Initial Population Of Genetic Algorithm Research Articles

Design of an Intelligent Shop Scheduling System Based on Internet of Things

In order to optimize the functionality of automated guidance vehicles (AGVs) in logistics workshops, a wireless charging and task-based logistics intelligent dispatch system was developed based on the Internet of Things. This system aimed to improve freight efficiency in the workshop’s logistics system. The scheduling system successfully addressed the round-trip scheduling issue between AGVs and multiple tasks through two degrees of improvement: the application of AGVs and task path planning. To handle conflict coordination and AGV cluster path planning, a shortest path planning algorithm based on the A* search algorithm was proposed, and the traffic control law was enhanced. The initial population of genetic algorithms, which used greedy algorithms to solve problems, was found to be too large in terms of task distribution. To address this, the introduction of a few random individuals ensured population diversity and helped avoid local optima. Numerical experiments demonstrated a significantly accelerated convergence rate towards the optimal solution.

An XGBoost-Based predictive control strategy for HVAC systems in providing day-ahead demand response

Heating, ventilation and air-conditioning systems (HVAC), at demand side, are regarded as promising candidates to provide demand response to smart power grids by adjusting indoor temperature setpoint. However, the optimal control of indoor temperature setpoint with the constraint of indoor environment is a key issue. In this study, an XGBoost-based predictive control strategy is proposed for HVAC systems in providing day-ahead demand response. XGBoost (Extreme Gradient Boosting) models are developed to predict the variation of power use and indoor temperature. Several innovations are implemented to improve the performance of the control strategy, including dedicated generation of the initial population of genetic algorithm (GA) in rolling optimization, the use of a step-decreasing control horizon, and the separation of the sampling period and control action period. The strategy is validated on a TRNSYS-Python co-simulation test platform. Results show that the developed XGBoost models can achieve a high prediction performance of power use (CV-RMSE of 4.52%) and indoor temperature (CV-RMSE of 0.40%) in the coming 10 min with only using one-week historical data. The computation cost is reduced by 78% due to the dedicated generation of GA initial population while the control performance is not affected. Compared with the traditional operation, the proposed control strategy can reduce the total cost by as much as 27.30% and 33.6% while the weighted indoor temperature only increases by 0.14 K and 0.62 K, respectively. In test cases, the highest Predicted Mean Vote (PMV) is 0.72 and the corresponding Predicted Percentage of Dissatisfied (PPD) is 15.9%.

A deep reinforcement learning based hybrid algorithm for efficient resource scheduling in edge computing environment

Edge computing can greatly decrease the delay between users and cloud servers, which can significantly improve system service performance. However, it remains challenging for more efficient scheduling and allocation of users’ application demands with dependence constraints to edge cloud servers. Due to the randomness of the initial population, traditional intelligent optimization algorithms have poor convergence speed in addressing resource scheduling. Therefore, to minimize the execution time of the application, this paper proposes a hybrid algorithm to solve the resource scheduling problem with parallelism and subtask dependency. To improve the convergence speed of the algorithm, this paper makes full use of the features of deep Q networks (DQN) and genetic algorithms (GA). The initial population of GA is generated using DQN. Finally, to evaluate the effectiveness of our proposed algorithm, this paper selects three real scientific workflows for experiments. The experimental results show that the hybrid algorithm can converge quickly and improve the optimization effect in a short time.

A Multi-objective Optimal Dispatch Method for Integrated Energy System Considering Multiple Loads Variations

This paper proposes a multi-objective optimization dispatching method for the Integrated Energy System (IES), in which cooling/heating/electricity multi-load demand are included, and IES’s requirements in terms of energy efficiency, economy, and emissions are considered. We established an IES model for multi-objective optimal dispatch, and simulated cooling/heating/electric multiple load variations in it; then we established multiple objective functions based on IES’s scheduling requirements in terms of energy efficiency, economy, environmental issues. The model is solved by finding the pareto front based on genetic algorithm (GA), so that the decision makers can filter out the scheduling plan based on their preference, in which single-objective optimization is used first to find out the initial population for GA. Finally, the proposed method’s effectiveness is verified by case studies.

