SSIEA: a hybrid evolutionary algorithm for supporting conceptual architectural design

A new method of designing BP neural networks based on evolutionary algorithm (EA) is proposed for Knowledge Management in Electricity Supply Industry. The mechanisms of diversity maintaining and antibody density regulation exhibited in evolutionary system are introduced into evolutionary algorithm (EA). The proposed algorithm overcomes the problems of EA on search efficiency, individual diversity and premature and enhances the convergent performance effectively. In order to solve the problem of random initial weights, neuro fuzzy system for diversity is used to initialize weight vectors, and the detailed design steps of the algorithm are given. Simulated results show that the BP neural networks designed by EA have better performance in convergent speed and global convergence compared with hybrid evolutionary algorithm and the method is more accurate than other ones.

Around 3/4 of global resources are currently consumed in urban settlements, with corresponding adverse environmental consequences. According to population forecasts this situation will worsen in the coming years. It is therefore imperative that we understand how to design less resource intensive urban settlements. Software for the modelling and optimisation of resource flows are of interest to support urban designers in achieving this objective. To this end we have, in the first instance, tackled the problem of optimising the layout and form of buildings for the utilisation of solar radiation by using a multi-objective evolutionary algorithm. This new methodology facilitates a considerably more exhaustive search of the parameter space than manual trial and error which has been favoured in past studies. However, multi-objective optimisers involve many redundant evaluations when only one objective (the energy consumption) and analytical constraints (the volume of the urban form has to remain within bounds) are taken into account. To resolve this, we developed a new evolutionary algorithm that avoids evaluations of potential solutions that violate constraints. This is a hybrid based of the CMA-ES and HDE evolutionary algorithms. This hybrid algorithm achieves 100% convergence to the global minimum relating to two highly multi-modal benchmark functions and good results compared to the multi-objective evolutionary algorithm previously used and to a hybrid of heuristic and direct search methods (PSO/HJ) for real-world problems. Key contributions have also been made to the development of a new urban energy modelling tool: CitySim. These contributions include a mono and multi-zone thermal model, which is linked to an HVAC model that takes into account the increased energy demands due to the use of air as medium for heating, cooling and fresh air supply. In order to complete the provision of heating/cooling as well as electricity, the most commonly used energy conversion systems are modelled. Outputs from these systems may also be coupled with a model of sensible/latent heat storage. Finally the new hybrid evolutionary algorithm was used to optimise the energy performance of a case study of 26 buildings in the Matthaus district (Basel) using CitySim as the energy modelling tool. The results indicated that air conditioning plant should not be necessary in Switzerland if occupants behave appropriately. Concrete strategies for minimising the primary energy demand for the case study were also identified subject to constrained investment capital. This demonstrated that optimally (environmentally) sustainable solutions can be found at the urban scale, to help guide urban designers' decisions.

The challenges in computer supported conceptual engineering design

An expert system for the conceptual design of bridges

The Swan Coastal Plain is situated along the Western Australian seaboard and accommodates a large number of permanently, seasonally and episodically flooded wetlands. Many of these wetlands are affected by eutrophication and hydrological changes. A systematic monitoring program has been conducted between 1989 and 1990 to assess the environmental status of 41 selected wetlands based on measurements of 19 physical-chemical attributes and the collection of 253 macroinvertebrate taxa samples (Davis et al. 1993). This study analysed 35 wetlands with consistent data collected in Nov 1989 and Nov 1990 by means of Gradient Forest (GF) and the Hybrid Evolutionary Algorithm (HEA). Whilst GF allows identifying macroinvertebrate taxa with the “strongest overall response” to gradients in “important” physical-chemical attributes, HEA allows to model population dynamics of the taxa depending on “important” attributes identified by GF along all 35 wetlands. HEA models are represented by IF-THEN-ELSE rules whereby IF-conditions disclose attribute thresholds that indicate changes in the species abundance across the wetlands. GF suggested different ranking of the attributes EC, TN and DIP for both years as well as different taxa assemblages for same attributes in 1989 and 1990. Since results for merged data were also different, only the year-by-year specific results have been taken into account. When inferential models have been built for the 4 species that responded “strongest” to EC, DIP and TN in 1989 and 1990 by HEA, the threshold conditions fall in the range of overall gradients of these attributes discovered by GF. GF and HEA proved to be complementary tools for identifying overall attribute gradients and species- and site-specific thresholds in complex ecological data sets. Davis, J.A., Rosich, R.S., Bradley, J.S., Growns, J.E., Schmidt, L.G. and F. Cheal, 1993. Wetlands of the Swan Coastal Plain. Vol. 6: Wetland classification on the basis of water quality and invertebrate community data. Water Authority of Western Australia.

