Multi-objective Genetic Approach Research Articles

A multi-objective dual dynamic genetic algorithm-based approach for thermoelectric optimization of integrated urban energy systems

In order to reduce the loss of energy in the process of conversion and utilization, and to improve the energy utilization effect and economic benefits of urban integrated energy system, a thermoelectricity optimization method of urban integrated energy system based on multi-objective double dynamic genetic algorithm is proposed. The method analyzes the operation characteristics of the urban integrated energy system through its operation model, constructs a multi-objective optimization model for the thermoelectricity of the urban integrated energy system, determines the optimization multi-objective function that minimizes the operation cost, minimizes the carbon emission, minimizes the technical dissatisfaction and the related constraints, and then solves the optimization model by using the dual dynamic genetic algorithm to output the results of the optimization of the thermoelectricity of the urban integrated energy system.The test results show that the algorithm has a crowding distance of 0.75 or above when solving, and the overall unit energy consumption cost reduction ratio is about 22.97 %. The state of charge results of the four energy storage power stations are all below 10 %, and the lowest convergence speed is only 7 seconds, effectively improving energy utilization efficiency.

VM queuing optimal scheduling in cloud using heuristic ant colony optimal based multi-objective genetic approach

The usages of cloud based applications are increasing tremendously. The cloud computing task distribution is an unknown polynomial time issue that is challengeable to find the optimal solution. In solve above mentioned issue with large amount user’s job requests, heuristic ant colony optimal based multi-objective genetic (HACOMOG) approach based job allocation and resource optimization is proposed. Utilization basis scheduler recognizes the task order and optimal resources to be scheduled. The primary contribution of the proposed technique is to develop several online techniques to find solution for the virtual machines (VM) Packing problem sharing-aware and for performing a comprehensive number of studies in order to assess their efficiency with online sharing algorithms. The proposed algorithm considers the utilization basis scheduler output and identified the optimzed task allocation technique based on job execution time, MakeSpan and throughput. The experimental outcomes show that the proposed HACOMOG Algorithm reduces 0.70 seconds job execution time (JET), 0.13 MakeSpan and improve 1.98 throughput on given parameters for 100, 200, and 500 tasks with conventional methodologies.

Hybrid Multi-Objective Optimization Approach With Pareto Local Search for Collaborative Truck-Drone Routing Problems Considering Flexible Time Windows

The collaboration of drones and trucks for last-mile delivery has attracted much attention. In this paper, we address a collaborative routing problem of the truck-drone system, in which a truck collaborates with multiple drones to perform parcel deliveries and each customer can be served earlier and later than the required time with a given tolerance. To meet the practical demands of logistics companies, we build a multi-objective optimization model that minimizes total distribution cost and maximizes overall customer satisfaction simultaneously. We propose a hybrid multi-objective genetic optimization approach incorporated with a Pareto local search algorithm to solve the problem. Particularly, we develop a greedy-based heuristic method to create initial solutions and introduce a problem-specific solution representation, genetic operations, as well as six heuristic neighborhood strategies for the hybrid algorithm. Besides, an adaptive strategy is adopted to further balance the convergence and the diversity of the hybrid algorithm. The performance of the proposed algorithm is evaluated by using a set of benchmark instances. The experimental results show that the proposed algorithm outperforms three competitors. Furthermore, we investigate the sensitivity of the proposed model and hybrid algorithm based on a real-world case in Changsha city, China.

One instruction set computer with optimised polarity‐tunable model of double gate CNTFETs

Emerging devices such as double gate carbon nanotube field effect transistors (DG CNTFETs) have opened up manifold possibilities for reconfigurable logic design. The thickness of gate oxide and the employment of inhomogeneous dielectrics over and under the carbon nanotubes (CNTs) impact the operation of DG CNTFETs. In this work, the dielectric constant and the thickness of the gate oxide are optimised to suppress the ambipolar conduction in CNTFETs. A multi-objective genetic algorithm-based approach is proposed to optimise these parameters. The contributions in this study are two-fold: firstly, a DG CNTFET is fabricated with the optimised parameters. By exploiting the ability to select the conduction behaviour using a second gate in the fabricated DG CNTFETs, the device is electrostatically programmed to behave as an n or p-type CNTFET. Secondly, a one instruction set computer is simulated with the optimised model of DG CNTFET. A universal static logic cell which implements 16 logic functions, a D-Latch and a D-FF were built with DG CNTFETs. The logic circuits with polarity-tunable DG CNTFETs outperform other logic structures. There is a 40% betterment in performance primarily due to the reduced number of logic levels and by 25% due to the reduced delay.

