evolutionary algorithms (EAs) based on the concept of Pareto dominance seem the most suitable technique for multiobjective optimisation. In multiobjective optimisation, several criteria (usually conflicting) need to b...
详细信息
evolutionary algorithms (EAs) based on the concept of Pareto dominance seem the most suitable technique for multiobjective optimisation. In multiobjective optimisation, several criteria (usually conflicting) need to be taken into consideration simultaneously to assess a quality of a solution. Instead of finding a single solution, a set of trade-off or compromise solutions that represents a good approximation to the Pareto optimal set is often required. This thesis presents an investigation on evolutionary algorithms within the framework of multiobjective optimisation. This addresses a number of key issues in evolutionary multiobjective optimisation. Also, a new evolutionary multiobjective (EMO) algorithm is proposed. Firstly, this new EMO algorithm is applied to solve the multiple 0/1 knapsack problem (a wellknown benchmark multiobjective combinatorial optimisation problem) producing competitive results when compared to other state-of-the-art MOEAs. Secondly, this thesis also investigates the application of general EMO algorithms to solve real-world nurse scheduling problems. One of the challenges in solving real-world nurse scheduling problems is that these problems are highly constrained and specific-problem heuristics are normally required to handle these constraints. These heuristics have considerable influence on the search which could override the effect that general EMO algorithms could have in the solution process when applied to this type of problems. This thesis outlines a proposal for a general approach to model the nurse scheduling problems without the requirement of problem-specific heuristics so that general EMO algorithms could be applied. This would also help to assess the problems and the performance of general EMO algorithms more fairly.
Voltage stability is one of the important issues in the existing power system network and in this paper presents the reactive power load management for voltage stability control in multi-area power system. Evolutionar...
详细信息
ISBN:
(纸本)9781509011780
Voltage stability is one of the important issues in the existing power system network and in this paper presents the reactive power load management for voltage stability control in multi-area power system. evolutionary programming (EP) as optimization technique was employed in order to manage the load with the aim to control the voltage stability. Pre-developed Fast Voltage Stability Index (FVSI) was utilized as an instrument for voltage security assessment. Single-load and multi-load increment were performed on several loading conditions to obtain maximum reactive power loading with FVSI range from 0 to 0.95 p.u. The proposed EP optimization technique managed to outperform the traditional voltage stability assessment (VSA) approach in terms of achieving accurate results.
Because of the revolution of power system structure nowadays, operation and control of generating unit must be modified. Energy price becomes an important parameter to make a decision in this restructured system. Unit...
详细信息
ISBN:
(纸本)0780373227
Because of the revolution of power system structure nowadays, operation and control of generating unit must be modified. Energy price becomes an important parameter to make a decision in this restructured system. Unit commitment (UC) in such a competitive environment is not the same as the traditional one anymore. The objective of UC is not to minimize production cost as before but to find the solution that produces a maximum profit for generation company (GENCO). This paper presents a new profit-based UC formulation under competitive environment considering both power and reserve generation. A hybrid method between Lagrange Relaxation (LR) and evolutionary programming (EP) is applied to solve this new UC problem. The proposed approach is applied to a test system. Simulation results are compared with those obtained from traditional UC.
Particle Swarm. is a novel optimization paradigm for real-valued functions, based on the social dynamics of group interaction. In this work, it is proposed its application to the training of Neural Networks. Comparati...
详细信息
ISBN:
(纸本)0780372786
Particle Swarm. is a novel optimization paradigm for real-valued functions, based on the social dynamics of group interaction. In this work, it is proposed its application to the training of Neural Networks. Comparative tests were carried out, for classification and regression tasks, being the results compared with other approaches.
Pupil plane filtering provides a convenient technique for modifying the point spread function. Such modifications are used in many practical applications that require enhancement of selective frequency band in images....
详细信息
ISBN:
(纸本)9780819482839
Pupil plane filtering provides a convenient technique for modifying the point spread function. Such modifications are used in many practical applications that require enhancement of selective frequency band in images. Also, in many new imaging paradigms, acquisition of 3D image information calls for tailoring of the 3D point spread function. This can be achieved by suitable pupil plane filtering, preferably by phase filters. By using a pupil plane filter with an array of concentric annuli, the point spread function can be tailored in a fashion such that a narrow central lobe is surrounded by neighboring lobes of low amplitude, with one or more lobes of high amplitude spaced far away from the center. In our study we intend to explore the use of phase annuli as pupil filters in tailoring of both transverse and axial resolution. Determination of such phase filters in accordance with a set of prespecified requirements for amplitude/intensity distribution around the focus constitutes a problem of nonlinear optimization. This paper reports some results of our preliminary investigations on an application of evolutionary programming in solving this problem to obtain globally or quasi-globally optimum solutions.
This paper initially describes how an inferred context-free (stochastic) grammar can be used to verify command transmissions and serve as a hedge against a successful cyber-attack. The remainder of the paper addresses...
