One of the most efficient algorithms for solving hydrothermal operation planning problems, which are large-scale multi-stage stochastic models, is the so-called stochastic dual dynamic programming (SDDP) algorithm. Operation planning of power systems aims to assess the value of the scarce resources (e.g. water) to feed short-term dispatch models used in the actual implementation of the decisions. When the planning model significantly deviates from the reality of the implemented operation, decision policies are said to be time-inconsistent. Recent literature has explored different sources of inconsistency such as time-inconsistent dynamic risk measures, inaccurate representation of the information process and simplifications in the network planning model. This work addresses the time-inconsistency due to simplifications in the network representation in the planning model extending the existing literature. The objective of this work is to propose a framework, comprised of a methodology and an open-source computational package, for testing the operative and economic impact of modeling simplifications over the network power-flow in hydrothermal power systems. Among the myriad of formulations available in the package, we focused on assessing the cost and operative performance of the following model approximations: the transportation network-flow model (NFA), currently in use by the Brazilian system operator; the second-order cone relaxation (SOC); the semidefinite programming relaxation (SDP); the DC power-flow approximation (DC); and the DC with line-loss power-flow approximation (DCLL). All the previously mentioned formulations are tested as approximations for the network model in the planning stage, where the cost-to-go function is built. Then, we evaluate each approximation by simulating the system s operation using an implementation model, which minimizes the immediate cost under AC power-flow constraints and the respective cost-to-go function. The comparison is made for two systems, one composed of a cycle and the other approximately radial.