Evolutionary Procedure Research Articles

A new hybrid approach for dynamic continuous optimization problems

A new hybrid approach for dynamic optimization problems with continuous search spaces is presented. The proposed approach hybridizes efficient features of the particle swarm optimization in tracking dynamic changes with a new evolutionary procedure. In the proposed dynamic hybrid PSO (DHPSO) algorithm, the swarm size is varied in a self-regulatory manner. Inspired from the microbial life, the particles can reproduce infants and the old ones die. The infants are especially reproduced by high potential particles and located near the local optimum points, using the quadratic interpolation method. The algorithm is adapted to perform in continuous search spaces, utilizing continuous movement of the particles and using Euclidian norm to define the neighborhood in the reproduction procedure. The performance of the new proposed approach is tested against various benchmark problems and compared with those of some other heuristic optimization algorithms. In this regard, different types of dynamic environments including periodic, linear and random changes are taken with different performance metrics such as real-time error, offline performance and offline error. The results indicate a desirable better efficiency of the new algorithm over the existing ones.

Towards Evolutionary Nonnegative Matrix Factorization

Nonnegative Matrix Factorization (NMF) techniques has aroused considerable interests from the field of artificial intelligence in recent years because of its good interpretability and computational efficiency. However, in many real world applications, the data features usually evolve over time smoothly. In this case, it would be very expensive in both computation and storage to rerun the whole NMF procedure after each time when the data feature changing. In this paper, we propose Evolutionary Nonnegative Matrix Factorization (eNMF), which aims to incrementally update the factorized matrices in a computation and space efficient manner with the variation of the data matrix. We devise such evolutionary procedure for both asymmetric and symmetric NMF. Finally we conduct experiments on several real world data sets to demonstrate the efficacy and efficiency of eNMF.

Memetic algorithm for non-identical parallel machines scheduling problem with earliness and tardiness penalties

We consider the problem of determining the production schedule of n jobs on m non-identical parallel machines so that the sum of earliness and tardiness penalties is minimised. In the problem, each job has its own due date, earliness penalty, and tardiness penalty. To reduce the size of possible solution set, the problem is reduced to the job sequencing problem. The evolutionary procedures are developed, as the search in the global optimisation phase, to properly assign jobs to each machine and determine job sequence. In the local optimisation phase, the optimal timing algorithm is applied to determine the best schedule of each job sequence. In order to evaluate the performance of the memetic algorithm (MA) at different penalty cost rate and problem size, the solution from the heuristic is compared with the optimal solution obtained from the simple branch and bound method. The results show that the MA is promising.

SOLVING A NEW BI-OBJECTIVE MODEL FOR A CELL FORMATION PROBLEM CONSIDERING LABOR ALLOCATION BY MULTI-OBJECTIVE PARTICLE SWARM OPTIMIZATION

Mathematical programming and artificial intelligence (AI) methods are known as the most effective and applicable procedures to form manufacturing cells in designing a cellular manufacturing system (CMS). In this paper, a bi -objective programming model is presented to consider the cell formation problem that is solved by multi-objective particle swarm optimization (MOPSO) algorithm. The model contains two conflicting objectives, namely optimal labor allocation and maximization of cell utilization. In order to verify its effectiveness of the MOPSO algorithm, the results are compared with those obtained from a well-known evolutionary procedure, called NSGA-II.

Evolutionary procedure based model to predict ground–level ozone concentrations

This study aims to predict the next day hourly average ozone (O3) concentrations using threshold autoregressive (TAR) models in which the threshold value and the threshold variable are defined by genetic algorithms. The procedure is also able to generate models with statistically significant regression parameters. The performance of TAR models was then compared to the one obtained with autoregressive (AR) and artificial neural network (ANN) models. Different TAR models were generated, corresponding to different threshold variables and values. For the training period, ANN model presented better results than TAR and AR models. However, in the test period, AR and one of the TAR models achieved better predictions of O3 concentrations than the ANN model. The distinction between the applied models became greater when they were evaluated in the prediction of the extreme values, for which the TAR model presented the best performance. The performance with respect to extreme values is a useful implication for the protection of public health as this model can provide more reliable early warnings about high O3 concentration episodes.

