scholarly journals A genetic algorithm with neural network fitness function evaluation for IMRT beam angle optimization

2013 ◽  
Vol 22 (3) ◽  
pp. 431-455 ◽  
Author(s):  
Joana Dias ◽  
Humberto Rocha ◽  
Brígida Ferreira ◽  
Maria do Carmo Lopes
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Fitness Function ◽  
Function Evaluation ◽  
Beam Angle ◽  
Beam Angle Optimization

scholarly journals Analyzing student performance using evolutionary artificial neural network algorithm

2018 ◽  
Vol 7 (2.26) ◽  
pp. 67 ◽  
Author(s):  
A S. Arunachalam ◽  
T Velmurugan
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Artificial Neural Network ◽  
Student Performance ◽  
Fitness Function ◽  
Research Work ◽  
Function Evaluation ◽  
Success Factor ◽  
Optimization Approach ◽  
Artificial Neural

Educational Data Mining (EDM) and Learning Systematic (LS) research have appeared as motivating areas of research, which are clarifying beneficial understanding from educational databases for many purposes such as predicting student’s success factor. The ability to predict a student’s performance can be beneficial in modern educational systems. This research work aims at developing an evolutionary approach based on genetic algorithm and the artificial neural network. The traditional artificial neural network lacks predicting student performance due to the poor modeling structure and the capability of assigning proper weights to each node under the hidden layer. This problem is overwhelmed with the aid of genetic algorithm optimization approach which produces appropriate fitness function evaluation in each iteration of the learning process. The performances gradually increase the accuracy of the prediction and classification more precisely.

scholarly journals PENENTUAN PARAMETER LEARNING RATE SELAMA PEMBELAJARAN JARINGAN SYARAF TIRUAN BACKPROPAGATION MENGGUNAKAN ALGORITMA GENETIKA

2020 ◽  
Vol 14 (2) ◽  
pp. 202-212
Author(s):  
Ade chandra Saputra
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Genetic Algorithms ◽  
Fitness Function ◽  
Learning Rate ◽  
Local Minima ◽  
Rate Parameter ◽  
Neural Network Learning ◽  
Value Of Learning ◽  
Backpropagation Learning

One of the weakness in backpropagation Artificial neural network(ANN) is being stuck in local minima. Learning rate parameter is an important parameter in order to determine how fast the ANN Learning. This research is conducted to determine a method of finding the value of learning rate parameter using a genetic algorithm when neural network learning stops and the error value is not reached the stopping criteria or has not reached the convergence. Genetic algorithm is used to determine the value of learning rate used is based on the calculation of the fitness function with the input of the ANN weights, gradient error, and bias. The calculation of the fitness function will produce an error value of each learning rate which represents each candidate solutions or individual genetic algorithms. Each individual is determined by sum of squared error value. One with the smallest SSE is the best individual. The value of learning rate has chosen will be used to continue learning so that it can lower the value of the error or speed up the learning towards convergence. The final result of this study is to provide a new solution to resolve the problem in the backpropagation learning that often have problems in determining the learning parameters. These results indicate that the method of genetic algorithms can provide a solution for backpropagation learning in order to decrease the value of SSE when learning of ANN has been static in large error conditions, or stuck in local minima

scholarly journals Determining optimal combination of roller compacted concrete pavement mixture containing recycled asphalt pavement and crumb rubber using hybrid artificial neural network–genetic algorithm method considering energy absorbency approach

2017 ◽  
Vol 44 (11) ◽  
pp. 945-955 ◽  
Author(s):  
Mansour Fakhri ◽  
Ershad Amoosoltani ◽  
Mona Farhani ◽  
Amin Ahmadi
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Network Model ◽  
Neural Network Model ◽  
Fitness Function ◽  
Crumb Rubber ◽  
Concrete Pavement ◽  
Genetic Operators ◽  
Efficiency Criterion ◽  
Roller Compacted Concrete

The present study investigates the effectiveness of evolutionary algorithms such as genetic algorithm (GA) evolved neural network in estimating roller compacted concrete pavement (RCCP) characteristics including flexural and compressive strength of RCC and also energy absorbency of mixes with different compositions. A real coded GA was implemented as training algorithm of feed forward neural network to simulate the models. The genetic operators were carefully selected to optimize the neural network, avoiding premature convergence and permutation problems. To evaluate the performance of the genetic algorithm neural network model, Nash-Sutcliffe efficiency criterion was employed and also utilized as fitness function for genetic algorithm which is a different approach for fitting in this area. The results showed that the GA-based neural network model gives a superior modeling. The well-trained neural network can be used as a useful tool for modeling RCC specifications.

