Ghazal Farhani
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Ghazal Farhani has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Mohamed H. Zaki
Department of Civil and Environmental Engineering, Western University, London, ON, Canada
6 shared publications - Taufiq Rahman
Department of Civil and Environmental Engineering National Research Council Canada Automotive and Surface Transportation Research Centre London Ontario Canada
6 shared publications - Robert J. Sica
Department of Physics and Astronomy, The University of Western Ontario, London, N6A 3K7, Canada
4 shared publications Alexander Haefele
Federal Office of Meteorology and Climatology, MeteoSwiss, Payerne, Switzerland
3 shared publications- Dominique Charlebois
Transport Ottawa Canada
3 shared publications - Sofiène Tahar
Concordia University
3 shared publications
Latest Publications
- Open Access
- 9 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 7, 1204-1222
String Stability of Vehicle Platoons Under Oscillatory Perturbations in Car-Following Models
Platooning is a key enabling technology for connected and automated vehicles, but its effectiveness depends critically on string stability, i.e., the ability to prevent disturbance amplification along the vehicle string. Although many car-following models have been proposed and analyzed, direct comparisons of their disturbance propagation behavior under identical conditions remain limited. This paper presents a comparative benchmarking study of four widely used car-following models-the Optimal Velocity Model (OVM), Intelligent Driver Model (IDM), General Motors Model (GMM), and Cooperative Adaptive Cruise Control (CACC)-subjected to unified lead-vehicle perturbations. Both discrete and continuous disturbances, including harmonic and non-harmonic velocity variations, are considered to reflect representative traffic scenarios. Model performance is evaluated using quantitative metrics such as spacing error amplification, velocity oscillation magnitude, and disturbance settling behavior across multiple platoon positions. For CACC, the impact of vehicle-to-vehicle communication and communication delay is explicitly examined, enabling comparison between sensing-only and communication-enabled platooning architectures. The results reveal trade-offs between robustness, responsiveness, and implementation complexity, and identify regimes in which simpler models achieve performance comparable to communication-based controllers, as well as scenarios where communication provides clear benefits. This work provides practical insights under identical evaluation conditions to support model and controller selection for vehicular platooning applications.
- 9 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 7, 1204-1222
Open AccessString Stability of Vehicle Platoons Under Oscillatory Perturbations in Car-Following Models
Platooning is a key enabling technology for connected and automated vehicles, but its effectiveness depends critically on string stability, i.e., the ability to prevent disturbance amplification along the vehicle string. Although many car-following models have been proposed and analyzed, direct comparisons of their disturbance propagation behavior under identical conditions remain limited. This paper presents a comparative benchmarking study of four widely used car-following models-the Optimal Velocity Model (OVM), Intelligent Driver Model (IDM), General Motors Model (GMM), and Cooperative Adaptive Cruise Control (CACC)-subjected to unified lead-vehicle perturbations. Both discrete and continuous disturbances, including harmonic and non-harmonic velocity variations, are considered to reflect representative traffic scenarios. Model performance is evaluated using quantitative metrics such as spacing error amplification, velocity oscillation magnitude, and disturbance settling behavior across multiple platoon positions. For CACC, the impact of vehicle-to-vehicle communication and communication delay is explicitly examined, enabling comparison between sensing-only and communication-enabled platooning architectures. The results reveal trade-offs between robustness, responsiveness, and implementation complexity, and identify regimes in which simpler models achieve performance comparable to communication-based controllers, as well as scenarios where communication provides clear benefits. This work provides practical insights under identical evaluation conditions to support model and controller selection for vehicular platooning applications.
- Open Access
- 13 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Intelligent Transportation Systems
- Vol. 7, 1091-1106
Comprehensive Assessment of LiDAR Evaluation Metrics: A Comparative Study Using Simulated and Real Data
For developing safe Autonomous Driving Systems (ADS), rigorous testing is required before they are deemed safe for road deployments. Since comprehensive conventional physical testing is impractical due to cost and safety concerns, Virtual Testing Environments (VTE) can be adopted as an alternative. Comparing VTE-generated sensor outputs against their real-world analogues can be a strong indication that the VTE accurately represents reality. Correspondingly, this work explores a comprehensive experimental approach to finding evaluation metrics suitable for comparing real-world and simulated LiDAR scans. The metrics were tested in terms of sensitivity and accuracy with different noise, density, distortion, sensor orientation, and channel settings. From comparing the metrics, we found that Density Aware Chamfer Distance (DCD) works best across all cases. In the second step of the research, a Virtual Testing Environment was generated using real LiDAR scan data. The data was collected in a controlled environment with only static objects using an instrumented vehicle equipped with LiDAR, IMU and cameras. Simulated LiDAR scans were generated from the VTEs using the same pose as real LiDAR scans. The simulated and LiDAR scans were compared in terms of model perception and geometric similarity. Actual and simulated LiDAR scans have a similar semantic segmentation output with a mIoU of 21% with corrected intensity and an average density aware chamfer distance (DCD) of 0.63. This indicates a slight difference in the geometric properties of simulated and real LiDAR scans and a significant difference between model outputs. During the comparison, density-aware chamfer distance was found to be the most correlated among the metrics with perception methods.