Optimal timing for generation investment with uncertain emission mitigation policy

In view of the fact that different mechanisms for mitigating the CO2 emission have been employed or proposed in different countries or regions, and those already implemented are still in an evolutionary procedure, the future CO2 emission prices would be highly uncertain. Given this background, an effort is made for investigating the problem of generation investment decision-making in electricity market environment with uncertainties from the climate change policy for limiting the CO2 emission. According to the changing characteristics of the uncertain factors, the models of the fuel prices, electricity prices, and CO2 emission prices are respectively presented first. Next, under the existing real option approach (ROA) based methodological framework for the generation investment decision-making problem, a mathematical model accommodating multiple kinds of uncertainties and an efficient solving method are developed. Finally, the proposed model and method are illustrated by a numerical example with different scenarios. Copyright © 2010 John Wiley & Sons, Ltd.

Cultural evolutionary design of adaptive wavelet filters based on lifting scheme for micro-instruments

An evolutionary procedure based on cultural algorithms for the optimal design of adaptive wavelet filters based on lifting scheme is proposed. Numerical results of characterization, based on statistical experiment design, as well as validation, based on the comparison with a genetic optimization algorithm, are presented. Experimental results of the validation on two case studies for reducing uncertainty arising from noise in on-field corrosion rate measurements are highlighted.

A scatter search algorithm for scheduling optimisation of job shop problems

The Job Shop Scheduling (JSS) problem is one of the most difficult NP-hard combinatorial optimisation problems. it is very difficult to solve by conventional optimisation techniques. There has been increasing interest in applying metaheuristic methods to solve such hard optimisation problems. In this work, a novel metaheuristic approach called Scatter Search (SS) is applied for the JSS problem, an NP-hard sequencing problem. The approach is used to find a schedule to minimise the makespan (Cmax), that is, the time required to complete all jobs. SS contrasts with other evolutionary procedures by providing a wide exploration of the search space through intensification and diversification. In addition, it has a unifying principle for joining solutions. It exploits the adaptive memory principle to avoid generating or incorporating duplicate solutions at various stages of the problem. The main aim of this research is to explore the potential of SS for scheduling JSS problems. SS provides unifying principles for joining solutions based on generalised path constructions and by utilising strategic designs where other approaches resort to randomisation. In this paper, 11 benchmark problems are taken from the literature. The results available for the various existing metaheuristic methods for the selected benchmark problems are compared with results obtained by the SS method. The proposed framework achieves better results for all 11 problems.

A single front genetic algorithm for parallel multi-objective optimization in dynamic environments

This paper proposes a new parallel evolutionary procedure to solve multi-objective dynamic optimization problems along with some measures to evaluate multi-objective optimization in dynamic environments. These dynamic optimization problems appear in quite different real-world applications with actual socio-economic relevance. In these applications, the objective functions, the constraints, and hence, also the solutions, can change over time and usually demand to be solved online whilst the size of the changes is unknown. Although parallel processing could be very useful in these problems to meet the solution quality requirements and constraints, to date, not many parallel approaches have been reported in the literature. Taking this into account, we introduce a multi-objective optimization procedure for dynamic problems that are based on PSFGA, a parallel evolutionary algorithm previously proposed by us for multi-objective optimization. It uses an island model where a process divides the population among the remaining processes and allows the communication and coordination among the subpopulations in the different islands. The proposed algorithm makes an exclusive use of non-dominating individuals for the selection and variation operator and applies a crowding mechanism to maintain the diversity and the distribution of the solutions in the Pareto front. We also propose a model to understand the benefits of parallel processing in multi-objective problems and the speedup figures obtained in our experiments.

