Improved snake optimization algorithm for parameter identification based on genetic algorithm

BackgroundMining epistatic loci which affects specific phenotypic traits is an important research issue in the field of biology. Bayesian network (BN) is a graphical model which can express the relationship between genetic loci and phenotype. Until now, it has been widely used into epistasis mining in many research work. However, this method has two disadvantages: low learning efficiency and easy to fall into local optimum. Genetic algorithm has the excellence of rapid global search and avoiding falling into local optimum. It is scalable and easy to integrate with other algorithms. This work proposes an epistasis mining approach based on genetic tabu algorithm and Bayesian network (Epi-GTBN). It uses genetic algorithm into the heuristic search strategy of Bayesian network. The individual structure can be evolved through the genetic operations of selection, crossover and mutation. It can help to find the optimal network structure, and then further to mine the epistasis loci effectively. In order to enhance the diversity of the population and obtain a more effective global optimal solution, we use the tabu search strategy into the operations of crossover and mutation in genetic algorithm. It can help to accelerate the convergence of the algorithm.ResultsWe compared Epi-GTBN with other recent algorithms using both simulated and real datasets. The experimental results demonstrate that our method has much better epistasis detection accuracy in the case of not affecting the efficiency for different datasets.ConclusionsThe presented methodology (Epi-GTBN) is an effective method for epistasis detection, and it can be seen as an interesting addition to the arsenal used in complex traits analyses.

Solution of Co-evolution Self-adaptive Genetic Algorithm to Production Scheduling Problem of Flexible Manufacturing System

The physical identification of aerodynamic parameters plays a decisive role in studying the characteristics of projectiles. During the development process of projectiles, it is an important task to accurately obtain the aerodynamic parameters of projectiles. In this paper, through a large number of experiments, the flight trajectory data of the projectile are measured, and the maximum likelihood estimation algorithm is used to physically identify the drag coefficient and lift coefficient of the projectile. First, the sensitivity coefficients of each parameter of the projectile trajectory are calculated and deduced. Second, based on the consistency and asymptotic characteristics of the maximum likelihood estimation algorithm, the sensitivity relationship between the velocity and the drag coefficient and between the position of the projectile’s center of mass and the lift coefficient is used to identify the aerodynamic physical parameters of the projectile. The results show that the maximum likelihood estimation algorithm has high identification accuracy, fast calculation speed, and low algorithm complexity, which can effectively identify the aerodynamic physical parameters of the projectile and meet practical engineering needs.

For the disadvantages of orthogonal wavelet transform multi-modulus blind equalization algorithm(WTMMA), such as slow convergence rate, large mean square error, and immerging in partial minimum easily, an orthogonal wavelet transform multi-modulus blind equalization algorithm based on Tabu search DNA genetic optimization algorithm(DNAGA-WTMMA) was proposed. Firstly, DNA populations were initialized by using double-stranded DNA form to select the single-stranded DNA sequences as the representative strands of individuals. Secondly, Tabu search strategy was introduced into crossover operations to avoid circuitous search, ensure the searching different paths, and escape from local optimum. Thirdly, in order to speed up the convergence rate and avoid premature convergence, the probability of crossover operation was decided by dynamic adaptive probability based on population prematurity degree index. Computer simulations show that, compared with multi-modulus blind equalization algorithm(MMA) and WTMMA, the proposed algorithm has the fastest convergence rate and the smallest mean square error. Introduction In wireless communication, blind equalization technology without sending the training sequence is one of important methods of suppressing the inter-symbol interference(ISI) caused by complicated communication channel and limited bandwidth. In the blind equalization algorithms, the constant modulus blind equalization algorithm(CMA) has good equalization performance for constant modulus signals, but it has large mean square error for high-order QAM signals. While the multi-modulus equalization algorithm(MMA) with the amplitude and phase information of output signals can reduce effectively the mean square error(MSE). Orthogonal wavelet multi-modulus blind equalization algorithm outperforms MMA in equalizing multi-modulus signals. However, the weight vector of the CMA and the MMA is updated by gradient search method, therefore, they are easy to fall into local convergence and difficult to obtain the global optimal solution[1]. Genetic algorithm(GA) has stronger robustness and global random search abilities and can find global optimal solution in complex, multi-peak large solution space. However, the traditional genetic algorithm is easy to fall into premature convergence and its convergence speed is slow[2]. DNA computing proposed by Adleman in 1994, has strong ability of parallel computing, can be used to search all potential solutions. Due to the nature connection with DNA computing and genetic algorithm, DNA computing is introduced into genetic algorithm to improve the performance of GA[3][4]5][6]. Tabu search algorithm(TS) is a global search algorithm with short-time memory, a tabu list (movements in the list were temporarily forbidden) and an aspiration criterion changing the content of tabu list are adopted[7][8]. Therefore, we introduce tabu search algorithm into DNA genetic algorithm to improve the performance of DNA genetic algorithm. Based on the above analysis, an orthogonal wavelet transform multi-modulus blind equalization algorithm based on tabu search DNA genetic optimization algorithm(DNAGA-WTMMA) is International Conference on Materials Engineering and Information Technology Applications (MEITA 2015) © 2015. The authors Published by Atlantis Press 129 proposed. This proposed algorithm uses cost function of the WTMMA as the fitness function of DNA-GA, then uses DNA-GA to solve the cost function of the equalizer to find optimal individual. Compared with multi-modulus blind equalization algorithm(MMA) and orthogonal wavelet transform multi-modulus blind equalization algorithm(WTMMA), the proposed algorithm has the fastest convergence rate and the smallest mean square error. Orthogonal Wavelet Transform Multi-modulus Blind Equalization Algorithm

