Virtual Flow Meter Kalman Filter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing flow meter technologies, both physical and virtual, face challenges such as high costs, reliability issues, power consumption problems, and accuracy errors in multiphase flow measurements, making them unsuitable for precise control in resource production contexts like oil and gas production.
Innovation Solution
A virtual flow meter algorithm that employs a Kalman filter framework to merge multiple sources of information, considering mass flow and pressure at every node, and uses pseudo-measurements to enhance estimation accuracy, allowing for recursive and computationally efficient implementation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical flow meters with complex sensors are used, then measurement capability is improved, but cost and reliability deteriorate
Solution Approach 1:
The patent replaces physical flow meters with complex mechanical/electronic sensors (microwave, electrical impedance, doppler ultrasound, gamma ray sensors) with a virtual flow metering system that uses a Kalman filter algorithm combining pressure and temperature measurements. This substitution eliminates the need for complex physical sensors while maintaining flow measurement capability through computational estimation.
Solution Approach 2:
The patent creates a virtual model (Kalman filter estimation) that copies the measurement function of physical flow meters without using their physical sensing components. The virtual flow meter algorithm processes pressure and temperature data to generate flow rate estimates, effectively creating a software-based copy of the measurement capability.
2Measurement precision
If physical flow meters with expensive sensors are used, then measurement capability is improved, but cost deteriorates
Solution Approach 1:
The patent replaces expensive physical flow meters with costly sensors (microwave, electrical impedance, doppler ultrasound, gamma ray sensors) with a computational algorithm (Kalman filter) that processes inexpensive pressure and temperature sensor data. This substitution dramatically reduces hardware costs while maintaining measurement capability.
Solution Approach 2:
The patent uses inexpensive pressure and temperature sensors instead of expensive flow meter sensors, accepting that these are simpler, cheaper components that can be easily replaced if needed, rather than investing in costly specialized flow measurement hardware.
3Device complexity
If virtual flow meters with simple sensors are used, then cost is reduced, but measurement accuracy deteriorates
Solution Approach 1:
The patent implements a Kalman filter algorithm that continuously processes pressure and temperature measurements with feedback loops, using prediction and update cycles to refine flow rate estimates. This feedback mechanism compensates for the simplicity of the sensors by dynamically adjusting estimates based on measured data patterns.
Solution Approach 2:
The patent makes the pressure and temperature sensors serve multiple functions: they not only monitor their primary parameters but also provide data for flow rate estimation through the Kalman filter algorithm. This multi-functionality extracts additional measurement capability from simple sensors through computational processing.
4Measurement precision
If complex data-fusion algorithms are used in virtual flow meters, then measurement capability is improved, but computational complexity and accuracy deteriorate
Solution Approach 1:
The patent replaces complex data-fusion algorithms with a Kalman filter implementation that, while computationally efficient, provides robust flow estimation. The Kalman filter's mathematical framework balances computational simplicity with estimation accuracy by using optimal filtering based on prediction and measurement update equations.
Data Source
Figure 1
Figure 2
AI summary
Approaches for a modeling and estimation approach for a virtual flow meter (VFM) are described. Certain aspects of the present virtual flow meter approaches relate to the manner in which multiple sources of information in the field are merged within a filter framework for estimation. In certain implementations, both mass flow and pressure at every node of the field are considered as part of the state estimated by the filter algorithm.