For problems involving uncertainties in design variables and parameters, a bi-objective evolutionary algorithm (EA) based approach to design optimization using evidence theory is proposed and implemented in this paper. In addition to a functional objective, a plausibility measure of failure of constraint satisfaction is minimized. Despite some interests in classical optimization literature, this is the first attempt to use evidence theory with an EA. Due to EA's flexibility in modifying its operators, nonrequirement of any gradient, its ability to handle multiple conflicting objectives, and ease of parallelization, evidence-based design optimization using an EA is promising. Results on a test problem and two engineering design problems show that the modified evolutionary multi-objective optimization algorithm is capable of finding a widely distributed trade-off frontier showing different optimal solutions corresponding to different levels of plausibility failure limits. Furthermore, a single-objective evidence-based EA is found to produce better optimal solutions than a previously reported classical optimization algorithm. Furthermore, the use of a graphical processing unit (GPU) based parallel computing platform demonstrates EA's performance enhancement around 160–700 times in implementing plausibility computations. Handling uncertainties of different types are getting increasingly popular in applied optimization studies and this EA based study is promising to be applied in real-world design optimization problems.

Microarray technology in the last decade has been widely applied to detect correlated biomarkers in biological processes. Due to the need for analyzing massive amounts of generated data in this technology, computational intelligence approaches are used increasingly in this field. Biclustering algorithms are one of the most important of these techniques in microarray analysis. Two aspects of search mechanisms and biologists' desired patterns are the most essential issues in the design and evaluation of these algorithms. Different patterns can be achieved by considering different metrics. In this paper, Transposed Virtual Error (VET) is used as main metric. Also a hybrid evolutionary algorithm is proposed based on Evo-Bexpa algorithm which is introduced with VET. The proposed method is developed based on Genetic Algorithm (GA) used in Evo-Bexpa and Asexual Reproduction Optimization (ARO). The results indicate that underlying algorithm against Evo-Bexpa is more efficient in finding biclusters.

Modeling and prediction for discharge lifetime of battery systems using hybrid evolutionary algorithms

In this paper, we present a performance-based approach to building configuration design to improve the urban ventilation potential at the conceptual design stage, and we demonstrate its application through a case study. The target performance optimized was the ventilation potential of a district, including a region of interest at a spatial scale of hundreds of meters. To estimate this performance, we used computational fluid dynamics (CFD), coupled with an evolutionary algorithm, to optimize the design alternatives to produce the building configuration most suitable for a given set of site conditions. Three calculation components must be assembled for a CFD-based design optimization: an optimizer, a geometry/mesh generator, and a CFD solver. To provide links between the calculation components, we utilized an in-house parametric design program. A case study was conducted to test the applicability of the proposed design method to identify the optimal solutions that minimize adverse effects on the ventilation potential of the surrounding area. For a configuration of buildings in a dense urban area, the proposed design method successfully improved the design alternatives. The results show that the urban ventilation potential in the case of the optimized building configuration is 16% greater than that of the initial building configuration.

A multi-objective feedback approach for evaluating sequential conceptual building design decisions

