-machine materials. The proposed approach first uses a back propagation neural network to formulate a fitness function for predicting the response parameters of the process. From the network output, the desirability method obtains a composite fitness function for further use in the genetic algorithm. The genetic algorithm predicts the optimal input parametric combinations and simultaneously optimizes the multiresponse characteristics of the process. Simulated results confirm the feasibility of this approach and show a good agreement with experimental results for a wide range of machining conditions.

A hybrid neural–genetic algorithm for reservoir water quality management

A combined neural network and genetic algorithm (GA) was developed for water quality management of Feitsui Reservoir in Taiwan. First, an artificial neural network (ANN) model was employed to simulate the behavior of nutrient loads into the reservoir. The data from watershed loads, precipitation in the watershed, and outflow were used in the ANN model to forecast the total phosphorus concentration in the reservoir. A 6-year (1992–97) record of water quality data was used for network training, and additional data collected in 1998–2000 were used for model verification. Further, a GA was used with this ANN model to optimize the control of nutrient loads from the watershed. The GA was used as a search strategy to determine the proper reduction rates of nutrient loads from the watershed so that the objective function could be as close to the optimal value as possible. The study results indicate that the ANN model can effectively simulate the dynamics of reservoir water quality. The GA is able to identify control schemes that reduce the in-reservoir total phosphorus concentration by as much as 60%, and water quality in the reservoir can be expected to achieve an oligotrophic (most of the time) or mesotrophic level if the watershed nutrient loads are reduced by 10–80%.

A Hybrid Neural-Genetic Algorithm for the Frequency Assignment Problem in Satellite Communications

A hybrid Neural-Genetic algorithm (NG) is presented for the frequency assignment problem in satellite communications (FAPSC). The goal of this problem is minimizing the cochannel interference between satellite communication systems by rearranging the frequency assignments. Previous approaches to FAPSC show lack of scalability, which leads to poor results when the size of the problem grows. The NG algorithm consists of a Hopfield neural network which manages the problem constraints hybridized with a genetic algorithm for improving the solutions obtained. This separate management of constraints and optimization of objective function gives the NG algorithm the properties of scalability required. We analyze the FAPSC and its formulation, describe and discuss the NG algorithm and solve a set of benchmark problems. The results obtained are compared with other existing approaches in order to show that the NG algorithm is more scalable and performs better than previous algorithms in the FAPSC.

A portable and scalable algorithm for a class of constrained combinatorial optimization problems

This paper presents a portable and scalable approach for a class of constrained combinatorial optimization problems (CCOPs) which requires to satisfy a set of constraints and to optimize and objective function simultaneously. In particular, this paper is focused on the class of CCOPs that admits a representation in terms of a square matrix of constraints C.The algorithm consists of a hybrid neural-genetic algorithm, formed by a Hopfield Neural Network (HNN) which solves the problem's constraints, and a Genetic Algorithm (GA) for optimizing the objective function. This separated management of constraints and optimization procedures makes the proposed algorithm scalable and robust. The portability of the algorithm is given by the fact that the HNN dynamics depends only on the matrix C of constraints.We show these properties of scalability and portability by solving three different CCOPs with our algorithm, the frequency assignment problem in a mobile telecommunications network, the reduction of the interference in satellite systems and the design of FPGAs with segmented channel routing architecture. We compare our results with previous approaches to these problems, obtaining very good results in all of them.

Conceptual design support system in a collaborative environment for injection moulding

The shortened time-to-market of new products requires experts from multiple disciplines to cooperate in product development. In this collaborative product development environment, an Internet-based conceptual design support system for Injection moulding is developed. It can help designers to put forward feasible design solutions quickly and provide strong support for collaborative design. The architecture, implementation and evaluation of this system are discussed in this paper. The proposed system makes fully use of component-based web technology and has a flexible distributed architecture. A hybrid neural network and genetic algorithm approach is also adopted for more accurate and efficient retrieval. Experiment results indicate it is superior to conventional systems in accuracy and time.