Search

Improving production efficiency while ensuring product surface quality is a constant focus of manufacturers. Cutting parameter optimization is an important technique for ensuring high-efficiency and high-quality production. In this paper, a novel method for cutting parameter optimization that integrates multi-agent reinforcement learning with a dual-drive virtual machining environment is proposed. First, a feature extraction, fusion and generation model for cutting simulation and experimental data is proposed to solve the problem of incomplete data acquisition in the production process. Second, a Markov decision model for optimizing cutting parameters is defined, and a virtual machining environment driven by both simulation and experimental data is constructed. Third, a novel multi-agent reinforcement learning method called Q-MIX-MATD3, in which a twin delay deep deterministic policy gradient, value function decomposition and a teacher model are combined, is proposed to explore the cutting parameter optimization policy by interacting with the virtual machining environment. Finally, the proposed method is verified on a commutator production line. Moreover, the results show that the accuracy of the virtual machining environment driven by both simulation and experiment increases by more than 5 %, response efficiency increases by 31 %, and Q-MIX-MATD3-based cutting parameter optimization method reduces time cost by 98 % and achieves the optimization effect of the classical optimization method. [ABSTRACT FROM AUTHOR]

• Uncertainty quantification based on Bayesian inference for SLDs was proposed. • FRFs of the tool tip during the cutting operation was considered as the prior distributions. • An advanced Morris method was used to select the most influential parameters. • An improved Ensemble MCMC algorithm was used to infer posterior distributions. To predict the stability boundary of cutting operation, it is necessary to establish process force model and dynamic mechanical behavior model. Uncertainty of model parameters may lead to significant differences between predicted and measured stability boundary. In this paper, a novel method for determining boundary of stability in milling, which performs the stability analysis on an uncertain dynamic milling model, is presented. Firstly, the input parameters of milling stability prediction model are considered as random variables. Subsequently, sensitivity analysis is used to determine the most influential parameters, the posterior distributions of all influential parameters are estimated using a Bayesian inference framework, which is built based on an improved Ensemble Markov Chain Monte Carlo (IGWMCMC) algorithm. Compared with the traditional method, IGWMCMC is not affected by affine transformations of space, which can achieve much better convergence on badly scaled problems and the computational costs are cheaper. Finally, the posterior distributions of uncertainties calculated using Bayesian inference are propagated through the computational model. The experimental results verify the reliability of the calculated boundaries of stability lobe diagrams. [ABSTRACT FROM AUTHOR]

• Two novel chatter indicators are employed as chatter detection indicators. • Angular synchronous averaging method is used to preprocess milling signals. • Gradient Tree Boostingis applied to diagnosis of the chatter severity levels. • A relative threshold is adopted under different cutting conditions. In the milling process, chatter, which results in poor surface quality, dimensional errors, reduced cutter and machine life, is one of the main limitations on performance. In this paper, a novel method of online chatter detection for milling processes is developed. In this method, first, the spindle revolution period component is obtained via angular synchronous averaging (ASA). Then, the residual part related to chatter information is calculated by subtracting the periodic component. Secondly, the multiscale permutation entropy (MPE) and multiscale power spectral entropy (MPSE) of the residual part are calculated, and the Laplacian score (LS) for feature selection is applied to select the optimal sensitive scale features with generalization. Online chatter detection that is based on selected sensitive scale features by splitting signal up into (overlapping) frames in milling process. Finally, a trained gradient tree boosting (GTB) can be used to intelligent diagnosis of the chatter severity level. The analysis results show that the proposed method can effectively detect the onset of chatter under stable cutting conditions and variable cutting conditions, which is simpler and more robust than the existing methods. [ABSTRACT FROM AUTHOR]

• A novel graph-based semi-supervised method is proposed to predict position- and speed-dependent tool tip dynamics with limited labeled data. • To avoid the over-fitting occurrence, the conventional GCN is extended by stacking a transposed GCN. • The accuracy of proposed method for predicting the milling stability is validated through chatter tests. The tool tip frequency response function (FRF), as one of the essential inputs to calculate the stability lobe diagram (SLD), varies with the position and speed of changes in the moving components within the machine tool work volume. How to predict the position- and speed-dependent tool tip dynamics accurately has become one of the most challenging tasks in obtaining SLDs while avoiding chatter. However, traditional finite element analyses or kinematic modelling-based methods are costly and time consuming. Data-driven machine learning methods require large amounts of labeled data to train a model, but labeled industrial data are limited and extremely valuable in real manufacturing industries. To minimize experimentation, an improved semi-supervised graph convolutional network (GCN) is proposed to predict position- and speed- dependent tool tip dynamics with limited labeled data. First, the inverse stability solution is applied to identify dominant modal parameters under cutting conditions to obtain labeled samples. Subsequently, both the limited labeled samples and large amounts of unlabeled samples are converted into graph data to train a GCN regression model integrated with a multilayer perceptron. To avoid overfitting under limited data conditions, the conventional GCN is extended by stacking a transposed GCN, which is utilized to reconstruct the initial node features. The results demonstrate an improved prediction performance by adding the unsupervised reconstruction error into the loss function. Compared with other machine learning methods, our proposed method has superior performance in predicting tool tip dynamics with only 20% of the labeled samples. Finally, the SLDs were reconstructed with predicted FRFs, and the accuracy of the SLDs was validated through chatter tests. [ABSTRACT FROM AUTHOR]