Engine Combustion Parameter Reconstruction Using Multi-Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing engine combustion control technologies face challenges in accurately measuring real-time combustion parameters due to the complexity of engine combustion models and the inability of prior calibration maps to reflect cycle-to-cycle and cylinder-to-cylinder variations, while relying on single sensing signals introduces noise interference and high costs for additional sensors.
Innovation Solution
A value-moving deviation method using multiple sensing signals, including crankshaft position, knock, and rail pressure signals, to reconstruct combustion parameters by establishing shape base models and calculating weighted fusion deviations, allowing for real-time acquisition of cycle-stage parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If engine combustion models are constructed to obtain combustion parameters, then combustion parameters can be obtained, but calculation is complex and cannot satisfy real-time requirements
Solution Approach 1:
The patent pre-establishes a comprehensive calibration map containing combustion parameters under multiple working conditions before actual engine operation. This preliminary action stores extensive combustion model calculations in advance, allowing the ECU to simply query pre-computed values during real-time control without performing complex calculations, thus resolving the contradiction between accuracy and real-time performance
Solution Approach 2:
The patent creates a simplified copy of the combustion model by using lookup tables that replicate the behavior of complex combustion models without requiring their full computational power. The calibration map serves as a copied representation of combustion characteristics that can be quickly accessed and interpolated, providing real-time parameter acquisition without complex model calculations
2Ease of operation
If prior calibration map is used to obtain combustion parameters, then parameters can be obtained under calibration working conditions, but it cannot reflect changes in combustion information between cycles and between different air cylinders
Solution Approach 1:
The patent merges multiple data sources including the prior calibration map with real-time sensor measurements from multiple cycles and cylinders. By combining the general trends from the calibration map with specific cycle-to-cycle and cylinder-to-cylinder variations captured by sensors, the system achieves both ease of operation and high precision in parameter acquisition
Solution Approach 2:
The patent transforms the static prior calibration map into a dynamic parameter acquisition system by continuously updating combustion parameters based on real-time sensor feedback from each cycle and cylinder. This dynamic adjustment allows the system to adapt to changing combustion conditions while maintaining the simplicity of map-based acquisition
3Measurement precision
If cylinder pressure sensor is installed to extract combustion information, then combustion parameters can be accurately obtained, but it is extremely expensive and the harsh combustion environment will greatly shorten service life
Solution Approach 1:
The patent replaces the expensive, harsh-environment cylinder pressure sensor with cheaper, more durable sensors that have shorter service lives but can be easily replaced. The system uses sensors that are not designed for direct cylinder pressure measurement but can still extract sufficient combustion information when combined with the calibration map approach, reducing cost and complexity while maintaining acceptable accuracy
Solution Approach 2:
The patent introduces the prior calibration map as an intermediary that translates measurements from inexpensive, easily installed sensors into accurate combustion parameters. The calibration map acts as a mediator that bridges the gap between simple sensor measurements and the complex combustion parameters that would otherwise require expensive direct measurement sensors
4Device complexity
If combustion parameters are reconstructed only by using a single sensing signal, then sensor cost is reduced, but accuracy of reconstruction relies heavily on sensor working state and is susceptible to noise interference
Solution Approach 1:
The patent makes the prior calibration map a universal reference that can be applied across multiple working conditions and sensor types. The calibration map serves multiple functions including noise filtering, parameter correlation, and accuracy enhancement, allowing the system to use simple single-sensor configurations while achieving high reconstruction accuracy through the multi-functional calibration data
Data Source
AI summary
A value-moving deviation method for acquiring cycle-stage combustion parameters of an engine based on multiple sensing signals includes: with prior values of combustion feature parameters under each piece of working condition information as a reference, reconstructing combustion feature parameter values based on signal features acquired by sensors of the engine, and calculating deviations between the combustion feature parameter values and the prior values and weights of the deviations; obtaining a fusion deviation of the combustion feature parameters as a transient fluctuation amount of the combustion feature parameters; correcting, based on the value-moving deviation method, the prior values according to the obtained transient fluctuation amount, so as to obtain actual values of the combustion feature parameters of a current cycle of the engine; and updating, according to a Weber equation, a combustion heat release rate curve to obtain each piece of transient combustion parameter information under a current working condition.