Sequence optimization of machining elements for process model based on the genetic algorithm of matrix constrained

Process sequence planning is the permutation and combination orienting to product manufacturing process, which plays an important role in improving the manufacturing produce efficiency and reducing the manufacturing cost. In this paper, an improved constraint matrix method is proposed to transform the relation between manufacturing features into the constraint relation between machining elements, and to determine the sequence coefficient of machining elements according to the correlation relation of manufacturing features. Through constructing the interactive associative relationship of sequence constraint matrix and combined constraint matrix, the initial population with high adaptability is selected out, and can improves the adaptability of initial population of genetic algorithm by using the process sequencing algorithm of improved constraint matrix. In order to improve the convergence speed, the accuracy and efficiency of process sequencing, the genetic algorithm of elite preserving is used.

Improving Initial Population for Genetic Algorithm using the Multi Linear Regression Based Technique (MLRBT)

Genetic algorithms (GAs) are powerful heuristic search techniques that are used successfully to solve problems for many different applications. Seeding the initial population is considered as the first step of the GAs.
 In this work, a new method is proposed, for the initial population seeding called the Multi Linear Regression Based Technique (MLRBT). That method divides a given large scale TSP problem into smaller sub-problems and the technique works frequently until the sub-problem size is very small, four cities or less. Experiments were carried out using the well-known Travelling Salesman Problem (TSP) instances and they showed promising results in improving the GAs' performance to solve the TSP.

Photoacoustic-guided Transcranial HIFU with Combined Time-reversal and Genetic Algorithm.

High intensity focused ultrasound (HIFU) is a noninvasive therapy used to induce tissue ablation for treating malignant tissues. Photoacoustic (PA) has recently been proposed as an alternative method to guide HIFU. In this paper, we present a method of HIFU guided by time-reversing the transcranial PA signals of an optically selective target in a nonselective background. To improve the focus performance on target area, we further propose to utilize the time-reversed PA signals as the initial population of Genetic Algorithm (GA) to optimize the focusing iteratively. In particular, we mimic both optical and acoustic parameters of the human brain and intracranial media in the simulation study. Experimental results show that the focusing accuracy of the proposed method has been significantly improved compared to just one-step PA time-reversal. At the same time, the combination of TR and GA makes the iteration time consumption of the optimization process less than other traditional algorithms without TR, showing its potential HIFU in clinical scenarios.

Genetic Algorithm-Based Anisotropic Diffusion Filter and Clustering Algorithms for Thyroid Tumor Detection

Medical imaging is a technique that has been used for diagnosis and treatment of a large number of diseases. Therefore it has become necessary to conduct a good image processing to extract the finest desired result and information. In this study, genetic algorithm (GA)-based clustering technique (K-means and Fuzzy C Means (FCM)) were used to segment thyroid Computed Tomography (CT) images to an extraction thyroid tumor. Traditional GA, K-means and FCM algorithms were applied separately on the original images and on the enhanced image with Anisotropic Diffusion Filter (ADF). The resulting cluster centers from K-means and FCM were used as the initial population in GA for the implementation of GAK-Mean and GAFCM. Jaccard index was used to study resemblance, dissimilarity, distance between two sets of images, and effect of ADF on enhancing the CT images. The results showed that ADF increases the segmentation accuracy, where the value of Jaccard index of similarity between the ground truth image and segmented image was increased for all segmentation algorithms, in particular for FCM and GAFCM where similarity percent was up to 88%.

Speeding up Travelling Salesman Problem Using Hybrid Algorithm

Traveling salesman problem (TSP) is one of the most popular optimization problems and attracts great attention from many researchers. Several studies have suggested different approaches to solving two-dimensional TSP. This research will present an enhancement for genetic algorithm by using a 2-opt method to solve TSP, such that we will use a 2-opt method to generate the initial population for genetic algorithm, and the roulette wheel choice strategy, the order crossover operator and 2-opt method for mutation operator will be highlighted.
 The experiments of TSP using five real TSP problems taken from the Traveling Salesman Problem Library (TSPLIB), including eil51, st70, pr76,eil76, rd100 (the numbers attached to the problem names represent the number of cities). The error rate of proposed algorithm is (2.233818, 0.030299, 0.057161, 0.084524, 0.123523) . While the error rate of the traditional GA is (2.300088, 0.052921, 0.064016, 0.091775, 0.14666). The results showed the proposed algorithm is acquired higher quality solutions within a reasonable computational time compared to the traditional genetic algorithm. The programming language Pythom3 was used in programming the proposed method.

