Hierarchical Strategies in Tool Wear Monitoring

Abstract

The paper presents a comparison of efficiency of tool wear monitoring strategies based on one signal feature, on a single neural network with several input signals, and on a hierarchical algorithm and a large number of signal features. In the first stage of the hierarchical algorithms, the tool wear was estimated separately for each signal feature. This stage was carried out using either simple neural networks or polynomial approximation. In the second stage, the results obtained in the first one, were integrated into the final tool wear evaluation. The integration was carried out by the use of either a neural network or averaging. The paper shows a considerable advantage of the hierarchical models over conventional industrial solutions (single signal feature) and typical laboratory solutions (single, large neural network).

Advanced monitoring of machining operations

“…Therefore, the calculation of a sufficient number of SFs related to the tool and/or process conditions [67][68][69][70] is a key issue in machining monitoring systems. This is obtained through signal processing methods that comprise the stages shown in Fig.…”

“…_ 1 1 ) T D $ F I G ] where Jemielniak et al [70,81] evaluated the correlation between SFs and used-up parts of tool life (ratio of cutting time to tool life DT = t/ T). Each SF was correlated with DT, using a 2nd order polynomial approximation, and the RMS error of this correlation was a measure of the SF applicability to tool wear monitoring.…”

Tool condition monitoring based on numerous signal features

“…Tool wear estimation is based on a hierarchical algorithm [4]. In the first step of the algorithm, the used-up portion of the tool life (ΔT) is evaluated using every selected signal feature separately [9].…”

Possibility of applying the RBF neural network to the on-line tool condition monitoring systems

“…Dzięki temu nie wprowadza się dodatkowych opóźnień (wynikających z czasu wykonywania obliczeń) w produkcji związanych ze stosowaniem układu nadzoru procesu skrawania. Dlatego w ramach pracy postanowiono przetestować właśnie tę sieć.Pierwsze porównanie algorytmu hierarchicznego i sieci neuronowej przedstawiono w [5]. Tam też znajduje się dokładny opis algorytmu hierarchicznego, do którego porównywano sieć RBF w niniejszej pracy.…”

“…Pierwsze porównanie algorytmu hierarchicznego i sieci neuronowej przedstawiono w [5]. Tam też znajduje się dokładny opis algorytmu hierarchicznego, do którego porównywano sieć RBF w niniejszej pracy.…”