The data fusion with synchro-phasor measurement unit (PMU) and supervisory control and data acquisition (SCADA) is the basis of improving the level of measurement and operational control of distribution network. In order to improve the accuracy of data fusion, a multi-source data fusion method based on multi-time scale is proposed to fuse PMU data and SCADA data. First of all, the high-precision PMU provides a large amount of accurate, high-frequency and multi-type data for the node state, and the topology network structure of open network is reconstructed by using the Depth-First-Search (DFS) algorithm to form an open network with PMU nodes. Secondly, the PMU measurement data is combined with the latest SCADA sampling data based on the sampling time, and the voltage and phase angle without PMU node are derived by power flow calculation. The state estimation based on least square method is used to obtain the state of the whole system. Finally, regularized residual is introduced to identify SCADA data for abnormal data. Numerical results show that the proposed model and method are effective to improve the accuracy of data fusion.