Genetic Algorithm Search Method Research Articles

Crack classification and quantitative evaluation based on dimensionality reduction optimization model of multifeature weak magnetic signal

In this paper, we propose a classification and quantitative method based on the optimization model of dimensionality reduction for detecting crack defects in superalloy turbine disks using multifeature weak magnetic signals. First, the experiment on magnetic weak nondestructive testing was conducted on the defects, resulting in the acquisition of 112 sets of preset magnetic anomaly eigenvalue data. The dimensionality of the multidimensional weak magnetic signal features was then reduced using the principal component analysis method to construct the weak magnetic feature sample library. On this basis, a support vector machine (SVM) model was established to classify and quantitatively evaluate the defects. Genetic algorithm and grid search method were used to optimize the model. Finally, the validity of the optimized SVM model was verified by adding artificial and natural crack defect specimens outside of the sample library. Results show that compared with the original model, the feature dimension reduction and optimization of the two methods enhance the classification accuracy of superalloy surface defects and the quantitative accuracy of defect length, width, and depth. The genetic algorithm significantly improved the classification accuracy, increasing it by 39.33 %. Specifically, the genetic algorithm improved the quantitative accuracy of length and width by 17.9 % and 28.56 % respectively. On the other hand, the grid search method improved the quantitative accuracy of depth the most, with an improvement of 49.83 %. For specimens with natural cracks, the optimized SVM model still demonstrated good classification and quantitative effectiveness in detecting defects, with an accuracy rate of over 85 %.

Electric Vehicle Routing Problem with Fuzzy Time Windows using Genetic Algorithm and Tabu Search


 
 
 
 The distribution of goods becomes a very calculated thing in the economic aspect, especially in the case of wide and complex distribution. The greater the range of distribution of goods, the more precise, fast, and accurate calculations are needed. Specifically, the calculation of the distribution required starts from mileage, total travel time, customer satisfaction level based on customer time windows, and operational costs. Vehicle Routing Problem (VRP) is a solution to the problem of distributing goods from the depot to its customers. This study aims to determine the optimal route. The methods used for VRP optimization are the Genetic Algorithm (GA) and Tabu Search (TS) methods. Fuzzy logic is used to provide leeway on the limitations of the time windows parameters, thus providing a time tolerance in the event of early arrival of the vehicle or delay in delivery. Data processing using the GA-TS combination was carried out as many as two types of trials, namely trials with the same dataset ten times and trials with various types of datasets ten times. The results of the first trial fitness value on E-VRPFTW average increased by 14.39% compared to the results of the E-VRPTW fitness value that did not use fuzzy. The results of the second trial also experienced an average increase of 8.49% compared to the results of the E-VRPTW fitness value that did not use fuzzy. Therefore, the addition of fuzzy logic has an effect in determining the optimum route of E-VRPTW.
 
 
 

Design Optimization of Angular Contact Ball Bearing Using Genetic Algorithm and Grid Search Method

Abstract The present work proposes an approach to design and optimize the internal geometry of angular contact ball bearings (ACBBs) with the help of the genetic algorithm (GA) and the grid search method (GSM). Rolling bearings have very complex internal geometry and have varied types. In literature among other types of bearings, very few authors attempted optimization of ACBBs, and subjected to very few constraints. Additionally, none of them considered the free contact angle as a design parameter, which is very important. So, the main motivation of the present work is to see the effect of inclusion of free contact angle in the optimum design of ACBBs. To optimize the geometry of ACBBs, in present study, we have taken a set of 11 design variables out of which 5 are geometric design variables and the rest are constraint parameters. The solution space has been bounded using a set of 20 realistic constraints. This way performing the same analysis with more realistic constraint model of the bearing, increases the feasibility and reliability of the optimized solutions. The results obtained from the genetic algorithm and the grid search method for the dynamic capacity of the bearing were compared to make a evaluate performance of two approaches. The simulation is performed on a series of selected ACBBs. Using the GA, for 7006 ACBB 83% improvement in its dynamic capacity over its rated values has been achieved, while using the GSM improvement is 43%. However, time taken by the GA over the GSM is one-sixth. A sensitivity analysis is done to check the most affecting parameters for the selected objective function. The present methodology helps the designers to reduce the design time, and select the optimum values of design parameters for manufacturing the custom bearing by prioritizing a specific objective.

