Powertrain Setpoint Control for Real-Time Transient Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional powertrain systems face challenges in optimizing transient performance due to the need for large-dimensional maps that exceed memory and processing power available in engine control units, limiting the ability to determine set points and trajectories in real-time effectively.

Innovation Solution

A powertrain controller is configured to determine set points and trajectories based on current operating conditions using a performance cost function, allowing for real-time adjustment of actuator positions to optimize conditions such as temperature, pressure, and flow, while maintaining constraints and minimizing fuel consumption and parasitic losses.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If static offline set points are computed as a function of disturbance variables for transient performance optimization, then the control accuracy is improved, but the memory and processing power requirements exceed the available resources in the engine control unit

Engineering Contradiction:
Improvecontrol accuracyVSAvoidmemory and processing power requirements
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The control approach is segmented into two distinct parts: an offline phase where a comprehensive map of set points is computed and stored, and an online phase where only simple lookup and interpolation operations are performed. This segmentation allows the computationally intensive optimization to be done beforehand when full processing power is available, while the runtime controller operates within its resource constraints.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The optimal set points and trajectories are pre-computed offline before actual operation. By performing the complex optimization calculations in advance and storing the results in lookup tables, the system eliminates the need for real-time computation of these values during engine operation, thus fitting within the control unit's memory and processing capabilities.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If large-dimensional maps are used for transient performance optimization, then the control precision is improved, but the device complexity increases beyond what can be implemented in online environment

Engineering Contradiction:
Improvecontrol precisionVSAvoiddevice complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The control system is divided into an offline optimization module that generates the comprehensive control maps and an online execution module that simply queries these pre-computed maps. This segmentation reduces online device complexity while preserving the precision benefits of large-dimensional optimization maps.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The complex optimization results are copied into pre-computed lookup tables that can be stored in memory. Instead of implementing the full optimization algorithm in the online controller, the system uses simplified copies of the optimization results that contain all necessary control information in an easily accessible format.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3239486B1A condition-based powertrain control system
Publication Date: 2024.01.03 GARRETT TRANSPORTATION I INC
  • EP3239486B1 patent drawingFigure 1
  • EP3239486B1 patent drawingFigure 2
  • EP3239486B1 patent drawingFigure 3

AI summary

A system and approach for development of setpoints for a controller of a powertrain system. The controller may be parametrized as a function of setpoints to provide performance variables that are considered acceptable by a user or operator for current operating conditions of the engine or powertrain. The controller may determine set point trajectories in real time during operation of the powertrain system and determine positions of manipulated variables do drive controlled variables to associated and determined set point trajectories. The present system and approach may determine set point trajectories for powertrain conditions on-line and in real time, whereas set point trajectories have previously been determined off-line for powertrain control.