- 13 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Intelligent Transportation Systems
- Vol. 7, 1091-1106
Open AccessComprehensive Assessment of LiDAR Evaluation Metrics: A Comparative Study Using Simulated and Real Data
For developing safe Autonomous Driving Systems (ADS), rigorous testing is required before they are deemed safe for road deployments. Since comprehensive conventional physical testing is impractical due to cost and safety concerns, Virtual Testing Environments (VTE) can be adopted as an alternative. Comparing VTE-generated sensor outputs against their real-world analogues can be a strong indication that the VTE accurately represents reality. Correspondingly, this work explores a comprehensive experimental approach to finding evaluation metrics suitable for comparing real-world and simulated LiDAR scans. The metrics were tested in terms of sensitivity and accuracy with different noise, density, distortion, sensor orientation, and channel settings. From comparing the metrics, we found that Density Aware Chamfer Distance (DCD) works best across all cases. In the second step of the research, a Virtual Testing Environment was generated using real LiDAR scan data. The data was collected in a controlled environment with only static objects using an instrumented vehicle equipped with LiDAR, IMU and cameras. Simulated LiDAR scans were generated from the VTEs using the same pose as real LiDAR scans. The simulated and LiDAR scans were compared in terms of model perception and geometric similarity. Actual and simulated LiDAR scans have a similar semantic segmentation output with a mIoU of 21% with corrected intensity and an average density aware chamfer distance (DCD) of 0.63. This indicates a slight difference in the geometric properties of simulated and real LiDAR scans and a significant difference between model outputs. During the comparison, density-aware chamfer distance was found to be the most correlated among the metrics with perception methods.
- Open Access
Bayesian Uncertainty Quantification with Anchored Ensembles for Robust EV Power Consumption Prediction
Accurate EV power estimation underpins range prediction and energy management, yet practitioners need both point accuracy and trustworthy uncertainty. We propose an anchored-ensemble Long Short-Term Memory (LSTM) with a Student-t likelihood that jointly captures epistemic (model) and aleatoric (data) uncertainty. Anchoring imposes a Gaussian weight prior (MAP training), yielding posterior-like diversity without test-time sampling, while the t-head provides heavy-tailed robustness and closed-form prediction intervals. Using vehicle-kinematic time series (e.g., speed, motor RPM), our model attains strong accuracy: RMSE 3.36 +/- 1.10, MAE 2.21 +/- 0.89, R-squared = 0.93 +/- 0.02, explained variance 0.93 +/- 0.02, and delivers well-calibrated uncertainty bands with near-nominal coverage. Against competitive baselines (Student-t MC dropout; quantile regression with/without anchoring), our method matches or improves log-scores while producing sharper intervals at the same coverage. Crucially for real-time deployment, inference is a single deterministic pass per ensemble member (or a weight-averaged collapse), eliminating Monte Carlo latency. The result is a compact, theoretically grounded estimator that couples accuracy, calibration, and systems efficiency, enabling reliable range estimation and decision-making for production EV energy management.