Evolutionary-based image encryption using RNA codons truth table

Symmetric image cryptography is a mechanism in which image pixels are encrypted into some meaningless format called cipher image. Only authorized users have access to the secret code for symmetric key encryption. In this paper a new symmetric image encryption method has been proposed using the concepts of ribonucleic acid (RNA) sequence and genetic algorithm (GA), called RNA-GA. The proposed method starts by generating specified number of initial cipher images using logistic map function. Then, the initial cipher images are converted to the corresponding one-dimensional binary sequences and relevant codons array using codons truth table. The codons array are then updated using encryption key and encryption RNA tables to form the initial population of genetic algorithm. Next, genetic algorithm optimizes the population using selection, crossover, and mutation operators. The results approve high resistance of the proposed method against well-known attacks and entropy of 7.9987 regarding standard image of size 256×256 after 30 repetitions.

An Optimal Fuzzy PID Controller Design Based on Conventional PID Control and Nonlinear Factors

This paper proposes an optimal fuzzy proportional–integral–derivative (PID) controller design based on conventional PID control and nonlinear factors. With the equivalence between fuzzy logic controllers (FLCs) and conventional PID controllers, a conventional PID controller design can be rapidly transformed into an equivalent FLC by defining the operating ranges of the input/output of the controller. The proposed nonlinear factors can further tune the nonlinearity of the membership functions (MFs) distributed in the operating ranges. In this manner, a fuzzy PID controller can be developed with less parameters and optimized by using the genetic algorithm (GA). In addition, the aforementioned equivalent FLC can act as one individual in the initial population of GA, and significantly enhances the GA efficiency. Simulation results demonstrate the feasibility of this technique. This resulted in an optimal fuzzy PID controller design with only eight parameters with a concise controller structure, and most importantly, the optimal fuzzy PID controller design is now more systematic.

A New Priority Rule for Solving Project Scheduling Problems

Priority rule-based scheduling technique is a scheduling method for constructing minimum feasible schedules when solving project scheduling problems. This approach is made up of two parts: a priority rule for determining the activity list and a schedule generation scheme which constructs the feasible schedule of the constructed activity list. Quite a number of priority rules are available, selecting the best one for a particular input problem is extremely difficult. Hence, we present a new priority rule which assembles a set of priority rules and uses machine learning to form a hybrid strategy out of the assembled strategies. The hybrid strategy operates by choosing the strategy with the best performance at every point in time to construct an activity list of a project. The one with better performance is used most frequently. This removes the problem of manually searching for the best priority rule amongst the dozens of rules that are available. Experimentally, we solved a fictitious single-mode resource-constrained project scheduling problem (single-mode RCPSP) which was solved with 13 different priority rules in Pm Knowledge Center. Our result showed that the total completion time of the project obtained with our approach competes favorably with the completion times gotten with the 13 priority rules. Additionally, we computed initial population for Genetic Algorithm in solving some multi-mode RCPSP. We compared our results with the initial solutions the authors started with, and our results competes favorably with their initial solutions, making our algorithm a good entry point for metaheuristic procedures.

Solving N-Queens Problem Using Subproblems based on Genetic Algorithm

<span lang="EN-US">Nowadays, permutation problems with large state spaces and the path to solution is irrelevant such as N-Queens problem has the same general property for many important applications such as integrated-circuit design, factory-floor layout, job-shop scheduling, automatic programming, telecommunications network optimization, vehicle routing, and portfolio management. Therefore, methods which are able to find a solution are very important. Genetic algorithm (GA) is one the most well-known methods for solving N-Queens problem and applicable to a wide range of permutation problems. In the absence of specialized solution for a particular problem, genetic algorithm would be efficient. But holism and random choices cause problem for genetic algorithm in searching large state spaces. So, the efficiency of this algorithm would be demoted when the size of state space of the problem grows exponentially. In this paper, the subproblems used based on genetic algorithm to cover this weakness. This proposed method is trying to provide partial view for genetic algorithm by locally searching the state space. This method works to take shorter steps toward the solution. To find the first solution and other solutions in N-Queens problem using proposed method: dividing N-Queens problem into subproblems, which configuring initial population of genetic algorithm. The proposed method is evaluated and compares it with two similar methods that indicate the amount of performance improvement.</span>

