Genetic programming and evolutionary programming are fields studying the application of artificial evolutionon evolving directly executable programs, in form of trees similar to Lisp expressions (GP-trees), or Finite ...
详细信息
ISBN:
(纸本)9781595936974
Genetic programming and evolutionary programming are fields studying the application of artificial evolutionon evolving directly executable programs, in form of trees similar to Lisp expressions (GP-trees), or Finite State Automata (FSA).In this exercise, we study the performance of these methods on several example problems, and draw conclusionson the suitability of the representations with respect to the task structure and properties. We investigate the roleof incremental evolution and its bias in the context of FSA representation. The experiments are performed in simulation and/or confirmed on real robots.
Classical evolutionary programming (CEP) and Fast evolutionary programming (FEP) have been applied to real-valued function optimisation. Both of these techniques directly evolve the real-values that are the arguments ...
详细信息
ISBN:
(纸本)9781595936974
Classical evolutionary programming (CEP) and Fast evolutionary programming (FEP) have been applied to real-valued function optimisation. Both of these techniques directly evolve the real-values that are the arguments of the real-valued function. lit this paper we have applied a form of genetic programming called Cartesian Genetic programming (CGP) to a number of real-valued optimisation benchmark problems. The approach we have taken is to evolve a computer program that controls a writing-head, which moves along and interacts with a finite set of symbols that are interpreted as real numbers, instead of manipulating the real numbers directly. In other studies, CGP has already been shown to benefit front a high degree of neutrality. We hope to exploit this for real-valued function optimisation problems to avoid being trapped oil local optima. We have also used an extended form of CGP called Embedded CGP (ECGP) which allows the acquisition, evolution and re-use of modules. The effectiveness of CGP and ECGP are compared and contrasted with CEP and FEP oil the benchmark problems. Results show that the new techniques are very effective.
Knowledge about evolutionary information is not made use of effectively in Genetic Algorithm. While traditional Cultural Algorithms with dual inheritance structure converge slowly because evolutionary programming is c...
详细信息
ISBN:
(纸本)9781424410910
Knowledge about evolutionary information is not made use of effectively in Genetic Algorithm. While traditional Cultural Algorithms with dual inheritance structure converge slowly because evolutionary programming is chosen for the population model and only mutation operator is adopted in the population space. A novel Cultural Algorithm based on Genetic Algorithm is proposed. Four kinds of knowledge are abstracted. Simulation results on the benchmark single-peak optimization functions indicate that the performance of this method is much better than traditional Cultural Algorithms especially for the "plain functions". Aiming at multi-peaks optimization problem, Multi-Windows Cultural Algorithm and Multi-Windows Cultural Algorithm based on Genetic Algorithm are introduced. Simulation results on benchmark multi-peaks function indicates that the latter is more effective in optimization performance than the former.
Mammography is an effective tool for the early detection of breast cancer;however, most women referred for biopsy based on mammographic findings do not, in fact, have cancer. This study is part of an ongoing effort to...
详细信息
ISBN:
(纸本)0819452831
Mammography is an effective tool for the early detection of breast cancer;however, most women referred for biopsy based on mammographic findings do not, in fact, have cancer. This study is part of an ongoing effort to reduce the number of benign cases referred for biopsy by developing tools to aid physicians in classifying suspicious lesions. Specifically, this study examines the use of an evolutionary programming (EP)-derived Support Vector Machine (SVM) with a modified radial basis function (RBF) kernel, and compares this with results using a normal Gaussian radial basis function kernel. Results demonstrate that the modified kernel can provide moderate performance improvements;however, due to its ability to create a more complex decision surface, this kernel can easily begin to memorize the training data resulting in a loss of generalization ability. Nonetheless, these methods could reduce the number of benign cases referred for biopsy by over half, while missing less than 5% of malignancies. Future work will focus on methods to improve the EP process to preserve SVMs which generalize well.
This paper presents the model of interleavers for the Interleave-Division Multiple-Access (IDMA) based on evolutionary algorithm. In all the previous works, interleavers are all generated independently and randomly wh...
详细信息
This paper presents the model of interleavers for the Interleave-Division Multiple-Access (IDMA) based on evolutionary algorithm. In all the previous works, interleavers are all generated independently and randomly which is simple but with good performance. Considered the difference between the model of interleavers and the traveling salesman problem(TSP), a specific fitness function based on covariance matrix is given and the optimum interleavers are computed by evolutionary algorithm. The simulation results show that the bit error ratio(BER) performance of the evolutionary interleavers(EI) is much better than other unrandom interleavers. The BER performance of independent and random interleavers is near to EI, it is a proof that EI is the theoretical optimum interleavers for IDMA.
