Ian C. Parmee

Evolutionary and adaptive computing in engineering design

1.1 Setting the Scene.- 1.2 Why Evolutionary/Adaptive Computing?.- 1.3 The UK EPSRC Engineering Design Centres.- 1.4 Evolutionary and Adaptive Computing Integration.- 1.4.1 The Design Process.- 1.4.2 Routine, Innovative and Creative Design.- 1.4.3 Complementary Computational Intelligence Techniques.- 1.5 Generic Design Issues.- 1.6 Moving On.- 2. Established Evolutionary Search Algorithms.- 2.1 Introduction.- 2.2 A Brief History of Evolutionary Search Techniques.- 2.3 The Genetic Algorithm.- 2.3.1 The Simple Genetic Algorithm.- 2.3.2 Binary Mapping and the Schema Theorem.- 2.3.3 Real Number Representation.- 2.3.4 The Operators.- 2.3.5 Elitism and Exploitation versus Exploration.- 2.3.6 Self-adaptation.- 2.4 GA Variants.- 2.4.1 The CHC Genetic Algorithm.- 2.4.2 The EcoGA.- 2.4.3 The Structured Genetic Algorithm.- 2.4.4 The Breeder GA and the Messy GA.- 2.5 Evolution Strategies.- 2.6 Evolutionary Programming.- 2.7 Genetic Programming.- 2.8 Discussion.- 3. Adaptive Search and Optimisation Algorithms.- 3.1 Introduction.- 3.2 The Ant-colony Metaphor.- 3.3 Population-based Incremental Learning.- 3.4 Simulated Annealing.- 3.5 Tabu Search.- 3.6 Scatter Search.- 3.7 Discussion.- 4. Initial Application.- 4.1 Introduction.- 4.2 Applying the GA to the Shape Optimisation of a Pneumatic, Low-head, Hydropower Device.- 4.3 The Design ofGas Turbine Blade Cooling Hole Geometries.- 4.3.1 Introduction.- 4.3.2 Integrating the Cooling Hole Model with a Genetic Algorithm.- 4.3.3 Further Work.- 4.4 Evolutionary FIR Digital Filter Design.- 4.4.1 Introduction.- 4.4.2 Coding Using a Structured GA.- 4.4.3 Fitness Function.- 4.4.4 Results.- 4.5 Evolutionary Design of a Three-centred Concrete Arch Dam.- 4.6 Discussion.- 5. The Development of Evolutionary and Adaptive Search Strategies for Engineering Design.- 5.1 Introduction.- 5.2 Cluster-oriented Genetic Algorithms.- 5.3 The GAANT (GA-Ant) Algorithm.- 5.4 DRAM and HDRAM Genetic Programming Variants.- 5.5 Evolutionary and Adaptive Search Strategies for Constrained Problems.- 5.6 Evolutionary Multi-criterion Satisfaction.- 5.7 Designer Interaction within an Evolutionary Design Environment.- 5.8 Dynamic Shape Refinement and Injection Island Variants.- 5.9 Discussion.- 6. Evolutionary Design Space Decomposition.- 6. I Introduction.- 6.2 Multi-modal Optimisation.- 6.3 Cluster-oriented Genetic Algorithms.- 6.4 Application of vmCOGA.- 6.4.1 Two-dimensional Test Functions.- 6.4.2 Engineering Design Domains.- 6.4.3 Single-objective/Continuous Design Space.- 6.4.4 Multi-level , Mixed-parameter Design Space.- 6.5 Alternative COGA Structures.- 6.5.1 Introduction.- 6.5.2 The COGA Variants.- 6.5.3 Summary of Results.- 6.5.4 Search Space Sampling.- 6.5.5 The Dynamic Adaptive Filter.- 6.6 Agent-assisted Boundary Identification.- 6.7 Discussion.- 7. Whole-system Design.- 7.1 Introduction.- 7.1.1 Whole-system Design.- 7.1.2 Designer Requirement.- 7.1.3 Design Environments.- 7.2 Previous Related Work.- 7.3 The Hydropower System.- 7.3.1 The System.- 7.3.2 The Model.- 7.4 The Structured Genetic Algorithm.- 7.4.1 The Algorithm.- 7.4.2 Dual Mutation Strategies.- 7.4.3 stGA Results.- 7.5 Simplifying the Parameter Representation.- 7.6 Results and Discussion.- 7.7 Thermal Power System Redesign.- 7.7.1 Introduction.- 7.7.2 Problem Definition.- 7.7.3 A Hybrid GA-SLP Algorithm.- 7.7.4 The Design Application.- 7.8 Discussion.- 8. Variable-length Hierarchies and System Identification.- 8.1 Introduction.- 8.2 Improving Rolls Royce Cooling Hole Geometry Models.- 8.2.1 Introduction.- 8.2.2 Simple Curve and Surface Fitting.- 8.2.3 Evolving Formulae to Determine the Friction Factor in Turbulent Pipe Flow.- 8.2.4 Eddy Correlations for Laminar Two-dimensional Sudden Expansion Flows.- 8.3 Discussion of Initial Application.- 8.4 Further Development of the GP Paradigm.- 8.4.1 Development of Node Complexity Ratings.