Traditional air conditioning load calculation methods suffer from overestimated results in steady-state computation, excessive computational burden in unsteady-state calculation, and low efficiency in numerical analysis. Such defects make it difficult to quickly and accurately satisfy the control requirements of multi-zone air conditioning loads under complex dynamic boundaries. To solve these problems, this paper proposes applying the Modelica-based RC thermal network heat transfer model to air conditioning load calculation, and carries out dynamic load simulation and verification with the multi-zone structure in the control cabin of an underwater vehicle as the research object. The results show that the proposed method can realize multi-zone dynamic load solution under dynamic boundaries. The dynamic response accuracy is verified against measured loads under surface and underwater navigation conditions. The average relative error rates of the total load are only 3.17% and 2.90%, respectively, and the maximum error rate is less than 10%. The model has high accuracy and provides an effective approach for multi-zone dynamic load calculation.