Two-Layer Predictive Controller for Bidirectional Inductive Power Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing control systems for bidirectional power transfer, such as vehicle-to-grid and grid-to-vehicle applications, face issues with settling time, overshoots, oscillations, and inaccuracies due to high frequency operations and harmonics, and require a supervisory controller, while being inefficient in computational processing and not considering misalignment effects.
Innovation Solution
A two-layer predictive controller system is introduced, comprising a first layer that communicates with infrastructures and decides on charge, discharge, or abstain modes, and a second layer that predicts control parameters for resonant converters to manage bidirectional inductive power transfer, including phase shifts and pulse-phase modulation for efficient power flow management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If classical PI or PID controllers are used for bidirectional power transfer, then the control structure is simple, but the system exhibits settling time, overshoots, oscillations, and cannot resist uncertainties and disturbances
Solution Approach 1:
The controller is divided into two independent layers: a supervisory controller that makes high-level decisions about charge/discharge modes, and a predictive controller that executes low-level power flow control. This segmentation allows each layer to specialize in its function, improving overall reliability while maintaining manageable complexity.
Solution Approach 2:
The predictive controller uses a mathematical model to predict future system states and pre-calculate optimal control parameters before disturbances occur. This preliminary action enables the system to proactively compensate for uncertainties and disturbances, eliminating overshoots and oscillations that plague reactive controllers like PID.
2Reliability
If fuzzy logic controllers are used to improve robustness, then the controller becomes more resistant to uncertainties, but computational processing time and effort increase significantly
Solution Approach 1:
The patent replaces the computational-intensive fuzzy logic approach with a model-based predictive control method that uses analytical mathematical models and closed-form solutions. This substitution maintains robustness to uncertainties while dramatically reducing computational processing time, making it suitable for real-time high-frequency power conversion applications.
3Device complexity
If existing single-layer controllers with fixed reference power-flow rate are used, then the control implementation is straightforward, but the system cannot operate independently and requires a supervisory controller
Solution Approach 1:
The predictive controller is designed to perform multiple functions within a single layer: it determines optimal power flow rates, calculates phase shift angles, and adjusts switching frequencies for both primary and secondary inverters. This multi-functionality enables autonomous operation without requiring a separate supervisory controller, while the modular architecture keeps implementation straightforward.
4Speed
If existing controllers operate at high frequency, then the power transfer speed is fast, but the operation becomes inaccurate due to large levels of harmonics
Solution Approach 1:
The controller dynamically adjusts switching frequencies of both primary and secondary inverters based on real-time operating conditions and misalignment detection. This dynamic frequency adjustment optimizes the resonant coupling at high frequencies while minimizing harmonic distortion, maintaining both fast power transfer speed and high control accuracy simultaneously.
5Device complexity
If misalignment effects are not considered in the controller, then the control algorithm is simpler, but the performance becomes inaccurate
Solution Approach 1:
The controller performs preliminary detection and estimation of misalignment between primary and secondary coils before executing power transfer. By calculating misalignment angles and compensating for them in advance through adjusted phase shift commands, the system maintains high performance accuracy without adding significant complexity to the control algorithm.
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 two-layer controller enables autonomous and efficient bidirectional power flow, reducing computational effort, improving stability, and considering misalignment effects, allowing electric vehicles to charge, discharge, or abstain based on psychological and retail prices, ensuring optimal driving performance and minimal grid impact.
Implementation Method 1
a wireless coupler coupling the primary inverter and the secondary inverter such that inductive power transfers bidirectionally between the primary inverter and the secondary inverter
Data Source
AI summary
A two layer predictive controller for bidirectional inductive power transfer can include: a first layer controller generating a mutual inductance and a reference active power; and a second layer controller receiving the mutual inductance and the reference active power, and generating a primary phase shift, a secondary phase shift, and a differential phase shift; the primary phase shift configured to manage a magnitude of an output voltage of a primary inverter; the secondary phase shift configured to manage a magnitude of an output voltage of a secondary inverter; and the differential phase shift being a phase difference between the output voltage of the primary inverter and the output voltage of the secondary inverter.