Machine Learning-Based Comparative Study For Heart Disease Prediction

Heart disease is one of the most common causes of death globally. In this study, machine learning algorithms and models widely used in the literature to predict heart disease have been extensively compared, and a hybrid feature selection based on genetic algorithm and tabu search methods have been developed. The proposed system consists of three components: (1) preprocess of datasets, (2) feature selection with genetic and tabu search algorithm, and (3) classification module. The models have been tested using different datasets, and detailed comparisons and analysis were presented. The experimental results show that the Random Forest algorithm is more successful than Adaboost, Bagging, Logitboost, and Support Vector machine using Cleveland and Statlog datasets.

Enhanced classification and regression tree (CART) by genetic algorithm (GA) and grid search (GS) for flood susceptibility mapping and assessment

Despite the high level of interpretability of the Classification and Regression Tree (CART) as one of the widely-used data mining models, its results usually are less accurate than other models. To address the low accuracy of the CART model, this paper developed two new and enhanced CART-based algorithms that can provide both high interpretability and high accuracy for flood susceptibility mapping. In the developed models, various hyperparameters of the CART were explicitly tuned using the genetic algorithm (GA) and grid search (GS) methods. Eleven variables and 222 flood locations were used in Sardabroud watershed in Iran for modelling. The results showed area under the receiver operating characteristic curve (AUROC) values of 0.884, 0.927 and 0.908, respectively, for the CART, CART-GS and CART-GA models in the validation run. The results showed that the developed models can produce the output that satisfies both accuracy and interpretability criteria.

A memetic algorithm for the inventory routing problem

In this article, we study an Inventory Routing Problem with deterministic customer demand in a two-tier supply chain. The supply chain network consists of a supplier using a single vehicle with a given capacity to deliver a single product type to multiple customers. We are interested in population-based algorithms to solve our problem. A Memetic Algorithm (MA) is developed based on the Genetic Algorithm (GA) and Variable Neighborhood Search methods. The proposed meta-heuristics are tested on small and large reference benchmarks. The results of the MA are compared to those of the classical GA and to the optimal solutions in the literature. The comparison shows the efficiency of using MA and its ability to generate high quality solutions in a reasonable computation time.

FACTORS AFFECTING THE THICKNESS OF REPLACEMENT LAYER ON MEDIUM CLAY

Soil replacement is a common technique that can be used to increase the soil bearing capacity and reduce the expected settlement. The thickness of replacement layer depends on many factors such as: the applied stress, original soil properties, material of replacement layer and the cost of foundation works. However, until now the practical thickness of replacement is usually selected based on soil experts’ experience. This study proposed an optimization model to assist geotechnical engineers in predicting the optimum thickness and material type of replacement layer that satisfy the main design requirements, i.e. bearing capacity, consolidation settlement and cost considerations at the same time. The Evolutionary solving method that uses a variety of genetic algorithm and local search methods was used to solve the research problem. Furthermore, the effect of the thickness and properties of clay layer and the depth of ground water table on determining optimum type and thickness of replacement soil were investigated. The study evaluated the relationship between the replacement layer thickness and the total direct cost of foundation works and found that, the notion of increasing replacement thickness to decrease cost limitlessly was not viable and an optimal thickness was usually achieved.

