Time-Sensitive Networking (TSN) is a set of protocol standards being developed by the TSN Task Group within the IEEE802.1 Working Group. These standards define time-sensitive mechanisms for Ethernet data transmission to ensure that Ethernet can provide a stable and consistent level of service for the transmission of critical data. In existing TSN static scheduling algorithms, either the reliability of traffic flows is not considered, or scheduling is performed using only backup path transmission schemes. This paper addresses the no-wait TSN scheduling problem by proposing an enhanced tabu search (ETS) algorithm based on multiple sets of redundant paths. The primary objective is to maximize the scheduling success rate, with minimizing the standard deviation of link load as a secondary objective. A mathematical model is established, and a method for calculating the reliability of traffic flows is proposed. Using a K-shortest Paths (KSP) algorithm, multiple sets of redundant paths are generated, and the global load is updated based on adaptive threshold values. Experimental results show that the proposed ETS effectively improves the reliability and success rate of traffic flows while reducing the variance of network load.