Fuel Pressure Estimation Using Pump and Flow Variables
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current engine systems require a pressure sensor in the fuel supply pipe to detect fuel pressure, which increases component count and cost; a method to estimate fuel pressure without a pressure sensor is needed.
Innovation Solution
A fuel pressure estimation system that uses a storage device with mappings to estimate fuel pressure based on input variables such as pump state, consumption flow rate, and fuel properties, allowing the system to calculate fuel pressure without a pressure sensor, utilizing machine learning or experimental data to establish relationships between these variables.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a pressure sensor is provided in the supply pipe to detect fuel pressure, then measurement precision of fuel pressure is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates a virtual copy of the pressure sensor function through software-based estimation. Instead of using a physical pressure sensor, the system uses a pressure estimation device that calculates fuel pressure based on pump current, pump rotation speed, and fuel consumption flow rate. This software-based copy replaces the hardware sensor, reducing component count while maintaining measurement capability.
Solution Approach 2:
The patent replaces the mechanical/electrical pressure sensor system with a computational estimation system. The pressure estimation device uses processing units to calculate fuel pressure from electrical parameters (pump current, rotation speed) and flow rate data, substituting the direct mechanical measurement approach with an indirect computational method.
2Measurement precision
If a pressure sensor is provided in the supply pipe to detect fuel pressure, then measurement precision of fuel pressure is improved, but cost increases
Solution Approach 1:
The patent creates a virtual copy of the pressure sensor function through software-based estimation. Instead of using a physical pressure sensor, the system uses a pressure estimation device that calculates fuel pressure based on pump current, pump rotation speed, and fuel consumption flow rate. This software-based copy replaces the hardware sensor, reducing component count while maintaining measurement capability.
Solution Approach 2:
The patent uses inexpensive electrical parameters (pump current, rotation speed) and flow rate data that are already available in the fuel injection system to estimate pressure. These parameters are essentially free to measure since they are part of the normal operation data, making the estimation method much cheaper than installing dedicated pressure sensing hardware.
3Adaptability or versatility
If machine learning is used to determine mappings, then adaptability of fuel pressure estimation is improved, but device complexity increases
Solution Approach 1:
The patent uses machine learning to automatically determine the optimal mapping relationships between input parameters (pump current, rotation speed, flow rate) and fuel pressure. The learning unit trains the system using actual pressure sensor data, allowing the estimation algorithm to adapt to specific system characteristics and operating conditions. This automatic parameter optimization improves accuracy without requiring manual calibration or complex manual setup.
Data Source
Figure 1
Figure 2~4
Figure 5~6
AI summary
A fuel pressure estimation system (70) estimates a fuel pressure variable on a fuel pressure in a supply pipe (53, 58) for an engine apparatus including an engine (12) having a fuel injection valve (25, 26), and a fuel supply device (50) having a fuel pump (52) that supplies the fuel in a fuel tank (51) to the supply pipe (53, 58), and includes a storage device (74) and an execution device (71). The storage device (74) stores a first mapping that receives, as an input, first input variables including a pump variable, a consumption flow rate variable on a consumption flow rate of the fuel, and a property variable on a property of the fuel, and outputs the fuel pressure variable. The execution device (71) acquires the first input variables and estimates the fuel pressure variable by applying the first input variables to the first mapping.