Development, optimization, and evaluation of a hybrid passive buoyancy compensation system for underwater gliders

Underwater glider (UG) is one of the most promising devices for ocean sampling. However, when the UG is operating in a vast ocean space, its performance in terms of motion stability, energy efficiency and endurance is affected due to the buoyancy variation caused by the change of displacement and seawater density. It is essential to compensate for the buoyancy variation for performance improvement of the UG. This paper proposes a hybrid passive buoyancy compensation system (HPBCS), integrating three standalone buoyancy compensation devices of spring accumulator, inflatable accumulator, and silicon oil bladder, to compensate for the buoyancy variation. The mathematic compensation model of this system is established and an effective optimization algorithm hybridizing genetic algorithm and tabu search method is adopted to optimize the design parameters. The case study of a Petrel-II underwater glider shows that the optimal HPBCS can significantly improve the buoyancy compensation. Then the motion stability, energy efficiency and endurance of the UG with HPBCS are evaluated and compared with those of other UGs. Results indicate that the UG with HPBCS has a more superior performance in motion stability and energy efficiency, but its endurance is affected by the mass brought by HPBCS.

An integrated optimum 3D transition orbit design procedure with guidance for a spacecraft

In this paper, an integrated optimal flight plan with a guidance plan in a general transfer orbit is considered. Due to the need to simultaneously reduce energy consumption and flight time in some space missions, the spacecraft trajectory is optimized based on two criteria to minimize fuel consumption and flight time in the orbital transfer process. For this purpose, the total velocity impulse on both sides of the transition orbit and the elapsed time are defined as the criteria of the objective function which includes the spacecraft’s insertion conditions and orbital parameters related to the shape of the trajectory. The optimal transition (coasting) trajectory is then obtained using the multi-objective genetic algorithm (MOGA) search method. Subsequently, the velocity-to-be-gained guidance steering scheme is integrated to track the trajectory onboard and reach the final orbit. The numerical results of the proposed algorithm are extracted in three case studies and compared with other references that show its ability to find the optimal transfer.

Comparison of the genetic algorithm and pattern search methods for forecasting optimal flow releases in a multi-storage system for flood control

This paper compares the well-known genetic algorithm (GA) and pattern search (PS) optimization methods for forecasting optimal flow releases in a multi-storage system for flood control. The simulation models used by the optimization models include (a) a batch of scripts for data acquisition of forecasted precipitation and their automated post-processing; (b) a hydrological model for rainfall-runoff conversion, and (c) a hydraulic model for simulating river inundation. This paper focuses on (1) demonstrating the application of the framework by applying it to the operation of a hypothetical eight-wetland system in the Cypress Creek watershed in Houston, Texas; and (2) comparing and discussing the performance of the two optimization methods under consideration. The results show that the GA and PS optimal solutions are very similar; however, the computational time required by PS is significantly shorter than that required by GA. The results also show that optimal dynamic water management can significantly mitigate flooding compared to the case without management.

Hybrid Genetic Algorithm With Haar Wavelet for Maximum Target Coverage Node Deployment in Wireless Sensor Networks

Wireless sensor networks (WSNs) are used in industrial applications and focused on target coverage and node connectivity based WSNs. The set of sensors and targets is placed in optimal position the target coverage and node connectivity achieving maximum with limited senor nodes. To resolve this problem, the proposed hybrid genetic algorithm combined with lifting wavelet multi-resolution principles for recognizing optimal position for sensors to cover entire targets present in the fields. The hybrid genetic algorithm randomly identifies each sensor position and 2D Haar lifting wavelet transform to improve the quality of target coverage by adjusting node position. The 2D Haar lifting decomposes the population matrix into the optimal position of sensors. Experimental results show the performance of the proposed hybrid genetic algorithm and fast local search method compared with available algorithms improves the target coverage and the number of nodes with varying and fixed sensing ranges with a different region.

VIDEO ANALYSIS AND 3D DETECTION SIMULATION OF JUMP SHOT PRECISION FOR BASKETBALL PLAYERS

ABSTRACT With the rapid development and application of computer technology, the application of computer science knowledge in basketball is also more and more extensive. Based on genetic algorithm and the background subtraction method, video analysis and 3D detection simulation model of shot jump action precision were constructed in this study. According to the genetic algorithm search method, jump shot precision was analyzed, and the problems encountered in the actual shooting process of basketball players were studied and solved. The results show that this study is necessary and feasible.

