Measurement of wellbore leakage in high‐pressure gas well based on the multiple physical signals and history data: Method, technology, and application

Leakage is one of the most serious challenges for the safe production of high‐pressure gas wells for its high risks, including abnormal annular pressure, natural gas accumulation, and environment pollution, but available methods can hardly accurately measure the leakage type and depth, which are the key parameters for the rigless leakage repair and risk assessment. Therefore, this paper proposes a method to measure the leakage based on the characteristics, which combines qualitative and quantitative measurement together. Qualitative measurement considers the annular pressure, tubing pressure, liquid level, cement quality, and workover history. Quantitative measurement is determined by noise logging, electromagnetic logging, pressure logging, and temperature logging. The logging should be optimized according to the qualitative measurement. The method was successfully applied in high‐pressure gas well belonging to Tarim Oilfield. Two potential leakage types are provided based on the annular pressure, liquid level, cement quality, and workover history, including tubing leakage and linger hanger leakage. Based on the potential leakage types, the pressure difference, logging devices string, stopping length, and time are optimized to make the engineering logging reliable. Through measurement, two leakage points are found in tubing string. One is tubing body crack at the depth of 2724 m and the other is tubing thread leakage at the depth of 5211.7 m, which well matches the production data.


| INTRODUCTION
2][3] The related risks include three aspects, as shown in Figure 1.First is abnormal annular pressure. 4,5As shown in Figure 1, A annulus is the space between tubing and production casing.B annulus is the space between the production casing and technique casing.Likeswise, C annulus is next to B annulus.High annular pressure can lead to wellhead movement, 6 cement sheath crack, 7,8 and casing or tubing damage. 9,10Taking one gas well in Sichuan Basin, China, for instance, the wellhead movement caused by high annular pressure is as large as 50 mm.Meanwhile, the low annular pressure can also lead to casing or tubing damage because of the highpressure difference.Second is the accumulation of flammable natural gas nearby wellhead.According to the modeling analysis, 11 the accumulation of natural gas nearby wellhead can reach thousands of cubic meters.This may cause fire or even explosion disaster.Such disaster used to happen in high-pressure injection well SS-25 of Aliso Canyon underground gas storage. 12Third is environment contamination, including air, water, and soil.The large volume of natural gas near wellhead has to be released periodically to keep the annular pressure under the allowable value.The greenhouse effect is aggravated during this process.Moreover, the natural gas may invade into the shallow water or soil through wellbore leakage path, like cement and casing.For example, natural gas appeared in a river near Well LJ-2 in Sichuan Basin due to wellbore leakage.Considering the above risks, many countries and companies are dedicated to measure the wellbore leakage.For example, Canada has published an act to requires regular determination of wellbore leakage.
Workover is a traditional method to measure and repair wellbore leakage by lifting the tubing string out of wellbore, but its cost is extreme high (usually over 5 million dollars) and the performance is poor. 13Therefore, workover is not with priority in the measurement of wellbore leakage.Diffrient from workover, quantitative risk assessment, 14,15 and rigless leakage repair 16,17 can overcome the disadvantages of workover.][20] However, it is crucial for the successful assessment and rigless leakage repiar to know the leakage type and depth.Otherwise, the risk assessment accuracy and leakage repair performance maybe fail to the expectation.
However, it is difficult to measure the wellbore leakage in high-pressure gas well.Due to high temperature, corrosive fluid, and high stress, well barriers have high probability to leak, including wellhead, 21 tubing string, 11 casing string, 22,23 hanger, 24 cement, 25 and packer or valve. 26What is more complicated, different well barriers may leak at the same time.Meanwhile, the number of leakage points or leakage path may be more than one or even over 10.Taking Well KS501 in Tarim Oilfield as instance, the tubing string, casing string, and cement sheath leaked at the same time.This made the annular pressure in A annulus raised to 71.1 MPa, B annulus raised to 42.6 MPa, and C annulus raised to 16.4 MPa.After the tubing string was lifted out of wellbore, three leakage points were found, respectively, in the depth of 1910, 6381, and 6391 m.Therefore, it can be concluded that the leakage of high-pressure gas well is Risks of wellbore leakage in high-pressure gas well.
