The arid and semi-arid areas of Northwest China are ecologically fragile areas, the construction of ecological networks can connect broken habitat patches and enhance connectivity between landscapes. The layout optimization of ecological nodes can reduce energy consumption and increase stability, which is of great significance to maintain the stability and security of regional ecological environment. Therefore, based on the typical ecologically vulnerable area—Dengkou County, remote sensing image interpretation data in 2015 was used as the research material. Eco existing network infrastructure, with improved H&V algorithm (force-directed model) optimized by optimizing the layout of ecological network node. The results showed that compared with the H&V algorithm, the optimized ecological node layout coverage of force-model optimization was 90.79%, which was increased by 4.08 percentage points. The average clustering coefficient was increased to 0.071, which was 1.4 times of that of the unmodified H&V algorithm. The distribution uniformity was reduced to 2.629, which was 0.629 lower than that of the unmodified H&V algorithm. By model optimization, the network structure was clear, the ecological node layout was uniform, and the node coverage was higher, indicating that the optimized ecological network structure was more stable.