The traditional transmission pricing method is based on the annual peak load model, which is difficult to adapt to the new development situation of the large-scale access of renewable energy and the deepening of the power market reform. Therefore, a new method based on the correlation load model is proposed. Using multi-variable kernel density estimation and multi-dimensional normal distribution sampling technology to build a data-driven correlation load model, sample the node load vector sample space, reduce the lack of random distribution artificial assumptions, and better simulate the system node load correlation according to the law of random change. On this basis, multiple typical operation modes are formed during the peak load period of the system, the source flow analysis method is used to evaluate the grid usage share of the single operation mode, and the transmission cost is comprehensively allocated with probability weighting to establish a fair transmission price. The IEEE RTS-79 node system calculation example verifies that this method can more truly and comprehensively reflect the use share of transmission equipment, which is conducive to achieving fair pricing and adapting to the new pattern of green and market-oriented power system development.