Dual-Fuel Engine Control System for Knocking Prevention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Dual-fuel engines face issues with abnormal combustion leading to knocking, which can damage the engine, and existing control systems are inadequate in predicting and preventing such events effectively.
Innovation Solution
A dual-fuel engine control system that utilizes a sensing unit, a main control unit, an EFI control unit, and a deep learning unit to analyze engine parameters and generate control signals for optimal fuel injection timing, predicting and preventing knocking by adjusting micro-pilot injection and gas fuel inlet valve operations in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If a dual-fuel engine uses diesel fuel as ignition source and natural gas for power generation, then fuel efficiency and emission reduction are improved, but the risk of abnormal combustion and knocking increases
Solution Approach 1:
The deep learning unit performs preliminary analysis of engine states and predicts potential knocking conditions before they occur. The system proactively adjusts fuel injection parameters and gas fuel inlet valve timing in advance to prevent abnormal combustion, rather than merely reacting after knocking is detected.
Solution Approach 2:
The sensing unit continuously monitors engine parameters and feeds this information back to the deep learning unit. The system uses this real-time feedback to dynamically adjust control signals for the micro-pilot injection valve and gas fuel inlet valve, creating a closed-loop control system that adapts to changing engine conditions to prevent knocking.
2Device complexity
If existing control systems are used for dual-fuel engines, then system simplicity is maintained, but the ability to predict and prevent knocking is insufficient
Solution Approach 1:
The deep learning unit acts as an intermediary between the sensing unit and the control valves. It processes complex sensor data and translates it into predictive insights about knocking risk, which then guides the adjustment of fuel injection and valve timing. This intermediary layer enables sophisticated knocking prevention without requiring complete redesign of the entire control system.
Solution Approach 2:
The system replaces traditional mechanical or rule-based control logic with a deep learning-based intelligent control system. The deep learning unit uses machine learning algorithms to analyze engine states and predict knocking conditions, substituting complex mechanical control mechanisms with software-based intelligent decision-making.
3Ease of operation
If fuel injection timing is not optimized according to load variation, then control simplicity is maintained, but engine performance and knocking prevention deteriorate
Solution Approach 1:
The control system dynamically adjusts fuel injection timing based on real-time load variations and engine conditions. The deep learning unit continuously processes sensor data and generates optimized control signals that adapt to changing operating conditions, transforming the static control system into a dynamic one that responds to load variations.
Solution Approach 2:
The system changes key operating parameters including fuel injection timing, micro-pilot injection value, and gas fuel inlet valve timing based on deep learning analysis of engine states. These parameter adjustments are made in real-time to optimize engine performance and prevent knocking under varying load conditions.
Data Source
AI summary
Provided is a dual-fuel engine control system including: a sensing unit for generating sensing information by sensing a parameter related to a dual-fuel engine; a main control unit for generating a control signal for controlling a micro-pilot injection value and a gas fuel inlet valve by analyzing a state of the dual-fuel engine on the basis of the sensing information; an EFI control unit for controlling the micro-pilot injection value and the gas fuel inlet valve on the basis of the control signal; and a deep learning unit for analyzing a state of the dual-fuel engine on the basis of the control signal transferred from the main control unit and the sensing information transferred to the main control unit at a time point of generating the control signal, and transferring an analysis result to the main control unit and the EFI control unit.