Applying Particle Swarm Optimisation Technique to Calculate Rod Torque Limits for Wells and Preventing Impending PCP Rod Failures

Abstract

Abstract Most of the producing oil wells require some form of artificial lift for pumping production fluids to the surface (Bates, Cosad et al. 2004). The Progressing cavity pump (PCP) is considered as one of the most robust types of artificial lift method. About 10 to 15 percent of almost one million wells worldwide produce oil with the help of PCPs. PCPs are very efficient and can handle a range of viscosities, gas-liquid ratios, and solids production. Over the years, one of the most pressing concern of operators using PCPs in their Oil-fields have been, high workover costs and subsequent failures. If the operators do not take care of the PCP installation and take care of the PCP whilst start up, then the PCP efficiency gradually declines over time and the PCP fails at some point of time. This can be attributed to High Torque, Pump Deterioration, Parting of the Rods and other such failures. Due to the failure the financial costs are substantial accompanied by lost production, workover rig costs, replacement costs and intervention costs. To reduce failures operators are investing in real-time surveillance systems to monitor PCP performance using downhole measurements and raise alarms in case of abnormal events such as tripping's or failures. This is done by monitoring various parameters and then action is taken based on the operator's ability to discern the impending failure based on some signature trends. However, more often than usual at this stage it is already too late to do anything that could save the pump or the whole PCP system. Thus, it is important to move from a reactive to a proactive approach. This paper offers such an approach. Specifically, a analytical solution is proposed to proactively calculate and suggest a Torque value within which the PCP should work to deliver the required deliverability from the reservoir. statistical techniques for dimensionality. This analytical solution combines engineering principles with mathematical models to detect a suitable torque long before the PCP crosses this value in an actual well startup scenario. Supervisory control and data acquisition (SCADA) systems and data historians bring about an increased value by optimizing the work procedure or Workflow. Particle swarm optimization (PSO) is one such technique that can be used in conjunction with the workflow to monitor an important parameter such as Rod torque to enable the production engineer to calculate the most suitable hydraulic torque limit for the oil and gas well where a PCP (Progressive cavity pump) is installed. Thus, if a situation arises where the Torque limit is not calculated correctly, this implementation of this analytical solution which is implemented in Python programming language can prevent the wrong Torque value from being calculated and result in saving the PCP pump.