Combine the AGA and L-M algorithms to predict the oil-water flow profile of production wells

Production logging is the main means of dynamic monitoring of oil wells at present. The interpretation method of production logging fluid profiles is the key to the dynamic evaluation of oil wells and is of great significance. A joint optimization technique of a AGA and L-M is proposed to deal with the profile data of oil-water two-phase flow. The method is based on the energy conservation of the fluid in the production unit space. According to the nonlinear weighted least squares principle and the error theory, the objective function of the minimum value is established by the difference between the theoretical fluid temperature and the measured value of the thermometer. According to the fluid volume conversion, the flow rate of each phase in the well and the flow log value at the bottom of the control body are constrained. In order to improve the interpretation accuracy, the global search method (AGA) of genetic algorithm and the direct local search method of L-M algorithm are combined to solve the optimal solution. Firstly, the adaptive genetic algorithm is used for heuristic global search, and then as a result of the initial parameters, the L-M method is applied to approach the optimal solution. The interpretation results from the joint inversion are stable and reliable. This method is not only suitable for vertical wells but also for inclined wells.

Trihedral lattice towers geometry optimization

The problem of trihedral lattice towers geometry optimization, the width of which varies linearly with height, has been considered. The variable parameters were the support width at the base and top point, as well as the cross-sectional areas of the lattice and chords. A restriction was introduced on the mass constancy. The objective function was the potential strain energy, the maximum displacement and the first frequency of natural vibrations. In the first and second cases the optimum corresponds to the minimum of the objective function, and in the third - to the maximum. The solution was performed by the finite element method in combination with four nonlinear optimization methods: the interior point method, surrogate optimization method, genetic algorithm and pattern search method. The efficiency of each of the listed optimization methods has been compared by the authors.

Congestion and Pollution, Vehicle Routing Problem of a Logistics Provider in Thailand

Aim and Objective: This study aims to minimise the travelling distance, operation cost in terms of fuel consumption, and CO2 emissions. It introduces the Time-Dependency Pollution-Routing Problem (TDPRP) with the implementation of the time-dependency and emission model, including constraints such as the limitation of vehicle capacity and vehicle’s speed during different time periods in Thailand. Furthermore, the time window constraint is applied for representing a more realistic model. The main objective is to minimise the total pollution generated because of transportation. Methods: The Genetic Algorithm (GA) and Tabu Search (TS) methods have been used to generate the optimal solution with a variety of experiments. The best solutions from all the experiments have been compared to the original solution in terms of the quality of the solution and the computation time. Results: The best solution was generated by using the TS method with 30,000 trials. The minimum of the total CO2 emissions was 183.9846 kilograms produced from all of the vehicles during transportation, nearly half from the current transportation plan, which produced 320.94 kilograms of CO2 emissions. Conclusion: The proposed model optimised both the route and schedules (multiple time periods) for a number of vehicles, for which the transportation during a fixed congestion period could be predicted to avoid traffic congestion and reduce the CO2 emission. Future research is suggested to add other specific algorithms as well as constraints in order to make the model more realistic.

A Multi-Objective Task Reallocating Method in Complex Product Design Process Considering Design Changes

In the highly dynamic design environments, task allocation is subject to considerable design changes. The task reallocating to cope with various unexpected design changes becomes an increasingly important issue in the complex product design (CPD). In this paper, we propose a task reallocating model based on an adaptive multi-objective genetic algorithm and Tabu search (AMOGATS) method to study the dynamic task allocating procedure considering design changes, which integrates the advantages of adaptive genetic algorithm and Tabu search algorithm. Three objectives, i.e. completion time, robustness and stability, are considered simultaneously to measure the task allocation performance. A real example is employed to test and evaluate the performance of proposed method. The computational results imply that the proposed AMOGATS performs better than the heuristic algorithms available in the literature, which has more advantages on the convergence rate and running efficiency than the other algorithms, along with a better solution. This work provides a useful decision support to carry out the task reallocating with high levels of flexibility and efficiency during the process of CPD.

