Abstract: Scientific planning of urban spatial layout and delimiting of urban growth boundary is the key to implement the “three control lines” of the Ministry of natural resources and the rapid development of cities with high quality, which is of great guiding significance for the optimization of urban space. Taking Hangzhou City as an example, this paper applies Euclidean distance to establish the influence factor model, and analyzes various factors affecting the urban growth boundary, such as urban spatial terrain factors, natural environment factors, current urban land classification factors, industrial and economic environment factors and modern service network layout.；Then the coupling neural network is applied to identify development potentials of urban space；Finally, a multi-dimensional simulation of the urban spatial expansion and growth boundary of Hangzhou City in 2030 is carried out by using the coupled neural network and cellular automata (ANN-CA), according to different constraints (strict constraints, basic restrictions, control area restrictions) and different scenarios constraints such as sustainable development, priority development of planning and construction land in the core area. The main conclusions are as follows: (1) From 2018 to 2030, Xiaoshan District, Jianggan District and Yuhang District have the largest urban area expansion, with the growth area of 83.75, 51.96 and 31.32 km² respectively, which are the regions with great development potential in Hangzhou. (2) Based on the spatial strategy of “city expansion to the East, tourism planning to the west, development along the river, and cross-river development”, Hangzhou will focus on the East and the West in the future. (3) Based on the principles of ecological protection and intensive economy, the future growth boundary of Hangzhou city is defined, which aims to provide reference for the land space planning and related practice of Hangzhou.