In this paper, first manufacturing scheduling is briefly discussed and later the problem studied is introduced. The optimal solution to minimizing the average flow time in single machine scheduling is obtained by the ...
详细信息
In this paper, first manufacturing scheduling is briefly discussed and later the problem studied is introduced. The optimal solution to minimizing the average flow time in single machine scheduling is obtained by the Shortest Processing Time rule if ready times are zero for all jobs. In the case of non-zero ready times, preemption plays a key role in the solution. Preemption allowed version is solved optimally by using the Shortest Remaining Processing Time procedure. However, the version of preemption not allowed is known as NP-hard and delay and nondelay strategies might be used in a hybrid fashion. This paper focuses on minimizing the average flow time in the presence of non-zero times and when preemption is not allowed. The proposed method is evolutionary programming (EP). The results indicate that EP produces near optimal and consistent results in a short period of time. (C) 2003 Elsevier Science Ltd. All rights reserved.
This paper presents a distributed scheduling architecture for a multi-service routing switch, based on evolutionary algorithms to solve a multi-objective optimization problem. The aim of the two-level scheduler is to ...
详细信息
ISBN:
(纸本)1880843498
This paper presents a distributed scheduling architecture for a multi-service routing switch, based on evolutionary algorithms to solve a multi-objective optimization problem. The aim of the two-level scheduler is to ensure better quality of service for individual flows and achieve near 100% throughput with minimal delay at the switch level under uniform traffic conditions. It is shown that evolutionary Algorithm provides an efficient scheduling mechanism, because the decision-making is dependent on the real traffic conditions. Simulation results show the performance of the distributed evolutionary scheduler is much better than the conventional fair queuing schemes and efficiently integrates flow and switch level scheduling. The scheduling scheme is simple to design and fairly inexpensive when implemented using FPGA technology.
A traditional mathematical model for maintenance scheduling of power generation systems may give an optimal schedule for a power system with known conditions. A change of the system condition due to uncertainties or s...
详细信息
A traditional mathematical model for maintenance scheduling of power generation systems may give an optimal schedule for a power system with known conditions. A change of the system condition due to uncertainties or sudden changes may tender the resulting optimal schedule unsuitable or inapplicable for the power system under study. This paper presents a fuzzy model and an evolutionary programming-based solution technique for the security-constrained maintenance scheduling (MS) problem of generation systems with uncertainties in the load and fuel and maintenance costs. The proposed technique results are fuzzy optimal cost range that reflects the problem uncertainties. The technique solves a decomposed maintenance model of two interrelated subproblems, namely the maintenance and the security-constrained economic dispatch problem. Test results on the IEEE 30-bus system with six generating units reported in this paper are quite encouraging. (C) 2003 Elsevier Science B.V. All rights reserved.
This paper presents an evolutionary programming (EP)-based technique to the unified model of the maintenance scheduling (MS) problem of power generation and transmission systems. In this paper, the Hill-Climbing techn...
详细信息
This paper presents an evolutionary programming (EP)-based technique to the unified model of the maintenance scheduling (MS) problem of power generation and transmission systems. In this paper, the Hill-Climbing technique (HCT) is used in conjunction with the EP to find a feasible solution in the neighborhood of the new infeasible solutions during the solution process. The EP search ability and the feasibility watch of the HCT motivate the sequential solution of the two interrelated subproblems of the MS problem. The paper reports test results of the proposed algorithm on the IEEE 30-bus system with six generating units and 41 transmission lines. (C) 2002 Elsevier Science B.V. All rights reserved.
The objective of the paper is to minimize the production cost of the thermal power generation. An elegant approach is presented in order to obtain the equivalent cost function of the participating non-fuel restricted ...
详细信息
The objective of the paper is to minimize the production cost of the thermal power generation. An elegant approach is presented in order to obtain the equivalent cost function of the participating non-fuel restricted units and the Economic Dispatch Calculations (EDC) are carried out along with fuel restricted units. The evolutionary programming (EP) technique is used for real power optimization with fuel restricted units. The optimal solution is obtained neglecting losses. The Fast Decoupled Load Flow (FDLF) analysis is conducted to find the losses by substituting the generation values. Then the loss is participated among all generating units using participation factor method. The load flow is conducted again and the voltage limit violation is checked. The Algorithm is tested on IEEE 6-bus system IEEE 30-bus system and a 66-bus utility system. The results obtained by this new approach are compared with those obtained using classical method. It is observed that the proposed method is more reliable and efficient. (C) 2003 Elsevier Ltd. All rights reserved.
暂无评论