Neural Network Model Design for Landscape Ecological Planning Assessment Based on Hierarchical Analysis

In this paper, an in-depth study and analysis of landscape ecological planning and evaluation are carried out using the analytic hierarchy process (AHP) algorithm that integrates neural networks. The application of AHP in the field of tree species planning and the introduction of quantitative analysis methods can effectively change the subjectivity of previous qualitative analysis in tree species selection and make it objective, scientific, and reasonable. The research can provide a reference for other urban tree species planning. From the connotation of landscape ecological service process and ecological space structure, the analysis of landscape ecological service process involves service supply area and service association area, which correspond to different key components of ecological space structure. With the help of the platform, based on the identification and identification methods and theories of ecological spatial structure, the key components of ecological spatial structure in different environments are identified and extracted by using the representation model, binary suitability model, weighted suitability model, and process model. The type of service is based on the different service processes supported by the key components of the ecological spatial structure, forming the ecological spatial structure under different service types. Spatial structure; on this basis, the basic characteristics of the key components of the ecological spatial structure are analyzed, and the correlation characteristics of the ecological spatial structure are analyzed based on the correlation classification system of ecological spatial structure. A backpropagation (BP) neural network-based state assessment method of the grid structure is established. The method takes the parameters of the autoregressive model constructed by the acceleration signals of different working conditions as the feature quantity and the results of the fuzzy hierarchical analysis method as the labels, divides the data set into a training set and a test set, and uses the BP neural network learning method and the training set to supervised train the BP neural network learning assessment model. The test set is used to test the effectiveness and accuracy of the BP neural network-based learning method. The study shows that the evaluation system established by the BP neural network structure is fast and accurate and can substantially reduce the cost of manual testing.