With the support of the Internet of Vehicles technology, UBI car insurance premium rate determination has certain guiding significance for realizing accurate pricing of freight vehicle car insurance rates and meeting the individual needs of users. The SSA algorithm is optimized, and the MSNSSA algorithm is proposed, which is used to build a model for determining the UBI rate of freight vehicles. The model first divides the original population into three sub-groups of leader, followers and salp chain evenly according to the fitness value from small to large, and perform different search tasks respectively. By adding a symbiosis strategy to the follower location update process, the development capability of the SSA algorithm can be effectively enhanced. It is proposed to add a non-uniform Gaussian mutation strategy to the chain with poor fitness value to enhance the diversity of the population. By establishing the MSNSSA-FCM algorithm, using the global optimization advantage of the MANSSA algorithm, the clustering center of the FCM algorithm is optimized, and then the sample weight and attribute weight are introduced to design the objective function of the FCM clustering algorithm to improve the clustering of the FCM algorithm performance. The empirical evidence shows that the UBI Freight Vehicle Car insurance rate rating model established based on the MSNSSA-FCM algorithm has reduced visual noise and reduced intra-cluster error variance. In the process of processing UBI Freight Vehicle Car insurance data, compared with FCM algorithm and SSA-FCM algorithm and SAGA-FCM algorithm have higher clustering accuracy, and provide decision-making basis for insurance companies to mine more accurate customer classification information, customer driving risk assessment and customer car insurance rate rating.