Classifier Design by a Multi-Objective Genetic Algorithm Approach for GPR Automatic Target Detection

GPR is an electromagnetic remote sensing technique, used for detection of relatively small objects in high noise environments. Data inversion requires a fitting procedure of hyperbola signatures, which represent the target reflections, sometimes producing bad results due to high resolution of GPR images. The idea proposed in this paper consists of narrowing down the position of hyperbolas to small regions, using a machine learning approach. A Multi-Objective Genetic Approach (MOGA) is used to design a Radial Basis Function classifier. High order statistic cumulants are employed as features to this framework. Due to the complexity of the formulated problem, feature selection can be done in two ways: either by MOGA alone, or acting on a reduced subset obtained using a mutual information approach. The chosen classifier was tested on experimental data, the results outperforming the one presented in literature, or achieving similar results with models of much lower complexity.

Synchronization control of oscillator networks using symbolic regression

Networks of coupled dynamical systems provide a powerful way to model systems with enormously complex dynamics, such as the human brain. Control of synchronization in such networked systems has far reaching applications in many domains, including engineering and medicine. In this paper, we formulate the synchronization control in dynamical systems as an optimization problem and present a multi-objective genetic programming-based approach to infer optimal control functions that drive the system from a synchronized to a non-synchronized state and vice-versa. The genetic programming-based controller allows learning optimal control functions in an interpretable symbolic form. The effectiveness of the proposed approach is demonstrated in controlling synchronization in coupled oscillator systems linked in networks of increasing order complexity, ranging from a simple coupled oscillator system to a hierarchical network of coupled oscillators. The results show that the proposed method can learn highly-effective and interpretable control functions for such systems.

Hull mixed-model assembly line balancing using a multi-objective genetic algorithm simulated annealing optimization approach

Hull assembly line balancing has significant impact on performance of shipbuilding system and is usually a multi-objective optimization problem. In this article, the primary objectives of the hull assembly line balancing are to minimize the number of workstations, to minimize the static load balancing index, to minimize the dynamic load balancing index between workstations, and to minimize the multi-station-associated complexity. Because this problem comes under combinatorial optimization category and is non-deterministic polynomial-time hard, an improved genetic algorithm simulated annealing is presented. In genetic algorithm simulated annealing, the task sequence numbers are used as chromosomes, and selection, crossover, and mutation operators only deal with the elements of task set instead of the ones of the problem space. In order to prevent the algorithm appearing early convergence or getting local optimal result, the simulated annealing algorithm is used to deal with the individuals. Meanwhile, the algorithm is embedded with the hierarchical scheduling tactics in order to solve the selection problem on optimal solution in the Pareto-optimal set. A number of benchmark problems are solved to prove the superior efficiency of the proposed algorithm. Finally, a case study of the optimization of a hull assembly line was given to illustrate the feasibility and effectiveness of the method.

Multiobjective Genetic Optimization Approach to Identify Pipe Segment Replacements and Inline Storages to Reduce Sanitary Sewer Overflows

Sanitary sewer overflows (SSOs) is the unintentional discharge of untreated sewage from the sanitary sewer system and pose serious risk to public health and to the environment. Rehabilitation plans to reduce SSOs involve increasing conveyance capacity and shaving peak flow using detention storages. Identifying the best location for rehabilitating the sanitary sewer network is a difficult task because of the great length of sanitary sewer systems. This study utilized single and multiobjective genetic algorithms (GAs) to design rehabilitation strategies for SSOs reduction in an existing sewer network. The Nondominated Sorting Genetic Algorithm II was linked to the EPA-SWMM to generate non-dominated sets of solutions that characterizes the tradeoffs between reduction in number of SSOs and cost (Case I), and the tradeoff between of volume of SSOs and cost (Case II). The results show that, when maximizing the reduction of number SSOs, the algorithm target first regions of the network with higher density of SSOs. When maximizing the reduction of volume of SSOs, the solutions prioritize the nodes with the largest overflow volumes. The tested approach provides a range of options to decision makers that seek to reduce or eliminate SSOs in an existing sanitary sewer system.

Optimised recrystallisation model using multiobjective evolutionary and genetic algorithms and k-optimality approach

The meta-models are constructed for static recrystallisation of dual phase steels using evolutionary neural nets (EvoNN). Four mutually conflicting objectives—(i) overall kinetics, (ii) grain size, (iii) the amount of strain and (iv) the precipitate volume fraction—are optimised simultaneously using an emerging k-optimal approach incorporated in the EvoNN, using a predator–prey genetic algorithm. The first objective involved minimisation of error with respect to experimental observation. The grain size and the amount of strain were minimised, whereas the precipitate volume fraction was maximised. The aim is to control the recrystallisation process in order to achieve desired material properties of dual phase steel during the final stages of heat treatment.

