Friction Model Parameter Identification via Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods face difficulties in identifying model parameters for friction models, especially when there are many parameters to be identified or non-linear characteristics are involved, and they struggle to separate the influence of position, displacement, velocity, and acceleration on friction characteristics effectively.
Innovation Solution
A method that measures and identifies parameters of position-dependent, displacement-dependent, velocity-dependent, and acceleration-dependent friction models by analyzing the relation between driving force and respective state amounts, allowing for the separation of friction characteristics and the creation of a single-input single-output system friction model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional identification algorithms are used to identify friction model parameters, then the identification process can be completed, but it becomes difficult and time-consuming when there are many parameters to be identified or when non-linear characteristics are involved
Solution Approach 1:
The patent segments the friction identification process into four independent single-input single-output models: position-dependent friction, displacement-dependent friction, velocity-dependent friction, and acceleration-dependent friction. Each model identifies one specific parameter independently, avoiding the complexity of identifying multiple parameters simultaneously. This segmentation reduces identification time while maintaining accuracy for each friction component.
Solution Approach 2:
The patent extracts and separates the influence of each state amount (position, displacement, velocity, acceleration) on friction characteristics into distinct models. By taking out each factor independently and identifying its specific friction model parameter separately, the method simplifies the overall identification process and makes it more efficient compared to conventional approaches that treat all parameters together.
2Adaptability or versatility
If conventional methods attempt to identify multiple friction parameters simultaneously, then comprehensive friction modeling is achieved, but the complexity of the identification process increases significantly
Solution Approach 1:
The patent divides the comprehensive friction model into four separate single-input single-output models, each handling one specific friction parameter. This segmentation maintains the versatility of comprehensive friction modeling while reducing the complexity of the identification process by treating each parameter independently through dedicated measurement and identification steps.
Solution Approach 2:
The patent applies local quality by creating specialized friction models for different aspects of friction behavior: position-dependent friction for static effects, displacement-dependent friction for hysteresis effects, velocity-dependent friction for viscous effects, and acceleration-dependent friction for dynamic effects. Each local model is optimized for its specific friction characteristic, achieving comprehensive coverage without overall process complexity.
3Ease of operation
If the influence of multiple state amounts on friction is not separated, then the identification process is simpler, but the accuracy of friction correction deteriorates
Solution Approach 1:
The patent segments the friction analysis into four distinct components based on different state amounts (position, displacement, velocity, acceleration). Each segment is measured and identified separately, maintaining process simplicity through systematic independent identification while achieving high accuracy by capturing the specific influence of each state amount on friction characteristics.
Solution Approach 2:
The patent introduces a new dimension of analysis by separating friction identification along the dimension of state amount influence. Instead of treating friction as a single unified parameter, the method adds dimensional separation based on position, displacement, velocity, and acceleration dependencies, enabling both simple systematic identification and high precision correction.
Data Source
AI summary
A friction identification method includes: measuring a relation between a driving force and a position of a driven object; identifying a parameter of a position-dependent friction model based on a relation between a driving force and a position of the driven object; measuring a relation between a driving force and a displacement of the driven object from a position at which a motion direction is reversed; identifying a parameter of a displacement-dependent friction model; measuring a relation between a driving force and a velocity of the driven object; identifying a parameter of a velocity-dependent friction model; measuring a relation between a driving force and an acceleration of the driven object; and identifying a parameter of an acceleration-dependent friction model.