RU2728116C1 - Method for mutual calibration of borehole fluid temperature sensors installed on a perforating column - Google Patents

Abstract

FIELD: measuring equipment.

SUBSTANCE: invention relates to measurement of pressure and temperature in well during perforation and subsequent testing of well. According to the method, a perforation string is lowered into the well. At the lower end of the lower section of the column there installed is a pressure sensor and a temperature sensor of the well fluid, and temperature transducers are installed on that part of the perforation string, which corresponds to the position of perforated productive formations. During lowering, temperature and pressure are measured and averaging of pressure transducer readings is performed. Average rate of run-in to the well of the perforation string and temperature gradient of the well fluid along the well length are calculated. A time interval is selected during descent of the column, during which the average descent rate and the temperature gradient of the well fluid remain constant. In the selected time interval, the temperature profiles measured by the sensors are shifted by temperature values proportional to the difference in recording times, so as to ensure the best match of all shifted temperature profiles with the temperature profile at the selected moment in time. Averaged readings of sensors for shifted temperature profiles are calculated. Selecting a temperature sensor, relative to which calibration will be performed, and constructing a calibration line passing through the selected temperature sensor and having a calculated steady temperature gradient along the descending column. Calibration corrections to results of temperature measurement by temperature sensors are calculated as deviation of calculated average readings of sensors from the constructed calibration line and mutual calibration of temperature sensors is performed relative to the selected temperature sensor.

EFFECT: technical result consists in providing mutual calibration of temperature sensors in the well before perforation, which in turn provides accuracy of measuring temperature of the well fluid during perforation and subsequent testing of the well.

4 cl, 11 dwg

Description

The invention relates to the field of measuring pressure and temperature in a well during perforation and subsequent testing of the well using sensors installed on the perforation string.

The results of measuring the pressure and temperature in the well during perforation and subsequent testing of the well using sensors installed on the perforation string can be used to assess the quality of perforation and determine the profile of the fluid flow into the well.

Quantitative differentiation of the inflow from different perforation intervals in terms of the magnitude and dynamics of temperature changes in the well is possible only if sufficiently high requirements for the accuracy of the mutual calibration of temperature sensors installed in the well are met. Since the temperature measured in the well is mainly determined by the adiabatic effect, the Joule-Thomson effect in the formation and the calorimetric effect of mixing flows in the well, the value of the useful temperature signal in the well is 0.05 ÷ 0.2 K. This means that for quantitative interpretation of non-stationary temperature data, it is necessary to ensure accuracy mutual calibration of temperature sensors not worse than 0.01 ÷ 0.02 K.

Under laboratory conditions, it is difficult to provide such an accuracy of intercalibration of a large number of temperature sensors [2016-2017 CALIBRATION TOOLS CATALOG, www.flukecal.eu]. In addition, the calibration of the sensors must be carried out just before running the perforation string into the well, which is very difficult.

In theory, temperature sensors can be calibrated in-situ using downhole measurements in an idle well where the temperature is geothermal to the required accuracy. On the one hand, this requires too long well holding times (several months), which is technologically unacceptable. On the other hand, in an idle well, fluid may move that distorts the geotherm and, in addition, the geotherm may differ from a straight line. All this can lead to uncontrollable errors in the intercalibration of temperature sensors, so calibration of temperature sensors in this way is practically impossible.

The technical result achieved by the implementation of the invention is to provide mutual calibration of temperature sensors in the well prior to perforation, which in turn ensures the accuracy of measuring the temperature of the well fluid during perforation and subsequent well testing.