This control decision addresses the Knock-Out Pot (KOP) Disposition KOP Processing System (KPS) conceptual design. The KPS functions to (1) retrieve KOP material from canisters, (2) remove particles less than 600 {micro}m in size and low density materials from the KOP material, (3) load the KOP material into Multi-Canister Overpack (MCO) baskets, and (4) stage the MCO baskets for subsequent loading into MCOs. Hazard and accident analyses of the KPS conceptual design have been performed to incorporate safety into the design process. The hazard analysis is documented in PRC-STP-00098, Knock-Out Pot Disposition Project Conceptual Design Hazard Analysis. The accident analysis is documented in PRC-STP-CN-N-00167, Knock-Out Pot Disposition Sub-Project Canister Over Lift Accident Analysis. Based on the results of these analyses, and analyses performed in support of MCO transportation and MCO processing and storage activities at the Cold Vacuum Drying Facility (CVDF) and Canister Storage Building (CSB), control decision meetings were held to determine the controls required to protect onsite and offsite receptors and facility workers. At the conceptual design stage, these controls are primarily defined by their safety functions. Safety significant structures, systems, and components (SSCs) that could provide the identified safety functions have been selected for the conceptual design. It is anticipated that some safety SSCs identified herein will be reclassified based on hazard and accident analyses performed in support of preliminary and detailed design.

As the description of design requirements at the earlier design stage is inaccurate and vague, it is difficult to figure out functional structure of a product and make sense product configuration. Therefore, it plays an important role to formally represent the process of design for product development in the conceptual design stage. Furthermore, port, as the location of intended interaction, is crucial to capture component concept and realize conceptual design for multi-solution generation. Agent is considered as an effective approach to collaboratively implementing design problem solving and reasoning. Combining both port and agent may be employed to generate new concepts of the product in order to customize product scheme varieties. In this paper, the product module attributes are firstly described. The objective is to implement modeling of design process for obtaining system new concepts to guide multi-solution generation. Secondly, an effective approach to decomposing design process is presented to describe the process of structure generations and product decomposition by formal representation. According to properties of modularity for product development and component connections, we can calculate the number of component connections and density of components. In addition, product module division and coupling degree analysis are conducted, and coupling degrees are calculated by considering the correspondence ratio and the cluster independence. A port-based knowledge building process is described for functional modeling. A port-agent collaborative design framework is given and describes different agent functions to help designers to obtain new design schemes. Finally, a case study is presented to describe the modeling process of conceptual design.

In this paper, a hybrid evolutionary algorithm (HEA) based on the approaches of the evolutionary algorithm and a local search (LS) is proposed to determine the gene signatures for predicting histologic response of chemotherapy on osteosarcoma patients, which is one of the most common malignant bone tumor in children. The HEA consists of a population of individuals but the evolution of individuals is conducted by a LS, rather than the crossover and mutation used in the traditional evolutionary algorithms. The proposed HEA can simultaneously optimize the feature subset and the classifier through a common solution coding mechanism. Experimental results indicate that HEA can obtain more accurate signatures than the other existing approaches in determining chemoresponse for osteosarcoma.

Water is an essential resource from an environmental, biological, economic or social point of view. In basin management, the irregular distribution in time and in space of this resource is well known. This issue is worsened by extreme climate conditions, generating drought periods or flood events. For both situations, optimal management is necessary. In one case, different water uses should be supplied efficiently using the available surface and groundwater resources. In another case, the most important goal is to avoid damages in flood areas, including the loss of human lives, but also to optimize the revenue of energy production in hydropower plants, or in other uses.

Presented in this paper is a hybrid algorithm simplex search enabled evolutionary algorithm (SSEA) which is fundamentally an evolutionary algorithm (EA) embedded with a local simplex search for unconstrained optimization problems. Evolutionary algorithms have been quite successful in solving a wide class of intractable problems and the non-dominated sorting genetic algorithm (NSGA-II) is a popular choice. However, like any other evolutionary algorithms, the rate of convergence of NSGA-II slows down with generations and often there is no improvement in the best candidate solution over a number of generations. The simplex search component comes into effect once the basic evolutionary algorithm encounters a slow rate of convergence. To allow exploitation around multiple promising regions, the simplex search is invoked from multiple promising regions of the variable space identified using hierarchical agglomerative clustering. In this paper, results are presented for a series of unconstrained optimization test problems that cover problems with a single minimum, a few minima and a large number of minima. Provided is a comparison of results with NSGA-II, fast evolutionary strategy (FES), fast evolutionary programming (FEP) and improved fast evolutionary programming (IFEP) where it's clear that SSEA outperforms all other algorithms for unimodal problems. On the suite of problems with large number of minima, SSEA performs better on some of them. For problems with fewer minima, SSEA performs better than FES, FEP and IFEP while demonstrating comparable performance to NSGA-II.