featured with various leakage types and multiple leakage points or pathways, which brings great challenge for the leakage measurement.
Avaiable methods can diagnose wellbore leakage by annular pressure, but can hardly satisfy the accurate measurement of wellbore leakage in high-pressure gas well, as shown in Table 1.First method is Bleed-Buildup test (known as B-B test). 27This method can judge whether wellbore leakage or not by recording the releasing-rising process of annular pressure, because the annular pressure caused by thermal expansion would not rebuildup after releasing, while the annular pressure caused by wellbore leakage will rebuildup. 14,28However, this method cannot measure the type and depth of wellbore leakage.Second method is locate the depth of tubing string leakage point based on U-tube principle. 29his method has been applied in offshore gas well 30 and deep gas well, 2 but this method can only locate the leakage point in tubing string.Moreover, it has been proved that this method would misjudge when the number of tubing string leakage points is over one. 31hird method is to locate the leakage point by receiving the leakage acoustic single at the wellhead during the annular pressure rising process. 32This method relies on the return signal of leakage sound reflected by annular liquid. 33As a result, it cannot measure the leakage under the annular liquid level or other well barriers.Also, the return signal of leakage sound becomes unclear when the number of leakage points is over one, which leads to large measurement error (Table 2).
Engineering logging is another way to measure the wellbore leakage, mainly including noise logging, 34 temperature logging, 35,36 electromagnetic flaw detection, 37 and multiarm caliper logging, 38,39 but it still needs to be improved.Noise logging has been applied in the wellbore leakage measurement since 1980s.Now, it can measure tubing leakage, casing leakage, and cement leakage, but not many successful cases were reported in high-pressure gas well.Temperature logging is based on the small temperature difference induced by fluid leakage.For high-pressure gas well, the tubing thread leakage is a common type, 40,41 but its leakage rate is so small that even high precision sensor cannot detect the temperature difference.Electromagnetic detection and multiarm caliper are usually combined together to measure the wall thickness of tubing or casing.Therefore, they can detect large leakage hole or deformation, but not all types of leakage points.Besides the above engineering logging techniques, distributed fiber acoustic sensing (DAS) 42,43 and distributed fiber temperature sensing (DTS) 44 can also measure the leakage noise and temperature difference, but the fiber cannot reach a depth over 4500 m due to limited tensile strength.][47] According to the above analysis, it is essential to develop a method applied in the deep gas well to measure the leakage type and depth.Based on available methods, this paper proposes a method based on multiple physical signals and production history to measure wellbore leakage in high-pressure gas well.The physical signals include noise, electromagnetic, temperature, and pressure.This can guarantee measurement effectiveness and accuracy.Not only the leakage location but also the type can be measured by this method.The production history includes production data, annular pressure, wellbore quality, and repair history, which can provide guidance and verification for the logging operation.The proposed method has been applied in Tarim Oilfield, China, and achieved good performance.

| PROPOSED METHOD
As shown in Figure 2, the proposed method can be divided into two interconnected steps.First step is to judge the potential leakage points or leakage path.Second step is to measure the leakage depth and type by engineering logging.The measurement result should be verified by production data.The measurement process can be optimized based on the first step.

| Find the potential leakage by production data
As stated in the introduction, B-B test can be used to judge the wellbore leakage.For the field operation, the remote control system 48 or related devices 49 can easily finish B-B test.Following B-B test, some other production data is used to find the potential wellbore barrier leakage, like tubing, casing, cement, or packer.Table 3 provides evidence to find the potential leakage.

| Tubing thread leakage/wellhead leakage
Annular pressure exists in A annulus and has the same change trend with tubing pressure when the leakage happens in tubing string thread.Liquid level remains stable because the tubing thread leakage is too minor for liquid to pass.Moreover, the tubing thread is easier to leakage if the thread is not air-sealed type.