The specified technical result is achieved by the fact that, in accordance with the proposed method of mutual calibration of the temperature sensors of the well fluid installed on the perforation string, a perforation string consisting of separate sections is lowered into the well by attaching the next lowered section to the perforation string sections already placed in the well , and running the string into the well for the length of this section. A pressure sensor and a well fluid temperature sensor are installed at the lower end of the lower section of the perforation string, and temperature sensors are installed on that part of the perforation string that corresponds to the position of the perforated productive formations. During the running of the perforation string, temperature measurements are carried out by means of temperature sensors installed on the perforation string and pressure measurements are carried out by means of a pressure sensor installed at the lower end of the lower section of the perforation string. The readings of the pressure sensor installed at the lower end of the lower section of the perforation string are averaged, and an average pressure versus time dependence is obtained. Using the obtained averaged dependence of pressure on time and temperature measured by a sensor installed at the lower end of the lower section of the perforation string, the average speed of the perforation string down the well and the temperature gradient of the well fluid along the well are calculated. A time interval is selected during running the perforation string into the well, during which the average speed of the perforation string running into the well and the temperature gradient of the well fluid remain constant. In the selected time interval, the temperature profiles measured by the sensors are shifted by temperature values proportional to the difference in the recording times, in such a way as to ensure the best match of all shifted temperature profiles with the temperature profile at the selected time point, and the average sensor readings for the shifted temperature profiles are calculated. Using the found average speed of perforation string running into the well and the temperature gradient of the well fluid, the steady-state temperature gradient along the running string is calculated. A temperature sensor is selected, relative to which the calibration will be carried out, and a calibration line is built, passing through the selected temperature sensor, and having a calculated steady-state temperature gradient along the running string. Calibration corrections to the temperature measurement results by temperature sensors are calculated as the deviation of the calculated average sensor readings from the constructed calibration straight line and mutual calibration of the temperature sensors relative to the selected temperature sensor is performed.

In accordance with one of the embodiments of the invention, the averaging of the readings of the pressure sensor installed at the lower end of the lower section of the perforation string is carried out by the sliding window method, using as the sliding window the time interval corresponding to the time of attachment of 3-4 sections to the perforating string.

In accordance with another embodiment of the invention, during shutdowns in the process of running the perforation string into the well, a short-term injection of fluid is performed, the temperature change is measured for each temperature sensor and, using the measured dynamics of the temperature increase of the sensors, the time constants of each sensor are determined for use in interpreting transient temperature data during perforation and well testing.

In accordance with another embodiment of the invention, short-term injection of fluid into the well is carried out after the completion of running the perforation string, the temperature change is measured for each sensor and, using the measured dynamics of the temperature increase of the sensors, the time constants of each sensor are determined, taking into account the position of the string relative to the wellbore walls, which will affect on sensor readings and interpretation of non-stationary temperature data during perforation and subsequent well testing.

The invention is illustrated by drawings, where FIG. 1 shows a diagram of a perforation string placed in a well with temperature and pressure sensors installed on its sections; in FIG. 2 shows graphs of pressure and temperatures measured by perforation string sensors during the running of the string into the well; in FIG. 3 shows the results of measuring the pressure during running the string into the well, the results of averaging the measured pressure over two time intervals and the difference between the averaged pressures; in FIG. 4. shows the dependence of the average speed of lowering the perforation string into the well on the distance measured along the well, calculated by averaging the pressure over two time intervals; in FIG. 5 shows the dependence of the temperature on the distance measured along the well, calculated by averaging the pressure over two time intervals; in FIG. 6a shows an axial 2D conductive-convective model of running a perforation string into the well, FIG. 6b - model of the running string in the reference system associated with the string; in FIG. 7a shows the results of calculating the surface temperature profile of the perforation string at various points in time after the start of running the perforation string, FIG. 7b - shifted temperature profiles; in FIG. 8 shows the average readings of the temperature sensors and the resulting calibration curve, FIG. 9 shows the errors of the temperature sensors remaining after the mutual calibration of the sensors at different values of the temperature gradient along the string; in FIG. 10 shows the results of modeling the radial temperature profile 20 m above the end of the string after running the string and the temperature profiles at the beginning and end of the injection; in FIG. 11 shows the dynamics of temperature changes at the surface of the perforation string and the temperature of the fluid at different distances from the surface of the string.

The proposed method is illustrated by two examples: field data on running a perforation string into the well with three temperature sensors and a pressure sensor installed on its sections and the results of numerical simulation of running a perforation string into the well with 30 temperature sensors. The first example confirms the possibility of obtaining the data required for the calibration in downhole conditions.

The data obtained is used for numerical simulation in the second example.