Multi-point optimization of transonic airfoils using an enhanced genetic algorithm

2017 ◽  
Vol 232 (7) ◽  
pp. 1347-1360 ◽  
Author(s):  
Narjes Timnak ◽  
Alireza Jahangirian
Keyword(s):  
Genetic Algorithm ◽  
Objective Function ◽  
Fitness Function ◽  
Single Point ◽  
Navier Stokes ◽  
Computational Time ◽  
Adaptive Mutation ◽  
Function Evaluation ◽  
New Techniques ◽  
New Methods

In this study, two new techniques are proposed for accelerating the multi-point optimization of an airfoil shape by genetic algorithms. In such multi-point evolutionary optimization, the objective function has to be evaluated several times more than a single-point optimization. Thus, excessive computational time is crucial in these problems particularly, when computational fluid dynamics is used for fitness function evaluation. Two new techniques of preadaptive range operator and adaptive mutation rate are proposed. An unstructured grid Navier–Stokes flow solver with a two-equation [Formula: see text] turbulence model is used to evaluate the objective function. The new methods are applied for optimum design of a transonic airfoil at two speed conditions. The results show that using the new methods can increase the aerodynamic efficiency of optimum airfoil at each operating condition with about 30% less computational time in comparison with the conventional genetic algorithm approach.

scholarly journals An Optimization Model for Environmental Ergonomics Assessment in Bioproduction of Food SMEs

2020 ◽  
Vol 27 (4) ◽  
pp. 296
Author(s):  
Mirwan Ushada ◽  
Hani Febri Mustika ◽  
Aina Musdholifah ◽  
Tsuyoshi Okayama
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Heart Rate ◽  
Artificial Neural Network ◽  
Optimization Model ◽  
Artificial Neural Network Model ◽  
Fitness Function ◽  
Research Result ◽  
Assessment System ◽  
Optimal Set

Environmental ergonomics in bioproduction of food Small Medium-sized Enterprises (SMEs) become a concern and need to be optimized. An optimization model was developed using a Genetic Algorithm (GA). The weight of an Artificial Neural Network Model was used as a fitness function for GA. The research objectives were: 1) To design an environmental ergonomic assessment system for bioproduction of Food SMEs, 2) To develop an optimization model for environmental ergonomic assessment using a Genetic Algorithm. GA is utilized to search optimal set points of environmental ergonomics based on the predicted fitness values. Each chromosome of GA represents the environmental ergonomics value. The parameters were heart rate, bioproduction temperature, distribution of bioproduction relative humidity and light intensity. The target of the optimization model was the bioproduction temperature set points. The research result indicated the model generated optimum values of environmental ergonomics parameter in bioproduction of food SMEs. The parameters could be used to provide standard workplace environment for the sustainability of food SMEs.

scholarly journals Mitigating Economic Denial of Sustainability (EDoS) in Cloud Environment using Genetic Algorithm and Artificial Neural Network

2019 ◽  
Vol 8 (10) ◽  
pp. 3415-3421
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Artificial Neural Network ◽  
Virtual Machines ◽  
Fitness Function ◽  
Medium Size ◽  
Cloud Environment ◽  
Edos Attacks ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

Economic Denial of Sustainability (EDoS) is a latest threat in the cloud environment in which EDoS attackers continually request huge number of resources that includes virtual machines, virtual security devices, virtual networking devices, databases and so on to slowly exploit illegal traffic to trigger cloud-based scaling capabilities. As a result, the targeted cloud ends with a consumer bill that could lead to bankruptcy. This paper proposes an intelligent reactive approach that utilizes Genetic Algorithm and Artificial Neural Network (GANN) for classification of cloud server consumer to minimize the effect of EDoS attacks and will be beneficial to small and medium size organizations. EDoS attack encounters the illegal traffic so the work is progressed into two phases: Artificial Neural Network (ANN) is used to determine affected path and to detect suspected service provider out of the detected affected route which further consist of training and testing phase. The properties of every server are optimized by using an appropriate fitness function of Genetic Algorithm (GA) based on energy consumption of server. ANN considered these properties to train the system to distinguish between the genuine overwhelmed server and EDoS attack affected server. The experimental results show that the proposed Genetic and Artificial Neural Network (GANN) algorithm performs better compared to existing Fuzzy Entropy and Lion Neural Learner (FLNL) technique with values of precision, recall and f-measure are increased by 3.37%, 10.26% and 6.93% respectively.

scholarly journals Performance Analysis of Simulated Annealing and Genetic Algorithm on systems of linear equations

F1000Research ◽  
2021 ◽  
Vol 10 ◽  
pp. 1297
Author(s):  
Md. Shabiul Islam ◽  
Most Tahamina Khatoon ◽  
Kazy Noor-e-Alam Siddiquee ◽  
Wong Hin Yong ◽  
Mohammad Nurul Huda
Keyword(s):  
Genetic Algorithm ◽  
Simulated Annealing ◽  
Real Time ◽  
Large Scale ◽  
Fitness Function ◽  
Random Search ◽  
Linear Equations ◽  
Simultaneous Equations ◽  
Function Evaluation ◽  
Solution Sets