Recognizing Surgically Altered Face Images and 3D Facial Expression Recognition

Altering Facial appearances using surgical procedures are common now days. But it raised challenges for face recognition algorithms. Plastic surgery introduces non linear variations. Because of these variations it is difficult to be modeled by the existing face recognition system. Here presents a multi objective evolutionary granular algorithm. It operates on several granules extracted from a face images at multiple level of granularity. This granular information is unified in an evolutionary manner using multi objective genetic approach. Then identify the facial expression from the face images. For that 3D facial shapes are considering here. A novel automatic feature selection method is proposed based on maximizing the average relative entropy of marginalized class-conditional feature distributions and apply it to a complete pool of candidate features composed of normalized Euclidian distances between 83 facial feature points in the 3D space. A regularized multi-class AdaBoost classification algorithm is used here to get the highest average recognition rate.

A multi-objective genetic optimization of interpretability-oriented fuzzy rule-based classifiers

Abstract The paper presents a multi-objective genetic approach to design interpretability-oriented fuzzy rule-based classifiers from data. The proposed approach allows us to obtain systems with various levels of compromise between their accuracy and interpretability. During the learning process, parameters of the membership functions, as well as the structure of the classifier's fuzzy rule base (i.e., the number of rules, the number of rule antecedents, etc.) evolve simultaneously using a Pittsburgh-type genetic approach. Since there is no particular coding of fuzzy rule structures in a chromosome (it reduces computational complexity of the algorithm), original crossover and mutation operators, as well as chromosome-repairing technique to directly transform the rules are also proposed. To evaluate both the accuracy and interpretability of the system, two measures are used. The first one – an accuracy measure – is based on the root mean square error of the system's response. The second one – an interpretability measure – is based on the arithmetic mean of three components: (a) the average length of rules (the average number of antecedents used in the rules), (b) the number of active fuzzy sets and (c) the number of active inputs of the system (an active fuzzy set or input means a set or input used by at least one fuzzy rule). Both measures are used as objectives in multi-objective (2-objective in our case) genetic optimization approaches such as well-known SPEA2 and NSGA-II algorithms. Moreover, for the purpose of comparison with several alternative approaches, the experiments are carried out both considering the so-called strong fuzzy partitions (SFPs) of attribute domains and without them. SFPs provide more semantically meaningful solutions, usually at the expense of their accuracy. The operation of the proposed technique in various classification problems is tested with the use of 20 benchmark data sets and compared to 11 alternative classification techniques. The experiments show that the proposed approach generates classifiers of significantly improved interpretability, while still characterized by competitive accuracy.

Parameter Space Representation of Pareto Front to Explore Hardware-Software Dependencies

Embedded systems design requires conflicting objectives to be optimized with an appropriate choice of hardware-software parameters. A simulation campaign can guide the design in finding the best trade-offs, but due to the big number of possible configurations, it is often unfeasible to simulate them all. For these reasons, design space exploration algorithms aim at finding near-optimal system configurations by simulating only a subset of them. In this work, we present PS, a new multiobjective optimization algorithm, and evaluate it in the context of the embedded system design. The basic idea is to recognize interesting regions—that is, regions of the configuration space that provide better configurations with respect to other ones. PS evaluates more configurations in the interesting regions while less thoroughly exploring the rest of the configuration space. After a detailed formal description of the algorithm and the underlying concepts, we show a case study involving the hardware/software exploration of a VLIW architecture. Qualitative and quantitative comparisons of PS against a well-known multiobjective genetic approach demonstrate that while not outperforming it in terms of Pareto dominance, the proposed approach can balance the uniformity and granularity qualities of the solutions found, obtaining more extended Pareto fronts that provide a wider view of the potentiality of the designed device. Therefore, PS represents a further valid choice for the designer when objective constrains allow it.

An adaptive variable neighborhood search for solving the multi-objective node placement problem

In this paper, we address the multi-objective node placement (MONP) problem that consists in extending an existing heterogeneous network while optimizing three conflicting objectives. Numerous devices' types are to be deployed in an area of interest in order to ensure the network connectivity and to satisfy users' demands. As the MONP problem is NP-Hard, we adapt a new variant of the multiobjective variable neighborhood search (MO-VNS) algorithm. The main idea is to automatically choose the most promising neighborhood structure by associating a rating to each one. This value is dynamically updated according to the number of so far generated non-dominated solutions. The empirical validation is done using a simulation environment called Inform Lab. A comparison to an existing multi-objective genetic approach is performed based on real instances of maritime surveillance application. The experimental results show that both methods have comparable performances based the unary hyper-volume metric (IH−). A slight improvement is noticed for the (MO-VNS) regarding medium sized instances.

Symmetry Based Automatic Evolution of Clusters: A New Approach to Data Clustering.