Along with the increase in population and industry in many countries, the fuel oil demand also increases. Petroleum exploration on a large scale will accelerate the depletion of petroleum reserves. One alternative to meet fuel needs is the discovery of biodiesel which is renewable alternative energy. Synthesis biodiesel is carried out through an enzymatic reaction. In the enzymatic reaction model making biodiesel, there are parameters that must be estimated. The estimated parameters of the enzymatic reaction model will determine the success of the reaction. The parameter estimation of the enzymatic reaction model can be done using local optimization or global optimization algorithms, but the local optimization algorithm has a major disadvantage, the optimal value obtained is the local optimal value. Genetic algorithms are global optimization algorithms that are capable of working on high-dimensional problems. The success of genetic algorithms is determined by chromosome models, crossover operations, and mutation operations. The use of improper crossover operations often produces local optimum solutions. There are various types of crossover operation, each of which has weaknesses and advantages. This paper studies the parameters estimation of the enzymatic reaction model for biodiesel synthesis by using genetic algorithms with some crossover operation.

The genetic Algorithm (GA) relies on the effective utilization of the crossover operator, which plays a crucial role in mitigating premature convergence, may occur when genetic operators dominate the population too quickly, leading to a loss of diversity and hindering the generation of optimal solutions. This study introduced a new crossover operator called the flip multi-sliced average crossover (FMSAX) operator with a rank selection function to address this problem. This new operator was more efficient for variable minimization in GA. The study used 2,403 records of student attrition from two private higher education institutions in the Philippines and found that the enhanced crossover outperformed the original crossover operator in GA. The improved GA using the FMSAX operator resulted in a 53.33% improvement in variable minimization, with 25% fewer variables eliminated than the original average crossover (AX) operator. Additionally, when integrated with the C4.5 algorithms, the enhanced GA using FMSAX produced an accuracy of the prediction model of 96.47%, higher than the original GA using AX and C4.5 algorithms with an accuracy of 93.85%. This study's prediction model using the enhanced GA and FMSAX operator helped private educational institutions design preventive measures to mitigate attrition rates in the academe.

Many evolutionary algorithms have been developed by the researcher. Genetic algorithms and memetic algorithms are most commonly used by researchers to solve various optimization problems. Genetic and Memetic algorithms have certain parameters that need to be modified so that the algorithm can produce better results. In the present paper, we are working on a genetic algorithm. Crossover operators and mutation operators play a very important role in the development of an efficient genetic algorithm (GA). The performance of the genetic algorithm depends upon its parameter setting. In the present paper, the basic conceptual features and unique characteristics of the different rates of crossover and mutation operators are discussed. The simulation environment was created to study the combinatorial problems, i.e., scheduling and sequencing problems. Currently, the researcher is focusing on the scheduling problem which is a NP-Hard problem. In the future study, the traveling salesman problem could be considered to investigate the additional parameters of GA. The current empirical study shows that the crossover and mutation rate have an impact on the GA's performance. This research helps us to understand how to choose the right crossover and mutation rate so that the GA's performance can be improved.

In order to obtain the aerodynamic parameters of a high-spinning projectile, a new parameter identification method is proposed based on the Cubature Kalman Filter (CKF). First of all, the motion equation of the spinning stabilized projectile is established by the 4D trajectory model. Second, unknown parameters are added to the state vector to obtain the augmented state vector. Next, a new filter is designed for the identification of the unknown parameters based on the basic theory of the CKF. Finally, the simulation results of the CKF are compared with those of the Extended Kalman Filter (EKF) and the Unscented Kalman Filter (UKF). The results of comparison show that the CKF method can effectively identify the aerodynamic parameters and the identification error is less than 1.2%. The CKF method has greater accuracy than the EKF and the UKF.

