Data Traffic Scheduling Algorithm for Wireless Smart Grid Communication Network Based on Deep Learning

One of the key technologies of smart grid is to provide an efficient, reliable and safe two-way communication system for power data acquisition. Using advanced power metering equipment (smart grid) with communication ability to form wireless mesh network to collect data, there is a challenge of application layer data traffic to network communication performance. When a large amount of data traffic bursts, smart grid close to local LAN will face greater communication pressure, which may lead to serious data congestion. Aiming at the shortcomings of current data scheduling algorithms in smart grid, and not fully considering the characteristics and types of data generated by the actual operation of smart grid, this paper proposes a new data scheduling algorithm for smart grid communication network, which considers the difference characteristics of bandwidth and time delay in multipath, and determines the main factors that affect the network performance through selection. Secondly, a new traffic scheduling algorithm is proposed. Finally, the network simulation is carried out. Compared with other algorithms, the algorithm proposed in this paper can greatly alleviate the system congestion at the time of data burst. No matter the data type sensitive to time delay or the data type less sensitive, the data scheduling algorithm proposed in this paper can reduce the time delay as much as possible.