Train speed profiles optimization using a genetic algorithm based on a random-forest model to estimate energy consumption

The most important part of the train's energy is consumed by the traction system. The tractive energy depends mainly on the driving behaviour. Improving driving strategies has great potential to enhance the energy efficiency. This paper presents a speed profile optimization approach based on a genetic algorithm. The objective of the genetic algorithm is to find, for each interstation, the best speed profile which minimizes the energy consumption. The optimized profile takes into account both the physical and the operational constraints such as the maximum allowed travel time, the speed limitations per section and the maximum allowed acceleration and jerk. The fitness function is based on a Random Forest model which is built using on-board measurements. The aim of the model is to estimate accurately the energy consumption corresponding to each speed profile. The initial population of genetic algorithm is mainly composed of the best realistic speed profiles extracted from the collected data.

Computing Regions of Stability for Limit Cycles of Piecewise Affine Systems

This paper proposes an improved algorithm to compute the regions of stability for limit cycles of piecewise affine systems. Instead of using convex optimization algorithms, such as solving linear matrix inequalities and sum-of-square programming, genetic algorithm (GA) is used to obtain the final results. With the help of GA, both the constraints and objectives can be nonconvex, leading to larger guaranteed regions of stability. Based on impact map and Lyapunov stability theory, the conditions of stability are analyzed. Algorithm-friendly criteria, both convex and nonconvex, are developed. Since randomly generated solutions are usually infeasible, initial population of GA is generated by convex optimization. To improve the diversity of the initial population, multiple convex objectives are used to generate different initial solutions. Other application-specific parts of GA, such as computing the fitness of solutions, are also introduced in detail. An example system is analyzed to illustrate the effectiveness of the proposed method.DOI: http://dx.doi.org/10.5755/j01.itc.46.4.16072

An improved genetic-fuzzy system for classification and data analysis

Interpretability of classification systems, which refers to the ability of these systems to express their behavior in an understandable way, has recently gained more attention and it is considered as an important requirement especially for knowledge-based systems. The main objective of this study is to improve the ability of a well-known fuzzy classifier proposed in Ishibuchi and Nojima (2007) to maximize the accuracy while preserve its interpretability. To achieve the above-mentioned objective, we propose two variants of the original fuzzy classifier. In the first variant classifier, the same components of the original classifier were used except NSGA-II which was replaced by an enhanced version called Controlled Elitism NSGA-II. This replacement aims at improving the ability of the first variant classifier to find non-dominated solutions with better interpretability-accuracy trade-off. In the second variant classifier, we further improve the first variant classifier by enhancing the selection method of the antecedent conditions of the rules generated in the initial population of genetic algorithm. Unlike the method applied in the original classifier and the first variant classifier, which uses a random selection of the antecedent conditions, we proposed a feature-based selection method to favor the antecedent conditions associated with the most relevant features. The results show that the two variant classifiers find more non-dominated fuzzy rule-based systems with better generalization ability than the original method which suggests that Controlled Elitism NSGA-II algorithm is more efficient than NSGA-II. In addition, feature-based selection method applied in the second variant classifier allowed this method to successfully obtain high-quality solutions as it has consistently achieved the best error rates for all the data sets compared to the original method and the first variant classifier.