Parameter Estimation for a Model of Ionizing Radiation Effects on Targeted Cells using Genetic Algorithm and Pattern Search Method

A mechanistic model has been used to explain the effect of radiation. The model consists of parameters which represent the biological process following ionizing radiation. The parameters in the model are estimated using local and global optimization algorithms. The aim of this study is to compare the efficiency between local and global optimization method, which is Pattern Search and Genetic Algorithm respectively. Experimental data from the cell survival of irradiated HeLa cell line is used to find the minimum value of the sum of squared error (SSE) between experimental data and simulation data from the model. The performance of both methods are compared based on the computational time and the value of the objective function, SSE. The optimization process is carried out by using the built-in function in MATLAB software. The parameter estimation results show that genetic algorithm is more superior than pattern search for this problem.

Improved Genetic Algorithm with Local Search for Satellite Range Scheduling System and its Application in Environmental monitoring

Satellites play an important role in such areas as environmental monitoring and disaster prediction that are related to human survival and development. Satellite monitoring and control is the key to the satellite's management and control of the ground to ensure its smooth implementation. Because of the imbalance of demand and available resources, satellite range scheduling has become particularly important. This paper analyzes the satellite range scheduling problem and sets up mathematical models and constraints. Afterwards, this paper proposes an efficient algorithm that combines improved genetic algorithm and local search method. The improved genetic algorithm is used to rapidly improve the quality of the planning scheme, and the neighborhood search is used for the subsequent small-scale optimization. In order to improve the speed of search, our algorithm uses a reorganization operation and a mutation operation adjusted with the number of iterations. In order to test the effectiveness of the algorithm, we conducted experimental verifications of the calculations of various types of satellites at different mission scales and compared them with other algorithms.

Combined data mining/NIR spectroscopy for purity assessment of lime juice

This paper reports the data mining study on the NIR spectrum of lime juice samples to determine their purity (natural or synthetic). NIR spectra for 72 pure and synthetic lime juice samples were recorded in reflectance mode. Sample outliers were removed using PCA analysis. Different data mining techniques for feature selection (Genetic Algorithm (GA)) and classification (including the radial basis function (RBF) network, Support Vector Machine (SVM), and Random Forest (RF) tree) were employed. Based on the results, SVM proved to be the most accurate classifier as it achieved the highest accuracy (97%) using the raw spectrum information. The classifier accuracy dropped to 93% when selected feature vector by GA search method was applied as classifier input. It can be concluded that some relevant features which produce good performance with the SVM classifier are removed by feature selection. Also, reduced spectra using PCA do not show acceptable performance (total accuracy of 66% by RBFNN), which indicates that dimensional reduction methods such as PCA do not always lead to more accurate results. These findings demonstrate the potential of data mining combination with near-infrared spectroscopy for monitoring lime juice quality in terms of natural or synthetic nature.

Comparison of Genetic Algorithm and Harmony Search Method for 2D Geometry Optimization.

There are many applications of Genetic Algorithm (GA) and Harmony Search (HS) Method for solving problems in civil engineering design. The question is, still, which method is better for geometry optimization of a steel structure. The purpose of this paper is to compare GA and HS performance for geometric optimization of a steel structure. This problem is solved by optimizing a steel structure using GA and HS and then comparing the structure’s weight as well as the time required for the calculation. In this study, GA produced a structural weight of 2308.00 kg to 2387.00 kg and HS scored 2193.12 kg to 2239.48 kg. The average computational time required by GA is 607 seconds and HS needed 278 seconds. It concludes that HS is faster and better than GA for geometry optimization of a steel structure.