Convergence towards, and diversity across the Pareto-optimal front are the two main requirements when optimising a multiobjectiveoptimisation problem (MOP) with conflicting objectives. Most established multiobjective...
详细信息
ISBN:
(纸本)9781424481262
Convergence towards, and diversity across the Pareto-optimal front are the two main requirements when optimising a multiobjectiveoptimisation problem (MOP) with conflicting objectives. Most established multiobjectiveevolutionary algorithms (MOEAs) have mechanisms that address these requirements. However, in many-objective optimisation, where the number of objectives is greater than 2 or 3, it has been found that these two requirements can conflict with one another, introducing problems such as dominance resistance and speciation. In this study, a previously introduced diversity management mechanism is deployed within a Progressive Preference Articulation (PPA) technique to optimise an 8-objective real-world problem of aircraft control system design. This paper illustrates the effective application of the new diversity management mechanism used in conjunction with the PPA technique when optimising a multiobjective real-world engineering problem.
In this paper, we propose a novel methodology to assist in identifying vulnerabilities in real-world complex heterogeneous industrial control systems (ICS) using two evolutionary multiobjective optimisation (EMO) algo...
详细信息
In this paper, we propose a novel methodology to assist in identifying vulnerabilities in real-world complex heterogeneous industrial control systems (ICS) using two evolutionary multiobjective optimisation (EMO) algorithms, NSGA-II and SPEA2. Our approach is evaluated on a well-known benchmark chemical plant simulator, the Tennessee Eastman (TE) process model. We identified vulnerabilities in individual components of the TE model and then made use of these vulnerabilities to generate combinatorial attacks. The generated attacks were aimed at compromising the safety of the system and inflicting economic loss. Results were compared against random attacks, and the performance of the EMO algorithms was evaluated using hypervolume, spread, and inverted generational distance (IGD) metrics. A defence against these attacks in the form of a novel intrusion detection system was developed, using machine learning algorithms. The designed approach was further tested against the developed detection methods. The obtained results demonstrate that the developed EMO approach is a promising tool in the identification of the vulnerable components of ICS, and weaknesses of any existing detection systems in place to protect the system. The proposed approach can serve as a proactive defense tool for control and security engineers to identify and prioritise vulnerabilities in the system. The approach can be employed to design resilient control strategies and test the effectiveness of security mechanisms, both in the design stage and during the operational phase of the system.
The possibility of performing computations with cellular automata (CAs) opens up new conceptual issues in emergent computation. Driven by this motivation, a recurring problem in this context is the automatic search fo...
详细信息
The possibility of performing computations with cellular automata (CAs) opens up new conceptual issues in emergent computation. Driven by this motivation, a recurring problem in this context is the automatic search for good one-dimensional, binary CA rules that can perform well in the density classification task (DCT), that is, the ability to discover which cell state outnumbers the other state. In the past, the most successful attempts to reach this target have relied on evolutionary searches in the space of possible rules. Along this line, a multiobjective, heuristic evolutionary approach, implemented as a distributed cooperative system, is presented here, which yielded outstanding results, including a rule that led to the characterisation of a class of four equivalent rules, all of them with the best performance currently available in the literature for the DCT. (c) 2006 Elsevier B.V. All rights reserved.
A solid-state, very high frequency (VHF) band, moving target detection air surveillance radar requires a complex pulse burst waveform to mitigate the visibility issues originating from the blind Doppler intervals and ...
详细信息
A solid-state, very high frequency (VHF) band, moving target detection air surveillance radar requires a complex pulse burst waveform to mitigate the visibility issues originating from the blind Doppler intervals and range eclipsing. The waveform employs multiple pulse repetition frequencies to mitigate the effects of the blind Doppler intervals and interleaves short and long pulses to mitigate range eclipsing. In the authors' previous works, they pointed out that the waveform design is a multi-objective optimisation problem and defined the mathematical model of the waveform optimisation problem. They also presented how the exact Pareto optimal (PO) set can be determined by means of exhaustive search. In this paper, they improve the mathematical model of the waveform optimisation problem by altering the way in which one of the objective functions is calculated and adding a new constraint, which eliminates meaningless solutions. Finally, they propose a solution method based on a multi-objective evolutionary algorithm. The performance evaluation test indicates that compared to the exhaustive search, the proposed method provides a solution that is insignificantly different. However, the proposed method is more scalable and requires over three orders of magnitude smaller number of comparisons to determine the PO set, which makes it more viable for the online waveform adaptation.
Image segmentation plays a key role in many fields such as image processing and recognition. Although various segmentation methods have been proposed in recent decades, most of these methods are based on only a single...
详细信息
Image segmentation plays a key role in many fields such as image processing and recognition. Although various segmentation methods have been proposed in recent decades, most of these methods are based on only a single feature space. How to combine various features to image segmentation is a challenge problem. To address this problem, the authors propose to combine different features based on evolutionary multiobjective optimisation. Two optimisation objectives, which are based on colour and texture features, respectively, are therefore designed for image segmentation. The experiments show that the author's method is able to combine multiple features for image segmentation successfully.
To contribute towards designing more cost-efficient, robust and flexible downstream processes for the manufacture of monoclonal antibodies (mAbs), a framework consisting of an evolutionarymultiobjective optimization ...
