In recent years

with the high proportion of power electronic equipment access

the data scale on the edge side of the power distribution system has increased sharply. Adaptive compressed sensing has gradually become one of the key technologies to alleviate the bandwidth pressure on the cloud edge. The corresponding distribution master station needs to reasonably decompress it to meet the delay requirements of differentiated business. Therefore

an adaptive allocation method of computing power resources for distribution master station adapted to the dynamic compressed sensing on the edge side is proposed. Firstly

a distribution master station decompression system model driven by computing power resources is established. Then

considering the differentiated business delay requirements

a computing power resource allocation revenue function is designed

and based on this

the corresponding optimization problem is designed. Finally

the original optimization problem is decoupled into the computing power resource allocation layer and the time section setting layer. The KKT(Karush-Kuhn-Tucker) condition and the variable step size reallocation mechanism are adopted to solve the corresponding sub-problems of each layer respectively

thereby achieving efficient solution of the original optimization problem. Simulation analysis shows that the proposed method can fully explore the utility of computing power resource allocation through a unique variable step size reallocation mechanism

flexibly adapt to the dynamic compression process on the edge side

and ensure the state perception ability of the distribution master station on the edge side.