Aiming at the problems of limited coverage and vulnerability to obstacles in Wireless Powered Communication Network （WPCN）， the resource allocation for Intelligent Reflecting Surface （IRS） assisted WPCN relay systems under communication blocking conditions was investigated. Specifically， in the downlink， IRS assisted users in harvesting energy from the Hybrid Access Point （HAP）； in the uplink， it facilitated information transmission from users to the HAP. Considering both energy transmission and information transfer， Time Division Multiple Access （TDMA） was used to partition time slots for energy harvesting， data communication， and data relay transmission. Based on the constructed system model and transmission strategy， an energy efficiency optimization problem was formulated with the constraints on user quality of service and energy consumption of sending information by the users， and total energy efficiency of the system was maximized by jointly optimizing the transmit power， IRS phase shift matrix and time scheduling. Due to the non-convex nature of the proposed problem， the Dinkelbach method was first applied to transform the fractional objective function into a non-fractional form. Subsequently， variable substitution and Semi-Definite Programming （SDP）were employed to convert the non-convex problem into a convex formulation， which was then solved suboptimally using CVX. Simulation results show that the proposed scheme not only extends system coverage， but also significantly improves energy efficiency. Compared with the average time allocation scheme and optimized time scheme with hybrid relay node， the proposed scheme achieves average energy efficiency improvements of 11.0% and 26.9% respectively.