In-network caching is one of the significant advantages of information-centric networking (ICN) that addresses many challenges to achieve high network efficiency. The previously proposed algorithms cannot effectively utilize the limited cache resources. Communication in ICN is driven by receivers, the current schemes cannot efficiently measure the influences of request for a router node. To solve these problems, we propose a novel cache placement model, FDCP (Fluid-dynamics cache placement), which uses fluid dynamics theory to reveal the time-evolving formulation process of request influences for a router. In this work, each network node corresponds to a container; containers are connected via unidirectional pipes, corresponding to the influence relations. We use the request service ratio approach to accurately calculate the consumer request and caching location, while using the FDCP rating system for cache placement decisions. The proposed solution has clear physical meaning and explains the cache placement decision-making process using fluid dynamics theory. The simulation results show that the proposed scheme can significantly improve the cache hit ratio, reduce the latency, especially improve cache hits ratio by up to 50% for limited cache resources.