PRODUCTION SYSTEM OPTIMIZATION FOR SUBMERSIBLE PUMP LIFTED WELLS: A CASE STUDY
A computer program has been written to perform production optimization in submersible pump lifted wells. Production optimization was achieved by the principles of Nodal Analysis Technique which was applied between the reservoir and the wellhead ignoring the surface choke and separator. Computer program has been written according to two lifting environment, which are: pumping with only liquid and pumping with both liquid and gas. Program played an important role in the study by overcoming difficult iterations existing in the pumping liquid and gas case due to variation of liquid volume between pump intake and discharge pressure. Hagedorn and Brown vertical multiphase flow correlation was utilized in the program to determine the pressure at required depth. However, Griffith Correlation was also used in the program since Hagedorn and Brown Correlation failed to give accurate results at bubble flow.
A case study was done by evaluating the 10 wells located in Diyarbakır-GK field which are all submersible pump lifted. Well, reservoir, fluid and lift-system data was transferred to already written computer program. Output of the computer program for both cases was used to calculate accurately the optimum production rates, required horsepower, number of pump stages and the relation between these parameters with each other. The sensitivity variable selected is the number of pump stages. At the end of the study, by comparing the actual operating data and the computer-based optimized data, it was observed that 3 wells: W-16, W-17, and W-24 were producing completely within their optimum range, 5 wells: W-07, W-08, W-25, W-27 and W-28 were not producing at their optimum range but their production parameters can said to be acceptable , 1 well: W-22 was producing inefficiently and should be re-designed to reach optimum conditions. It was realized that W-15 has insufficient data to make necessary interpretations.
The electrical submersible pumping system can said to be an attractive artificial lift technique in reservoirs having high water-cut and low gas-oil ratio. Currently, it is considered as an effective and economical means of lifting large volumes of fluid from great depths under a variety of well conditions. Pumping equipment is capable of producing as high as 60,000 b/d and as low as 200 b/d. The oil cut may also vary within very wide limits, from negligible amounts to 100 %. The pump performs at highest efficiency when pumping liquid only; it can handle free gas with the liquid but high volumes of free gas causes inefficient operation and gas lock problems. The first submersible pumping unit was installed in an oil well in 1928 and since that time the concept has proven itself throughout the oil-producing world1. A submersible pumping unit consists of an electric motor, a seal section, an intake section, a multistage centrifugal pump, an electric cable, a surface installed switchboard, a junction box and transformers. Additional miscellaneous components also present in order to secure the cable alongside the tubing and wellhead supports. Pressure sentry for sensing bottom-hole pressure, check and bleeder valves are the optional equipment that can be taken into consideration. Under normal operating conditions, submersible pumping unit can be expected to give from 1 to 3 years of good operating life with some units operating over 10 years. Despite this advantage, many submersible pump lifted oil and gas wells produce at rates different than optimum.
This fact makes necessary to apply production optimization techniques to wells having low production rates. Nodal Analysis has been applied to artificial lift method for many years to analyze the performance of the systems composed of interacting components. It is a process of determining the effect of each component in the production system on the total system performance. The analysis can improve the completion design, well productivity and producing efficiency, all of which lead to increased profitability from oil and gas investments. The Nodal analysis technique is essentially a simulator of the producing well system. The system includes all flow between the reservoir and the separator. As the entire system is simulated, each of the components is modelled using various correlations or equations to determine the pressure loss through that component as a function of flow rate. The summation of these individual losses make up the total pressure loss through the entire system for a given flow rate. The production rate or deliverability of a well can be severely restricted by the poor performance of just one component in the system. If the effect of each component on the performance of the total system can be isolated, the efficiency of the system can be optimized in the most economical way. When performing a Nodal analysis, we divide the production system into its components, i.e., reservoir, perforations, tubing, surface choke, flowline and separator. Then we pick a problem area in this production system as a node. This node acts as the intersection point between the inflow and outflow performances. Different inflow and outflow performance curves intersect on the same plot and give the design considerations for different arrangements2. Optimization and design of submersible pump lifted wells pumping only liquid are generally straight-forward however pumping gas with the liquid is complicated because of the high compressibility of gas. In this case, volume of the produced fluid rate shows a significant variation between the pump intake and discharge pressures, consequently considerable amount of iterations should be performed to determine the volume factor at any pressure between the intake and discharge pressures. Thus, computer program should be written to overcome these iterations. Optimization of wells with Nodal Analysis requires pressure gradient correlation in order to reach a solution so it is necessary to use a vertical multiphase flow correlation method in the computer program. In this study, Hagedorn and Brown vertical multiphase flow correlation3 has been used to determine the pressure and pressure losses at required depth. However, during the study it was observed that Hagedorn and Brown Correlation failed to give accurate output at bubble flow. Thus, Griffith Correlation4 was constructed at bubble flow to obtain accurate results.
The purpose of this study was to write a general computer program that gives simultaneously the possible production rates for submersible pump lifted wells and also the optimum required horsepower and number of pump stages at these possible rates both considering pumping liquid and pumping gas with liquid. In addition to that objective, comparison made by using the production data of wells located in the GK field will assist us in suggesting optimum pump operating conditions.
Production optimization, nodal system analysis technique, electrical submersible pump, artificial lift, Hagedorn and Brown correlation, Griffith correlation