Application of Bidirectional Long Short-Term Memory Neural Networks to Determine Cutting Tool Wear for Machines with Computer Numerical Control in Operation Mode

During the operation of machines with computer numerical control (CNC), a fairly large number of parameters are controlled, including the position and parameters of the equipment used, the temperature of the machine components, readings from vibration and force sensors. However, there are a number of parameters that cannot be tracked during the manufacturing process using the machine. One of these parameters is the amount of wear on the cutting tool, which can be measured only during periods of idle time of the machine tool. Tool wear significantly affects the quality of the resulting surface. The operation of the tool with high wear leads to an increase in vibration, noise, additional load on other parts of the machine. To solve the problem of assessing the state of wear of a cutting tool, it makes sense to use the available operational information as an indicator of the amount of wear. The article proposes the implementation of such an approach by assessing the amount of wear of the cutting tool according to the data of vibration of the spindle and cutting forces. To forecast this dependence, it was proposed to use bidirectional networks of long short-term memory, since this type of neural networks is one of the most effective in the problem of processing large time series data. By checking the trained model on a test dataset, it was found that the proposed model makes it possible to determine tool wear with an accuracy of 97.5 %. The proposed approach and model for assessing the wear of the cutting tool can be used as part of control systems for CNC machines.

1. Ivashchenko A.P. Metody i sredstva kontrolya sostoyaniya rezhushchego instrumenta. Mezhdunarodnyj zhurnal prikladnyh i fundamental'nyh issledovanij. 2015;12(3):393–396. (In Russ.)

2. Masalimov K.A., Munasypov R.A., Fecak S.I., Kudoyarov R.G. Diagnostika sostoyaniya rezhushchego instrumenta metallorezhushchih stankov s ispol'zovaniem dvunapravlennyh rekurrentnyh nejronnyh setej s dolgoj kratkosrochnoj pamyat'yu. Nauchno-tekhnicheskij zhurnal STIN. 2020;12:12–16. (In Russ.)

3. Ivanov V.A., Feshchenko A.A. Osobennosti podhodov k tekhnicheskomu obsluzhivaniyu i remontu oborudovaniya v nepreryvnom proizvodstve. Vestnik Permskogo nacional'nogo issledovatel'skogo politekhnicheskogo universiteta. Mashinostroenie, materialovedenie. 2018;20(3):82–89. (In Russ.)

4. Fomina A.V., Muhin K.Yu. Industriya 4.0. Osnovnye ponyatiya, preimushchestva i problemy. Ekonomicheskij vektor. 2018;3(14):33–38. (In Russ.)

5. Kokorev D.S., Yurin A.A. Cifrovye dvojniki: ponyatie, tipy i preimushchestva dlya biznesa. Colloquium-journal. 2019;10(34):31–35. DOI: https://www.doi.org/10.24411/2520-6990-2019-10264. (In Russ.)

6. Omelchak A., Fecak S.I., Idrisova U.V. Dynamic Processes in a Мachine-Tool at High-Speed. DAAAM International Scientific Book. 2016:175–182.

7. Li Guofa et al. Tool wear state recognition based on gradient boosting decision tree and hybrid classification RBM. The International Journal of Advanced Manufacturing Technology. 2020;110:511–533. DOI: https://doi.org/10.1007/s00170-020-05890-x

8. Bergs Thomas et al. Digital image processing with deep learning for automated cutting tool wear detection. Procedia Manufacturing. 2020;48:947–958. DOI: https://doi.org/10.1016/j.promfg.2020.05.134.

9. Oo HtunHtun et al. Tool wear monitoring system in belt grinding based on image-processing techniques. The International Journal of Advanced Manufacturing Technology. 2020;111:2215-2229. DOI: https://doi.org/10.1007/s00170-020-06254-1

10. Lin Cheng-Jian et al. Using an Interval Type-2 Fuzzy Neural Network and Tool Chips for Flank Wear Prediction. IEEE Access. 2020;8:122626–122640. DOI: https://doi.org/10.1109/ACCESS.2020.3006849

11. Gajewski Jakub et al. Classification of wear level of mining tools with the use of fuzzy neural network. Tunnelling and Underground Space Technology. 2013;35:30–36. DOI: https://doi.org/10.1016/J.TUST.2012.12.002

12. Siami-Namini Sima et al. The Performance of LSTM and BiLSTM in Forecasting Time Series. 2019 IEEE International Conference on Big Data (Big Data). 2019:3285–3292. DOI: https://doi.org/10.1109/BigData47090.2019.9005997