To address the limited accuracy of existing linearized power flow models in large-scale power system analysis and computation

this paper proposes an approximate linearized power flow model based on a logarithmic-power transformation of voltage magnitude. First

an adjustable power parameter is introduced into the logarithmic transformation of voltage magnitude to construct a generalized transformation form

effectively enhancing the flexibility of the linearized power flow model in fitting the nonlinear characteristics of power systems. Second

a power parameter optimization method is proposed

which determines the optimal power parameter based on the boundary information of voltage magnitude from historical operational data of the power system

thereby optimizing the transformation space of state variables and improving the computational accuracy of the model under different operating conditions. Finally

simulation results based on standard test systems of varying scales demonstrate that the proposed optimal logarithmic-power transformation model significantly reduces the approximation errors in multiple electrical quantities

including voltage magnitude

phase angle

and branch power flow

compared to existing mainstream linearized power flow models

validating its effectiveness and superiority in enhancing computational accuracy.