The law of network propagation is one of the important research topics of complex networks and complexity sciences. In the financial field, network transmission is manifested as risk transmission in a complex network composed of enterprises and their relations, also known as risk transmission. The current research work is mainly focused on the law of transmission under a single relationship, while in the real world, risk transmission under complex relationships is the main form. Therefore, researches on risk propagation algorithms based on complex relationships are very important. In the traditional risk propagation model, relevant scholars have proposed a nonlinear point-to-point risk propagation method. The disadvantage is that it does not consider the network structure of complex multidimensional relationships. If the relationship type is complex, its propagation intensity cannot be calculated quantitatively. Therefore, the risk micropropagation model and relationship fitness under multidimensional and complex relationships are introduced, and a super-network structure system is established. Based on the accumulation idea of compound series, a microscopic superposition algorithm of relations in hyper-network space is given, which maps the multi-dimensional compound relationship to a single dimension, making its comprehensive influence tend to a critical value, and then simulates the risk propagation under the multi-dimensional relationship.