- 8.4.2 Constrained-complexity Crossover.- 8.4.3 Steady-state GP.- 8.4.4 Injection Mutation.- 8.5 Symbolic Regression with HDRAM-GP.- 8.6 Dual-agent Integration.- 8.7 Return to Engineering Applications.- 8.7.1 Introduction.- 8.7.2 Explicit Formula for Friction Factor In Turbulent Pipe Flow.- 8.7.3 Eddy Correlations for Laminar Two-dimensional Sudden Expansion Flows.- 8.7.4 Thermal Paint Jet Turbine Blade Data.- 8.8 Discussion.- 9. Evolutionary Constraint Satisfaction and Constrained Optimisation.- 9.1 Introduction.- 9.2 Dealing with Explicit Constraints.- 9.2.1 The Fault Coverage Test Code Generation Problem.- 9.2.2 The Inductive Genetic Algorithm.- 9.2.3 Application to the Problem.- 9.3 Implicit Constraints.- 9.4 Defining Feasible Space.- 9.4.1 Introduction.- 9.4.2 The Problem Domain.- 9.4.3 Fixing a Feasible Point.- 9.4.4 Creating a Feasible Subset.- 9.4.5 Establishing the Degree of Constraint Violation.- 9.4.6 Results and Discussion.- 9.5 Satisfying Constraint in the Optimisation of Thermal Power Plant Design.- 9.6 GA/Ant-colony Hybrid for the Flight Trajectory Problem.- 9.6.1 The Problem Domain.- 9.6.2 The Ant-colony Model for Continuous-space Search.- 9.6.3 A Hybrid Search Framework.- 9.7 Other Techniques.- 9.8 Discussion.- 10. Multi-objective Satisfaction and Optimisation.- 10.1 Introduction.- 10.2 Established Multi-objective Optimisation Techniques.- 10.2.1 Weighted-sum-based Optimisation.- 10.2.2 Lexicographic Order-based Optimisation.- 10.2.3 The Pareto Method.- 10.2.4 Pareto Examples.- 10.2.5 The Vector-evaluated Genetic Algorithm.- 10.2.6 Comparison of the Various Techniques.- 10.3 Interactive Approaches to Multi-objective Satisfaction/Optimisation.- 10.4 Qualitative Evaluation ofGA-generated Design Solutions.- 10.4.1 Introduction.- 10.4.2 The Design Model.- 10.4.3 Adaptive Restricted Tournament Selection.- 10.4.4 Assessing the Qualitative Fitness of High-performance Solutions.- 10.4.5 Knowledge Representation.- 10.4.6 Typical Results.- 10.4.7 Further Work.- 10.5 Cluster-oriented Genetic Algorithms for Multi-objective Satisfaction.- 10.6 Related Work and Further Reading.- 10.7 Discussion.- 11. Towards Interactive Evolutionary Design Systems.- 11.1 Introduction.- 11.2 System Requirements.- 11.3 The Design Environment and the IEDS.- 11.4 The Rule-based Preference Component.- 11.4.1 Introduction.- 11.4.2 Preferences.- 11.4.3 Example Application.- 11.5 The Co-evolutionary Environment.- 11.5.1 Introduction.- 11.5.2 Initial Methodology.- 11.5.3 The Range Constraint Map.- 11 .5.4 Sensitivity Analysis.- 11.5.5 Results.- 11.6 Combining Preferences with the Co-evolutionary Approach.- 11.7 Cluster-oriented Genetic Algorithm s as Information Gathering Processes.- 11.7.1 Introduction.- 11.7.2 Extraction and Processing of COGA-generated Data.- 11.8 Machine-based Agent Support.- 11.8.1 Introduction.- 11.8.2 Interface Agents.- 11.8.3 Communication Agents.- 11.8.4 Search Agents.- 11.8.5 Information Processing Agents.- 11.8.6 Negotiating Agents.- 11.9 Machine-based Design Space Modification.- 11.9.1 Introduction.- 11.9.2 The Developed EcoGA Framework.- 11.9.3 Determining Direction and Extent of Design Space Extension.- 11.10 Discussion.- 12. Population-based Search, Shape Optimisation and Computational Expense.- 12.1 Introduction.- 12.2 Parallel , Distributed and Co-evolutionary Strategies.- 12.3 Introducing the Problem and the Developed Strategies.- 12.4 The Evaluation Model.- 12.5 Initial Results.- 12.6 Dynamic Shape Refinement.- 12.6.1 Introduction.- 12.6.2 Stand-alone CHC and DSR CHC.- 12.7 The Injection Island GA.- 12.8 Dynamic Injection.- 12.9 Distributed Search Techniques.- 12.9.1 Introduction.- 12.9.2 Co-operative Search.- 12.10 Discussion.- 13. Closing Discussion.- 13.1 Introduction.- 13.2 Difficulties Facing Successful Integration ofEC with Engineering Design.- 13.3 Overview of the Techniques and Strategies Introduced.- 13.4 Final Remarks.- Appendix A. Some Basic Concepts.- References.