An adaptive neuro-fuzzy sliding mode based genetic algorithm control system for under water remotely operated vehicle

This study presents an adaptive neuro-fuzzy sliding-mode-based genetic algorithm (ANFSGA) control system for a remotely operated vehicle (ROV) with four degrees of freedom (DOF)s. In many applications, ROVs will need to be capable of maneuvering to any given point, following object, and to be controllable from the surface. Therefore, an ANFSGA control system is introduced for tracking control of the ROV to achieve a high precision position control. Since the dynamic of ROVs are highly nonlinear and time varying, an ANFSGA control system is investigated according to direction-based genetic algorithm (GA) with the spirit of sliding mode control and adaptive neuro-fuzzy sliding mode (ANFS) based evolutionary procedure. In this way, on-line learning ability is employed to deal with the parametric uncertainty and disturbance by adjusting the ANFS inference parameters. In this proposed controller a GA control system is utilized to be the major controller, and stability can be indirectly insured by the concept of sliding mode control system without strict constraints and detailed system knowledge.

Strength of Ferritic Steels: Neural Networks and Genetic Programming

An analysis is presented of a complex set of data on the strength of steels as a function of chemical composition, heat treatment, and test temperature. The steels represent a special class designed to resist deformation at elevated temperatures (750–950 K) over time periods in excess of 30 years, whilst serving in hostile environments. The aim was to compare two methods, a neural network based on a Bayesian formulation, and genetic programming in which the data are formulated in an evolutionary procedure. It is found that in the present context, the neural network is able more readily to capture greater complexity in the data whereas a genetic program seems to require greater intervention to achieve an accurate representation.

Evolutionary design of structural systems with time-variant performance

This paper presents an evolutionary approach to the optimal design of structural systems with time-variant performance. The complexity of a multidimensional design is handled by introducing a set of quality indices, which represent the multiple targets of design, and by a systemic vision of the structural model, which can be viewed either as a whole having its own emerging properties, or as composed of elementary parts with their own specific characteristics. An overall measure of lifetime structural quality, able to also take the time-variant structural performance into account, is then obtained through a weighted integration of the quality indices over the expected service life of the structure. In addition, in order to overcome the limits of mathematical optimization methods in solving design problems of high complexity and multidimensionality, the optimal structural morphology is searched for through a two-level heuristic approach based on biologically inspired evolutionary procedures. Finally, the application of the optimal design to a cable-stayed bridge is presented. The results of this evolutionary design process highlight the important role played by suitable measures of structural quality and show that the optimal configurations strongly depend on the time-variant structural performance.

Computationally Mapping Sequence Space To Understand Evolutionary Protein Engineering

Evolutionary protein engineering has been dramatically successful, producing a wide variety of new proteins with altered stability, binding affinity, and enzymatic activity. However, the success of such procedures is often unreliable, and the impact of the choice of protein, engineering goal, and evolutionary procedure is not well understood. We have created a framework for understanding aspects of the protein engineering process by computationally mapping regions of feasible sequence space for three small proteins using structure-based design protocols. We then tested the ability of different evolutionary search strategies to explore these sequence spaces. The results point to a non-intuitive relationship between the error-prone PCR mutation rate and the number of rounds of replication. The evolutionary relationships among feasible sequences reveal hub-like sequences that serve as particularly fruitful starting sequences for evolutionary search. Moreover, genetic recombination procedures were examined, and tradeoffs relating sequence diversity and search efficiency were identified. This framework allows us to consider the impact of protein structure on the allowed sequence space and therefore on the challenges that each protein presents to error-prone PCR and genetic recombination procedures.

A study on ranked bidirectional evolutionary structural optimization (R-BESO) method for fully stressed structure design based on displacement sensitivity

