Injection Valve Timing Model Training via Difficulty Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in accurately determining the opening and closing points in time of injection valves in internal combustion engines due to complex dependencies on instantaneous operating conditions, leading to uncertainty in fuel injection timing and efficiency.
Innovation Solution
A method for training a data-based point in time determination model using sensor signals from piezo sensors, where training data sets are classified by difficulty and augmented to improve model reliability, allowing for precise adaptation of injection valve activation based on pressure changes and fuel pressure variations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If training data is collected from measurements on a test stand, then the training data reflects real operating conditions, but the determination accuracy of opening and closing points in time is limited by uncertainty in instantaneous operating conditions
Solution Approach 1:
The patent applies preliminary action by classifying training data into difficulty classes before training the model. Training data sets are pre-processed and organized by difficulty level, with easier cases trained first to establish baseline performance, then progressively harder cases are incorporated. This staged approach allows the model to build reliable predictions systematically, improving overall timing determination accuracy while maintaining model reliability across varying operating conditions.
2Measurement precision
If difficulty-based classification of training data is implemented, then model training becomes more systematic, but the complexity of the training process increases
Solution Approach 1:
The patent applies segmentation by dividing the training data into multiple difficulty classes based on the consistency of time indications. Each difficulty class represents a segment of the overall training task, allowing the model to learn from simpler cases first and progressively handle more complex scenarios. This segmentation transforms a single complex training problem into manageable segments, improving timing determination accuracy while organizing the training process into structured stages.
Solution Approach 2:
The patent applies dynamics by making the training process adaptive rather than static. The difficulty classification creates a dynamic training sequence where the model progresses from easier to harder cases based on performance. This dynamic approach allows the training process to adjust automatically, focusing computational resources on difficult cases that benefit most from additional training, thereby improving accuracy without requiring uniformly complex training procedures throughout.
3Adaptability or versatility
If new training data sets are ascertained based on difficulty classes, then coverage of operating conditions is improved, but the time and resources required for data processing increase
Solution Approach 1:
The patent applies partial action by selectively processing and augmenting training data based on difficulty classes. Rather than uniformly processing all training data with the same level of augmentation, the method applies data augmentation techniques preferentially to difficult cases where they provide the most benefit. This selective approach improves coverage of operating conditions and model adaptability while avoiding unnecessary processing time spent on easier cases that require less enhancement.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the accuracy and reliability of determining opening and closing points in time, improving fuel injection precision and engine efficiency by adapting the activation of injection valves according to real-time conditions.
Implementation Method 1
sensor signals from piezo sensors, where training data sets are classified by difficulty and augmented to improve model reliability, allowing for precise adaptation of injection valve activation based on pressure changes and fuel pressure variations
Data Source
AI summary
A computer-implemented method for training a data-based point in time determination model for ascertaining an opening or closing point in time of an injection valve of an internal combustion engine, based on a sensor signal. The method includes:providing a set of training data sets from a measurement of the internal combustion engine by scanning the sensor signal of a sensor of the injection valve on a test stand, the training data sets assigning a time indication of the opening or closing point in time to an evaluation point time series; assigning a difficulty value to each training data set; classifying the training data sets into a number of difficulty classes corresponding to their respective difficulty value; ascertaining new training data sets as a function of the training data sets assigned to each difficulty class; training the model with the set of new training data sets.