In accordance with the proposed method, a perforating string consisting of separate sections is lowered into the well by attaching the next lowered section to the perforating string sections already placed in the well. FIG. 1 shows a diagram of the lower part of the perforation string 1 installed in the well for subsequent perforation of layers 2 and 3. At the lower end of the perforation string 1, a pressure sensor 4 (P ₀ ) and a temperature sensor 5 (T ₀ ) are installed. On the upstream sections of the perforation string, above the respective perforation zones, sensors 6 and 7 for temperatures T ₁ and T _{2 of the} well fluid are installed. These sensors 6 and 7 are located at a distance of z ₁ = 21 m and z ₂ = 33 m from the lower end of the perforation string. All temperature sensors installed on the sections of the perforation string, before perforation during the running of the perforation string, actually measure the surface temperature of the string, and the sensor 5, installed at the lower end of the perforation string 1, during the run measures the temperature that the well fluid had before the start of running the string. into the well. Later, after calibration, during perforation and subsequent testing of the well, temperature sensors will measure the temperature of the well fluid.

During the running of the perforation string, temperature measurements are carried out by means of temperature sensors 5, 6 and 7 installed on the perforation string, and pressure measurements are carried out by means of a pressure sensor 4 installed at the lower end of the perforation string. The pressures and temperatures measured by perforation-mounted sensors while running the perforation string sections are shown in FIG. 2. Temperature sensor 5, installed at the lower end of the perforation string 1, measures the temperature T _{0 of the} well fluid in the well before perforation (rocks near the well), not distorted by the movement of the string (curve 8). Curves 9 and 10 show the results of measuring the surface temperature of the perforation string by sensors 6 and 7. Curve 11 shows the change in pressure P ₀ (t) measured by the pressure sensor 4. Eliminating the time from the measured dependences P ₀ (t) and T ₀ (t), the relationship between the measured temperature and the pressure T = T (P) is found, which will be used below to determine the dependence of temperature on the distance to the surface y, measured along the well: T = T (y).

A relatively simple procedure for processing the temperature data obtained during the running of the perforated string into the well is possible only with quasi-continuous running of the string, that is, when running the string with approximately the same time spent on attaching the next section to the string and running the string into the well for the length of this section.

Thus, quasi-continuous running into the well of perforating string sections with temperature and pressure sensors installed on it is carried out.

Then, the readings of the pressure sensor 4 installed at the lower end of the perforating string are averaged. Pressure averaging can be carried out, for example, by the sliding window method; For this, as a sliding window, for example, time intervals of 5 and 10 minutes can be used, which corresponds to the time of joining 3-4 sections to the perforation string. The pressure values measured in these intervals are summed up and divided by the number of measurements and this time interval; the resulting value is assigned to the midpoint of the interval. Then the window moves one point and the operation is repeated. FIG. 3 shows the results of measuring the pressure by the sensor 4 during the running of the string into the well (curve 12), the results of pressure averaging in a sliding time interval of 5 minutes, pos. 13, in a time interval of 10 minutes, pos. 14 and the difference between the average pressures (curve 15). Referring to FIG. 3, the maximum amplitude of short-term pressure variations (pressure waves in the well) reaches 2 bar, however, after averaging the pressure in time intervals of 5 minutes and 10 minutes, the pressure variations decrease very significantly. In particular, in the selected time interval (390 ÷ 420 min), the difference between the average pressures does not exceed 0.04 bar.

The next step is to estimate the average rate of running the string into the well using the pressure measurements. The estimation of the average speed of running the string is carried out after averaging (smoothing) the measured pressure.

The average speed of _{running the} string into the well (V _RIH ) is calculated by the formula:

where y is the distance from the surface, measured along the borehole.

The relationship between y and depth h is determined by the trajectory of the well, which is calculated from the results of the well logging. In the considered interval of depths, in our case, this relationship is determined by the formula y = 1.0006⋅ (39.15 + h). In turn, the change in h with time is associated with the measured pressure by the formula:

- плотность скважинного флюида. В рассматриваемом случае

where h _BH is the maximum depth to which the perforation string was lowered, P _BH is the well pressure measured at this depth,

is the density of the borehole fluid. In the case under consideration

The average speed of running the string, calculated according to these formulas with averaging the pressure over time intervals of 5 and 10 minutes, is shown in Fig. 4 (over an interval of 5 minutes - curve 16, over an interval of 10 minutes - curve 17). In the depth interval 2720 ÷ 2810 m (corresponding to the selected time interval for calibration), the average descent speed is 2.8 ± 0.2 m / min.

