With the development of equipment towards complexity,integration and intelligence,how to quickly and accurately construct high-fidelity digital twin models is the fundamental task and biggest technical obstacle for intelligent operation and maintenance of complex equipment.A strategy based on the design structure matrix (DSM) for task coupling optimizing and intelligent distribution was proposed to realize comprehensive dynamic supervision of the multi-point modeling of complex equipment.An improved genetic algorithm was proposed based on the characteristics of the DSM matrix,by which the optimization process was improved and task processing speed was increased.Finally,a web-based distributed collaborative modeling platform was developed to achieve collaborative construction of the digital twin entity model of the screw experiment platform.The experimental result shows that the collaborative modeling efficiency of the platform is significantly improved,providing an engineering case for the rapid construction of the digital twin model of complex equipment,and providing technical support for the intelligent operation and maintenance of complex equipment.