Power demand response is an important means to maintain the balance of power supply and demand in the peak period, and peak clipping is the main way of smart grid to achieve power demand response. In order to enable enterprises using hybrid ?ow-shop production to better participate in peak clipping demand response(PCDR) and optimize production scheduling, this paper introduces PCDR in the hybrid ?ow-shop scheduling problem, establishes a new problem model, and proposes an improved Sparrow Search Algorithm(ISSA) for model solving. Aiming at the problem that the standard SSA is prone to local optimization, the ISSA improves the local search ability of SSA by adding KMeans clustering replacement strategy. The experimental results show that the proposed model and algorithm can provide a better peak clipping production scheduling scheme, meeting the needs of enterprises to implement PCDR scheduling.