We present a multiobjective genetic clustering approach, in which data points are assigned to clusters based on new line symmetry distance. The proposed algorithm is called multiobjective line symmetry based genetic clustering (MOLGC). Two objective functions, first the Davies-Bouldin (DB) index and second the line symmetry distance based objective functions, are used. The proposed algorithm evolves near-optimal clustering solutions using multiple clustering criteria, without a priori knowledge of the actual number of clusters. The multiple randomized K dimensional (Kd) trees based nearest neighbor search is used to reduce the complexity of finding the closest symmetric points. Experimental results based on several artificial and real data sets show that proposed clustering algorithm can obtain optimal clustering solutions in terms of different cluster quality measures in comparison to existing SBKM and MOCK clustering algorithms.

Developing a multi-objective genetic optimisation approach for an operational design of a manual mixed-model assembly line with walking workers

A walking worker assembly line (WWAL), in which each cross-trained worker travels along the line to carry out all required tasks, is an example of lean system, specifically designed to respond quickly and economically to the fluctuating nature of market demands. Because of the complexity of WWAL design problems, classical heuristic approaches are not capable of solving problematic design characteristic of WWAL of very large design space. This paper presents a new genetic approach to address the mixed model walking worker manual assembly line optimisation design problem with multiple objectives. The aim is to select a set of operational variables to perform to the required demand for two product models. The goal is to produce the required models at the lowest cost possible, whilst keeping within an ergonomically balanced operation. Genetic algorithms are developed to tackle this problem. This paper describes the fundamental structure of this approach, as well as the influence of the crossover probability, the mutation probability and the size of the population on the performance of the genetic algorithm. The paper also presents an application of a developed algorithm to the operational design problem of plastic electrical box assembly line.

A multi-objective genetic approach to domestic load scheduling in an energy management system

In this paper a multi-objective genetic algorithm is used to solve a multi-objective model to optimize the time allocation of domestic loads within a planning period of 36 h, in a smart grid context. The management of controllable domestic loads is aimed at minimizing the electricity bill and the end-user’s dissatisfaction concerning two different aspects: the preferred time slots for load operation and the risk of interruption of the energy supply. The genetic algorithm is similar to the Elitist NSGA-II (Nondominated Sorting Genetic Algorithm II), in which some changes have been introduced to adapt it to the physical characteristics of the load scheduling problem and improve usability of results. The mathematical model explicitly considers economical, technical, quality of service and comfort aspects. Illustrative results are presented and the characteristics of different solutions are analyzed.

Analysis of Various Multiobjective Genetic Approaches in Association Rule Mining

mining is used now days by companies with a strong consumer focus. It enables these companies to know the relationships among internal factors such as, product positioning, price or staff skills, and external factors such as indicators, economic, competition, and customer demographics. The overall aim of the data mining process is to extract information from a data set and transform it into an understandable structure for further use. In this paper, the multi-objective genetic approach for the result Comparison of Pittsburgh and Michigan approach using multi-objective genetic algorithm has been proposed, and it is shown that using Pittsburgh approach is much better than the Michigan approach. KeywordsPittsburgh, Multi-objective, Genetic Algorithm.

Recognizing Surgically Altered Face Images Using Multiobjective Evolutionary Algorithm

Widespread acceptability and use of biometrics for person authentication has instigated several techniques for evading identification. One such technique is altering facial appearance using surgical procedures that has raised a challenge for face recognition algorithms. Increasing popularity of plastic surgery and its effect on automatic face recognition has attracted attention from the research community. However, the nonlinear variations introduced by plastic surgery remain difficult to be modeled by existing face recognition systems. In this research, a multiobjective evolutionary granular algorithm is proposed to match face images before and after plastic surgery. The algorithm first generates non-disjoint face granules at multiple levels of granularity. The granular information is assimilated using a multiobjective genetic approach that simultaneously optimizes the selection of feature extractor for each face granule along with the weights of individual granules. On the plastic surgery face database, the proposed algorithm yields high identification accuracy as compared to existing algorithms and a commercial face recognition system.

Multi Objective Genetic Approach for Solving Vehicle Routing Problem

Multi Objective Genetic Approach for Solving Vehicle Routing Problem

A Genetic Approach for Optimization Design of Gear System in Co-Rotating Twin Screw Extruder

According to the structure features of gear system in co-rotating twin screw extruder, the model of the gear system is built, and multi-objective optimization genetic approach based on the Pareto-rank is proposed to optimize the parameters of the gear system. The advanced optimal design method for the gear unit in co-rotating twin screw extruder is studied. A typical gear system in co-rotating twin screw extruder is designed, and the results show that the gear system meets the requirements in all aspects. This design method of the gear system is synthetic optimization for the gear system in co-rotating twin screw extruder.