EV Thermal Management MPC for Multi-Route Uncertainty
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing thermal management systems for electric vehicles face challenges in handling multiple route scenarios due to uncertainties in vehicle speed and thermal load, leading to potential robustness issues and inefficiencies in energy consumption and constraint violations.
Innovation Solution
A stochastic model predictive control framework that accounts for multiple possible routes by minimizing energy consumption and ensuring favorable battery and cabin thermal conditions, using a cost function that adapts to predicted constraint violations and adjusts coolant split ratios and flow rates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If model predictive control with long horizon is used to handle slow thermal dynamics, then thermal management performance is improved, but computational effort increases significantly
Solution Approach 1:
The patent pre-calculates and stores thermal models, cost functions, and optimization parameters before real-time operation. By preparing lookup tables and pre-processing route information, the system reduces the computational burden during actual thermal management control while maintaining long-horizon predictive capabilities.
Solution Approach 2:
The patent implements a dynamic prediction horizon that adapts to different operating conditions and route characteristics. Instead of using a fixed long horizon always, the system adjusts the prediction horizon length based on current thermal states, vehicle operation mode, and route uncertainty, thereby reducing computational effort when full long-horizon optimization is not necessary.
2Ease of operation
If deterministic model predictive control is used with predetermined route assumption, then control simplicity is maintained, but robustness deteriorates when route changes occur
Solution Approach 1:
The patent extends the model predictive control framework to handle multiple route scenarios simultaneously. The cost function and optimization algorithm are designed to evaluate and select from multiple possible routes, making the control system universal enough to handle both predetermined and uncertain route conditions without requiring separate control strategies.
Solution Approach 2:
The system pre-processes multiple potential route information and incorporates route uncertainty into the prediction model before control execution. By preparing route probability distributions and pre-calculating thermal impacts of different routes, the system maintains robustness when actual routes differ from predictions.
3Reliability
If thermal management system operates with high energy consumption, then thermal constraints are better satisfied, but overall vehicle efficiency decreases
Solution Approach 1:
The patent dynamically adjusts thermal management parameters such as coolant flow rates, pump speeds, and valve positions based on real-time vehicle operation and thermal states. By optimizing these parameters continuously rather than maintaining fixed high-level settings, the system satisfies thermal constraints only when necessary, thereby reducing overall energy consumption and improving vehicle efficiency.
Solution Approach 2:
The model predictive control operates in discrete time steps with receding horizon optimization, periodically adjusting thermal management actions based on current state. This periodic control approach allows the system to apply energy-intensive cooling or heating only when thermal constraints are at risk, rather than continuously operating at high energy consumption levels.
Data Source
AI summary
A stochastic Model Predictive Control approach is developed to efficiently optimize the thermal management of electric vehicles and accommodate scenarios with multiple routes. To account for the uncertainties, the cost function is constructed to minimize the expected cost across all possible routes over the prediction horizon. Thermal constraints are treated as soft constraints using slack variables. This approach allows for flexibility in satisfying the constraints while optimizing the performance. Through simulations, the performance of the proposed method is evaluated using a fleet of vehicles. In this way, the proposed method achieves a good trade-off between multiple competing performance metrics. Furthermore, an adaptation strategy is introduced, which dynamically adjusts the penalty weight value. This adaptive approach eliminates the need for offline calibration and further enhances performance. The results indicate that the time-varying penalty weight significantly reduces the total constraint violations by up to 20% without impacting the performance on energy consumption.