A parts surface roughness and cutting tool wear prediction method based on multi-task learning, relating to the technical field of machining. Firstly, vibration signals in the machining process are collected; next, the parts surface roughness and a wear condition of a cutting tool are measured, and the measured results respectively correspond to vibration signals; secondly, sample expansion is performed, and features are extracted and normalized; then, a multi-task prediction model based on a deep belief network is constructed, the parts surface roughness and the cutting tool wear condition serve as model output, features are extracted as input, and a multi-task DBN network prediction model is established; and finally, test verification is performed, the vibration signals are inputted into the multi-task prediction model, and the surface roughness and the cutting tool wear condition are predicted. The method is mainly advantaged in that: online prediction of the parts surface roughness and the cutting tool wear is achieved by means of one-time modeling, hidden information contained in monitoring data is fully utilized, and the workload and model building costs are reduced.
展开▼