In view of the inefficiency problem, of the cache replacement strategy for current content-centric networks, we propose a cache replacement algorithm based on potential energy cooling (PEC-Rep), which combines the “natural coding” phenomenon with the reference of the concept of “potential energy”. The PEC-Rep can evaluate content value according to the access number and time interval. And it deletes the content with the minimum value when it performs content replacement so that the node can maintain maximum value to satisfy the subsequent requests of users. Simulation results show that the PEC-Rep can effectively improve the domain-cache hit ratio, reduce the server load and improve the overall performance of the CCN.