A Study of the Ultrasonic Measurements for Logging Behind a Steel Pipe: Expanding the proessing to improve the differentiation between heavy fluids and light solids
MetadataVis full innførsel
In petroleum applications using acoustic wave theory, the most used practice can be found in borehole geophysical methods, where acoustic logging has long been the foremost source for cement evaluation. Due to stricter regulatory requirements and more complex construction designs over the last few decades, the hardware and interpretation techniques have evolved to counterbalance the needs. When the oil well is drilled and the casing (steel pipe) is lowered down into the hole to prevent the leakage of fluids from the surrounding formation, cement needs to be squeezed behind the casing and the formation to seal off this zone (the annulus). The condition of the well barrier is a serious safety concern, since hydrocarbons from the reservoir can migrate into the annulus and up to the surface. To highlight this, the sonic method, often referred to as the cement bond log (CBL), was introduced in 1963 for investigating the cement/formation behind the casing. The sonic tool consists of a transducer and two receivers, where the receivers are distanced 3 ft. and 5 ft. away from the transducer. A pressure wave is transmitted with oblique incidence at the casing. By measuring the time arrival and dampening of the wave, and by comparing the waveform between the two receivers, one can evaluate the area outside the casing. Since the transducer operates at sonic frequencies (1-20 kHz), it can investigate a relatively long distance, but with the consequence of poor resolution. Therefore, to increase the resolution, the ultrasonic borehole logging was introduced, and today, it is an essential component of wireline logging in the field. The first ultrasonic technique was the pulseecho method developed in 1979. This method uses a transreceiver, where a pressure wave is transmitted with normal incidence at the casing and measures the reflected waveform. It has its main strength in estimating the casing ovality (casing surface), the casing thickness and the acoustic impedance of the material outside the casing. Studies that are more recent have shown that it can be used to calculate the thickness of a possible fluid-filled annulus between the casing and the formation. However, this is difficult and requires very precise measurements and complex processing. As the industry desired new methods to verify the quality of the cement bonding, the ultrasonic pitch-catch technique (commonly referred to as the flexural attenuation technique) was presented in 2003. The tool was introduced as a supplement to the traditional CBL and the ultrasonic pulse-echo method, but in recent years, it has become necessary when the cement job is critical. The ultrasonic pitch-catch technique can resemble the CBL but has a much higher resolution and, therefore, the ability to log in the azimuthal direction. This is done by exciting a guided wave propagating upwards within the casing. The guided wave leaks off energy as it propagates, and the leaked off pressure wavefront is measured at receivers positioned further up the well. The high resolution enables the log interpreters to detect if there is a fluid-filled annulus between the casing and the formation, or if there is a channel in the cement so that fluid can migrate upwards to the surface. The thesis consists of several, individual manuscripts where a laboratory setup is built to perform ultrasonic measurements that resemble logging in the field. The pulse-echo method and the pitch-catch technique were tested experimentally, where the first task was to study the wave propagation and the different modes excited in the steel pipe. Furthermore, a model was established to quantify how different materials behind the steel pipe affect the ultrasonic measurements. From this research, one could estimate if the accuracy is sufficient to conclude what the annulus consists of. In addition, numerical models were built to resemble the laboratory setup. The numerical data is used to verify the results from the laboratory and to investigate the magnitude of the parameters affecting the measurements. The final study in this thesis was an expansion of the processing of the flexural wave in the frequency domain to improve differentiation in critical scenarios. The results from the laboratory measurements and numerical simulations show that one can distinguish materials behind the casing. However, if the acoustic impedance of the annulus material is similar (±0.5 MRayls), the conventional processing struggles to separate the two since the difference is too small compared to the accuracy in the data. To distinguish the different materials, instead of estimating the acoustic impedance from the attenuation, one can separate based on the P-wave velocity. In the time domain, one can observe a modification of the flexural pulse, where the pulse is smeared out as the P-wave velocity of the annulus (V a,P) exceeds the casing flexural phase velocity (V Ao, Ƥ). The modification of the pulse is believed to happen in the transition of the wave in the annulus from a propagating wave (V a,P) < (V Ao, Ƥ) to an evanescent wave (V a,P) < (V Ao, Ƥ). In the frequency domain, one can observe a perturbation in the spectrum, and it is believed that the frequency at which this occurs is related to velocity where V a,P = V Ao, Ƥ. By plotting an analytical solution of the casing flexural phase velocity, the speed in the dispersion curve at this frequency is believed to be the same as the P-wave velocity of the annulus material.
Består avTore Sirevaag, Tonni Franke Johansen, Idar Larsen and Rune Martin Holt. A Study of the Flexural Attenuation Technique through Laboratory Measurements and Numerical Simulations
Tore Sirevaag, Tonni Franke Johansen, Idar Larsen and Rune Martin Holt. Evaluation of the pulse-echo technique for logging behind casing through laboratory experiments and numerical modelling
Tore Sirevaag, Tonni Franke Johansen, Rune Martin Holt. Expanding the processing of the flexural wave to improve the differentiation between heavy fluids and light solids behind a steel pipe
Tore Sirevaag, Tonni Franke Johansen, Rune Martin Holt. Processing of field data for comparison with laboratory measurements and detection of P-wave velocity in cement
Lozovyi, S., Sirevaag T., and Szewczyk, D. Non-elastic effects in static and dynamic rock stiffness. Presented at 51st US Rock Mechanics / Geomechanics Symposium