| Tubing body leakage
Annular pressure exists in A annulus and has the same change trend with tubing pressure when tubing body leaks.Annular liquid level decreases first and then remains unchanged.The tubing body is easier to leak if the annular liquid is corrosive or the natural gas contains CO 2 /H 2 S.

| Packer leakage/liner hanger leakage
For packer leakage, annular pressure rises in A annulus and does not change with the tubing pressure.
The annular liquid level continues decreasing until the annulus is empty.For liner hanger leakage, annular pressure may exist in A annuls and does not change with tubing pressure.The annular liquid level continues decreasing.The annular pressure may also decrease if the leakage rate of the annular liquid is large enough.F I G U R E 2 Method to measure the wellbore leakage in high-pressure gas well.
ZHANG ET AL.Annular pressure exists in the casing annulus and does not change with tubing pressure.Cement bond logging may showed poor cement quality.Moreover, the cement sheath is easier to leak under hydraulic fracture 50 or low displacement efficiency. 51Another type of cement sheath leakage is microannuli in the casing and cement interface. 52,53The microannuli is usually caused by high stress or cyclic load and temperature.Fluid may also leak into cement sheath from the formation.Annular pressure exists in A or casing annulus and does not change with the tubing pressure.The cement quality is poor.This kind of leakage is easier to happen when the formation has high-pressure brine layer or salt paste layer, 54 because the casing may be collapse under the high pressure induced by the brine layer or salt paste layer.After the casing collapse happens, the cement sheath immediately faces with high-pressure brine layer or salt paste layer.Moreover, the casing perhaps is worn to leakage during workover.
One more situation is that fluid leaks through casing, cement sheath, and tubing together.Annular pressure exists in both A and B annuls, which has the same change trend with the tubing pressure.The annular liquid level may decrease at the same time.

| Measure the leakage by engineering logging
The wellbore leakage of high-pressure gas well is so complex that single engineering logging method cannot satisfy the requirements.The leakage measurement of high-pressure gas well should satisfy the following requirements, including resistance of high-pressure high temperature, ability to measure multiple leakage points/ paths, ability to measure leakage under annular liquid level, and small leakage.Moreover, not only the leakage location but also the leakage type should be measured.
There are two challenges to meet the above requirements.One is how to get reliable leakage signals.The other is how to determine where the signal comes from.To solve the above challenges, a method is proposed by combining several engineering logging technologies.The proposed method consists of noise logging, electromagnetic logging, temperature, and pressure logging.The noise logging captures the leakage sound to locate the depth of the leakage point or path.The electromagnetic logging judges leakage type by detect casing and tubing wall thickness around the leakage point to help determine the leakage type.The temperature logging records the temperature difference caused by leakage as assistance to locate the leakage depth.
The pressure logging collects the pressure profile to ensure that the leakage is activated.

| Noise logging
The wellbore noise is induced by fluid flow and its spectrum is related to the size of the leakage hole, fluid properties, pressure, temperature, and flow rate. 55herefore, the source of noise can be identified by the spectrum characteristics, as follows.
(1) Noise generated by gas flow in the tubing.The frequency of this noise is usually less than 1 kHz.It distributes the whole wellbore from bottom to wellhead, so it belongs to background noise.(2) Noise generated by tubing, casing, and packer leakage.Due to the difference in the size and shape of the leakage point, the frequency of this noise is not fixed.Moreover, it is an obvious prominent point in the acoustic spectrum of the whole wellbore, because the depth of the leakage point is at a specific depth.
In some situations, this noise can be masked by normal wellbore flow with similar frequency and amplitude, so it needs to be compared, identified, and analyzed.(3) Noise generated by leakage in the cement sheath behind the casing.The fluid would flow through the cement sheath when the cement sheath loses its sealed integrity, which may be fractures 54 or microannuli. 55Such flow has obvious starting and ending, so the noise also have clear boundaries in the wellbore longitudinal direction, which displays as as a narrow vertical strip.(4) Noise of reservoir flow.This noise is generated by the particles' vibration, pore channels, and fractures during reservoir fluid flow.It also has clear vertical boundaries (top and bottom), but it usually appears in the depth of the reservoir, which is under the packer.Its frequency depends on the reservoir properties.In tight gas reservoir, the frequency can be as high as 30 kHz.
According to the noise characteristic and highpressure gas well complex structure, a device is selected to capture the wellbore noises.The parameters are in Table 4.