Considering the above-mentioned relationship between the temperature measured at the lower end of the perforation string and the pressure T = T (P) (from Fig. 2), using these formulas, it is easy to find the dependence of temperature on the distance measured along the well: T = T (y). This again uses pressure data averaged over time intervals of 5 and 10 minutes.

FIG. 5 shows the dependence of the temperature on the distance measured along the well, calculated by averaging the pressure over time intervals of 5 minutes (curve 18) and 10 minutes (curve 19). Depth 20 corresponds to the time of running the string into the hollow 390 minutes, depth 21 - 420 minutes.

The temperature gradient of the fluid in the well before perforation along the length of the well D is calculated by the formula:

In the considered interval of depths, it is Г = 0.0310 ± 0.0005 K / m.

Next, mathematical modeling of the process of lowering the perforation string into the well is carried out. Modeling is carried out for the field values of the radii of the perforation string and the casing string and for the found values of the temperature gradient and the rate of running the string into the well.

FIG. 6a shows an axial 2D conductively convective perforation string run model, and FIG. 6b - a model of the running string in the reference frame associated with the string. The main assumptions used in the construction of a numerical model for running the string into the well (Fig.6a):

1) in the considered interval of depths, the temperature gradient along the length of the well G is constant,

2) casing _running speed V _{RIH is} constant.

3) the initial temperature of the string, the temperature in the annular gap between the casing and perforation strings and the temperature of the surrounding rocks are determined by the temperature gradient G.

The first two assumptions correspond to the requirements described above for the selection of the time interval for running the string, which can be used to calibrate temperature sensors.

The descent into the well of the perforation string is modeled in a frame of reference associated with the string (Fig. 6b).

Let T _f0 be the temperature of the rocks at the lower end of the lower section of the string at the time of the beginning of _{running the} string into the well with a velocity V _RIH (Fig. 6a). During the run, the temperature T _{fb of the} well fluid and rocks at the end of the string is determined by the formula:

During the running of the string, the velocity V _w0 of fluid lifting in the annular gap is determined by the inner radius of the casing r _ci and the outer radius of the perforated string r _st :

In the frame of reference associated with the running string, the velocity V _{w of} the fluid flow in the annular gap is determined by the sum (Fig.6b):

V _w = V _RIH + V _w0

The unsteady temperature field in the column, annular gap and rock, determined by running the column, is modeled using an axial 2D conductive convective model:

Within this model, we neglect the radial velocities (V _r = 0), and the vertical velocity V _z is determined by the formula:

with the boundary condition:

The following calculation results for this model were performed using the commercial COMSOL Multiphysics package.

FIG. 7a shows the results of calculating the profile of the surface temperature of the perforation string T _st (t, z) at different times (in minutes) after the start of running. The calculations were carried out for the following parameter values: Г = 0.025 K / m, V _RIH = 2.8 m / min, r _ci = 7.8 cm, r _st = 5.7 cm, the borehole fluid is water, the thermal conductivity of rocks is 2 W / m / K. The same temperature profiles, but shifted so that the temperatures at the lower end of the column coincide, are shown in FIG. 7b.

From FIG. 7b, it can be seen that, in the case under consideration, ~ 20 min after the start of the descent and at a distance of more than 10 m from the end of the column, all temperature profiles coincide and are straight lines. This fact is proposed to be used to construct a calibration line.

и среднеквадратичным отклонением s(ΔT)=0.3K. Значение s(ΔT)=0.3K было выбрано как типичное для датчиков, которые используются в нефтегазовой промышленности. Погрешность самого нижнего калибруемого датчика температуры принималась равной нулю, т.е. калибровка проводилась относительно этого датчика.The proposed procedure for constructing a calibration line and intercalibration of temperature sensors is illustrated in the synthetic data of FIG. 7a. It is assumed that in the interval z = 10 ÷ 30 m there are ng = 30 sensors, the readings of which differ from the temperature of the casing surface by the sum of the systematic error of this sensor ΔT _j and the random error δT (t _i , z _j ). The systematic error ΔT _{j is} assumed to be a normally distributed random variable with zero mean

and the standard deviation s (ΔT) = 0.3K. The value s (ΔT) = 0.3K was chosen as typical for sensors used in the oil and gas industry. The error of the lowest calibrated temperature sensor was taken to be zero, i.e. calibration was carried out with respect to this sensor.

