With the increasing scale of distributed power generation and non-linear load, power grid harmonic pollution is becoming more and more serious. However, for economic reasons, it is unrealistic to monitor harmonic current of all feeders of a substation. Given this, this paper proposes a harmonic spectrum estimation method of feeder lines based on multi-source data. This method uses multi-source data obtained from the power quality monitoring, dispatching and marketing business application systems, and can estimate the harmonic current content of each feeder by only monitoring data of bus harmonic voltage and incoming line harmonic current. First, the objective function of spectrum estimation is constructed according to the phasor operation relationship of feeder harmonic current and power relationship. Then, the improved particle swarm optimization algorithm is used to analyze the objective function based on multi-source data, and after that the harmonic current content of each feeder and the phase angle difference between the harmonic currents of the feeders are obtained. Finally, the feasibility of the proposed method is verified by simulation and measurement examples. The examples show that this method only requires the installation of a power quality monitoring terminal at the busbar of the substation. By combining basic electrical data from existing business systems, it can estimate the harmonic spectrum of the feeder, helping to save the configuration cost of substation monitoring devices.