| Electromagnetic logging
The attenuation of the electromagnetic pulse can identify the metal loss of tubing and casing, so the wall thickness of tubing or casing can be known.As shown in Table 5, different metal loss rates mean different damage degrees.It can be supposed that there is a leakage hole in the tubing when the metal loss rate is over 20%.Once the leakage depth is determines by noise logging, the electromagnetic logging can tell the leakage type by the following instruction.(1) Large metal loss rate in the casing means the casing may leak.Likewise, tubing may leak if there is a large metal loss.The leakage type can be seen as a hole.(2) The electromagnetic logging shows there is a thread at the leakage depth.The leakage type is thread leakage.(3) Small metal loss rate means leakage type may be fracture in casing or tubing body (Table 6).
The electromagnetic device should be able to measure multiple layers of casing and tubing due to the complex structure of high-pressure gas well.Figure 6 shows the parameters used in the wellbore leakage measure of high-pressure gas well.It can seen that the device diameter allows it runs inside tubing string, so it is not necessary to uplift the tubing string.

| Temperature and pressure logging
The fluid leakage would impact the temperature distribution around the leakage point or along the leakage pathway.For gas leakage, the Joule-Thompson effect is one of the reasons for the temperature redistribution, as expressed by the below equation.
where h is enthalpy, J; T is temperature, K; p is pressure, Pa; ρ is density, kg/m 3 ; C p is specific heat capacity, J/(kg K); v is gas specific volume, m 3 /kg.
Generally speaking, the temperature difference caused by leakage is related to the fluid properties and the pressure difference.Therefore, to identify the temperature change at the leakage point, it is necessary to establish an appropriate pressure difference.It should be noted that leakage does not mean that there must be a temperature difference, such as small leakage or liquid leakage.Another function of temperature logging is to know the working temperature of noise and electromagnetic devices, so as to ensure the effectiveness of the measurement signal.
As analyzed above, the pressure difference is one of the key parameters for the successful measurement.Reliable pressure differences should higher than the minimum required pressure difference.Only when the pressure difference is over the minimum required pressure difference, can the measured signal of noise be seen as reliable, as expressed by below equation.
where, p d is pressure difference, MPa; p dm is minimum required pressure difference, MPa; pi is pressure inside tubing, MPa; p o is pressure outside tubing, MPa.The pressure inside tubing can be obtained by pressure logging, which is a mature method.The pressure outside tubing consists of fluid column pressure and annular pressure, as expressed by below equation.
where p a is annular pressure, MPa; ρ g is density of annular gas, g/cm 3 ; g is gravitational acceleration, m/s 2 ; h L is depth of annular liquid level, m; h is depth of measurement point, m; ρ L is density of annular liquid, g/cm 3 .

| APPLICATION AND ANALYSIS
The proposed method is applied in Well KS201, Tarim Oilfield, China.Its depth is 6792.00m and the packer is in the depth of 6358 m.

| Analysis of wellbore leakage based on production data
With the help of production data, two potential leakage situations are given.One is tubing string leakage, which may be leakage hole or combination of leakage hole and thread leakage.The other is the liner hanger and cement leakage.

| Annular pressure and liquid level
The annular pressure in A annulus rose to 33.4 MPa on August 19, 2014, and decreased to 13.83 MPa within 25 days.
Meanwhile, the tubing pressure also decreased from 69 to 65 MPa.And large volume of natural gas accumulated in A annulus.It can be known that the A annular pressure has the same change trend with tubing pressure, so the tubing must leak and natural gas entered into A annulus.Moreover, the A annular pressure periodically declines after the annular liquid is reinjected.Figure 3 showed the A annular pressure recorded from February 2020 to April 2020.It can be seen that the A annular pressure decreased from 26.22 to 18.51 MPa within 35 days.It is reasonable to doubt that annular liquid also leaked.The annular liquid level also decreased periodically, which is another strong proof of annular liquid leakage.Beside liquid leakage into tubing, the casing may be damaged by high annular pressure so that the annular liquid leaks.On the other hand, the annular pressure in B annulus only changes with temperature.And no longer rises after releasing.So B annular pressure is caused by thermal expansion.Therefore, there is no leakage in the casing.