The first calibration step, as described above, is to determine the average _run velocity V _RIH and the temperature gradient G of the wellbore fluid using the pressure and temperature measurements at the lower end of the lower section of the string (FIG. 4 and FIG. 5).

The next step is to select a time interval for calibrating temperature sensors. The time interval for calibration should be removed from a prolonged stoppage of running the casing sections (for example, 325 ÷ 350 min in Fig. 2) or from the moment of a significant change in the running speed of the casing. In this case, it can be expected that the heat transfer regime between the perforation string lowered into the well and the surrounding rocks will be quasi-stationary and the rate of temperature rise measured by different sensors will be the same. The fulfillment of this condition can be checked directly using the results of temperature changes by various sensors.

The time interval selected for calibration must be sufficiently long. The number nt of temperature measurements made during this time must be at least nt = 10. Increasing the number of measurements increases the accuracy of the calibration. In the case under consideration, a time interval of 390 ÷ 420 min was chosen to calibrate the temperature sensors, in which the rates of temperature change measured by different sensors are practically the same.

In addition, the duration of the selected time interval should be sufficient to stabilize the temperature distribution along the length of the running string and to conduct measurements. In the case under consideration, this is about 60 minutes. Moreover, in this time interval, the average descent speed V _RIH and the temperature gradient Г should be constant.

Using the numerical model described above (Fig. 6) of running the string into the well and the found values of V _RIH and G, the steady-state temperature gradient G _st along the running string is calculated. In the case under consideration, Г _st = 0.022 K / m, while the initial temperature gradient of the borehole fluid is Г = 0.025 K / m.

The processing of the temperature measurement results obtained in the selected time interval is carried out. One of the temperature sensors and a time point in the selected time interval are selected and the temperature profiles of the remaining temperature sensors obtained in the selected time interval are shifted to temperature intervals proportional to the difference in recording times, so as to ensure the best match of all shifted temperature profiles with the temperature profile of the selected sensor ( for example, the lowest) at the selected time.

That is, the processing is carried out by shifting the temperature profiles to a certain, for example, the last t _nt time instant (t _i is the time of the i-th measurement, t ₁ is the time of the first measurement, T _{i, j is the} temperature measured by the j-th sensor in time t _i ):

where DT is determined from the condition

In the case under consideration, DT = 2.1 K. This procedure makes it possible to reduce the influence of a random error on the calibration accuracy.

The average temperature profile Tm _{j is} calculated by the formula:

The required calibration straight line is:

where z is the distance from the lower end of the column, z ₁ is the distance from the lower end of the column of the lowest of the calibrated sensors relative to which the calibration is performed.

23 для рассмотренного примера.FIG. 8 shows the averaged readings of the Tm _j - 22 sensors and the calibration line

23 for the considered example.

Corrections ΔTc _j to the readings of the sensors, which ensure their mutual calibration, are calculated as the deviation of the average readings of the sensors from the calibration straight line:

The difference between the calculated values of the corrections ΔTc _j from the above specified sensor errors ΔT _{j is} shown in FIG. 9. For the calculated value of the column temperature gradient Г _st = 0.022 K / m (curve 24), the systematic errors of sensors with a standard deviation s (ΔT) = 0.3K were reduced to s≈0.02 K. The error became less than the random error δT of an individual sensor (s (δT) = 0.05K). This is because the calibration error is inversely proportional to the square root of the number of measurements nt. It is obvious that by increasing the number of measurements, it is possible to practically eliminate the influence of random measurement errors on the calibration accuracy. Thus, the proposed calibration method allows achieving an accuracy sufficient for quantitative interpretation of downhole changes.

When the initial fluid temperature gradient in the well G = 0.025 K / m (curve 25) is used as the temperature gradient of the column Г _{st, the} calibration error increases with distance from the reference (lower) temperature sensor and reaches ~ 0.08 K. If you use a significantly underestimated (0.015 K / m (26)) or overestimated (0.03 K / m (27)) value of the column temperature gradient, the calibration error increases to -0.1 K and +0.15 K, respectively.

