Engine Torque Estimation via Phase Compensation and Weighted Averaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating engine torque in gas turbine engines are often inaccurate, computationally complex, and resource-intensive, particularly in real-time and transient conditions, necessitating a more efficient and accurate approach.
Innovation Solution
A method involving the generation of corrected variable values from initial engine parameters using theta and delta corrections, followed by phase compensation and weighted averaging to provide a redundant torque estimation, which is less computationally complex and more accurate.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple parameters are used to estimate engine torque, then accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the torque estimation process into distinct computational stages: calculating intermediate variables from sensor inputs, applying correction factors for operating conditions, and combining results through weighted averaging. This segmentation allows complex multi-parameter processing to be broken into manageable steps that improve accuracy while controlling computational complexity at each stage.
Solution Approach 2:
The patent applies preliminary corrections to sensor variables before final torque calculation. Correction factors are pre-determined based on engine operating conditions (temperature, pressure, altitude), and these corrections are applied in advance to the raw sensor data. This preliminary action ensures that the main torque calculation uses already-corrected values, improving accuracy without adding significant computational burden during real-time operation.
2Measurement precision
If computationally complex techniques are used, then torque estimation accuracy may improve, but computational time and resource consumption increase
Solution Approach 1:
The system uses readily available sensor data from the engine's existing monitoring system (fuel flow, air flow, temperature, pressure sensors) to perform torque estimation. Rather than requiring additional specialized sensors or complex external measurements, the method serves itself using the engine's own operational parameters, thereby achieving accurate torque estimation without excessive computational overhead or additional hardware complexity.
3Reliability
If redundant torque estimation methods are implemented, then reliability is improved, but system complexity increases
Solution Approach 1:
The patent implements a feedback mechanism where the estimated torque value is continuously compared with the actual torque sensor reading. The system uses weighted averaging that can adjust based on the agreement between estimated and measured values. This feedback loop provides redundancy and reliability by cross-validating the torque measurement, while maintaining relatively simple system architecture by using the same sensor inputs for both estimation and comparison.
Data Source
AI summary
A system and method of estimating the torque value of an engine are provided. The method includes generating corrected variable values from engine measured parameters and using a plurality of steady state tables that output corrected engine torque estimates based on various corrected variable values as inputs. In a preferred embodiment, the method also includes a phase compensation technique that converts each steady state table torque estimate into a dynamic torque estimate that closely matches the torque sensor measurements during both transient and steady state engine operations. In addition, also in a preferred embodiment, the method further includes a weighted averaging scheme that combines multiple torque estimates with weighting factors that are optimized based on the accuracy attributes of each torque estimate.