AI Power Divider for Hybrid Truck Energy Recovery and Low Drag
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current heavy truck propulsion systems are inefficient for highway drive cycles due to limited electrical power assist and mechanical drag, and lack the ability to effectively capture and distribute terrain potential and kinetic energy, leading to reduced fuel efficiency and increased mechanical stress.
Innovation Solution
An AI-controlled multi-channel power divider/combiner system that utilizes sensor data and machine learning algorithms to optimize power management in hybrid electric vehicles, enabling real-time adjustments in power distribution and storage based on operating conditions, such as terrain and altitude, to enhance energy harvesting and thermal dissipation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If a diesel engine is directly connected to the transmission for mechanical drive power, then the system is simple and durable, but it cannot collect and distribute terrain potential energy or kinetic energy effectively
Solution Approach 1:
The power train is segmented into separate mechanical and electrical power paths. The mechanical path handles direct diesel engine power delivery through the transmission, while the electrical path independently captures and manages terrain potential energy and kinetic energy through generators and energy storage systems, allowing each segment to optimize its function without interfering with the other
Solution Approach 2:
The electrical power system serves multiple functions simultaneously: it captures terrain potential energy during descent, recovers kinetic energy during braking, provides auxiliary power for vehicle systems, and can supplement propulsion when needed. This multi-functional electrical system resolves the contradiction by adding energy recovery capabilities without requiring complete redesign of the mechanical power train
2Loss of energy
If a parallel hybrid configuration is used with mechanical power assist, then some energy recovery is possible, but mechanical drag increases when the diesel engine is the sole power source
Solution Approach 1:
The energy recovery and storage functions are extracted from the mechanical power train and placed in a separate electrical system. This allows the mechanical components to remain simple and non-intrusive during diesel-only operation, eliminating mechanical drag, while the electrical system independently handles energy capture and management without adding resistance to the mechanical power path
Solution Approach 2:
The patent replaces mechanical energy storage and recovery mechanisms with an electrical system. Instead of using mechanical flywheels, springs, or complex gear-based energy recapture devices that would add drag to the mechanical power train, the system uses generators, power electronics, and electrical energy storage to achieve the same energy recovery function without mechanical interference
3Reliability
If DC-to-DC inverters are used to supply regulated battery power, then power management is achieved, but the system becomes inefficient and prone to failure due to high switching times and currents
Solution Approach 1:
The power management system uses dynamic control strategies that adapt to real-time operating conditions. The controller monitors vehicle state, energy storage levels, and power demands to dynamically adjust power flow paths, selecting the most efficient routing for each condition rather than using fixed high-frequency switching, thereby reducing energy losses and improving reliability
Solution Approach 2:
The system changes operational parameters based on conditions, using different power management modes for different scenarios. Instead of maintaining constant high-frequency switching, the system adjusts switching frequencies, power flow directions, and component engagement based on real-time needs, reducing unnecessary energy losses and thermal stress on components
Data Source
AI summary
A method is provided for controlling power in a hybrid electric vehicle. The method may include receiving sensor input data in a computer-implemented artificial intelligence neural network operatively associated with the vehicle. The sensor input data may be generated in response to a travel condition or an operating state associated with the vehicle. The method may also include generating condition-based awareness signals with the artificial intelligence neural network; processing the condition-based awareness signals with control algorithms; and adjusting a power-related operating state of the vehicle in response to the processing performed by the control algorithm.