Optimised design of energy efficient building façades via Evolutionary Neural Networks

Zemella, Giovanni; March, Davide De; Borrotti, Matteo; Poli, Irene

doi:10.1016/j.enbuild.2011.10.006

Cited by 72 publications

(26 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Tresidder et al [33] and Zemella et al [34], make use of surrogates in EAs for discrete building design problems. Bajer and Holeňa [35] also used a RBFN surrogate, for a mixed continuous and discrete problem.…”

Section: Related Workmentioning

confidence: 99%

Constrained, mixed-integer and multi-objective optimisation of building designs by NSGA-II with fitness approximation

Brownlee

Wright

2015

Applied Soft Computing

105

View full text Add to dashboard Cite

a b s t r a c tReducing building energy demand is a crucial part of the global response to climate change, and evolutionary algorithms (EAs) coupled to building performance simulation (BPS) are an increasingly popular tool for this task. Further uptake of EAs in this industry is hindered by BPS being computationally intensive: optimisation runs taking days or longer are impractical in a time-competitive environment. Surrogate fitness models are a possible solution to this problem, but few approaches have been demonstrated for multi-objective, constrained or discrete problems, typical of the optimisation problems in building design. This paper presents a modified version of a surrogate based on radial basis function networks, combined with a deterministic scheme to deal with approximation error in the constraints by allowing some infeasible solutions in the population. Different combinations of these are integrated with NonDominated Sorting Genetic Algorithm II (NSGA-II) and applied to three instances of a typical building optimisation problem. The comparisons show that the surrogate and constraint handling combined offer improved run-time and final solution quality. The paper concludes with detailed investigations of the constraint handling and fitness landscape to explain differences in performance.

show abstract

Section: Related Workmentioning

confidence: 99%

Constrained, mixed-integer and multi-objective optimisation of building designs by NSGA-II with fitness approximation

Brownlee

Wright

2015

Applied Soft Computing

105

View full text Add to dashboard Cite

show abstract

“…For the solution of this problem several methods have been proposed in the past [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19][20], which refer to almost all factors influencing the thermal behavior of a building, including occupant behavior and control and to methods of optimal design. The later are mostly based on multivariate analysis of data and on mathematical optimization methods.…”

Section: Introductionmentioning

confidence: 99%

Building Retrofitting using Hierarchical Optimization and Principal Component Analysis

Kanarachos¹,

Kanarachos²

2014

AJS

View full text Add to dashboard Cite

In view of the needs for optimization (performance, costs) of building retrofitting, several methods have been proposed in the past. These methods are mostly based on multivariate analysis of data and on mathematical optimization methods. However, the complexity of the task reveals the need for further methodological developments. In this paper a new method based on Hierarchical Optimization (HO) and principal component analysis is proposed. Hierarchical optimization is concerned with decision making problems that involve multiple decision makers ordered within a hierarchical structure and has proved already its usefulness. It helps transforming a global optimization problem into a number of local ones, which also consort with engineering knowledge and practice. In this sense a principal component analysis helps greatly the definition of efficient sub-optimization problems. The above methodology is presented in the following and a case study illustrates its usefulness.

show abstract

“…This iterative process, the improvement generation by generation, the exploitation of the more informative areas (according to a fitness measure) are driven by a model selection procedure that identifies the best neural network at each generation of the algorithm. Previous works showed that an ENN‐Design approach compares positively in high dimensional problem domains against an optimization based on a simple genetic algorithm and efficiently explore solutions in multi‐objective problems . Moreover, ENN‐Design has been shown to be effective in the sequential design of experimental points to be collected in an adaptive way .…”

Section: Introductionmentioning

confidence: 99%

T‐optimality and neural networks: a comparison of approaches for building experimental designs

Berni

March

Stefanini

2012

Appl Stoch Models Bus & Ind

Self Cite

View full text Add to dashboard Cite

This paper deals with optimal experimental design criteria and neural networks in the aim of building experimental designs from observational data. It addresses the following three main issues: (i) the introduction of two radically different approaches, namely T-optimal designs extended to Generalized Linear Models and Evolutionary Neural Networks Design; (ii) the proposal of two algorithms, based on model selection procedures, to exploit the information of already collected data; and (iii) the comparison of the suggested methods and corresponding algorithms by means of a simulated case study in the technological field.Results are compared by considering elements of the proposed algorithms, in terms of models and experimental design strategies. In particular, we highlight the algorithmic features, the performances of the approaches, the optimal solutions and the optimal levels of variables involved in a simulated foaming process. The optimal solutions obtained by the two proposed algorithms are very similar, nevertheless, the differences between the paths followed by the two algorithms to reach optimal values are substantial, as detailed step-by-step in the discussion.Optimal designs are basically model dependent; therefore, model discrimination has received great attention in literature also following different approaches, which can be grouped according to the methodological approach. The first class of contributions deals with model discrimination together with other issues, in particular the parameter estimation. In this regard, following the seminal contributions by [14,15] and [7], constrained optimality based on D-optimality is introduced, and the concept of canonical moments is expounded in order to find an optimal design and to choose the best model. In [16], an optimal robust design for polynomial regression models is developed by eliciting priors onˇ-coefficients as weight, as previously introduced in [15].In addition, in [17], generalization of previous optimality measures, such as in [4], is developed; while in [18], further issues are studied for the Fourier regression model and the power of the F -test. It must be noted that in [18], for discriminating among designs/models, the weights and the power of tests are evaluated by considering D-optimal restricted criterions.Undoubtedly, modeling discrimination and parameter estimation compared with T-optimality, as stated in [19], share common features and properties but, at the same time, the sequential nature of T-optimality may be viewed as a specific characteristic that is not a fundamental feature when applying D-optimality [1,5]. Nonetheless, as in [20] and [21], the T-optimality algorithm works through an adaptive sequential scheme, and the power of the F -test is affected by this selection process.A different approach for efficiently designing experimental strategies is to adopt an evolutionary paradigm [22][23][24][25][26][27]. This approach is fundamentally led by an optimization process that iteratively evolves the initial design toward some op...

show abstract

Optimised design of energy efficient building façades via Evolutionary Neural Networks

Cited by 72 publications

References 8 publications

Constrained, mixed-integer and multi-objective optimisation of building designs by NSGA-II with fitness approximation

Constrained, mixed-integer and multi-objective optimisation of building designs by NSGA-II with fitness approximation

Building Retrofitting using Hierarchical Optimization and Principal Component Analysis

T‐optimality and neural networks: a comparison of approaches for building experimental designs

Contact Info

Product

Resources

About