Abstract: The planar and time-varying models (geomagnetic index K_p≤5) of the daily variation of the geo-electric field in two regional geo-electric networks in East China and North China are built up based on the Taylor polynomial and surface spline function methods, respectively. Then the fitting results are compared and analyzed. The results show that although the geo-electric field variation is more complex, the models of the daily variation fitted by the two methods are identical to each other, and the two model curves are consistent with the measured sample curves with the variation of time. The two model curves could be used to describe the measured daily variation on the same lunar date in other years with the similar geomagnetic index K_p. The daily variation results based on the two models on the condition of a low geomagnetic index are better than that on a high geomagnetic index. The average model curves based on multi-day sample curves more clearly show the daily variation features of geo-electric field such as the twice fluctuation waveform in some day and its amplitude and the extreme value concentrating on before and after noon. It is important to establish the daily variation models of geo-electric field for the different lunar date in the different geomagnetic indices in each regional network. The models can be used to quantitatively evaluate the quality of observation data of regional geo-electric field networks, and the geo-electric field anomaly measured by the networks can be picked up based on these models, which can be applied to prediction of disasters like earthquakes and monitoring the changes in the electromagnetic environment.