Drilling Rig Mud Balance Modeling for Kick and Loss Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for downhole fluid gain and loss detection in drilling operations are expensive, unreliable, and prone to false alarms due to delays and human error, particularly in pit-volume monitoring techniques, which struggle to differentiate between surface events and downhole fluid changes.
Innovation Solution
A method and system for monitoring and controlling mud flow in drilling rigs using real-time modeling and automatic recalibration of mud volume balance, combined with segmentation algorithms to identify and adjust for transient mud volumes and transfers between pits, enabling accurate detection of fluid gains or losses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If pit-volume monitoring is used for kick detection, then ease of deployment is improved, but detection speed and reliability deteriorate due to long delays between downhole events and pit level changes
Solution Approach 1:
The patent introduces an intermediary computational model that directly correlates pump operation parameters with expected pit volume changes. This model acts as a mediator between the pump control system and volume monitoring, enabling real-time detection of abnormal volume changes without waiting for delayed pit level measurements. The model predicts what the pit volume should be under normal conditions, allowing immediate identification of kicks or losses.
Solution Approach 2:
The patent replaces the mechanical/waiting-based pit level measurement approach with a computational model-based detection system. Instead of physically measuring pit volume changes that occur slowly, the system uses mathematical models to predict expected volume changes based on pump operations and compares these predictions with actual measurements in real-time, substituting mechanical delay with computational speed.
2Ease of operation
If pump stop is used as baseline reference for pit-volume monitoring, then detection simplicity is improved, but reliability deteriorates when flow conditions differ from baseline
Solution Approach 1:
The patent transforms the static baseline reference (pump stop) into a dynamic reference that adapts to changing flow conditions. The computational model continuously adjusts expected pit volume predictions based on current pump rate, hole depth, mud properties, and other dynamic parameters. This allows the system to maintain reliability across varying operational conditions rather than relying on a fixed baseline from pump stop conditions.
Solution Approach 2:
The patent changes multiple parameters to create an adaptive reference system. Instead of using a single fixed baseline, the model incorporates varying parameters including pump rate, hole depth, mud density, and annular geometry to dynamically calculate expected volume changes. This multi-parameter approach ensures detection accuracy maintains reliability regardless of flow conditions.
3Adaptability or versatility
If manual operator monitoring is used to distinguish surface events from downhole events, then detection flexibility is improved, but productivity deteriorates due to human error and reaction time delays
Solution Approach 1:
The patent implements a self-service automated detection system that independently distinguishes between surface events and downhole events without human intervention. The computational model automatically analyzes volume changes, correlates them with pump operations and known surface activities, and generates alerts only for genuine downhole events. This eliminates human reaction time delays and errors while maintaining the flexibility to handle various operational scenarios.
Solution Approach 2:
The patent incorporates feedback mechanisms where the system continuously monitors actual pit volume changes against model predictions, and when discrepancies indicate potential kicks or losses, it generates automated alerts. The system learns from operational patterns and refines its ability to distinguish surface events from downhole events through continuous feedback from actual drilling operations, improving both accuracy and response efficiency.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for monitoring and controlling a mud flow system in a drilling rig includes measuring an active mud volume in an active mud pit and an inactive mud volume in an inactive mud pit, modeling a modeled active mud volume in the active mud pit, determining a mud volume balance by calculating a difference between the measurement of the active mud volume and the modeled active mud volume, detecting a transfer of mud from the inactive mud pit to the active mud pit based on a combination of a change in the measurement of the inactive mud volume in the inactive mud pit and a change in the mud volume balance, and detecting downhole gains and losses automatically based on the mud volume balance.