Challenges to Autonomous and Continuous Fluid Level Sensing

Size: px

Start display at page:

Download "Challenges to Autonomous and Continuous Fluid Level Sensing"

Jody Lester
5 years ago
Views:

1 10 th Annual Sucker Rod Pumping Workshop Renaissance Hotel Oklahoma City, Oklahoma September 16-19, 2014 Challenges to Autonomous and Continuous Fluid Level Sensing Dr. Duncan Earl 1, Michael Smith 2 1 Reservoir Management Services, 2 NOV

2 Need For Continuous Monitoring Real-time / accurate measurement of fluid level is critical feedback for automatic and optimal control of a well More Data = More Certainty Big data. Data mining of automation data is already being leveraged to improve how wells and fields are monitored / evaluated. Many indirect measurements (such as well noise and casing gas density) are measured in determining fluid depth. Continuous monitoring of these variables can be valuable indicators of well performance. Sept , Sucker Rod Pumping Workshop 2

Surface-based Fluid Level Monitoring Long history of using reflected gas pulses (time-domain reflectometry) to sense fluid surface Fluid Distance = ½ x [Round Trip Time] x [Speed of Sound] Convenient

3 Surface-based Fluid Level Monitoring Long history of using reflected gas pulses (time-domain reflectometry) to sense fluid surface Fluid Distance = ½ x [Round Trip Time] x [Speed of Sound] Convenient solution as it does not require down-hole equipment and is easily installed/maintained at surface Echometer s snapshot measurements widely used and accepted in the industry Capturing reflected signals with Echometer product Sept , Sucker Rod Pumping Workshop 3

measurements Too expensive Most techniques based on well sounding (explosion, implosion, etc.

4 Continuous/Autonomous Surface-based FLM Many attempts at continuous, autonomous fluid level monitoring (FLM), but no commercial traction Common complaints are: Does not work on majority of wells Complex installation/software Short Mean-Time-Between-Failures Erroneous measurements Too expensive Most techniques based on well sounding (explosion, implosion, etc.) Well Sonic Surveyor III Hytech Gen II Fluid Level Monitor Why is continuous/autonomous fluid sensing so difficult? TruDepth Instrument Module Sept , Sucker Rod Pumping Workshop 4

5 Three Major Challenges Three biggest challenges to continuous/autonomous surface-based fluid level sensing include: 1. Noise management 2. Signal loss 3. Autonomous signal processing Sept , Sucker Rod Pumping Workshop 5

6 Noise Management Noise is caused by mechanical vibrations, gas flow, seismic/tube waves, jewelry in the well, and electrical interference. Noise can often look like signal Every well s noise is different! Extremely difficult to apply a standard set of classical filters (i.e. high pass/low pass) to isolate reflected peak. Well burping is a good example of how noise can look like signal. Sept , Sucker Rod Pumping Workshop 6

Signal Loss A pressure pulse will reduce in intensity when it s kinetic energy is converted into heat (through inelastic collisions) Air flow, perforations, and foam are the largest killers of signal

7 Signal Loss A pressure pulse will reduce in intensity when it s kinetic energy is converted into heat (through inelastic collisions) Air flow, perforations, and foam are the largest killers of signal These factors can be extremely variable (might have a good signal one moment and no signal the next) Many wells have these features Interestingly, depth is only a minor factor in signal loss. Pushing a pressure pulse against gas flow is like barking into the wind Sept , Sucker Rod Pumping Workshop 7

Autonomous Signal Processing If Google can make a self driving car, shouldn t we be able to automatically process a reflected waveform? Yes, but the algorithms are still sophisticated.

8 Autonomous Signal Processing If Google can make a self driving car, shouldn t we be able to automatically process a reflected waveform? Yes, but the algorithms are still sophisticated. They constitute a difficult pattern recognition problem. This problem requires CS specialists, DSP specialists, and firmware developers to solve (which many companies do not have access to) A balance must be struck between computational power, cost, processing time, error rate, and measurement yield. Google s Self-Driving Car Which peak is the fluid surface? Sept , Sucker Rod Pumping Workshop 8

A new solution Over the past 6 months, we have been testing a new fluid level sensor design System incorporate four important improvements: A double pulse Decision-making based on

9 A new solution Over the past 6 months, we have been testing a new fluid level sensor design System incorporate four important improvements: A double pulse Decision-making based on signal-to-noise ratio Rule-based peak detection Third party integration MaxPro5000 Fluid Level Monitor from Reservoir Management Services Sept , Sucker Rod Pumping Workshop 9

10 Noise Management (D-PER) Two narrow pressure pulses emitted 300 msec apart. Double pulse provides a fingerprint for reflected pressure pulse D-PER (Double Pulse Echo Reflectometry) filter significantly reduces noise and pulls out signal Greatly reduces the probability that random noise spikes will be confused for signal Unit attached to well. Sept , Sucker Rod Pumping Workshop 10

11 D-PER Noise Filter (Example) Single peak, due to gas bubble, has been eliminated. Sept , Sucker Rod Pumping Workshop 11

12 Signal Loss (SNR qualifier) Internal compressor compresses 500 in 3 STP of casing gas to 200 PSI above casing Quality of reflected signal is measured relative to noise (signal-to-noise ratio) Mathematically, SNR > 10 is a real peak A test involving 100 different wells showed that a little added buffer (SNR > 30) is needed to ensure 99% measurement reliability Signal-To-Noise Ratio (SNR) = Signal / Noise Sept , Sucker Rod Pumping Workshop 12