Evolutionary Structural Optimization(ESO) method is well known as one of several topology optimization methods and has been applied to a lot of optimization problems. While ESO method evolves the given model into an optimum by subtracting several elements, in AESO method elements are added in a previous step of the evolutionary procedure. And in BESO(Bidirectional ESO) method, some elements are either generated or eliminated from a previous model of evolutionary procedure. In this paper, Ranked Bidirectional Evolutionary Structural Optimization(R-BESO) method is introduced as one of the topology optimization methods using an evolutionary algorithm and is applied to several optimization problems. The method can get optimum topologies of the structures throughout fewer iterations comparing with previous several methods based on ESO. R-BESO method is similar to BESO method except that elements are generated near a candidate element according to the rank calculated by sensitivity analyses. The displacement sensitivity analysis was adopted by the nodal displacements of a candidate element in order to determine a rank on the free edges for two dimensional model or the free surfaces for three dimensional model. In this paper, R-BESO method is proposed as another useful design tool like the previous ESO and BESO method for the two bar frame problem, the Michell type structure problem and the three dimension short cantilever beam problem, which had been used to verify reasonability of ESO method family. For the three dimensions short cantilever beam problem an optimized topology could be obtained with much fewer iterations with respect to the results of other ESO methods.

A Memetic Fingerprint Matching Algorithm

Minutiae point pattern matching is the most common approach for fingerprint verification. Although many minutiae point pattern matching algorithms have been proposed, reliable automatic fingerprint verification remains as a challenging problem, both with respect to recovering the optimal alignment and the construction of an adequate matching function. In this paper, we develop a memetic fingerprint matching algorithm (MFMA) which aims to identify the optimal or near optimal global matching between two minutiae sets. Within the MFMA, we first introduce an efficient matching operation to produce an initial population of local alignment configurations by examining local features of minutiae. Then, we devise a hybrid evolutionary procedure by combining the use of the global search functionality of a genetic algorithm with a local improvement operator to search for the optimal or near optimal global alignment. Finally, we define a reliable matching function for fitness computation. The proposed algorithm was evaluated by means of a series of experiments conducted on the FVC2002 database and compared with previous work. Experimental results confirm that the MFMA is an effective and practical matching algorithm for fingerprint verification. The algorithm is faster and more accurate than a traditional genetic-algorithm-based method. It is also more accurate than a number of other methods implemented for comparison, though our method generally requires more computational time in performing fingerprint matching.

Evaluation of scatter-search approach for scheduling optimization of flexible manufacturing systems

Many optimization problems from the industrial engineering world (in particular manufacturing systems) are very complex in nature and quite hard to solve by conventional optimization techniques. There has been increasing interest to apply meta-heuristic methods to solve such kinds of hard optimization problems. In this work, a meta-heuristic approach called scatter-search (SS) is applied for scheduling optimization of flexible manufacturing systems by considering multiple objectives, i.e., minimizing the idle time of the machine and minimizing the total penalty cost for not meeting the due date concurrently. Scatter search (SS) contrasts with other evolutionary procedures in that it provides a wide exploration of the search space through intensification and diversification. In addition, it has a unifying principle for joining solutions and they exploit adaptive memory principle to avoid generating or incorporating duplicate solutions at various stages of the problem. In this paper, various meta-heuristic methods are used for solving three different sizes of scheduling problems taken from the literature. The results available for the various existing meta-heuristic methods are compared with results obtained by the scatter-search method. The proposed framework achieves better results for all the three problems and also achieves an average deviation of 16.67% from the best results obtained by other methods.

Performance evaluation of the scatter search method for permutation flowshop sequencing problems

Many optimization problems from the industrial engineering world, in particular manufacturing systems, are very complex in nature and are quite hard to solve by conventional optimization techniques. There has been increasing interest to apply metaheuristic methods to solve such kinds of hard optimization problems. In this work, a novel metaheuristic approach called scatter search (SS) is applied for the n/m/P/C max problem, an NP-hard sequencing problem, which is used to find a processing order of n different jobs to be processed on m machines in the same sequence with minimizing the makespan. SS contrasts with other evolutionary procedures by providing a wide exploration of the search space through intensification and diversification. In addition, it has a unifying principle for joining solutions and they exploit the adaptive memory principle to avoid generating or incorporating duplicate solutions at various stages of the problem. In this paper, various metaheuristic methods and best heuristics from the literature are used for solving the well-known benchmark problem set of Taillard (Eur J Oper Res 64:278–285, 1993). The results available for the various existing metaheuristic and heuristic methods are compared with the results obtained by the SS method. The proposed framework achieves better results for 4 of 12 benchmark problems and also achieves an average deviation of 1.003% from the benchmark problem set of Taillard (Eur J Oper Res 64:278–285, 1993). The computational results show that SS is a more effective metaheuristic for the n/m/P/C max problem.

