tailieunhanh - Machining and Monitoring Strategies Part 12

Tham khảo tài liệu 'machining and monitoring strategies part 12', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Machining and Monitoring Strategies 541 algorithm - in its learningphase Fig. 260a . Fundamentally this network operates in the following manner the input layer neurons transmit signals via the hidden layers to the ouput layer. At this juncture in its operation the desired and actual outputs are compared to evaluate the system error - usually Euclidean error. Hence this error value is employed to adjust the strengths of the connectivity amongst the network s neurons. The algorithm utilised to perform this undertaking is normally known as error-back-propagation which is available in many forms. The hidden layer s nodes have their capabilities developed during training this is achieved in such a manner that the extracted features are better suited for the classification task . Tool - Condition Monitoring System The ANN described above was utilised on a two-axis slant-bed turning centre Fig. 261b which was equipped with sensors and associated equipment allowing on-line data capture during a comprehensive run of machining trials. Three sensors were utilised in this work for monitoring the cutting process fitted onto a specially-manufactured platform situated on the tool turret Figs. 261a and c . The three sensors were a Kistler force dynamometer model 9275B - which sensed measured the cutting forces in three perpendicular axes . X Y and Z AE sensor - Physical Acoustics type WDI vibrational sensor - Vibrometer type CE501 M101 miniature accelerometer. The force and acceleration signals were amplified and then sampled at 50 kHz while the AE sampling was undertaken separately by a digital storage adaptor at a sampling rate of 1 MHz. All of this information was then stored on a suitably- fast PC. The schematic layout of the monitoring hardware is illustrated in Fig. 260b. Prior to utilising by the neural network a preprocessing procedure was operated to reduce what is termed its dimensionality of the signals. This action was achieved by computing the power spectral .