| Cement quality and workover
The cement quality is poor from the depth of 6200 to 6455 m because of liquid channeling.This is mainly caused by the casing eccentricity.Liner hanger is in this depth range, so this is a potential path of annular liquid leakage.
Two workovers used to conduct in this well.First, a wellbore leakage measurement was conducted in 2012 and a casing leakage point was found at depth of 9.36 m.This leakage point was repaired by plugging.Second, at F I G U R E 3 Annular pressure and production rate of the case well.
least two leakage points were found in the original tubing string in 2012 during the workover, as shown in Figure 4.One is leakage hole and the other one is thread fracture.Although the original tubing string was replaced by new one, there is still possibility for similar leakage.Moreover, the stress corrosion crack is easy to happen in the tubing string of high-pressure gas wells of Tarim Oilfield, 23,56 as shown in Figure 5.

| Optimization of the measurement process
As shown in Figure 6, the proposed engineering logging method was used for detection after wellbore leakage analysis.The measurement process was optimized according to the analysis results.

Pressure difference
The larger the pressure difference is, the more accurate the measurement is.However, the wellbore may be damaged under too large pressure difference.The lab experiment shows that the minimum required pressure difference pressure is 3 MPa for noise logging device to get a reliable signal.Therefore, the annular pressure was increased by injecting water to form a pressure difference no less than 3 MPa.Considering the density of the annular liquid is heavier than the density of fluid inside the tubing, the reliable noise signal can be assured in the whole wellbore.

Logging devices string
The logging devices string consists of noise device, electromagnetic device, temperature, and pressure device, so the noise, electromagnetic, temperature, and pressure signals can be collected during one logging.Moreover, centralizer is installed in the string to center the electromagnetic flaw detector and to avoid strong noise caused by the collision between the string and tubing.At least two noise devices are installed in the string, so the noise signal can be seen as reliable when two devices collect the similar signal.

Measurement parameters
The device string stops every 3 m and stays still for 30 s to receive the leakage signal.The experiment shows that under the pressure difference of 3 MPa, the receiving length in the longitudinal direction of the wellbore is 3 m.

| Noise caused by fluid flow
Figure 7 is the background noise measured in the range of 1580-2100 m.It can be seen that the noise is mainly concentrated in the range of low frequency 0-2 kHz.There is no obvious noise in the high-frequency range.Although some weak high-frequency signals appeared around the depth of 1850, 1900, and 1935 m, they did not reappear on No. 2 noisy device.This noise should be caused by the scratch of the centralizer or the device vibration.Based on the above analysis, it can be judged that there is no leakage point within this range.The background noise is mainly caused by fluid flow inside tubing, and the frequency is mainly in the range of 0-2 kHz.This provides the comparative basis to measure the depth of leakage point through noise.

| Noise of no. 1 leakage point
As shown in Figure 8, there is a strong high-frequency noise around the depth of 2724 m compared with the  background noise.The frequency of this noise is from 0 to 60 kHz, but it is much more stronger in the range of 4-18 kHz.Since there is no starting point and ending point of leakage noise and CCL shows there is no tubing thread at this depth.Moreover, this leakage noise is captured by both two noise devices.Therefore, it can be judged this noise is caused by tubing body leakage.

| Noise of no. 2 leakage point
As shown in Figure 9, one more leakage noise is measured at the depth of 5211.7 m by two noise devices.This noise has a wide frequency range, but it is stronger in the range of 0-20 kHz.Likewise, there is no starting point and ending point of this leakage noise, so the cement and casing do not leak.Different from No. 1 leakage point, there is a tubing ZHANG ET AL. thread at the depth of 5211.7 m according to CCL measurement.Therefore, No. 2 leakage point can be regarded as tubing thread leakage, but this should be double-verified and explained by electromagnetic logging.

