Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Discovery Grants Program - Individual

The University of British Columbia

$83,000.00

Hybrid physics-based and data-driven machine tool monitoring

FY 2024-25

Summary
For 37 years, the applicant has focused on developing mathematical models for machining processes, machine tool kinematics and control, and the complex interactions between machine vibrations and cutting processes. These digital models are well-regarded in academia and widely used in global industry. The research goal is to refine these models by incorporating sensory data to improve key physical parameters characterized by uncertainties. These models encompass coupled relationships with nonlinear dependencies, considering operating conditions, machine tool structural dynamics, and work material properties. The research is proposed to establish a framework for a hybrid physics and data-driven model towards developing a digital twin of machining system as a long-term research objective. The proposed approach involves an inverse solution methodology, blending physics-based simulations, sensory data, and machine learning. Initially, advanced digital models will simulate the process, varying key parameters within uncertainty ranges. An artificial neural network architecture will be designed, optimizing predicted measurable states (e.g., cutting forces, vibration spectra, deflection errors, contouring errors) as inputs and uncertain parameters as outputs. The architecture will be refined further using experimental machining data. Ultimately, the artificial neural network will track machine and process parameters during machining, mapped to the tool path for manufacturing process and machine health monitoring. This approach will be applied to two digital models: 5-axis turbine blade machining: Developed in the applicant's laboratory, this model utilizes efficient substructuring to identify blade structural dynamics as mass is removed during cutting. It predicts deflection marks and assesses chatter stability to minimize costly physical trials in the gas turbine industry. The network will identify uncertain parameters, such as position-dependent natural frequencies, process damping, cutting force coefficients, and Computer Numerical Control (CNC) parameters. Feed drive dynamics: A five-axis research Computer Numerical Control (CNC) system developed in the laboratory predicts motor torque, velocity, displacement, tracking errors, and vibrations under cutting loads during multi-axis motions. Bearing damage, guideway, ball screw, and nut wear will be modeled and incorporated into the dynamic model of the feed drives. A recurrent neural network will be trained to predict faults using simulation data, followed by fine-tuning with experimental data, enabling fault detection within the feed drive. This research is foundational for creating physics-based digital twins of manufacturing processes and machines, driving advancements in the field. Physics-based digital twins will enable the industry to simulate and monitor manufacturing operations more accurately, reduce cost and cycle times by as much as 40%, and improve the accuracy of manufactured parts.
Program
Discovery Grants Program - Individual
Location
British Columbia, Canada
Recipient Type
Academia
Type
Grant
Source
NSERC's Awards Data

As published

Application id
RGPIN-2024-03820
Name nom
Altintas, Yusuf
Department département
Mechanical Engineering Department
Organization id
2
Institution établissement
The University of British Columbia
Province en
British Columbia
Province fr
Colombie-Britannique
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
83000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1512
Committee name en
Mechanical Engineering
Committee name fr
Génie mécanique
Area of application code
703
Area of application group en
Transportation systems and services
Area of application group fr
Services et systèmes de transport
Area of application en
Aerospace
Area of application fr
Aérospatial
Research subject code
1702
Research subject group en
Design and manufacturing
Research subject group fr
Conception et fabrication
Research subject en
Advanced manufacturing
Research subject fr
Méthodes de fabrication avancées
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Hybrid physics-based and data-driven machine tool monitoring
Keywords
machine tools; vibrations; dynamics; control; metal cutting; monitoring; chatter; spindle; feed drive; cnc
Application summary
For 37 years, the applicant has focused on developing mathematical models for machining processes, machine tool kinematics and control, and the complex interactions between machine vibrations and cutting processes. These digital models are well-regarded in academia and widely used in global industry. The research goal is to refine these models by incorporating sensory data to improve key physical parameters characterized by uncertainties. These models encompass coupled relationships with nonlinear dependencies, considering operating conditions, machine tool structural dynamics, and work material properties. The research is proposed to establish a framework for a hybrid physics and data-driven model towards developing a digital twin of machining system as a long-term research objective. The proposed approach involves an inverse solution methodology, blending physics-based simulations, sensory data, and machine learning. Initially, advanced digital models will simulate the process, varying key parameters within uncertainty ranges. An artificial neural network architecture will be designed, optimizing predicted measurable states (e.g., cutting forces, vibration spectra, deflection errors, contouring errors) as inputs and uncertain parameters as outputs. The architecture will be refined further using experimental machining data. Ultimately, the artificial neural network will track machine and process parameters during machining, mapped to the tool path for manufacturing process and machine health monitoring. This approach will be applied to two digital models: 5-axis turbine blade machining: Developed in the applicant's laboratory, this model utilizes efficient substructuring to identify blade structural dynamics as mass is removed during cutting. It predicts deflection marks and assesses chatter stability to minimize costly physical trials in the gas turbine industry. The network will identify uncertain parameters, such as position-dependent natural frequencies, process damping, cutting force coefficients, and Computer Numerical Control (CNC) parameters. Feed drive dynamics: A five-axis research Computer Numerical Control (CNC) system developed in the laboratory predicts motor torque, velocity, displacement, tracking errors, and vibrations under cutting loads during multi-axis motions. Bearing damage, guideway, ball screw, and nut wear will be modeled and incorporated into the dynamic model of the feed drives. A recurrent neural network will be trained to predict faults using simulation data, followed by fine-tuning with experimental data, enabling fault detection within the feed drive. This research is foundational for creating physics-based digital twins of manufacturing processes and machines, driving advancements in the field. Physics-based digital twins will enable the industry to simulate and monitor manufacturing operations more accurately, reduce cost and cycle times by as much as 40%, and improve the accuracy of manufactured parts.
The University of British Columbia - $83,000 Grant (2… | Canada Spends