Optimal conduction pathways for cooling a heat-generating body: A comparison exercise

The objectives of this paper are two: (i) propose an evolutionary algorithm for optimizing the topology of a heat-generating area cooled by a high thermal conductivity material in contact with a heat sink, and (ii) compare the evolutionary procedure and optimal structures found to the ones predicted by constructal theory. The evolutionary algorithm starts with an initial random topology, for which control volumes are repositioned based on temperature and heat flux information in order to minimize the hot spot temperature. Significant diminution of the hot spot temperatures have been achieved by the algorithm. Results revealed three families of geometrical configurations for the optimal conductive pathways generated by our algorithm, depending on the values of the conductivity ratio and amount of high thermal conductivity material: few radial blades growing on the heat sink, tree-shaped networks, and networks with loops. The optimal structures are compared with the heat tree architectures generated by constructal theory, and several similarities in terms of performances and geometries are revealed.

Design of Total Sliding-Mode-Based Genetic Algorithm Control for Hybrid Resonant-Driven Linear Piezoelectric Ceramic Motor

This study presents a total sliding-mode-based genetic algorithm control (TSGAC) system for a linear piezoelectric ceramic motor (LPCM) driven by a newly designed hybrid resonant inverter. First, the motor configuration and driving circuit of an LPCM are introduced, and its hypothetical dynamic model is represented by a nonlinear function with unknown system parameters. In the hybrid resonant drive system, it has the merits of the high voltage gain from a parallel-resonant current source, and the invariant output characteristic from a two-inductance two-capacitance resonant driving circuit. Since the dynamic characteristics and motor parameters of the LPCM are highly nonlinear and time varying, a TSGAC system is therefore investigated based on direction-based genetic algorithm with the spirit of total sliding-mode control (TSC) and fuzzy-based evolutionary procedure to achieve high-precision position control under a wide operation range. In this control scheme, a GAC system is utilized to be the major controller, and the stability can be indirectly ensured by the concept of TSC without strict constraints and detailed system knowledge. In addition, the effectiveness of the proposed drive and control system is verified by numerical simulations and experimental results in the presence of uncertainties

Optimization for large scale process based on evolutionary algorithms: Genetic algorithms

Abstract This work has as objective the development of an optimization methodology, using Genetic Algorithms (GAs), as evolutionary procedure coupled with the concepts of evolutionary. As case study a large scale multiphase catalytic reactor is considered. The reactor is tubular in shape and is built-up with concentric tubes using the same concept of the auto-thermal reactors, with coolant fluid flow in the external annular. The mathematical equations of the deterministic model are based on conservation principles (mass, energy and momentum) for the reactants and for the coolant fluid and validated with real operational data. The model represents the steady-state with the plug-flow assumption which is quite reasonable due to the large flow rates usually found in industrial reactors. The desired product is a specific cyclical alcohol (CA), and the minimization of the by-products is required for economical and environmental reasons. For that it is necessary to optimize some important operational parameters. This problem is of difficult solution since the reactor is a large scale system with complex behavior and conventional optimization tools as Successive Quadratic Programming tends to fail in such situation since local minima may be achieved. In this work is shown that the Genetic Algorithms technique can be useful to CA production maximization, obtaining good results with operational improvements (reduction in the catalyst rate, as well as in the undesired product rate—cycloalkane (C)). The GA parameters used for the process optimization are population size, crossover types with variation of crossover rates. The used coding was the binary form. The results are quite good, showing high performance in the CA productivity (considerable increase CA production) with changes in the operational parameters analyzed and showing that this optimization procedure is very robust and efficient. The results point out that this technique is very promising to deal with large scale system with complex behavior due to non-linearity and variable interactions.