| Noise around liner hanger
According to the well structure, the liner hanger is around the depth of 6129.00 m.As shown in Figure 10, there is no strong or high-frequency noise at the depth of 6129.00 m.Only background noise appears in the low-frequency and high-frequency range.Therefore, there is no leakage in the liner hanger and the cement sheath behind the liner.
3.3 | Verification of electromagnetic, pressure, and temperature

| Leakage type judged by electromagnetic logging
As shown in Figure 11, the electromagnetic logging shows that the largest metal loss rate of tubing string is only 6.9% at 3249.6 m, which means no obvious damage.
For the depth of No. 1 leakage point, the metal loss rate is 3.3%.Therefore, the leakage point cannot be caused by large corrosion damage.Considering the original tubing string has the phenomenon of stress corrosion crack, the leakage point can be also small cracks in the tubing body.
As shown in Figure 12, the electromagnetic logging shows dense alternating color bars in the range of 5100-5320 m.This can be explained as tubing buckling, because the buckling behavior makes the electromagnetic logging device eccentric in the wellbore.This would lead to the increase of the measured thickness of the tubing wall.The measured thickness is not the actual thickness, but it can judge the tubing buckling or not.This can well explained the tubing thread leakage at 5211.7 m in that buckling is one of the reasons for the tubing thread leakage. 57Moreover, tubing buckling is not rare to see in high-pressure gas wells due to high stress and high temperature.141.8°C and 67.7 MPa.Therefore, all the logging devices are in well condition during measurement.Both temperature and pressure are suitable for the logging device to work.
For the temperature variation caused by leakage, Figure 13 shows little temperature variations at both two leakages, but this is not strange.First, the leakage medium is water.Water cannot lead to large temperature variations compared with gas.Second, the leakage rate is very low.During the logging, water was injected into the A annulus to increase the pressure, thus creating a suitable pressure difference for noise logging.According to the statistics of injection water, the total leakage rate of the A annulus is as low as 0.38 L/min.Therefore, the little temperature variations do not impact the leakage measurement.

| Reliability verified by pressure difference
According to Equation (2), the pressure difference is determined by pressures outside and inside the leakage point.The pressure inside the tubing has been obtained by pressure logging.To verify the pressure difference, the pressure outside the leakage point must be calculated.It can be known from Equation (2), the pressure outside the leakage point is related to annular liquid density and annular pressure.Annular pressure can be measured at the wellhead, so the pressure outside the leakage point only changes with the annular liquid.
Equation ( 2) can help draw a curve to describe the change law of pressure outside the leakage point, as show in Figure 14.After the pressure outside leakage is known, the pressure difference can also be calculated.For No. 1 leakage point, the pressure outside the leakage point increases as the annular liquid increases.The pressure difference can be kept over 3 MPa when the annular liquid is heavier than 1.1 g/cm 3 .For No. 2 leakage point, the pressure outside leakage point also increases as the annular liquid increases.The pressure difference is always over 3 MPa when the annular liquid is heavier than 1 g/cm 3 .
The initial density of annular liquid in Well KS201 is 1.4 g/cm 3 .Considering the formation of water may enter into the annulus, the annular liquid density may change.However, the formation water density is highpressure brine and 59 and its density is usually over 1.1 g/cm 3 , so the annular liquid density has no possibility to be lower than 1.1 g/cm 3 .Accordingly, the pressure difference of No. 1 leakage point is higher than 3 MPa.Therefore, pressure differences of the two leakage points are both qualified to ensure the reliability of the noise signal.

