Throttle Valve Clogging Learning Using Least-Squares Approximation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional control apparatuses for internal combustion engines face high computational load and memory capacity issues when learning the degree of throttle valve clogging, which affects fuel efficiency and energy efficiency.
Innovation Solution
A control apparatus that calculates the clogging parameter using a first and second flow rate function, with sample points averaged for each position range, and coefficients of an approximate polynomial equation calculated by the least-squares method, reducing computational load and memory requirements while accurately learning throttle valve clogging over a wide range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all estimated KTHs calculated at each predetermined time during operation of the internal combustion engine are stored as sample points, then the accuracy of throttle valve clogging learning is improved, but the memory capacity required becomes enormous
Solution Approach 1:
The patent extracts only the necessary information from the large set of sample points by calculating representative points that characterize the throttle valve clogging state. Instead of storing all estimated KTH values, the system extracts representative characteristics through least-squares method calculations, retaining only the essential data needed for accurate clogging learning while discarding redundant information.
Solution Approach 2:
The patent transforms the raw sample point data into a different parameter representation through mathematical processing. By applying the least-squares method to calculate representative points and update the approximate function coefficients, the system changes the data parameters from raw KTH values to processed coefficients that equally represent the clogging state with much smaller memory requirements.
2Measurement precision
If the least-squares method is applied to a large number of stored sample points to calculate representative points, then the accuracy of throttle valve clogging learning is improved, but the computational load increases
Solution Approach 1:
The patent segments the computational process into two distinct phases: an offline learning phase where the least-squares method is applied to calculate representative points and update coefficients, and an online operation phase where the pre-calculated approximate function is used for rapid clogging rate calculation. This segmentation allows complex calculations to be performed only when necessary, reducing overall computational load during engine operation.
Solution Approach 2:
The patent performs preliminary calculations of the approximate function coefficients during the learning phase before actual engine operation requires clogging rate calculations. By pre-calculating and storing the coefficients a, b, and c of the approximate function, the system eliminates the need to perform computationally intensive least-squares calculations during real-time operation, significantly reducing computational load.
3Measurement precision
If representative points are calculated for each position range of the throttle valve using the least-squares method, then the accuracy of clogging learning over wide position range is improved, but the computational complexity increases
Solution Approach 1:
The patent segments the throttle valve position range into multiple predetermined position ranges and calculates representative points separately for each range. This segmentation allows the system to capture the non-linear characteristics of throttle valve clogging at different positions more accurately, while the modular structure of calculating representative points for each range independently makes the overall process more manageable and less computationally complex than a single global calculation.
Data Source
AI summary
In the present invention, a clogging rate is calculated using a first flow rate function when the degree of clogging of the throttle valve is in the reference state and a second flow rate function estimated based on the intake air volume. For each predetermined period, sample points, which are combinations of the second flow rate function and throttle valve position, are obtained, and the learning points are calculated by averaging a plurality of sample points for each predetermined position range. Based on the plurality of learning points, coefficients a to c of the approximate function of the second flow rate function characteristic are calculated by the least-squares method, and the clogging rate is calculated based on the second flow rate function characteristic and the first flow rate function approximated by the approximate function using the coefficients a to c.