Virtual Calibration of Integrated Computational Elements
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The calibration of integrated computational elements (ICE) in optical computing devices is a time-consuming and costly process, requiring multiple calibration fluids, temperature, and pressure data points, which can take weeks to complete, especially in field operations where complex PVT equipment is often not readily available.
Innovation Solution
An agile light source is used to simulate the spectral responses of all calibration fluids at desired temperature, pressure, and gas-oil-ratio (GOR) points, allowing for virtual calibration without the need for extensive laboratory time, reducing the calibration time from weeks to hours and eliminating the requirement for complex PVT equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional calibration methods using multiple calibration fluids and PVT equipment are used, then measurement precision is improved, but loss of time increases significantly
Solution Approach 1:
The patent creates a virtual copy of the complex PVT calibration system by training an artificial neural network to replicate the calibration fluid responses. The trained ANN serves as a virtual calibration system that can be deployed in field devices, eliminating the need for physical PVT equipment and multiple calibration fluids while maintaining calibration accuracy.
Solution Approach 2:
The patent replaces the mechanical PVT equipment and physical calibration fluid system with a software-based artificial neural network. This substitution eliminates the need for physical equipment, reduces calibration time from weeks to hours, and enables field deployment without complex laboratory equipment.
2Measurement precision
If multiple calibration fluids and gas charges are used to improve calibration accuracy, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent creates a virtual copy of the complex PVT calibration system by training an artificial neural network to replicate the calibration fluid responses. The trained ANN serves as a virtual calibration system that can be deployed in field devices, eliminating the need for physical PVT equipment and multiple calibration fluids while maintaining calibration accuracy.
Solution Approach 2:
The patent develops a universal calibration approach where a single artificial neural network can handle multiple calibration scenarios (different fluids, temperatures, pressures, gas charges) that traditionally required separate calibration systems. This universal ANN model simplifies the calibration process while maintaining accuracy across diverse conditions.
3Measurement precision
If extensive calibration routines are performed to ensure accuracy, then measurement precision is improved, but productivity decreases
Solution Approach 1:
The patent creates a virtual copy of the complex PVT calibration system by training an artificial neural network to replicate the calibration fluid responses. The trained ANN serves as a virtual calibration system that can be deployed in field devices, eliminating the need for physical PVT equipment and multiple calibration fluids while maintaining calibration accuracy.
Solution Approach 2:
The patent performs the extensive calibration routine once during the training phase to create the artificial neural network model. This preliminary action captures all the calibration information needed, allowing rapid deployment and calibration of multiple sensors without repeating the time-consuming physical calibration process for each device.
4Measurement precision
If PVT equipment and multiple calibration fluids are used, then measurement precision is improved, but capital expenditures increase
Solution Approach 1:
The patent creates a virtual copy of the complex PVT calibration system by training an artificial neural network to replicate the calibration fluid responses. The trained ANN serves as a virtual calibration system that can be deployed in field devices, eliminating the need for physical PVT equipment and multiple calibration fluids while maintaining calibration accuracy.
Solution Approach 2:
The patent replaces expensive, complex PVT equipment with a software-based artificial neural network that can be deployed on inexpensive field devices. This substitution dramatically reduces capital expenditures while maintaining calibration accuracy, making the technology accessible for widespread deployment.
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
This approach significantly reduces calibration time and capital expenditures, enhances flexibility, and improves safety by minimizing the handling of hazardous materials, while ensuring standardized ICE sensor calibration across all devices.
Implementation Method 1
an agile light source may be programmed with each spectrum derived from each calibration fluid... The programmed agile light source may then be used to virtually calibrate additional sets of ICE
Data Source
AI summary
Disclosed are systems and methods for calibrating integrated computational elements. One method includes measuring with a spectrometer sample interacted light comprising spectral data derived from one or more calibration fluids at one or more calibration conditions, the one or more calibration fluids circulating in a measurement system, programming a virtual light source based on the spectral data, simulating the spectral data with the virtual light source and thereby generating simulated fluid spectra corresponding to the spectral data, conveying the simulated fluid spectra to the one or more ICE and thereby generating corresponding beams of optically interacted light, and calibrating the one or more ICE based on the corresponding beams of optically interacted light.