Open AccessBayesian Uncertainty Quantification with Anchored Ensembles for Robust EV Power Consumption Prediction
Accurate EV power estimation underpins range prediction and energy management, yet practitioners need both point accuracy and trustworthy uncertainty. We propose an anchored-ensemble Long Short-Term Memory (LSTM) with a Student-t likelihood that jointly captures epistemic (model) and aleatoric (data) uncertainty. Anchoring imposes a Gaussian weight prior (MAP training), yielding posterior-like diversity without test-time sampling, while the t-head provides heavy-tailed robustness and closed-form prediction intervals. Using vehicle-kinematic time series (e.g., speed, motor RPM), our model attains strong accuracy: RMSE 3.36 +/- 1.10, MAE 2.21 +/- 0.89, R-squared = 0.93 +/- 0.02, explained variance 0.93 +/- 0.02, and delivers well-calibrated uncertainty bands with near-nominal coverage. Against competitive baselines (Student-t MC dropout; quantile regression with/without anchoring), our method matches or improves log-scores while producing sharper intervals at the same coverage. Crucially for real-time deployment, inference is a single deterministic pass per ensemble member (or a weight-averaged collapse), eliminating Monte Carlo latency. The result is a compact, theoretically grounded estimator that couples accuracy, calibration, and systems efficiency, enabling reliable range estimation and decision-making for production EV energy management.
- Open Access
Comprehensive Assessment of LiDAR Evaluation Metrics: A Comparative Study Using Simulated and Real Data
For developing safe Autonomous Driving Systems (ADS), rigorous testing is required before they are deemed safe for road deployments. Since comprehensive conventional physical testing is impractical due to cost and safety concerns, Virtual Testing Environments (VTE) can be adopted as an alternative. Comparing VTE-generated sensor outputs against their real-world analogues can be a strong indication that the VTE accurately represents reality. Correspondingly, this work explores a comprehensive experimental approach to finding evaluation metrics suitable for comparing real-world and simulated LiDAR scans. The metrics were tested in terms of sensitivity and accuracy with different noise, density, distortion, sensor orientation, and channel settings. From comparing the metrics, we found that Density Aware Chamfer Distance (DCD) works best across all cases. In the second step of the research, a Virtual Testing Environment was generated using real LiDAR scan data. The data was collected in a controlled environment with only static objects using an instrumented vehicle equipped with LiDAR, IMU and cameras. Simulated LiDAR scans were generated from the VTEs using the same pose as real LiDAR scans. The simulated and LiDAR scans were compared in terms of model perception and geometric similarity. Actual and simulated LiDAR scans have a similar semantic segmentation output with a mIoU of 21\% with corrected intensity and an average density aware chamfer distance (DCD) of 0.63. This indicates a slight difference in the geometric properties of simulated and real LiDAR scans and a significant difference between model outputs. During the comparison, density-aware chamfer distance was found to be the most correlated among the metrics with perception methods.
Open AccessComprehensive Assessment of LiDAR Evaluation Metrics: A Comparative Study Using Simulated and Real Data
For developing safe Autonomous Driving Systems (ADS), rigorous testing is required before they are deemed safe for road deployments. Since comprehensive conventional physical testing is impractical due to cost and safety concerns, Virtual Testing Environments (VTE) can be adopted as an alternative. Comparing VTE-generated sensor outputs against their real-world analogues can be a strong indication that the VTE accurately represents reality. Correspondingly, this work explores a comprehensive experimental approach to finding evaluation metrics suitable for comparing real-world and simulated LiDAR scans. The metrics were tested in terms of sensitivity and accuracy with different noise, density, distortion, sensor orientation, and channel settings. From comparing the metrics, we found that Density Aware Chamfer Distance (DCD) works best across all cases. In the second step of the research, a Virtual Testing Environment was generated using real LiDAR scan data. The data was collected in a controlled environment with only static objects using an instrumented vehicle equipped with LiDAR, IMU and cameras. Simulated LiDAR scans were generated from the VTEs using the same pose as real LiDAR scans. The simulated and LiDAR scans were compared in terms of model perception and geometric similarity. Actual and simulated LiDAR scans have a similar semantic segmentation output with a mIoU of 21\% with corrected intensity and an average density aware chamfer distance (DCD) of 0.63. This indicates a slight difference in the geometric properties of simulated and real LiDAR scans and a significant difference between model outputs. During the comparison, density-aware chamfer distance was found to be the most correlated among the metrics with perception methods.