详细信息
To contribute towards designing more cost-efficient, robust and flexible downstream processes for the manufacture of monoclonal antibodies (mAbs), a framework consisting of an evolutionarymultiobjective optimization algorithm (EMOA) linked to a biomanufacturing process economics model is presented. The EMOA is tuned to discover sequences of chromatographic purification steps and column sizing strategies that provide the best trade-off with respect to multiple objectives including cost of goods per gram (COG/g), robustness in COG/g, and impurity removal capabilities. Additional complexities accounted for by the framework include uncertainties and constraints. The framework is validated on industrially relevant case studies varying in upstream and downstream processing train ratios, annual demands, and impurity loads. Results obtained by the framework are presented using a range of visualization tools, and indicate that the performance impact of uncertainty is a function of both the level of uncertainty and the objective being optimized, and that uncertainty can cause otherwise optimal processes to become suboptimal. The optimal purification processes discovered outperform the industrial standard with, e.g. savings in COG/g of up to 10%. Guidelines are provided for choosing an optimal purification process as a function of the objectives being optimized and impurity levels present. (C) 2014 Published by Elsevier B.V.
Energy efficiency (EE) is considered as one of the pivotal uplink (UL) performance metrics for 5G dense networks. It consists of two conflicting objectives that are recognised as benefit-cost ratio: spectral efficienc...
详细信息
Energy efficiency (EE) is considered as one of the pivotal uplink (UL) performance metrics for 5G dense networks. It consists of two conflicting objectives that are recognised as benefit-cost ratio: spectral efficiency and energy consumption. Accordingly, network design tradeoff is the key challenge for future networks. In this paper, we aim to jointly maximise network spectral efficiency and minimise power consumption with respect to the following design parameters: base station (BS) density, users' number, equipped antennas' number, and signal-to-noise power ratio without losing service quality. The performance of the ultra-dense network is characterised on the basis of the Pareto optimality concept through the following benchmarks: (i) studying impact of exhausted power on the deployed hardware elements. (ii) Validating the total EE performance through Monte-Carlo simulation within a low cost of processing time. (iii) The proposed approach is compared against single objective scheme to show the significance of the design tradeoff. Furthermore, we will introduce a detailed mathematical analysis for UL power policy and channel estimation for reliable pilot reusing. Simulation results will show that our proposed solution guarantee remarkable EE performance via reducing the number of deployed BSs without scarificing service quality.
Runtime analysis has recently been applied to popular evolutionary multi-objective (EMO) algorithms like NSGA-II in order to establish a rigorous theoretical foundation. However, most analyses showed that these algori...
详细信息
ISBN:
(纸本)9798400704949
Runtime analysis has recently been applied to popular evolutionary multi-objective (EMO) algorithms like NSGA-II in order to establish a rigorous theoretical foundation. However, most analyses showed that these algorithms have the same performance guarantee as the simple (G)SEMO algorithm. To our knowledge, there are no runtime analyses showing an advantage of a popular EMO algorithm over the simple algorithm for deterministic problems. We propose such a problem and use it to showcase the superiority of popular EMO algorithms over (G)SEMO: ONETRAPZEROTRAP is a straightforward generalization of the well-known TRAP function to two objectives. We prove that, while GSEMO requires at least n(n) expected fitness evaluations to optimise ONETRAPZEROTRAP, popular EMO algorithms NSGA-II, NSGA-III and SMS-EMOA, all enhanced with a mild diversity mechanism of avoiding genotype duplication, only require O(n log n) expected fitness evaluations. Our analysis reveals the importance of the key components in each of these sophisticated algorithms and contributes to a better understanding of their capabilities.
Runtime analysis has produced many results on the efficiency of simple evolutionary algorithms like the (1+1) EA, and its analogue called GSEMO in evolutionary multiobjective optimisation (EMO). Recently, the first ru...
详细信息
ISBN:
(纸本)9798400701191
Runtime analysis has produced many results on the efficiency of simple evolutionary algorithms like the (1+1) EA, and its analogue called GSEMO in evolutionary multiobjective optimisation (EMO). Recently, the first runtime analyses of the famous and highly cited EMO algorithm NSGA-II have emerged, demonstrating that practical algorithms with thousands of applications can be rigorously analysed. However, these results only show that NSGA-II has the same performance guarantees as GSEMO and it is unclear how and when NSGA-II can outperform GSEMO. We study this question in noisy optimisation and consider a noise model that adds large amounts of posterior noise to all objectives with some constant probability.. per evaluation. We show that GSEMO fails badly on every noisy fitness function as it tends to remove large parts of the population indiscriminately. In contrast, NSGA-II is able to handle the noise efficiently on LeadingOnesTrailingZeroes when.. < 1/2, as the algorithm is able to preserve useful search points. We identify a phase transition at.. = 1/2 where the expected time to cover the Pareto front changes from polynomial to exponential. This is the first proof that NSGA-II can outperform GSEMO and the first runtime analysis of NSGA-II in noisy optimisation.
The proximity of an approximation set to the Pareto-optimal front of a multiobjectiveoptimisation problem and the diversity of the solutions within the approximation set are two essential requirements in evolutionary...
详细信息
ISBN:
(纸本)9783642010194
The proximity of an approximation set to the Pareto-optimal front of a multiobjectiveoptimisation problem and the diversity of the solutions within the approximation set are two essential requirements in evolutionary multiobjective optimisation. These two requirements may be found to be in conflict with each other in many-objective optimisation scenarios deploying Pareto-dominance selection alongside active diversity promotion mechanisms. This conflict is hindering the optimisation process of some of the most established MOEAs and introducing problems such as the problem of dominance resistance and speciation. In this study, a diversity management operator (DMO) for controlling and promoting the diversity requirement in many-objective optimisation scenarios is introduced and tested on a set of test functions with increasing numbers (6 to 12) of objectives. The results achieved by the proposed strategy outperform results achieved by a reputed and representative MOEA in terms of both criteria: convergence and diversity.
暂无评论