详细信息
ISBN:
(纸本)9781538615621
This paper initially describes how an inferred context-free (stochastic) grammar can be used to verify command transmissions and serve as a hedge against a successful cyber-attack. The remainder of the paper addresses a computational problem not amenable to closed-form solution;namely, the hard real-time (similar to 57 usec) synthesis of a desired waveform through the adaptive modification of a carrier wave. This effectively increases the signal to noise ratio - ensuring better UAV communications. Here, the modulation of the primary waveform is under user control and is of strictly positive amplitude. The primary waveform induces a secondary waveform having delayed leading and trailing edges and expanded rise and fall times. There is, in general, a direct relation between the period of the primary waveform and the amplitude of the secondary waveform. The relation between the primary and secondary waveforms may be characterized by trigonometric functions or even interpolating polynomials. However, response time will be minimized where the primary waveforms are discretized and stored in the form of array-based cases. The tertiary (target) wave may be any periodic trigonometric function, but is taken to be a simple sine wave without loss of generality. The task of the adaptive program is to minimize parallel to s(t) - g(t)parallel to 2, where f(t) -> g(t) and f(t) is the primary waveform at time t, g(t) is the secondary waveform at time t, and s(t) is the tertiary waveform at time t. A computationally efficient algorithm is provided for solving this task in real time. Moreover, an evolutionary program (EP) is provided for automatic case acquisition. Primary waveforms are mutated in accordance with a normal distribution.
作者:
Yao, SusuASTAR
Inst Infocomm Res Singapore 138632 Singapore
Image registration is an important preprocessing technique in high dynamic range (HDR) image synthesis. This paper proposed a robust image registration method for aligning a group of low dynamic range images (LDR) tha...
详细信息
ISBN:
(纸本)9780819484079
Image registration is an important preprocessing technique in high dynamic range (HDR) image synthesis. This paper proposed a robust image registration method for aligning a group of low dynamic range images (LDR) that are captured with different exposure times. Illumination change and photometric distortion between two images would result in inaccurate registration. We propose to transform intensity image data into phase congruency to eliminate the effect of the changes in image brightness and use phase cross correlation in the Fourier transform domain to perform image registration. Considering the presence of non-overlapped regions due to photometric distortion, evolutionary programming is applied to search for the accurate translation parameters so that the accuracy of registration is able to be achieved at a hundredth of a pixel level. The proposed algorithm works well for under and over-exposed image registration. It has been applied to align LDR images for synthesizing high quality HDR images.
The Estimation of Distribution Algorithms (EDAs) is a novel class of evolutionary algorithms which is motivated by the idea of building probabilistic graphical model of promising solutions to represent linkage informa...
详细信息
ISBN:
(纸本)9780769536347
The Estimation of Distribution Algorithms (EDAs) is a novel class of evolutionary algorithms which is motivated by the idea of building probabilistic graphical model of promising solutions to represent linkage information between variables in chromosome. Through learning of and sampling from probabilistic graphical model, new population is generated and optimization procedure is repeated until the stopping criteria are met. In this paper, the mechanism of the Estimation of Distribution Algorithms is analyzed. Currently existing EDAs are surveyed and categorized according to the probabilistic model they used.
An efficient evolutionary-based approach, termed as Differential Evolution (DE), is presented for the solution of Optimal Power Flow (OPF) with the continuous variables. The continuous control variables are unit-activ...
详细信息
ISBN:
(纸本)9781424417636
An efficient evolutionary-based approach, termed as Differential Evolution (DE), is presented for the solution of Optimal Power Flow (OPF) with the continuous variables. The continuous control variables are unit-active power outputs and generator-bus voltage magnitudes, transformer tap settings and switchable shunt devices. The differential evolution is illustrated for two case studies of IEEE-30 bus system. Both conventional and non-conventional cost characteristics are considered for the optimal power flow solution. The feasibility of the proposed method is compared with a simple evolutionary, programming algorithm. The algorithm is computationally faster, in terms of the number of load flows executed, and provides better results than other heuristic techniques.
In this paper, a new particle swarm optimization (PSO) algorithm namely Turbulent Crazy Particle swarm Optimization (TRPSO) is introduced to solve multi-constrained optimal reactive power dispatch in power system. Opt...
详细信息
ISBN:
(纸本)9781424450534
In this paper, a new particle swarm optimization (PSO) algorithm namely Turbulent Crazy Particle swarm Optimization (TRPSO) is introduced to solve multi-constrained optimal reactive power dispatch in power system. Optimal reactive power dispatch problem is a multi-objective optimization problem that minimizes bus voltage deviations and transmission loss. The feasibility of the proposed algorithm is demonstrated for IEEE 30-bus system and it is compared to other well established population based optimization techniques like conventional PSO, general passive congregation PSO (GPAC), local passive congregation PSO (LPAC), coordinated aggregation (CA) and Interior point based OPF (IP-OPF). A comparison of simulation results indicates that the proposed algorithm can produce better solution than other optimization techniques.
暂无评论