Correction of readings of temperature sensors is carried out according to the formula:

Tcor _{i, j} = T _{i, j} -ΔTc _j

The proposed method of mutual calibration of temperature sensors allows ensuring the accuracy of the measuring system in quasi-static conditions. However, the processes during perforation and during subsequent testing of the well are significantly unsteady, therefore, it is necessary to estimate the thermal inertia (time constant) for all sensors installed on the perforation string, and take this data into account when quantitatively processing the measurement results. This means that, in addition to quasi-static intercalibration of temperature sensors, it is necessary to estimate the in-situ response of each sensor to a sharp change in the temperature of the well fluid, that is, to estimate the time constant τ of each sensor.

The present invention proposes to carry out dynamic calibration of temperature sensors during shutdowns when lowering sections of the perforation string into the well or after the string has reached a predetermined depth by short-term injection of fluid into the well. The injection of the fluid provides an increase in the pressure ΔР of the fluid in the well and a corresponding increase in the temperature ΔТ _{а of the} well fluid due to the effect of adiabatic compression:

ΔT _a = η _a ⋅ΔP.

where η _a is the adiabatic coefficient of the fluid. For water, it is approximately equal to η _a ≈0.005K / bar.

The time constant τ of the sensor is found as a value that allows the best approximation of the temperature change of the sensor T _m (t) measured during the adiabatic test by solving the differential equation

with the initial condition T (t = 0) = T ₀ (T ₀ is the temperature measured before the start of injection). Here, T _fl (t) is the fluid temperature calculated from the dynamics of fluid pressure change P (t) (P (t = 0) = P ₀ ) measured in the well:

The adiabatic test should be carried out a few minutes after the casing has been stopped to stabilize the pressure and temperature fields in the well, distorted by the string running operations.

. Если скважина заполнена водой (со сжимаемостью β_w=5⋅10^-5 бар) темп закачки Q=500 м³/день, то для увеличения давления в скважине на ΔР продолжительность t_p закачки определяется формулой

. Для ΔР=100 бар имеем: t_p=30 с.Let us estimate the injection parameters required for the adiabatic test. In the example considered above, the outer radius of the string r _st = 5.7 cm, the inner radius of the casing r _ci = 7.8 cm, the well depth L = 4000 m and the fluid volume in the well:

... If the well is filled with water (with compressibility β _w = 5⋅10 ^-5 bar), the injection rate is Q = 500 m ³ / day, then to increase the pressure in the well by ΔР, the duration of the injection t _p is determined by the formula

... For ΔР = 100 bar we have: t _p = 30 s.

The proposed method for dynamic in-situ calibration can be illustrated using the numerical model described above, which should be supplemented with a casing shutdown model and an increase in wellbore pressure.

The corresponding energy equation can be written as:

where the term J describes the release of energy due to the adiabatic compression of the fluid:

The unsteady pressure in the well and, accordingly, the source term J are written as:

where t ₁ is the time elapsed after stopping the string and P ₀ is the wellbore pressure (at the considered depth) before fluid injection.

с (кривая 31). Во всех случаях теплопроводность окружающих пород принималась равной λ_ƒ=2 Вт/м/К, плотность и удельная теплоемкость ρ_ƒ=2700 кг/м³ и с_ƒ=1000 Дж/кг/К, внутренний радиус перфорационной колонны r_sti=4.7 см, теплофизические свойства материала колонны λ_st=45 Вт/м/К, ρ_st=7800 кг/м³, c_st=450 Дж/кг/К. Предполагалось, что внутреннее пространство перфорационной колонны заполнено воздухом.FIG. 10 shows the results of modeling the radial temperature profile 20 m above the end of the string after running the string for 1000 s (time elapsed after stopping t _SI = 0 s (curve 28)), the temperature profile at the start of injection (t _SI = t ₁ = 500 s ( curve 29)), at the end of injection (t _SI = t ₁ + t _p = 530 s (curve 30)) and

s (curve 31). In all cases, the thermal conductivity of the surrounding rock taken equal to λ _ƒ = 2 W / m / K, density and specific heat ρ _ƒ = 2700 kg / m ³ and _ƒ = 1000 J / kg / K, the inner radius of the perforating string r _sti = 4.7 cm , thermophysical properties of the column material λ _st = 45 W / m / K, ρ _st = 7800 kg / m ³ , c _st = 450 J / kg / K. It was assumed that the interior of the perforation string was filled with air.