- Open Access
3D Roadway Scene Object Detection with LIDARs in Snowfall Conditions
Because 3D structure of a roadway environment can be characterized directly by a Light Detection and Ranging (LiDAR) sensors, they can be used to obtain exceptional situational awareness for assitive and autonomous driving systems. Although LiDARs demonstrate good performance in clean and clear weather conditions, their performance significantly deteriorates in adverse weather conditions such as those involving atmospheric precipitation. This may render perception capabilities of autonomous systems that use LiDAR data in learning based models to perform object detection and ranging ineffective. While efforts have been made to enhance the accuracy of these models, the extent of signal degradation under various weather conditions remains largely not quantified. In this study, we focus on the performance of an automotive grade LiDAR in snowy conditions in order to develop a physics-based model that examines failure modes of a LiDAR sensor. Specifically, we investigated how the LiDAR signal attenuates with different snowfall rates and how snow particles near the source serve as small but efficient reflectors. Utilizing our model, we transform data from clear conditions to simulate snowy scenarios, enabling a comparison of our synthetic data with actual snowy conditions. Furthermore, we employ this synthetic data, representative of different snowfall rates, to explore the impact on a pre-trained object detection model, assessing its performance under varying levels of snowfall
Open Access3D Roadway Scene Object Detection with LIDARs in Snowfall Conditions
Because 3D structure of a roadway environment can be characterized directly by a Light Detection and Ranging (LiDAR) sensors, they can be used to obtain exceptional situational awareness for assitive and autonomous driving systems. Although LiDARs demonstrate good performance in clean and clear weather conditions, their performance significantly deteriorates in adverse weather conditions such as those involving atmospheric precipitation. This may render perception capabilities of autonomous systems that use LiDAR data in learning based models to perform object detection and ranging ineffective. While efforts have been made to enhance the accuracy of these models, the extent of signal degradation under various weather conditions remains largely not quantified. In this study, we focus on the performance of an automotive grade LiDAR in snowy conditions in order to develop a physics-based model that examines failure modes of a LiDAR sensor. Specifically, we investigated how the LiDAR signal attenuates with different snowfall rates and how snow particles near the source serve as small but efficient reflectors. Utilizing our model, we transform data from clear conditions to simulate snowy scenarios, enabling a comparison of our synthetic data with actual snowy conditions. Furthermore, we employ this synthetic data, representative of different snowfall rates, to explore the impact on a pre-trained object detection model, assessing its performance under varying levels of snowfall
- Open Access
A Comparative Study of Oscillatory Perturbations in Car-Following Models
As connected and autonomous vehicles become more widespread, platooning has emerged as a key strategy to improve road capacity, reduce fuel consumption, and enhance traffic flow. However, the benefits of platoons strongly depend on their ability to maintain stability. Instability can lead to unsafe spacing and increased energy usage. In this work, we study platoon instability and analyze the root cause of its occurrence, as well as its impacts on the following vehicle. To achieve this, we propose a comparative study between different car-following models such as the Intelligent Driver Model (IDM), the Optimal Velocity Model (OVM), the General Motors Model (GMM), and the Cooperative Adaptive Cruise Control (CACC). In our approach, we introduce a disruption in the model by varying the velocity of the leading vehicle to visualize the behavior of the following vehicles. To evaluate the dynamic response of each model, we introduce controlled perturbations in the velocity of the leading vehicle, specifically, sinusoidal oscillations and discrete velocity changes. The resulting vehicle trajectories and variations in inter-vehicle spacing are analyzed to assess the robustness of each model to disturbance propagation. The findings offer insight into model sensitivity, stability characteristics, and implications for designing resilient platooning control strategies.
Open AccessA Comparative Study of Oscillatory Perturbations in Car-Following Models
As connected and autonomous vehicles become more widespread, platooning has emerged as a key strategy to improve road capacity, reduce fuel consumption, and enhance traffic flow. However, the benefits of platoons strongly depend on their ability to maintain stability. Instability can lead to unsafe spacing and increased energy usage. In this work, we study platoon instability and analyze the root cause of its occurrence, as well as its impacts on the following vehicle. To achieve this, we propose a comparative study between different car-following models such as the Intelligent Driver Model (IDM), the Optimal Velocity Model (OVM), the General Motors Model (GMM), and the Cooperative Adaptive Cruise Control (CACC). In our approach, we introduce a disruption in the model by varying the velocity of the leading vehicle to visualize the behavior of the following vehicles. To evaluate the dynamic response of each model, we introduce controlled perturbations in the velocity of the leading vehicle, specifically, sinusoidal oscillations and discrete velocity changes. The resulting vehicle trajectories and variations in inter-vehicle spacing are analyzed to assess the robustness of each model to disturbance propagation. The findings offer insight into model sensitivity, stability characteristics, and implications for designing resilient platooning control strategies.
