Virtual Friction Model for MDPS Column Torque Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional motor driven power steering systems face challenges in accurately estimating frictional torque during stop or low-speed driving conditions, leading to severe vibration and requiring repeated tuning operations to achieve desired steering performance.
Innovation Solution
A virtual friction model is set for the column connected to the steering wheel, calculating frictional torque using steering angular speed as an input parameter, and a target steering torque is calculated based on this model, incorporating damping and stiffness torques as nonlinear functions, along with stiction and Stribeck effects, to improve control accuracy and reduce vibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional friction models are used for column friction estimation, then the model is simple to implement, but the frictional torque estimation accuracy deteriorates in stop or low-speed driving conditions
Solution Approach 1:
The patent changes the parameters of the friction model by introducing a virtual friction model that incorporates stiction and Stribeck effects, using steering angular speed as an input parameter to accurately estimate frictional torque across different driving conditions, particularly improving accuracy in stop and low-speed conditions
Solution Approach 2:
The patent introduces a virtual friction model as an intermediary component between the steering system and control algorithm. This virtual model acts as a mediator that estimates frictional torque based on steering angular speed, allowing the control system to compensate for friction effects without directly measuring them
2Productivity
If open-loop control is used for MDPS, then the control implementation is simple, but the steering performance varies according to hardware distribution and requires repeated tuning
Solution Approach 1:
The patent implements feedback control by using a virtual steering system model that predicts target steering torque based on steering angular speed and frictional torque estimation. This feedback mechanism allows the system to adapt to different hardware configurations and achieve consistent steering performance without repeated tuning
Solution Approach 2:
The patent performs preliminary action by pre-establishing a virtual steering system model that incorporates friction characteristics. This model allows the system to predict and compensate for friction effects in advance, improving both development efficiency and performance consistency
3Reliability
If feedback control with lookup table is used, then control robustness and tuning efficiency are improved, but the performance prediction capability in initial design step is lost
Solution Approach 1:
The patent introduces a virtual steering system model as an intermediary that maintains performance prediction capability while enabling feedback control. This virtual model serves as a simulator that allows designers to predict steering performance in the initial design phase while also supporting robust feedback control implementation
4Ease of operation
If conventional steering system model is used, then the model works adequately in general driving conditions, but severe vibration occurs in stop or low-speed driving condition
Solution Approach 1:
The patent changes the friction model parameters by incorporating stiction and Stribeck effects that are particularly relevant at low speeds and stop conditions. This parameter adjustment eliminates the severe vibration problems that occur with conventional models in these operating conditions
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enhances the accuracy of frictional torque estimation, reduces steering system vibration, and allows for diverse target steering torque generation, improving development efficiency and steering performance prediction.
Implementation Method 1
The damping torque may be set as ahyperbolic tangent function, in which the steering angular speed of the steering wheel is an input parameter
Implementation Method 2
The stiffness torque may be set as a polynomial function of two degrees or more, in which the torsional displacement of the column is an input parameter
Implementation Method 3
The setting of the virtual friction model may include setting the virtual friction model to reflect a stiction and Stribeck effect of the column
Implementation Method 4
The setting of the virtual friction model may include setting the virtual friction model to reflect a stiction and Stribeck effect of the column
Data Source
AI summary
A motor driven power steering control method may include setting a virtual friction model to a column connected between a steering wheel and a rack gear; calculating a frictional torque of the column by taking a steering angular speed of the steering wheel as an input parameter in the set virtual friction model; and calculating a target steering torque on the basis of a virtual steering system model using the frictional torque of the column as a parameter.