Injection Valve Timing Detection Using Shifted Sensor Signal Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing injection valves in internal combustion engines face challenges in accurately determining the opening and closing times due to structural differences and varying operating conditions, leading to uncertainties in fuel injection, which affects combustion efficiency and emissions.
Innovation Solution
A data-based time determining model is trained using a combination of labeled and unlabeled analysis point time series, employing shifting functions and loss functions to accurately predict opening and closing times, which is then refined through retraining in engine operation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a predetermined control curve is used to control the injection valve, then the control process is simple, but the actual opening and closing times cannot be determined accurately due to varying valve movements
Solution Approach 1:
The patent applies feedback by using a piezo sensor to detect actual pressure changes in the fuel supply caused by valve movements. The sensor signal is processed to determine actual opening and closing times, which are then fed back to adjust the control signal. This closed-loop feedback mechanism resolves the contradiction by maintaining simple control while achieving accurate timing determination through continuous measurement and adjustment.
Solution Approach 2:
The patent replaces direct mechanical measurement of valve position with a sensor-based electrical measurement system. Instead of mechanically tracking valve movement, a piezo sensor detects pressure changes in the fuel supply, which are then processed through signal analysis to determine timing. This substitution enables accurate measurement while maintaining control simplicity.
2Measurement precision
If a piezo sensor is used to monitor valve movement by detecting pressure changes, then opening and closing times can be determined, but the sensor signal is noisy and depends on fuel pressure variations
Solution Approach 1:
The patent extracts the relevant timing information from the noisy sensor signal by focusing analysis on specific features. The signal processing identifies characteristic points (opening and closing times) by detecting specific patterns in the pressure changes, effectively separating the useful timing information from the noisy background and fuel pressure variations.
Solution Approach 2:
The patent introduces an intermediate signal processing step between the piezo sensor and the timing determination. The sensor signal is processed through analysis that identifies characteristic features and patterns, acting as an intermediary that filters out noise and fuel pressure dependencies while preserving the actual valve timing information.
3Measurement precision
If training data is collected on a test rig to train the time determining model, then initial model training is possible, but the model needs retraining during engine operation to maintain accuracy
Solution Approach 1:
The patent applies preliminary action by collecting training data on a test rig before engine operation to initially train the time determining model. This pre-training provides a good starting point, and the system is designed to accept additional training data during operation to refine the model, combining advance preparation with ongoing optimization.
Solution Approach 2:
The patent makes the model training process dynamic by enabling retraining during engine operation. The system can continuously or periodically update the model with new data collected from actual operation, allowing the model to adapt to changing conditions and maintain accuracy throughout the engine's lifecycle rather than being static after initial training.
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
The model enhances the accuracy of fuel injection control, improving engine efficiency and reducing pollutant emissions by precisely determining the opening and closing times of the injection valve.
Implementation Method 1
a piezo sensor is provided in the injection valves and is designed as a pressure sensor in order to sense the pressure changes of a fuel pressure
Data Source
AI summary
A computer-implemented method for training a data-based time determining model for determining an opening or closing time of an injection valve based on a sensor signal. The method includes: providing an unlabeled analysis point time series by sampling the sensor signal of a sensor of the injection valve; training the data-based time determining model to assign a time specification which represents a specific opening or closing duration to an analysis point time series, the training process being carried out using a first shifting function to time-shift the analysis point time series and a second shifting function in order to time-shift the time specification. A consistency loss function is used for the training process.