| RESULTS OF LEAKAGE MEASUREMENT
As shown in Table 7, it can be concluded that there are two leakage points in the tubing string while the liner hanger does not leak.According to the statistics of injection water during logging, the total leakage rate of the A annulus is as low as 0.38 L/min.It should be noted that this leakage rate is not the leakage rate of a single leakage point.
The measurement results can well match the production data.The annular liquid leaks into the tubing through leakage point No. 1 and gas enters the tubing-casing annulus through leakage point No. 2. Therefore, the annular liquid level decreases and there is large volume of natural gas in the annulus.The annular pressure decreases instead of rising for two reasons.First, the gas is not enough to increase annular pressure.This well produces 85 m 3 water and about 1.5 × 10 4 m 3 /day.Moreover, The gas leakage rate through tubing thread is very low, which also should account for the decrease of annular pressure.Second, the annular gas reduces the thermal expansion of the annular liquid, 60 so the annular pressure caused by thermal expansion is also no obvious.

| 7 T A B L E 3 4 |
Find the potential leakage by production data.Leakage through cement sheath

T A B L E 4
Parameters of device to capture wellbore noise.

F I G U R E 4
Leakage points in the original tubing string.F I G U R E 5 Typical stress corrosion crack of the tubing string.

F I G U R E 6
Field operation of engineering logging to measure wellbore leakage.F I G U R E 7 Background noise measured in the range of 1580-2100 m.

| 13 F
I G U R E 8 Leakage noise measured at the depth of 2724 m.

3. 3 . 2 | 15 F
Reliability verified by temperature and pressure As shown in Figure 13, the temperature and pressure at No. 1 leakage point are 89.3°C and 48.1 MPa.The temperature and pressure at No. 2 leakage point are F I G U R E 9 Leakage noise measured at the depth of 5211.7 m.F I G U R E 10 Noise around liner hanger.ZHANG ET AL. | I G U R E 11 Largest metal loss rate of tubing string at 3249.6 m.F I G U R E 12 Tubing buckling in the range of 5100-5320 m.

F
I G U R E 13 Pressure and temperature at leakage points.(A) Pressure and temperature No. 1. (B) Pressure and temperature No. 2. F I G U R E 14 Pressure and temperature at leakage points.(A) Pressure difference at No. 1 leakage point.(B) Pressure difference at No. 2 leakage point.

T A B L E 7
Results of leakage measurement.
Engineering logging methods to measure wellbore leakage.
T A B L E 2 Not able to measure small hole or thread leakage; Not able to measure cement sheath leakage 4 DAS + DTS Locate the depth of wellbore leakage The measurement depth cannot over 4500 m; The measurement cost is expensive Abbreviations: DAS, distributed fiber acoustic sensing; DTS, distributed fiber temperature sensing.
Wellbore leakage can bring about abnormal annular pressure, accumulation of natural gas, and environment pollution.The leakage of high-pressure gas well is characterized with high probability, various types, and multiple points or paths, so the leakage measurement should satisfy the following requirements, including high-pressure high temperature, multiple leakage points/paths, leakage under liquid level, and minor leakage.(2)The proposed method combines qualitative analysis and quantitative measurement.The qualitative measure is mainly based on annular pressure, tubing pressure, liquid level, cement quality, and workover.The method provides the characteristics of different wellbore leakage types.Through the qualitative measurement, the following leakage types can be judged, including tubing thread leakage, tubing body leakage, packer leakage, leakage through cement sheath, leakage through casing and cement sheath, leakage of the casing, cement sheath, and tubing together or liner hanger leakage.Qualitative measurement can guide and verify quantitative measurement.(3) The quantitative measurement relies on engineering logging.By combining noise logging, electromagnetic logging, pressure logging and temperature logging together, the measurement overcomes the challenges of getting reliable leakage signals and determining the where the signal comes from.The noise logging captures the leakage sound to locate the leakage depth.The electromagnetic logging judge leakage type.The temperature logging records the temperature difference caused by leakage as assistant method to locate the leakage depth; The pressure field collects the pressure profile to ensure that the leakage is activated.(4) The application in case well found two leakage points.One is tubing body leakage at the depth of 2724 m and the other is tubing thread at the depth of 5211.7 m.The measurement result well matches the qualitative analysis.Based on the qualitative analysis, the process of engineering logging is optimized, including pressure difference, logging devices string, stopping length, and time.The logging devices string should have at least two noise devices.The noise signal can be seen as reliable only when two devices collect the same signal and the pressure difference meets the required value.