Predictive Control for Vehicle Regenerative Braking Oscillations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Motor vehicles with energy recovery systems experience oscillations in the torque transmission chain during braking, leading to passenger discomfort and increased wear on vehicle components, due to delays in controlling the energy recovery means which are caused by measurement, calculation, and control delays.
Innovation Solution
A control method that uses a predictive mathematical model to estimate a forecast value of the dynamic characteristic of the energy recovery system, allowing for instantaneous counteraction of oscillations by incorporating this predicted value into the control setpoints, thereby reducing the onset of oscillations before they are detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If a conventional control method is used to control the energy recovery means during braking, then the energy recovery system can operate, but oscillations occur in the torque transmission chain due to control delays
Solution Approach 1:
The patent applies preliminary action by using a predictive mathematical model to estimate the dynamic characteristic of the energy recovery means before actual oscillations occur. The predictive filter calculates a predicted value based on current and past measurements, allowing the control system to prepare counteracting torque commands in advance, thereby preventing oscillation onset rather than reacting after delays detect them.
Solution Approach 2:
The patent implements preliminary anti-action by calculating a second intermediate setpoint that actively counteracts predicted oscillations before they manifest. The corrective filter uses the predicted dynamic characteristic to generate opposing torque commands that neutralize harmful oscillations in the torque transmission chain before they affect passenger comfort or cause component wear.
2Ease of operation
If the control unit reacts only after oscillations are detected, then the control logic is simple, but the reaction is too slow due to measurement, calculation, and control delays
Solution Approach 1:
The patent eliminates control delay by performing preliminary calculations through the predictive filter. Instead of waiting for oscillations to be detected and then calculating the response, the system continuously computes predicted values using the mathematical model, which incorporates current and past measurements to forecast future states, enabling instantaneous counteraction when needed.
3Strength
If the torque transmission chain components have non-zero rigidity and damping, then the system is mechanically realistic, but oscillations are generated during braking
Solution Approach 1:
The patent implements feedback by continuously measuring the dynamic characteristic of the energy recovery means and using these measurements as inputs to the predictive mathematical model. The predictive filter processes current and past measurements to generate predicted values that feed into the control calculation, creating a closed-loop system that actively maintains torque transmission stability despite the inherent oscillatory nature of rigid-damped mechanical components.
Data Source
Figure 1~4
Figure 2
Figure 5~6
AI summary
The invention relates to a method for controlling a vehicle (1) equipped with a means (10) for recuperating the energy generated under braking, comprising steps of acquiring a braking instruction, calculating a first intermediate setpoint using a preventive filter which filters said braking instruction at the resonant frequency of the drive train (20) of the vehicle (1), calculating a second intermediate setpoint, formulated so as to damp out the oscillations in the drive train (20) using a curative filter, controlling said regenerative means (10) on the basis of said first and second intermediate setpoints, and controlling the brakes (60, 70). According to the invention, the second intermediate setpoint is calculated as a function of a forecast value of a dynamic characteristic of the recuperative means, estimated using a predictive mathematical model.