<p>Superalloy GH4169G is widely used in the aerospace field. Residual stress profile generated during the processing of superalloy GH4169G has a significant impact on the quality of the workpiece. It is necessary to predict the residual stress profile generated during the processing of superalloy GH4169G. However, at present, most of the research on the prediction of residual stress profile adopts static process parameters as input, and rarely considers dynamic factors. This can’t meet the requirements of real-time prediction. Therefore, this paper proposes a prediction method of residual stress profile for milling superalloy GH4169G based on multi-source signals. Firstly, obtain multi-source signal data and residual stress data through experiments. Then, the time-domain and frequency-domain features of multi-source signals are extracted, and the Principal Component Analysis (PCA) is established to automatically determine the key features. Next, the residual stress profile is fitted by the Exponential Decay Cosine (EDC) function with Firefly algorithm (FA), and the fitting average value of <i>R</i>² is 95.4%. Finally, the Gaussian Process Regression (GPR) model is adopted to establish the mapping relationship between the key features and the EDC function coefficients. By predicting the EDC function coefficients, the residual stress profile is obtained. The running time of the method is about 0.2s and the predicting average value of <i>R</i>² for residual stress profile is 94.5%. The proposed method provides a basis for the online monitoring of residual stress profile and meets the requirements of intelligent manufacturing.</p>