FIG. 11 shows the dynamics of temperature change at the surface of the perforation string (curve 32) and fluid temperature at a distance of 1 mm (curve 33) and 2 mm (curve 34) from the surface. In all cases, during the running of the string, the calculation was carried out for the thermal conductivity of the rocks 2 W / m / K, and after stopping it was taken equal to 1 W / m / K (dashed curves), 2 W / m / K (solid curves) and 3 W / m / K (points). This is consistent with the averaged impact of rock properties during the running of the casing and possibly different rock properties at different depths after the casing stopped.

FIG. 11 shows that within 300 ÷ 500 s after the adiabatic test, possible vertical variations in the thermal properties of rocks practically do not affect the measured temperature. This means that the temperature change ΔT (t) = T (t) -T (t ₁ ), caused by a sharp increase in pressure in the well, should be the same for all sensors installed on the perforation string.

The proposed dynamic in-situ calibration can be performed at different depths, during shutdowns of the perforation string. However, the most informative is the adiabatic test after the column reaches its greatest depth - in the perforation zone. In this case, the test result is influenced precisely by the position of the string relative to the wellbore walls, which will affect the readings of the sensors and the interpretation of non-stationary temperature data during perforation and subsequent well testing. As a result of processing the dynamics of the behavior of each sensor during the adiabatic test, for each sensor its time constant is determined, which will be used to interpret the results of perforation and well testing.

Claims (14)

1. A method for inter-calibration of well fluid temperature sensors installed on a perforation string, according to which: - a perforation string consisting of separate sections is lowered into the well by attaching the next lowered section to the perforation string sections already placed in the well and running the string into the well for the length of this section, while a pressure sensor is installed at the lower end of the lower section of the perforating string and a temperature sensor of the well fluid, and temperature sensors are installed on that part of the perforation string, which corresponds to the position of the perforated productive formations, - during the running of the perforation string, temperature measurements are carried out by means of temperature sensors installed on the perforation string and pressure measurement by means of a pressure sensor installed at the lower end of the lower section of the perforation string, - the readings of the pressure sensor installed at the lower end of the lower section of the perforation string are averaged, and the average dependence of the pressure on time is obtained, - calculate the average speed of the perforation string running into the well and the temperature gradient of the well fluid along the length of the well using the obtained averaged pressure versus time dependence and the temperature measured by the sensor installed at the lower end of the lower section of the perforation string, - select the time interval during running the perforating string into the well, during which the average speed of running the perforating string into the well and the temperature gradient of the well fluid remain constant, - in the selected time interval, the temperature profiles measured by the sensors are shifted by temperature values proportional to the difference in recording times, so as to ensure the best coincidence of all shifted temperature profiles with the temperature profile at the selected time, - calculate the average sensor readings for shifted temperature profiles, - using the found average speed of the perforation string running into the well and the temperature gradient of the well fluid, the steady-state temperature gradient along the running string is calculated, - select a temperature sensor, relative to which the calibration will be carried out, and build a calibration line passing through the selected temperature sensor and having a calculated steady-state temperature gradient along the running string, - Calibration corrections to the results of temperature measurement by temperature sensors are calculated as the deviation of the calculated averaged sensor readings from the constructed calibration straight line and mutual calibration of temperature sensors relative to the selected temperature sensor is performed. 2. The method according to claim 1, according to which the averaging of the readings of the pressure sensor installed at the lower end of the lower section of the perforation string is carried out by the sliding window method, using the time interval corresponding to the time of joining 3-4 sections to the perforating string as a sliding window. 3. The method according to claim 1, in accordance with which short-term injection of fluid into the well is carried out during shutdowns in the process of lowering the perforation string into the well, the temperature change is measured for each temperature sensor and, using the measured dynamics of the temperature increase of the sensors, the time constants of each sensor are determined for use in interpreting non-stationary temperature data during perforation and well testing. 4. The method according to claim 1, in accordance with which a short-term injection of fluid into the well is carried out after the completion of lowering the perforation string into the well, the temperature change is measured for each temperature sensor and, using the measured dynamics of the temperature increase of the sensors, the time constants of each sensor are determined, taking into account the position casing against the borehole wall, which will affect the readings of the sensors and the interpretation of non-stationary temperature data during perforation and subsequent well testing.

Images