Problem solving and modelling in traditional substitution methods at large scale for systems using sets of simultaneous equations is time consuming. For such large scale global-optimization problem, Simulated Annealing (SA) algorithm and Genetic Algorithm (GA) as meta-heuristics for random search technique perform faster. Therefore, this study applies the SA to solve the problem of linear equations and evaluates its performances against Genetic Algorithms (GAs), a population-based search meta-heuristic, which are widely used in Travelling Salesman problems (TSP), Noise reduction and many more. This paper presents comparison between performances of the SA and GA for solving real time scientific problems. The significance of this paper is to solve the certain real time systems with a set of simultaneous linear equations containing different unknown variable samples those were simulated in Matlab using two algorithms-SA and GA. In all of the experiments, the generated random initial solution sets and the random population of solution sets were used in the SA and GA respectively. The comparison and performances of the SA and GA were evaluated for the optimization to take place for providing sets of solutions on certain systems. The SA algorithm is superior to GA on the basis of experimentation done on the sets of simultaneous equations, with a lower fitness function evaluation count in MATLAB simulation. Since, complex non-linear systems of equations have not been the primary focus of this research, in future, performances of SA and GA using such equations will be addressed. Even though GA maintained a relatively lower number of average generations than SA, SA still managed to outperform GA with a reasonably lower fitness function evaluation count. Although SA sometimes converges slowly, still it is efficient for solving problems of simultaneous equations in this case. In terms of computational complexity, SA was far more superior to GAs.

Genetic Algorithm (GA) and Internal Energy Transfer Calculations Using Neural Network (NN) Methods

2012 ◽  
Author(s):  
Lionel Raff ◽  
Ranga Komanduri ◽  
Martin Hagan ◽  
Satish Bukkapatnam
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Neural Networks ◽  
Objective Function ◽  
Fitness Function ◽  
Solution Space ◽  
Large Ensemble ◽  
Nonlinear Functional ◽  
Squared Error ◽  
Time Required

Genetic algorithms (GA), like NNs, can be used to fit highly nonlinear functional forms, such as empirical interatomic potentials from a large ensemble of data. Briefly, a genetic algorithm uses a stochastic global search method that mimics the process of natural biological evolution. GAs operate on a population of potential solutions applying the principle of survival of the fittest to generate progressively better approximations to a solution. A new set of approximations is generated in each iteration (also known as generation) of a GA through the process of selecting individuals from the solution space according to their fitness levels, and breeding them together using operators borrowed from natural genetics. This process leads to the evolution of populations of individuals that have a higher probability of being “fitter,” i.e., better approximations of the specified potential values, than the individuals they were created from, just as in natural adaptation. The most time-consuming part in implementing a GA is often the evaluation of the objective or the fitness function. The objective function O[P] is expressed as sum squared error computed over a given large ensemble of data. Consequently, the time required for evaluating the objective function becomes an important factor. Since a GA is well suited for implementing on parallel computers, the time required for evaluating the objective function can be reduced significantly by parallel processing. A better approach would be to map out the objective function using several possible solutions concurrently or beforehand to improve computational efficiency of the GA prior to its execution, and using this information to implement the GA. This will obviate the need for cumbersome direct evaluation of the objective function. Neural networks may be best suited to map the functional relationship between the objective function and the various parameters of the specific functional form. This study presents an approach that combines the universal function approximation capability of multilayer neural networks to accelerate a GA for fitting atomic system potentials. The approach involves evaluating the objective function, which for the present application is the mean squared error (MSE) between the computed and model-estimated potential, and training a multilayer neural network with decision variables as input and the objective function as output.

Optimal Design of Pre-Forging for Gear Blank Using BP Neural Network and Genetic Algorithm

2011 ◽  
Vol 179-180 ◽  
pp. 801-806
Author(s):  
Ming Yue Zhang ◽  
Xin Yun Wang ◽  
Ju Chen Xia
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Fitness Function ◽  
Back Propagation ◽  
Nonlinear Problems ◽  
Back Propagation Neural Network ◽  
Highly Nonlinear ◽  
Die Stress ◽  
Trial And Error Method ◽  
Quality Of Products

The design of pre-forging is very important during multistage forging of producing gear blank. It directly affects the behavior of metal flowing, filled situation of die cavity of finish forging, quality of products and die life. Most designs of pre-forging for gear blank are based on trial and error method. This paper presents a suitable method for practical designation of pre-forging for gear blank by proposing an improved algorithm, which combines Back Propagation Neural Network and Genetic Algorithm. Firstly, the mathematical model between the size parameters of pre-forging and forming force and maximum die stress of finish forging was established by using Back Propagation Neural Network which has the feature of processing highly nonlinear problems. Secondly, the established model was set as the fitness function of Genetic Algorithm. At last, the most superior pre-forging shape and the size parameters were solved by using the Genetic Algorithm with the function of overall situation optimization. These can lead to lower cost and time in the stages of designing pre-forging for gear blank.

Sign in / Sign up

Export Citation Format

Share Document