In order to solve the problem that it is difficult to accurately locate the collector line after short-circuit of hybrid connection in wind farm, a fault location method based on improved deep denoising auto-encoder (DDAE) network is presented in this paper. By analyzing the zero-sequence current on collector line faults, it is known that the transient current value, steady-state current amplitude, steady-state current phase and fault distance are strongly non-linear, and the precise location of collector line is achieved by deep learning mining this complex relationship. A distance regression output port is added to the deep denoising auto-encoder network, and joint training is used to improve the accuracy, noise resistance and robustness of the positioning network. Firstly, the hub model is built with PSCAD/EMTDC, and fault zero-sequence current sequence and corresponding distance in a given time window are used as fault samples to simulate different failure cases to generate sample sets. Then, an improved deep auto-encoder network is trained on the training set to obtain an optimal network for precisely measuring the fault distance. With the help of the zero-sequence current amplitude relationship of each measurement point, the fault area can be determined first, and the precise location of the fault can be determined by feeding the fault samples into the trained network. This method proposed in this paper has a good adaptability to multi-branch and hybrid short lines of collector lines. Location performance is significantly better than traditional machine learning algorithms, as well as less affected by transition resistance, sampling rate, noise, fault phase angle.