Interactive, Evolutionary Search in Upstream Object-Oriented Class Design

Although much evidence exists to suggest that early life cycle software engineering design is a difficult task for software engineers to perform, current computational tool support for software engineers is limited. To address this limitation, interactive search-based approaches using evolutionary computation and software agents are investigated in experimental upstream design episodes for two example design domains. Results show that interactive evolutionary search, supported by software agents, appears highly promising. As an open system, search is steered jointly by designer preferences and software agents. Directly traceable to the design problem domain, a mass of useful and interesting class designs is arrived at which may be visualized by the designer with quantitative measures of structural integrity, such as design coupling and class cohesion. The class designs are found to be of equivalent or better coupling and cohesion when compared to a manual class design for the example design domains, and by exploiting concurrent execution, the runtime performance of the software agents is highly favorable.

Improving problem definition through interactive evolutionary computation

Poor definition and uncertainty are primary characteristics of conceptual design processes. During the initial stages of these generally human-centric activities, little knowledge pertaining to the problem at hand may be available. The degree of problem definition will depend on information available in terms of appropriate variables, constraints, and both quantitative and qualitative objectives. Typically, the problem space develops with information gained in a dynamical process in which design optimization plays a secondary role, following the establishment of a sufficiently well-defined problem domain. This paper concentrates on background human–computer interaction relating to the machine-based generation of high-quality design information that, when presented in an appropriate manner to the designer, supports a better understanding of a problem domain. Knowledge gained from such information combined with the experiential knowledge of the designer can result in a reformulation of the problem, providing increased definition and greater confidence in the machine-based representation. Conceptual design domains related to gas turbine blade cooling systems and a preliminary air frame configuration are introduced. These are utilized to illustrate the integration of interactive evolutionary strategies that support the extraction of optimal design information, its presentation to the designer, and subsequent human-based modification of the design domain based on knowledge gained from the information received. An experimental iterative designer or evolutionary search process resulting in a better understanding of the problem and improved machine-based representation of the design domain is thus established.

Elegant Object-Oriented Software Design via Interactive, Evolutionary Computation

Design is fundamental to software development but can be demanding to perform. Thus, to assist the software designer, evolutionary computing is being increasingly applied using machine-based, quantitative fitness functions to evolve software designs. However, in nature, elegance and symmetry play a crucial role in the reproductive fitness of various organisms. In addition, subjective evaluation has also been exploited in interactive evolutionary computation (IEC). Therefore, to investigate the role of elegance and symmetry in software design, four novel elegance measures are proposed which are based on the evenness of distribution of design elements. In controlled experiments in a dynamic IEC environment, designers are presented with visualizations of object-oriented software designs, which they rank according to a subjective assessment of elegance. For three out of the four elegance measures proposed, it is found that a significant correlation exists between elegance values and reward elicited. These three elegance measures assess the evenness of distribution of 1) attributes and methods among classes; 2) external couples between classes; and 3) the ratio of attributes to methods. It is concluded that symmetrical elegance is in some way significant in software design, and that this can be exploited in dynamic, multiobjective IEC to produce elegant software designs.

