Research On Chip Removal And Cutting Tool Condition Monitoring Technology In BTA Deep Hole Drilling

Deep hole drilling has been extensively applied in manufacturing industry. The condition of tool wear and chip removal is a key factor that determines whether deep-hole drilling is successfully processed or not. It is essential to research on tool wear and chip removal condition monitoring.Taking BTA deep-hole drilling system as objective, the characteristic of deep hole drilling and ordinary fault type were analyzed.The monitoring system which is used to monitor chip block and tool condition by cutting power and oil pressure signal in deep hole drilling was eatablished. The feature extraction of power and oil pressure signal as well as condition recognition was researched through theoretical analysis and experiments.Dynamic characteristic of oil system was analyzed in deep-hole drilling. Simplized model of the oil system was established, and simulation of flow field of deep-hole drilling oil system was carried out with the FLUENT software, the law of oil pressure changing with hole depth in deep hole drilling was obtained, which provided foundation for monitoring whether chips are blocked or not in drilling process.Aim at the problems in feature extraction and condition recognition caused by nonstationary feature of sensor signal and intensive impulse interference, wavelet packet based anti-impulse interference moving average de-noising method was proposed, and was applied to de-noising of power and oil pressure signals. The results of simulation and experiment showed that SNR of the monitored signal was greatly improved with the method used.The wavelet transform based nonstationary signal processing approaches were analyzed, the relation between oil pressure signal and chip removal was researched as well as power signal and tool condition. The results showed that oil pressure signal was closely related with chip removal condition as well as power signal frequency band energy characteristics and tool wear. Additionally, aiming at oil pressure signal characteristic, chip removal condition wasmonitored through calculating pressure changing rate.The application of HOS (Higher order statistics) to tool condition characteristic extracting was discussed, high cumulant and averaging energry of sensitive frequency band of bispectrum were used to describe the tool condition characteristic information. The experiment results showed that HOS could be used to distinguish different tool wear styles.The RBF neural network model of tool wear and chip removal condition recognition in deep hole drilling was established, and mapping from power singal characteristic to tool wear condition was realized as well as oil pressure signal characterisitic to chip removal condition.The results showed that the model can effectively recognize the two typical fault conditions in deep hole drilling.