- Open Access
Weather-Dependent Variations in Driver Gaze Behavior: A Case Study in Rainy Conditions
Rainy weather significantly increases the risk of road accidents due to reduced visibility and vehicle traction. Understanding how experienced drivers adapt their visual perception through gaze behavior under such conditions is critical for designing robust driver monitoring systems (DMS) and for informing advanced driver assistance systems (ADAS). This case study investigates the eye gaze behavior of a driver operating the same highway route under both clear and rainy conditions. To this end, gaze behavior was analyzed by a two-step clustering approach: first, clustering gaze points within 10-second intervals, and then aggregating cluster centroids into meta-clusters. This, along with Markov transition matrices and metrics such as fixation duration, gaze elevation, and azimuth distributions, reveals meaningful behavioral shifts. While the overall gaze behavior focused on the road with occasional mirror checks remains consistent, rainy conditions lead to more frequent dashboard glances, longer fixation durations, and higher gaze elevation, indicating increased cognitive focus. These findings offer valuable insight into visual attention patterns under adverse conditions and highlight the potential of leveraging gaze modeling to aid in the design of more robust ADAS and DMS.
Open AccessWeather-Dependent Variations in Driver Gaze Behavior: A Case Study in Rainy Conditions
Rainy weather significantly increases the risk of road accidents due to reduced visibility and vehicle traction. Understanding how experienced drivers adapt their visual perception through gaze behavior under such conditions is critical for designing robust driver monitoring systems (DMS) and for informing advanced driver assistance systems (ADAS). This case study investigates the eye gaze behavior of a driver operating the same highway route under both clear and rainy conditions. To this end, gaze behavior was analyzed by a two-step clustering approach: first, clustering gaze points within 10-second intervals, and then aggregating cluster centroids into meta-clusters. This, along with Markov transition matrices and metrics such as fixation duration, gaze elevation, and azimuth distributions, reveals meaningful behavioral shifts. While the overall gaze behavior focused on the road with occasional mirror checks remains consistent, rainy conditions lead to more frequent dashboard glances, longer fixation durations, and higher gaze elevation, indicating increased cognitive focus. These findings offer valuable insight into visual attention patterns under adverse conditions and highlight the potential of leveraging gaze modeling to aid in the design of more robust ADAS and DMS.
- Open Access
Fuel Consumption in Platoons: A Literature Review
Platooning has emerged as a promising strategy for improving fuel efficiency in automated vehicle systems, with significant implications for reducing emissions and operational costs. While existing literature on vehicle platooning primarily focuses on individual aspects such as aerodynamic drag reduction or specific control strategies, this work takes a more comprehensive approach by bringing together a wide range of factors and components that contribute to fuel savings in platoons. In this literature review, we examine the impact of platooning on fuel consumption, highlighting the key components of platoon systems, the factors and actors influencing fuel savings, methods for estimating fuel use, and the effect of platoon instability on efficiency. Furthermore, we study the role of reduced aerodynamic drag, vehicle coordination, and the challenges posed by instability in real-world conditions. By compiling insights from recent studies, this work provides a comprehensive overview of the latest advancements in platooning technologies and highlights both the challenges and opportunities for future research to maximize fuel savings in real-world scenarios.
Open AccessFuel Consumption in Platoons: A Literature Review
Platooning has emerged as a promising strategy for improving fuel efficiency in automated vehicle systems, with significant implications for reducing emissions and operational costs. While existing literature on vehicle platooning primarily focuses on individual aspects such as aerodynamic drag reduction or specific control strategies, this work takes a more comprehensive approach by bringing together a wide range of factors and components that contribute to fuel savings in platoons. In this literature review, we examine the impact of platooning on fuel consumption, highlighting the key components of platoon systems, the factors and actors influencing fuel savings, methods for estimating fuel use, and the effect of platoon instability on efficiency. Furthermore, we study the role of reduced aerodynamic drag, vehicle coordination, and the challenges posed by instability in real-world conditions. By compiling insights from recent studies, this work provides a comprehensive overview of the latest advancements in platooning technologies and highlights both the challenges and opportunities for future research to maximize fuel savings in real-world scenarios.