Agent-based support within an interactive evolutionary design system

This paper describes the use of software agents within an interactive evolutionary conceptual design system. Several different agent classes are introduced (search agents, interface agents, and information agents) and their function within the system is explained. A preference modification agent is developed and an example is given illustrating the use of agents in preference modeling.

A cross-disciplinary technology transfer for search-based evolutionary computing: from engineering design to software engineering design

Although object-oriented conceptual software design is difficult to learn and perform, computational tool support for the conceptual software designer is limited. In conceptual engineering design, however, computational tools exploiting interactive evolutionary computation (EC) have shown significant utility. This article investigates the cross-disciplinary technology transfer of search-based EC from engineering design to software engineering design in an attempt to provide support for the conceptual software designer. Firstly, genetic operators inspired by genetic algorithms (GAs) and evolutionary programming are evaluated for their effectiveness against a conceptual software design representation using structural cohesion as an objective fitness function. Building on this evaluation, a multi-objective GA inspired by a non-dominated Pareto sorting approach is investigated for an industrial-scale conceptual design problem. Results obtained reveal a mass of interesting and useful conceptual software design solution variants of equivalent optimality—a typical characteristic of successful multi-objective evolutionary search techniques employed in conceptual engineering design. The mass of software design solution variants produced suggests that transferring search-based technology across disciplines has significant potential to provide computationally intelligent tool support for the conceptual software designer.

Integrating aesthetic criteria with evolutionary processes in complex, free-form design – an initial investigation

This research is a continuation of previous work by the authors relating to the inclusion of aesthetic criteria within an interactive evolutionary design system. The work described extends the design system re the manipulation of a more complex design problem with increased importance placed on aesthetic criterion. The paper initially introduces the previous work before describing more recent research and positioning this in terms of previously published work. Finally, initial results from the further developed design system are presented.

Single and multi-objective genetic operators in object-oriented conceptual software design

This poster paper investigates the potential of single and multi-objective genetic operators with an object-oriented conceptual design space. Using cohesion as an objective fitness function, genetic operators inspired by genetic algorithms and evolutionary programming are compared against a simple case study. Also, using both cohesion and coupling as objective fitness functions, multi-objective genetic operators inspired by a non-dominated sorting algorithm have been developed. Cohesion and coupling values achieved are similar to human performed designs and a large number and variety of optimal solutions are arrived at, which could not have been produced by the human software engineer. We conclude that this mass of optimal design variants offers significant potential for design support when integrated with user-centric, computationally intelligent tools.

Building Compact Rulesets for Describing Continuous-Valued Problem Spaces Using a Learning Classifier System

Learning Classifier Systems have previously been shown to have some application in deducing the characteristics of complex multi-modal test environments to a suitable level of accuracy. In this study, the issue of presenting human-readable rulesets to a potential user is addressed. In particular, two existing ruleset compaction algorithms originally devised for rulesets with an integer-valued representation are applied to rulesets with a continuous-valued representation. The algorithms are used to reduce the size of rulesets evolved by the XCS classifier system. Following initial testing, both algorithms are modified to take into account problems associated with the new representation. Finally, the modified algorithms are shown to outperform the originals.

User-centered, evolutionary search in conceptual software design

Although much evidence exists to suggest that conceptual software engineering design is a difficult task for software engineers to perform, current computationally intelligent tool support for software engineers is limited. While search-based approaches involving module clustering and refactoring have been proposed and show promise, such approaches are downstream in terms of the software development lifecycle - the designer must manually produce a design before search-based clustering and refactoring can take place. Interactive, user-centered search-based approaches, on the other hand, support the designer at the beginning of, and during, conceptual software design, and are investigated in this paper by means of a case study. Results show that interactive evolutionary search, supported by software agents, appears highly promising. As an open system, search is steered jointly by designer preferences and software agents. Directly traceable to the design problem domain, a mass of useful and interesting conceptual class designs are arrived at which may be visualized by the designer with quantitative measures of structural integrity such as design coupling and class cohesion. The conceptual class designs are found to be of equivalent or better coupling and cohesion when compared to a manual conceptual design of the case study, and by exploiting concurrent execution, the performance of the software agents is highly favorable.