Model-Based Engine Control Reducing Adjustment Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for model-based control and regulation of internal combustion engines are limited by high adjustment complexity and information loss due to linearization, particularly in representing the entire engine rather than just the gas path.
Innovation Solution
A method that uses an optimizer to read emission classes and maximum mechanical component loads from libraries to set binding parameters for combustion and gas path models, calculating injection and gas path setpoint values to minimize a quality measure within a prediction horizon, thereby adjusting the engine's operating point efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If characteristic lines and characteristic fields are used for engine control, then the engine can be controlled based on power requirements, but the adjustment outlay becomes high due to the plurality of characteristic lines/fields and their correlations
Solution Approach 1:
The patent replaces the mechanical/tabular approach of characteristic lines and fields with a mathematical model-based system. The engine control unit uses a mathematical model that directly calculates actuating variables from measured operating parameters, eliminating the need for extensive pre-programmed characteristic data and reducing adjustment complexity.
Solution Approach 2:
The patent changes the fundamental parameter representation from fixed characteristic lines/fields to dynamic mathematical model parameters. The system uses measured operating parameters (pressure, temperature, flow rates) to dynamically calculate optimal actuating variables, allowing continuous adaptation without reprogramming characteristic data.
2Device complexity
If a linearized gas path model is used for model-based regulation, then the regulation can be simplified, but information loss is unavoidable due to linearization
Solution Approach 1:
The patent implements a dynamic, non-linear gas path model that adapts to changing operating conditions. The system continuously updates model parameters based on measured operating variables (pressures, temperatures, flow rates), maintaining accuracy across the entire operating range without requiring linearization approximations.
Solution Approach 2:
The system uses feedback from measured operating parameters to continuously update the gas path model and optimize actuating variables. The control unit compares measured values with model predictions and adjusts parameters to minimize deviations, maintaining model accuracy without linearization.
3Adaptability or versatility
If trained data are mapped using neuronal networks for adaptation of injection parameters, then only learned data from test stand runs need to be mapped, but the approach is limited to specific trained datasets
Solution Approach 1:
The patent creates a universal mathematical model that can handle various operating conditions and engine types without requiring separate training datasets. The gas path model and optimization algorithm work across different operating ranges (partial load, full load, transient conditions) using the same fundamental equations, eliminating the need for extensive test stand training for each scenario.
Solution Approach 2:
The system uses real-time measured operating parameters to automatically update and optimize its control strategy without external training input. The optimization algorithm continuously adapts to current operating conditions using live sensor data, eliminating dependency on pre-collected training datasets while maintaining adaptability.
Data Source
AI summary
A method for model-based control and regulation of an internal combustion engine. An emission class for operating the engine is read from a first library by an optimizer; a maximum mechanical component load is read from a second library by the optimizer using the engine type; and the emission class and the component load are set as mandatory for a combustion model and a gas path model. Injection system target values for actuating injection system actuators are calculated using the combustion model based on a target torque. Gas path target values for actuating gas path actuators are calculated using the gas path model based on the target torque. A quality measurement is calculated by the optimizer based on the injection system and gas path target values. The quality measurement is minimized by the optimizer by changing the injection system and gas path target values within a prediction horizon. The injection system and gas path target values are set as decisive for adjusting the operating point of the engine by the optimizer using the minimized quality measurement.