Hybrid genetic algorithm to solve resource constrained assembly line balancing problem in footwear manufacturing

This paper aims to develop a hybrid genetic algorithm (HGA) to solve the resource constrained assembly line balancing problem (RCALBP) in the sewing line of a footwear manufacturing plant. Sewing, which is the most critical process in footwear manufacturing, has a series of processes, such as punching, trimming, attaching shoelaces. RCALBP in the sewing line considers not only the precedence constraints of product assembly but also the resource constraints, such as operators and equipment. A novel HGA that includes two stages is proposed to optimize the resources in the sewing line. The first stage uses the priority rule-based method (PRBM) to determine the feasible solutions of assigning tasks and machines to workstations. The solutions of PRBM are used to construct the initial population of genetic algorithm (GA) in the second stage. To ensure that the solution of GA is feasible, a two-point-order crossover with the new technique of searching feasible solution patterns is proposed. Moreover, the mutation procedure of GA is modified to avoid the building block from breaking, which may cause unfeasible solutions in RCALBP. A self-tuning method is also applied recursively to exclude unfeasible solutions. The proposed HGA is compared with the manual procedure adopted practically in factories, the existing heuristic model in the literature, and the traditional GA. Based on actual data from a footwear factory, computational results demonstrate that the proposed HGA can achieve better results than the other algorithms.

A Hybrid Approach Based on ACO and Ga for Multi Objective Mobile Robot Path Planning

In this study, we developed an Ant Colony Optimization (ACO) - Genetic Algorithm (GA) hybrid approach for solving the Multi objectives Optimization global path planning (MOPP) problem of mobile robot. The ACO optimization algorithm is used to find the sub-optimal collision free path which then used as initial population for GA. In the proposed modified genetic algorithms, specific genetic operator such as deletion operator is proposed, which is based on domain heuristic knowledge, to fit the optimum path planning for mobile robots. The objective of this study is improving GA performance for efficient and fast selection in generating the Multi objective optimal path for mobile robot navigation in static environment. First we used the proposed approach to evaluate its ability to solve single objective problem in length term as well as we compared it with traditional ACO and simple GA then we extended to solve Pareto optimality ideas based on three criteria: length, smoothness and security, and making it Multi objective Hybrid approach. The proposed approach is tested to generate the single and multi objective optimal collision free path. The simulation results show that the mobile robot travels successfully from one location to another and reaches its goal after avoiding all obstacles that are located in its way in all tested environment and indicate that the proposed approach is accurate and can find a set Pareto optimal solution efficiently in a single run.

Development and analysis of hybrid genetic algorithms for flow shop scheduling with sequence dependent setup time

This paper deals with the development and analysis of hybrid genetic algorithms for flow shop scheduling problems with sequence dependent setup time. A constructive heuristic called setup ranking algorithm is used for generating the initial population for genetic algorithm. Different variations of genetic algorithm are developed by using combinations of types of initial populations and types of crossover operators. For the purpose of experimentation, 27 group problems are generated with ten instances in each group for flow shop scheduling problems with sequence dependent setup time. An existing constructive algorithm is used for comparing the performance of the algorithms. A full factorial experiment is carried out on the problem instances developed. The best settings of genetic algorithm parameters are identified for each of the groups of problems. The analysis reveals the superior performance of hybrid genetic algorithms for all the problem groups.

Network Partitioning Domain Knowledge Multiobjective Application Mapping for Large-Scale Network-on-Chip

This paper proposes a multiobjective application mapping technique targeted for large-scale network-on-chip (NoC). As the number of intellectual property (IP) cores in multiprocessor system-on-chip (MPSoC) increases, NoC application mapping to find optimum core-to-topology mapping becomes more challenging. Besides, the conflicting cost and performance trade-off makes multiobjective application mapping techniques even more complex. This paper proposes an application mapping technique that incorporates domain knowledge into genetic algorithm (GA). The initial population of GA is initialized with network partitioning (NP) while the crossover operator is guided with knowledge on communication demands. NP reduces the large-scale application mapping complexity and provides GA with a potential mapping search space. The proposed genetic operator is compared with state-of-the-art genetic operators in terms of solution quality. In this work, multiobjective optimization of energy and thermal-balance is considered. Through simulation, knowledge-based initial mapping shows significant improvement in Pareto front compared to random initial mapping that is widely used. The proposed knowledge-based crossover also shows better Pareto front compared to state-of-the-art knowledge-based crossover.