Turbomachine Process Quantity Estimation Using Error-Weighted Data Subsets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurately estimating process quantities in turbomachines, such as mass flow and efficiency, is challenging due to the need for complex and expensive measurement equipment, and existing methods may compromise accuracy based on assumptions that are not always valid, especially when measuring conditions vary.
Innovation Solution
A method that calculates estimates by weighting sub-sets of data values based on error estimates, where more weight is given to sub-sets with smaller error estimates, using a scalar product of sensitivity and variance vectors, and employing techniques like Kalman filters to minimize errors and adapt to varying conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex measurement equipment is used to measure process quantities like mass flow with sufficient accuracy, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces complex mechanical measurement equipment with a computational system that calculates mass flow and other process quantities using mathematical models and readily available sensor data. The processor computes process quantities based on measured values from simple sensors and pre-stored reference data, eliminating the need for complex direct measurement devices.
2Device complexity
If assumptions are made in calculation methods to simplify the estimation process, then device complexity is reduced, but measurement precision and reliability deteriorate when assumptions are not valid
Solution Approach 1:
The patent implements a dynamic selection mechanism that automatically chooses the most appropriate calculation method based on current operating conditions. The system evaluates multiple calculation approaches and selects the one with the smallest estimated error, adapting to varying suction conditions and gas compositions without requiring fixed assumptions.
Solution Approach 2:
The patent changes the parameter being optimized from simplicity to accuracy by selecting calculation methods based on their estimated error margins. The system calculates error estimates for different methods and dynamically adjusts which method is used, ensuring highest possible accuracy under current conditions.
3Measurement precision
If multiple data values are used to calculate estimates, then measurement precision can be improved through error minimization, but device complexity increases due to additional processing requirements
Solution Approach 1:
The patent implements a self-optimizing system that automatically evaluates multiple calculation methods and selects the best one without external intervention. The processor autonomously calculates error estimates, compares different approaches, and chooses the method with the smallest error, making the system self-adjusting and reducing the need for complex external control mechanisms.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for obtaining an estimate for a process quantity of a turbomachine comprises measuring (301) first process quantities and calculating (302) an estimate for a second process quantity, for example mass flow, on the basis of the measured first process quantities and a reference table of interrelations between process quantities prevailing in reference inlet conditions corresponding to a pre- defined inlet fluid temperature, a pre-defined inlet fluid pressure, and pre-defined fluid properties at the inlet. The reference table and the measured first process quantities represent such a set of data values that the estimate is obtainable on the basis of different sub-sets of the data values. Hence, the estimation task is over-posed in the sense that there is more data than necessary for obtaining the estimate. In the calculation of the estimate, more weight is given to such a part of the data which has a smaller error contribution.