In modern society, with the continuous growth of electricity demand, enhancing the transmission capacity of power transmission lines is crucial. The dynamic capacity expansion technology of power transmission lines can utilize real-time meteorological parameter data to tap the hidden capacity and boost the transmission capacity. However, currently, the existing analytical calculation methods and finite-element analysis methods have difficulty in balancing calculation accuracy and real-time performance, which impacts the efficient application of the dynamic capacity expansion technology. This paper innovatively puts forward a method based on finite-element simulation data. By making use of the powerful non-linear mapping ability of the BP neural network, it deeply mines meteorological parameter data and constructs an accurate surrogate model between meteorological parameters and temperature for real-time prediction. Tests demonstrate that the prediction results of the surrogate model are highly consistent with those of the calculation methods recommended by national standards, with an error controlled within 5%. When the predicted current value is input into the finite-element model, the model temperature is close to the preset temperature limit value, verifying the accuracy and feasibility of the proposed scheme.