Automatic tool changer system (ATCS) and drawbar mechanism (DM) are two of key basic parts in machining centers for realizing automatic tool-changing cycle. In the condition monitoring, fault diagnosis and failure warning of the ATCS and DM, the dynamic force is an important characteristic signal. However, there is little research about the specific dynamic force measurement system in this regard. Thus, a novel dynamic force measurement system (DFMS) is developed and implemented. Based on the BT40 toolholder, a resistance strain gauge-based force senor is used to convert the dynamic force signal into electrical signal. The real-time dynamic force acquiring system is controlled via an 8-bit RISC microcontroller. Digital measurements are obtained from the 24-bit sigma-delta analog-to-digital converter with a programmable gain array, which then are transmitted to the upper computer software via a wireless transceiver for display and storage. Finally, a Teager energy operator based dual-threshold two sentences endpoint detection method is proposed to extract the maximum dynamic force and the duration time. Experimental results show that the DFMS is reliable and can be easily used to detect the dynamic force for the ATCS and DM.

Tool wear during machining has a great influence on the quality of machined surface and dimensional accuracy. Tool wear monitoring is extremely important to improve machining efficiency and workpiece quality. Multidomain features (time domain, frequency domain and time-frequency domain) can accurately characterise the degree of tool wear. However, manual feature fusion is time consuming and prevents the improvement of monitoring accuracy. A new tool wear prediction method based on multidomain feature fusion by attention-based depth-wise separable convolutional neural network is proposed to solve these problems. In this method, multidomain features of cutting force and vibration signals are extracted and recombined into feature tensors. The proposed hypercomplex position encoding and high dimensional self-attention mechanism are used to calculate the new representation of input feature tensor, which emphasizes the tool wear sensitive information and suppresses large area background noise. The designed depth-wise separable convolutional neural network is used to adaptively extract high-level features that can characterize tool wear from the new representation, and the tool wear is predicted automatically. The proposed method is verified on three sets of tool run-to-failure data sets of three-flute ball nose cemented carbide tool in machining centre. Experimental results show that the prediction accuracy of the proposed method is remarkably higher than other state-of-art methods. Therefore, the proposed tool wear prediction method is beneficial to improve the prediction accuracy and provide effective guidance for decision making in processing.

An automatic tool-changing system (ATCS) is one of the key sub-systems for realizing automatic tool changing in machining centers. Each step in a tool-changing cycle tends to result in impacts, and thus generates transients in the vibration signal. The impact features often reflect important operational information related to the ATCS dynamics, and a crucial problem for impact-feature extraction is how to effectively represent the transients. A novel method for extracting impact features from an ATCS is proposed, based on sparse representation theory. A parametric multiple-impulse dictionary is constructed by the unit impulse-response function of a damped multiple-degree-of-freedom system, whose modal order, amplitudes, natural frequencies, relative damping ratios and initial phases are directly identified from the vibration signal by an improved state-space method. This leads to high similarity between atoms and impact-induced transients. To improve the calculation speed, a split augmented Lagrangian shrinkage method is used to obtain optimal sparse coefficients. With the proposed method, both the moments of impact occurrence and the time intervals between transients can be effectively identified, and thus the impact features can be extracted. The effectiveness of the proposed method is validated by simulated signals as well as practical ATCS vibration signals. A comparison study shows that the proposed method is superior to empirical-mode decomposition, ensemble-empirical-mode decomposition and variational-mode decomposition when used for impact-feature extraction.

The NC machine tool are subjected to a variety of working load during the actual machining process. The influence of different kinds of working loads on the NC machine tool’s reliability level are different. Therefore, this paper proposes an importance evaluation method of NC machine tool working loads based on analytic hierarchy process (AHP) -fuzzy comprehensive evaluation to judge the influence of different kinds of working loads on the reliability level. Firstly, the hierarchical structure of the importance evaluation of NC machine tool working loads is established, and then the structure of discrimination matrix and the consistency test of the matrix are determined. Secondly, the fuzzy membership R of the fuzzy subset is established, and then the weight vector M of the evaluation factor and the evaluation result vector S of the fuzzy comprehensive evaluation are determined to realize the quantitative analysis of the fuzzy object. Finally, the importance of calculating the machining center load is ranked as cutting force, feed rate, torque, tool changing frequency, temperature and noise in the example.

No abstract