13 Signal-to-Noise Ratio (SNR) SNR allows us to quantify the quality of a signal. Experimentally, we found that an SNR of > 30 produced long-term reliable measurements. Sept , Sucker Rod Pumping Workshop 13

14 Autonomous Signal Processing Standardized, rule-based, peak detection and decision making algorithm allows internal processor to reliably detect peak Rule-based Algorithm: Rule 1: Apply standardized low pass, high pass, and D-PER filter to raw waveform. Rule 2: Calculate the SNR of the four largest peaks. Eliminate any peaks below an SNR of 30. Rule 3: With remaining peaks, eliminate those that are an integer multiple of previous peaks (if, and only if, temporally later peak is smaller than earlier peak) Rule 4: With remaining peaks, temporally latest peak is fluid surface Sept , Sucker Rod Pumping Workshop 14

15 Peak Detection Algorithm (Example) Sept , Sucker Rod Pumping Workshop 15

16 Integration with 3 rd Parties Can be integrated with: Well managers External sensors Motor driven valves SCADA systems Greatly enhances potential to eliminate issues and better understand well (i.e. automatic build up tests, foam suppression tests, close-in casing to reduce noise, etc.) Feature Fluid Level Monitor Proven Technology Yes Equipment Down-Hole No Accuracy ±2.5 ft Output 4-20mA Maintenance Activity Only at surface Max Depth Fluid Level 10,000+ ft Measurement Rate Every 10 minutes Installation Time 30 minutes Certification Class I Div 2, CSA Price $15,350 Sept , Sucker Rod Pumping Workshop 16

17 Recent Data Currently installed on wells in Bakersfield and Wyoming. Have successfully controlled wells using NOV Guardian II and MaxPro5000 FLM Gen II product was tested over 1,000,000 shots. Gen III is just getting started ( 10,000 shots) Have measured fluid levels from 400 ft. to 9500 ft. Wells with high gas flow (>40 MCF) are difficult. Usually requires back casing valve to be closed. Works equally well on ESP, PCP, and Rod Beam pumps (though noise levels are quite different) Sept , Sucker Rod Pumping Workshop 17

18 Recent Data (Example) Sept , Sucker Rod Pumping Workshop 18

19 Upcoming Testing and Development We recognize that we are still between prototype and product. Goal over next 6-months is to quantify performance of system (accuracy, depth, and MTBF) Answer quantitatively On what wells does it work, how well does it perform, how much service does it need, and what can I do with it? Testing a new NACE-Compliant Compressor Working with 3 rd parties for device integration and testing Limited number of installations in Canada, Bakersfield, and Texas planned this year Looking for assistance and partners in proving this technology Sept , Sucker Rod Pumping Workshop 19

20 Copyright Rights to this presentation are owned by the company(ies) and/or author(s) listed on the title page. By submitting this presentation to the Sucker Rod Pumping Workshop, they grant to the Workshop, the Artificial Lift Research and Development Council (ALRDC), and the Southwestern Petroleum Short Course (SWPSC), rights to: Display the presentation at the Workshop. Place it on the web site, with access to the site to be as directed by the Workshop Steering Committee. Place it on a CD for distribution and/or sale as directed by the Workshop Steering Committee. Other use of this presentation is prohibited without the expressed written permission of the author(s). The owner company(ies) and/or author(s) may publish this material in other journals or magazines if they refer to the Sucker Rod Pumping Workshop where it was first presented. Sept , Sucker Rod Pumping Workshop 20

21 Disclaimer The following disclaimer shall be included as the last page of a Technical Presentation or Continuing Education Course. A similar disclaimer is included on the front page of the Sucker Rod Pumping Web Site. The Artificial Lift Research and Development Council and its officers and trustees, and the Sucker Rod Pumping Workshop Steering Committee members, and their supporting organizations and companies (here-in-after referred to as the Sponsoring Organizations), and the author(s) of this Technical Presentation or Continuing Education Training Course and their company(ies), provide this presentation and/or training material at the Sucker Rod Pumping Workshop "as is" without any warranty of any kind, express or implied, as to the accuracy of the information or the products or services referred to by any presenter (in so far as such warranties may be excluded under any relevant law) and these members and their companies will not be liable for unlawful actions and any losses or damage that may result from use of any presentation as a consequence of any inaccuracies in, or any omission from, the information which therein may be contained. The views, opinions, and conclusions expressed in these presentations and/or training materials are those of the author and not necessarily those of the Sponsoring Organizations. The author is solely responsible for the content of the materials. The Sponsoring Organizations cannot and do not warrant the accuracy of these documents beyond the source documents, although we do make every attempt to work from authoritative sources. The Sponsoring Organizations provide these presentations and/or training materials as a service. The Sponsoring Organizations make no representations or warranties, express or implied, with respect to the presentations and/or training materials, or any part thereof, including any warrantees of title, non-infringement of copyright or patent rights of others, merchantability, or fitness or suitability for any purpose. Sept , Sucker Rod Pumping Workshop 21

Automation Breakout Session

14 th Annual Sucker Rod Pumping Workshop Renaissance Hotel Oklahoma City, Oklahoma September 12 13, 2017 Automation Breakout Session Anthony Allison Artificial Lift Advisor Sucker Rod Pumping Occidental