Abstract
A blowout preventer (BOP) is critical safety equipment in drilling operations, designed to prevent the uncontrolled flow of liquids and gases. BOP systems require inspection and testing at least every 14 days, with a pressure test conducted to ensure their condition. This test currently involves manually decoding pressure chart data from a mechanical pressure chart recorder, a time-consuming process. Our solution automates this by capturing the image of the pressure chart and decoding it using AI image analysis.
The application integrates with an existing BOP Inspection application. The drilling supervisor takes a picture of the pressure chart, which is then processed by an image processing algorithm. The solution includes techniques like grid and trace splitting, color detection, trace pixel classification, template detection, text extraction and recognition, template matching, outlier removal and pressure value analysis.
This automation saves an average of 90 minutes per test for the drilling supervisor and improves the accuracy of pressure test reporting. Applied across ADNOC Onshore rigs, it has shown an estimated 300% ROI. The extracted data is exported in JSON format for integration with existing applications.
This paper presents a novel application of AI in computer vision for automating this pressure chart data extraction, demonstrating significant efficiency gains and a high ROI, adding value to the petroleum industry's body of knowledge.
