To achieve the effective access management and load balancing under the trend of intensive access

multi access points（APs） collaborative network architecture has become a feasible solution for electric power wireless local area networks (WLAN).However

there may be mutual interference among multi APs

and the contention and collision of trigger frames will affect the performance of uplink orthogonal random access (UORA) and cause additional bower consumption.Thus

this paper proposes an power consumption optimization method for uplink random access in electric power WLAN with the core of reducing the collision probability of multi APs trigger frames.Firstly

an energy consumption optimization model based on orthogonal frequency division multiple access(OFDMA)and competition window(CW) adjustment is established

which reveals the intrinsic correlation between OCW adjustment and overall energy consumption based on the probability of successful trigger frame transmission.Secondly

an OCW adaptive dynamic adjustment algorithm based on deep Q-network (DQN) is proposed to reduce overall energy consumption caused by collision retransmission.Simulation results show that the proposed method can reduce overall energy consumption while ensuring the performance of throughput and latency

achieving reliable operation and control of power wireless local area networks access.