- Open Access
- 1 August 2024
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 7, 5815-5823
Enhancing Semantic Clarity: Discriminative and Fine-grained Information Mining for Remote Sensing Image-Text Retrieval
Remote sensing image-text retrieval is a fundamental task in remote sensing multimodal analysis, promoting the alignment of visual and language representations. The mainstream approaches commonly focus on capturing shared semantic representations between visual and textual modalities. However, the inherent characteristics of remote sensing image-text pairs lead to a semantic confusion problem, stemming from redundant visual representations and high inter-class similarity. To tackle this problem, we propose a novel Discriminative and Fine-grained Information Mining (DFIM) model, which aims to enhance semantic clarity by reducing visual redundancy and increasing the semantic gap between different classes. Specifically, the Dynamic Visual Enhancement (DVE) module adaptively enhances the visual discriminative features under the guidance of multimodal fusion information. Meanwhile, the Fine-grained Semantic Matching (FSM) module cleverly models the matching relationship between image regions and text words as an optimal transport problem, thereby refining intra-instance matching. Extensive experiments on two benchmark datasets justify the superiority of DFIM in terms of retrieval accuracy and visual interpretability over the leading methods.
- 1 August 2024
- Published by International Joint Conferences on Artificial Intelligence
- Vol. 7, 5815-5823
Open AccessEnhancing Semantic Clarity: Discriminative and Fine-grained Information Mining for Remote Sensing Image-Text Retrieval
Remote sensing image-text retrieval is a fundamental task in remote sensing multimodal analysis, promoting the alignment of visual and language representations. The mainstream approaches commonly focus on capturing shared semantic representations between visual and textual modalities. However, the inherent characteristics of remote sensing image-text pairs lead to a semantic confusion problem, stemming from redundant visual representations and high inter-class similarity. To tackle this problem, we propose a novel Discriminative and Fine-grained Information Mining (DFIM) model, which aims to enhance semantic clarity by reducing visual redundancy and increasing the semantic gap between different classes. Specifically, the Dynamic Visual Enhancement (DVE) module adaptively enhances the visual discriminative features under the guidance of multimodal fusion information. Meanwhile, the Fine-grained Semantic Matching (FSM) module cleverly models the matching relationship between image regions and text words as an optimal transport problem, thereby refining intra-instance matching. Extensive experiments on two benchmark datasets justify the superiority of DFIM in terms of retrieval accuracy and visual interpretability over the leading methods.
- Open Access
- 27 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 105096-105107
Unlocking Dual Utility: 1D-CNN for Milling Tool Health Assessment and Experimental Optimization
In a novel application of 1D Convolutional Neural Networks (1D-CNN), this study pioneers a tri-class classification framework for accurately forecasting the Remaining Useful Life (RUL) of milling tools. By harnessing the 1D-CNN’s innate capability to analyze raw time-series data, we eliminate the traditional bottleneck of extensive feature engineering. Our model undergoes rigorous validation using a leave-one-out cross-validation method, catering to the constraints of a limited dataset. When optimized, the model delivers compelling performance metrics: average accuracy, precision, and recall scores stand at 0.90 ± 0.02, 0.85 ± 0.12, and 0.87 ± 0.08, respectively. What sets this work apart is its dual utility: not only does it excel in tool health assessment, but its output also serves as a diagnostic tool for experimental setups. For instance, anomalies detected in the model’s predictions can act as early warnings for potential sensor malfunctions. Additionally, the model’s performance metrics offer invaluable guidance in optimizing experimental parameters, such as choosing the most efficient sampling rate. In summary, this study not only establishes the robustness of 1D-CNNs in assessing milling tool health but also unveils their untapped potential as diagnostic aids for fine-tuning experimental setups.
- 27 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 105096-105107
Open AccessUnlocking Dual Utility: 1D-CNN for Milling Tool Health Assessment and Experimental Optimization
In a novel application of 1D Convolutional Neural Networks (1D-CNN), this study pioneers a tri-class classification framework for accurately forecasting the Remaining Useful Life (RUL) of milling tools. By harnessing the 1D-CNN’s innate capability to analyze raw time-series data, we eliminate the traditional bottleneck of extensive feature engineering. Our model undergoes rigorous validation using a leave-one-out cross-validation method, catering to the constraints of a limited dataset. When optimized, the model delivers compelling performance metrics: average accuracy, precision, and recall scores stand at 0.90 ± 0.02, 0.85 ± 0.12, and 0.87 ± 0.08, respectively. What sets this work apart is its dual utility: not only does it excel in tool health assessment, but its output also serves as a diagnostic tool for experimental setups. For instance, anomalies detected in the model’s predictions can act as early warnings for potential sensor malfunctions. Additionally, the model’s performance metrics offer invaluable guidance in optimizing experimental parameters, such as choosing the most efficient sampling rate. In summary, this study not only establishes the robustness of 1D-CNNs in assessing milling tool health but also unveils their untapped potential as diagnostic aids for fine-tuning experimental setups.