Online Monitoring and Analysis of Water Quality in Offshore Oil & Gas Production

70% of the world’s oil production derives from matured hydrocarbon fields. Water is together with oil trapped in the reservoirs and becomes a byproduct of the oil production, which steadily increases over time. In the Danish sector alone, the oil to water ratio (water cut) reached 80% in 2015, and the required amount of energy to treat the volumes of produced water before ocean discharge to comply with legislation increases. Another solution is to reinject the produced
water into the reservoir to gain more sustainable production and comply with discharge legislation. Produced water reinjection has gained increasing attention in the last decades as it can increase yield and reduce oil discharge. However, the presence of solid particles, crude oil, and chemicals in the produced water, when reinjecting into the reservoir, introduces other complications in the injection water treatment process and the reservoir. Ideally, injection water should be sterile, non-scaling, free of suspended solids and oil to prevent plugging the reservoir rock’s pores and reduce permeability. Furthermore, corrosion due to the presence of dissolved gases and microbially influenced corrosion bacteria should be prevented. A treatment process for achieving these ideal conditions is challenged to yield economic payoffs. The current solutions for measuring water quality are based on manual sampling, which suffers from being time-consuming and may not suffice to retain a high and consistent injection water quality. Despite the long history of implementing online quality monitors for measuring particle and oil droplet concentrations and sizes, there is no consensus regarding which method provides the most reliable estimate leaving petroleum engineers with an array of choices.
An extensive review study has been carried out of how different water quality issues can add to the suspended solids concentration that can potentially cause injectivity decline in the injection water treatment facility or, even worse,
cause formation damage. Furthermore, the complexity of selecting quality monitors that reliably measure the concentration and sizes of suspended solids in the process has been heavily reviewed. Through the review process, online
microscopy analyzers were deemed the best candidates for measuring single-particle properties within an injection water treatment process. Two different microscopy analyzers for measuring suspended particles have been validated, including four identical fluorescence-based monitors for measuring oil-in-water concentrations. For each monitor type, a thorough calibration validation has been proposed to investigate their accuracy and reliability. The four identical fluorescence-based monitors for measuring oil-in-water concentrations revealed that the calibration procedure using a weighted least square yields higher reproducibility compared to ordinary least square due to the heteroskedastic behavior. The experimental validation results of the four fluorescence-based monitors showed a high precision between each other. However, due to its high sensitivity to fluorescent substances (fluorophores), they will be challenged in a dynamic separation facility with continuous changes of the fluorophores. The two different types of microscopy analyzers showed promising results in estimating known particle sizes and were able to measure oil-in-water concentrations with high precision in steady-state. A trailing moving average window of 1min was proposed resulting as a feasible solution for continuous measurements. It is acknowledged that an issue will arrive at shallow concentrations as the required number of oil droplets will not be captured within the expected relative error to represent the size distribution. Additionally, the classification procedure for both microscopy analyzers needs further validation.
Lastly, a connection between an industrial human-machine interface software and an academic used software were established to reduce the gap between academic research and industrial implementation. The connection was successfully
established, and a case study of a model-predictive control method in the industrial used software was presented. The industrial human-machine interface software connection is a beneficial tool when, i.e., the presentation of water
quality should be furtherly investigated.