Title: Ecological assessment of low carbon design of garden based on optimal BP neural network

Authors: Min Yu; Yahui Zhang; Fangrong Yang; Panpan Jiao

Addresses: College of Information Technology and Urban Construction, Luoyang Polytechnic, Luoyang, 471000, China ' Project Department, Zhengzhou Xindaguan Real Estate Co., Ltd, Zhengzhou, 450000, China ' College of Landscape Architecture and Art, Henan Agricultural University, Zhengzhou, 450000, China ' College of Art and Design, Zhengzhou University of Economics and Business, Zhengzhou, 450000, China

Abstract: The contemporary ecological effect evaluation methods are mostly applied to a number of building industries in large municipal areas, and the ecological evaluation criteria for small area-wide buildings such as gardens are not very clear, leading to a lack of relevant evaluation guidance for garden design and architecture. The error of the proposed model is compared with the traditional algorithm, and the application effect of the model is analysed with the actual garden samples. The experimental results show that compared with the traditional BP neural network, the three output errors of the back propagation-genetic algorithm neural network model are reduced by 7.1%, 4.09% and 2.6%. At the same time, the low carbon design of the gardens reduced the concentration of SO₂, NO₂ harmful gases in the air of the town area by 0.193 mg/m³ and 0.263 mg/m³, respectively.

Keywords: landscape design; back propagation neural network; genetic algorithm; ecological environment assessment; low carbon effect.

DOI: 10.1504/IJESD.2024.137785

International Journal of Environment and Sustainable Development, 2024 Vol.23 No.2/3, pp.158 - 175

Received: 05 Aug 2022
Accepted: 19 Dec 2022
Published online: 05 Apr 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article