Detecting and quantifying liquid pools in hydrocarbon fluid pipelines
Pressure-inducing devices and pressure transducers can be used to detect and quantify liquid pools in hydrocarbon fluid pipelines. Pressure fluctuations can be detected by a pressure transducer, where the pressure fluctuations are the response of a pressure-inducing device outputting a pressure signal in a pipe carrying hydrocarbons. Variation in a pipe diameter caused by pooling or deposition can be estimated using an inverse model. The pooling or depositions can be classified by applying a machine-learning model to the pressure fluctuations. The variation in pipe diameter can be converted to an equivalent liquid volume for pooling locations. A pooling or deposition location and volume can be output and used for determining an action on the pipe to remove the pooling or deposition.
1. An assembly comprising:
a pipe for carrying hydrocarbon fluid;
a pressure-inducing device positioned in a portion of the pipe to output a pressure wave in an inner area of the portion of the pipe; and
a pressure transducer in fluid communication with the portion of the pipe to detect pressure fluctuations of the pressure wave and output a signal indicative of the pressure fluctuations to a computing device that determines (i) a pooling or deposition location and (ii) a volume in the portion of the pipe, the computing device including program code that is executable by a processing device for causing the processing device to:
estimate a localized variation in a diameter of the pipe caused by a pooling or deposition using an inverse model;
identify and classify the pooling or deposition by applying a machine-learning model to the pressure fluctuations; and
convert the localized variation in the diameter of the pipe to equivalent liquid volume for pooling locations.
2. The assembly of claim 1 , wherein the signal indicative of the pressure fluctuations is usable by the computing device to:
output the pooling or deposition location and volume that is usable for determining an action on the pipe to remove the pooling or deposition.
3. The assembly of claim 1 , wherein the inverse model is usable to estimate, by the computing device, the localized variation in the diameter of the pipe caused by pooling or deposition by:
generating an estimated value of the localized variation in the diameter of the pipe;
generating a simulated pressure profile using a forward model based on the pressure wave and estimated value of the localized variation in the diameter of the pipe;
determining an error value between the simulated pressure profile and the pressure fluctuations; and
updating the forward model with an updated estimated value of the localized variation in the diameter of the pipe and repeating the steps of generating the estimated value, generating the simulated pressure profile, and determining the error value until the error value is below a threshold value.
4. The assembly of claim 1 , wherein the pressure-inducing device comprises a valve, a fluid-injector, or a pump.
5. The assembly of claim 1 , wherein the pressure transducer is positioned proximate to the pressure-inducing device.
6. The assembly of claim 1 , wherein the pressure transducer is configured to detect the pressure fluctuations corresponding to an interaction of the pressure wave against a pooling liquid volume at the pooling or deposition location.
7. A non-transitory computer-readable medium including program code that is executable by a processing device for causing the processing device to:
receive a signal indicative of pressure fluctuations detected by a pressure transducer in a pipe for carrying hydrocarbon fluid in response to a pressure-inducing device of the pipe outputting a pressure signal;
estimate a localized variation in a diameter of the pipe caused by a pooling or deposition using an inverse model;
identify and classify the pooling or deposition by applying a machine-learning model to the pressure fluctuations;
convert the localized variation in the diameter of the pipe to equivalent liquid volume for pooling locations; and
output (i) a pooling or deposition location and (ii) a volume that are usable for determining an action on the pipe to remove the pooling or deposition.
8. The non-transitory computer-readable medium of claim 7 , wherein the non-transitory computer-readable medium includes program code that is executable by the processing device for causing the processing device to estimate the localized variation in the diameter of the pipe caused by the pooling or deposition using the inverse model by:
generating an estimated value of the localized variation in the diameter of the pipe;
generating a simulated pressure profile using a forward model based on the pressure signal and estimated value of the localized variation in the diameter of the pipe;
determining an error value between the simulated pressure profile and the pressure fluctuations; and
updating the forward model with an updated estimated value of the localized variation in the diameter of the pipe and repeating the steps of generating the estimated value, generating the simulated pressure profile, and determining the error value until the error value is below a threshold value.
9. The non-transitory computer-readable medium of claim 7 , wherein the non-transitory computer-readable medium includes program code that is executable by the processing device for causing the processing device to:
transmit a command to the pressure-inducing device to cause the pressure-inducing device to generate the pressure signal, wherein the pressure fluctuations detected by the pressure transducer are signal reflections of the pressure signal.
10. The non-transitory computer-readable medium of claim 7 , wherein the pressure transducer is located proximate to the pressure-inducing device.
11. The non-transitory computer-readable medium of claim 7 , wherein the non-transitory computer-readable medium includes program code that is executable by the processing device for causing the processing device to:
classify the pooling or deposition at the pooling or deposition location as a liquid or a solid.
12. The non-transitory computer-readable medium of claim 7 , wherein the pressure-inducing device comprises a valve, a fluid-injector, or a pump.
13. The non-transitory computer-readable medium of claim 7 , wherein the localized variation in the diameter of the pipe is measured as a distance between a surface of the pooling or deposition and a surface of an opposite end of the pipe.
14. A computer-implemented method comprising:
receiving pressure fluctuations detected by a pressure transducer in a pipe for carrying hydrocarbon fluid in response to a pressure-inducing device of the pipe outputting a pressure signal;
estimating a localized variation in a diameter of the pipe caused by pooling or deposition using an inverse model;
identifying and classifying the pooling or deposition by applying a machine-learning model to the pressure fluctuations;
converting the localized variation in the diameter of the pipe to equivalent liquid volume for pooling locations; and
outputting (i) a pooling or deposition location and (ii) outputting a volume that are usable for determining an action on the pipe to remove the pooling or deposition.
15. The computer-implemented method of claim 14 , further comprising:
transmitting a command to the pressure-inducing device to cause the pressure-inducing device to generate the pressure signal, wherein the pressure fluctuations detected by the pressure transducer are signal reflections of the pressure signal.
16. The computer-implemented method of claim 14 , wherein estimating the localized variation in the diameter of the pipe caused by the pooling or deposition using the inverse model further comprises:
generating an estimated value of the localized variation in the diameter of the pipe;
generating a simulated pressure profile using a forward model based on the pressure signal and estimated value of the localized variation in the diameter of the pipe;
determining an error value between the simulated pressure profile and the pressure fluctuations; and
updating the forward model with an updated estimated value of the localized variation in the diameter of the pipe and repeating the steps of generating the estimated value, generating the simulated pressure profile, and determining the error value until the error value is below a threshold value.
17. The computer-implemented method of claim 14 , wherein the pressure transducer is located proximate to the pressure-inducing device.
18. The computer-implemented method of claim 14 , wherein the localized variation in the diameter of the pipe is measureable as a distance between a surface of the pooling or deposition and a surface of an opposite end of the pipe.
19. The computer-implemented method of claim 14 , wherein classifying the pooling or deposition includes classifying the pooling or deposition as a liquid or a solid.
20. The computer-implemented method of claim 14 , wherein the pressure-inducing device comprises a valve, a fluid-injector, or a pump.