Tool remaining useful life prediction using bidirectional recurrent neural networks (BRNN)

Abstract

Nowadays, new challenges around increasing production quality and productivity, and decreasing energy consumption, are growing in the manufacturing industry. In order to tackle these challenges, it is of vital importance to monitor the health of critical components. In the machine tool sector, one of the main aspects is to monitor the wear of the cutting tools, as it affects directly to the fulfillment of tolerances, production of scrap, energy consumption, etc. Besides, the prediction of the remaining useful life (RUL) of the cutting tools, which is related to their wear level, is gaining more importance in the field of predictive maintenance, being that prediction is a crucial point for an improvement of the quality of the cutting process. Unlike monitoring the current health of the cutting tools in real time, as tool wear diagnosis does, RUL prediction allows to know when the tool will end its useful life. This is a key factor since it allows optimizing the planning of maintenance strategies. Moreover, a substantial number of signals can be captured from machine tools, but not all of them perform as optimum predictors for tool RUL. Thus, this paper focuses on RUL and has two main objectives. First, to evaluate the optimum signals for RUL prediction, a substantial number of them were captured in a turning process and investigated by using recursive feature elimination (RFE). Second, the use of bidirectional recurrent neural networks (BRNN) as regressive models to predict the RUL of cutting tools in machining operations using the investigated optimum signals is investigated. The results are compared to traditional machine learning (ML) models and convolutional neural networks (CNN). The results show that among all the signals captured, the root mean squared (RMS) parameter of the forward force ($${F}_{y}$$ F y ) is the optimum for RUL prediction. As well, the bidirectional long-short term memory (BiLSTM) and bidirectional gated recurrent units (BiGRU), which are two types of BRNN, along with the RMS of $${F}_{y}$$ F y signal, achieved the lowest root mean squared error (RMSE) for tool RUL, being also computationally the most demanding ones.

Unlocking Dual Utility: 1D-CNN for Milling Tool Health Assessment and Experimental Optimization

“…This underutilization highlights a missed opportunity to harness CNN and LSTM models to fully capture temporal dependencies in time series data. Hence, [17] implemented a BiLSTM model that directly processed time series data from a CNC machine, achieving a root mean squared error of less than 10%.…”

“…We benchmark our model against state-of-the-art models detailed in the literature. Notably, given the frequent application of BiLSTM in processing time series data, we have implemented the algorithm as suggested by [17]. Additionally, we compare our model with the SUBLSTM model, as detailed in [10], which employs bidirectional, uni-directional, and 1D-CNN layers.…”

“…Accuracy Precision Recall F1 score BiLSTM [17] 0.85 ± 0.07 0.72 ± 0.14 0.75 ± 0.07 0.70 ± 0.05 SUBLSTM [10] 0.81 ± 0.08 0.85± 0.07 0.73 ± 0.09 0.75 ± 0.07 Attention 1 [19] 0.86 ± 0.05 0.73 ± 0.10 0.70 ± 0.05 0.70 ± 0.08 Attention 2 [21] 0.86 ± 0.07 0.75 ± 0.12 0.74 ± 0.10 0.75 ± 0.10 Proposed 0.90 ± 0.04 0.81 ± 0.12 0.82 ± 0.05 0.82 ± 0.08 Upon detailed examination of how the models perform on individual tools, we observe that the more complex models exhibit lower per-class precision and recall. While these models effectively identify class 1 (healthy state), they struggle to differentiate between transition and failure modes.…”

Utilising Machine Learning for Tool Condition Monitoring of Diamond-Coated Burrs with Acoustic Emission

“…Physicsbased models require in-depth knowledge of the system to generate models based on the fundamental failure mechanisms. Data-driven models require significant data in order for the models to be trained, but requires little expertise about the process [7]. Due to the random nature of grinding compared to traditional machining processes, statistical or data-driven models are more commonly employed and successful [8].…”

Generative and self‐supervised ensemble modeling for multivariate tool wear monitoring

“…The ML‐based deep modeling has extended this performance further by enabling on‐line wear determination without need for hand‐crafted data features engineering 12 . Different deep modeling based algorithms have been adopted for the wear estimation task, such as convolutional neural networks (CNNs), 13,14 recurrent cells 15,16 and even varying combinations of the two, 17,18 among other algorithms 19,20 . The accuracy of these models' predictions is generally anchored on the architecture used and optimized development.…”