Fuzzy connectives based crossover operators to model genetic algorithms population diversity

The aim of this study was to develop a hybrid optimization model for solving the routing problem identified at Zoomlion Ghana Limited in Shama district in the Western Region of Ghana. Two main optimization models were considered: Particle Swarm Optimization (PSO) and Genetic Algorithm (GA). A hybrid algorithm was developed from the two by merging crossover and mutation operators of GA with PSO. A sample of 20 breakpoints was run through 10,000 iterations for all the models and the results of the proposed hybrid model was compared with PSO and GA separately. The optimal results of PSO, GA and the proposed models are 1160.6km, 1190.3km and 1132.3km respectively. The proposed model’s results were also compared with other hybrid models to test the robustness of the new model. This result was achieved because the new model eliminates the low convergence rate in PSO and also prevents it from easily falling into local optimum in high-dimensional space and the inclusion of crossover and mutation operators of GA improves the diversity of the iterations. After the iterations, PSO reduced a field distance of 1700 km to 1160.6 km within 780.4098 seconds. GA on the other hand reduced the same field distance of 1700 km to 1190.3km within 397.3308 seconds. The proposed hybrid model reduced the same field distance from 1700 km to 1132.3 km within 550.2527 seconds. This indicates that the proposed hybrid model performed better than PSO and GA separately. A performance test between the proposed hybrid model and other hybrid models showed that merging crossover and mutation operators into PSO gives a better optimal result. MATLAB was used for the iterations.

Genetic Algorithm Difficulty and the Modality of Fitness Landscapes

The precise identification of aerodynamic parameters for aircraft is increasingly critical in aircraft design endeavors, particularly for furnishing accurate dynamic models for ground and flight simulation. To mitigate the dependence on traditional identification methods that rely heavily on aircraft models, this work introduces an aerodynamic parameters identification methodology based on Physics-informed Neural Networks (PINNs). By utilizing the six degrees of freedom motion equation as the physical constraint with the neural network, and the aerodynamic parameters to be identified as the neural network variables, the neural network model is trained to serve as a surrogate for the aircraft model. As a case study, the longitudinal motion of the aircraft is employed to identify and analyze the aerodynamic parameters. A comprehensive comparison is conducted among a method based on the genetic algorithm, a conventional neural network-based approach, and the proposed PINN-based methodology. The results obtained demonstrate that the proposed method can effectively mitigate system and data errors, exhibiting high precision in parameter identification and anti-interference capabilities with regard to noise in data. This innovative approach holds the potential to substantially decrease the reliance on flight test data for parameter identification purposes.

The paper presents the problem of parametric identification of induction motor mathematical model with the use of genetic algorithm (GA). The influence of population replacement strategy on the identification results is analyzed. The identified parameters of the model were determined as a result of minimization of performance index defined as the mean-square error of stator current and angular velocity. The experimental investigations were performed for induction motor powered from voltage inverter. The work deals with the problem of identification, tournament selection, operators of crossover and mutation, replacement strategies and stop criterion. The steady-state genetic algorithm was analyzed with regard to convergence and accuracy of the identification process and the time of numerical calculations.

Firefly algorithm (FA) is an easily implementable, robust, simple and flexible technique, but the major drawback associated with this technique is the imbalanced exploration and exploitation during firefly position changing stage. This imbalanced relation degrades the solution quality which ultimately results in either skipping the most optimal solution even present in the vicinities of the current solution or trapping the solution in the local optima. In this paper, this issue is resolved by introducing genetic algorithm (GA) operators namely selection, mutation and crossover operators in the firefly position stage of the standard FA. The performance of the proposed approach has been tested on energy consumption optimization and user comfort management inside smart building and has been compared with the standard FA, GA, artificial bee colony (ABC) and ant colony optimization (ACO) algorithm in terms of temperature, illumination, air quality and total power consumption minimization and user comfort maximization. The minimum, average, maximum and total power consumption and minimum, maximum and average user comfort were the performance evaluation parameters. The least amount of 145.39 kilowatt hour (kWh) of total power consumed for temperature control was observed for the proposed approach followed by ACO, ABC, FA and GA where the power consumed for temperature was observed as 173.68 kWh, 179.27 kWh, 181.93 kWh and 188.95 kWh, respectively. Similarly, for illumination control, the consumed power for the proposed model was 118.30 kWh followed by FA, ABC, ACO and GA where the power consumed was 146.93 kWh, 162.96 kWh, 169.28 kWh and 193.53 kWh, respectively. For air quality control, the minimum power of 186.94 kWh was found for the proposed algorithm followed by FA, ABC, ACO and GA where the power consumed was 229.01 kWh, 234.38 kWh, 240.47 kWh and 244.76 kWh, respectively. Likewise, maximum user comfort was observed for the proposed technique with the value of 0.94004/1 followed by ACO, ABC, FA and GA where the user comfort recorded was 0.939655/1, 0.93878/1, 0.938314/1 and 0.937896/1, respectively. The statistical analysis shows the efficiency of the proposed model for power consumption minimization and user comfort maximization.

A genetic algorithm with some improvement is proposed to avoid the local optimum for job-shop scheduling problem (JSP). There is recurrent searching process of genetic operation in the improved genetic algorithm. The improved crossover operation can shake current population from local optimum in genetic algorithm. The recurrent crossover operation and mutation operation can inherit excellent characteristics from parent chromosomes and accelerate the diversity of offspring. Both benchmark FT(6times6) and LA1(10times5) job-shop scheduling problems are used to show the effectiveness of the proposed method. Experimental results demonstrate that the proposed genetic algorithm does not get stuck at a local optimum easily, and it is fast in convergence, simple to be implemented