A Method and System for Through-Bus Resistivity Measurement Based on Multi-Source Spatiotemporal Joint Optimization

Abstract

This invention belongs to the field of geophysical logging technology. It relates to a method and system for measuring resistivity through casing based on multi-source spatiotemporal joint optimization, comprising two stages: offline table construction and downhole logging. In the offline stage, the target observation time window is simulated and calibrated. Using the formation signal proportion as the fitness function, a genetic algorithm is employed to optimize the number of turns of each transmitting coil and the baseline transmission delay time. Further optimization is performed under various casing environmental conditions. The double logarithmic slope of the induced electromotive force in the very early time window is extracted as an environmental feature fingerprint. A multidimensional discrete mapping table is constructed by binding the transmission delay time, and an apparent resistivity interpretation chart of the induced electromotive force amplitude and formation resistivity is established through forward modeling. During downhole logging, the environmental feature fingerprint is extracted in real time, the transmission time sequence is obtained by looking up the table and dynamically reconfigured. The target observation time window signal is collected and substituted into the apparent resistivity interpretation chart to quickly invert the formation apparent resistivity. This invention solves the environmental mismatch problem under complex and variable conditions, balancing logging accuracy and real-time performance.

Description

Technical Field

[0001] This invention belongs to the field of geophysical logging technology, and relates to electrical logging technology. Specifically, it relates to a method and system for measuring resistivity through casing based on multi-source spatiotemporal joint optimization. Background Technology

[0002] Through-casing transient electromagnetic resistivity logging has become an emerging technology for residual oil evaluation and dynamic monitoring because it enables continuous measurement without contacting the casing wall. However, the high conductivity and magnetic permeability of the metal casing can cause severe shielding and aliasing effects on formation signals, and extracting weak formation responses is a key bottleneck for the practical application of this technology.

[0003] Currently, the industry mainly relies on spatial multi-receiver coil combinations to cancel casing interference through spatial differential algorithms. This type of method heavily depends on the detection mode of the static differential array, and its underlying logic is based on the prior assumption of "absolute stability of the baseline casing environment." In real deep well complex conditions, the casing environment is dynamically changing. Strong geostress concentration can cause nonlinear distortion of the casing's relative permeability, and high-mineralization formation fluids can induce casing wall corrosion and thinning. These variations alter the system's characteristic time constant and spatial eddy current diffusion rate, causing the pre-fixed suppression window to drift or even fail, and background noise instantly engulfing weak formation signals. Summary of the Invention

[0004] This invention addresses the problems existing in the prior art by providing a method and system for measuring resistivity through casing based on multi-source spatiotemporal joint optimization. By dynamically adjusting the temporal degrees of freedom of the transmitter, it solves the environmental mismatch problem under complex and variable working conditions, while taking into account both logging accuracy and real-time performance.

[0005] In a first aspect, the present invention provides a method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization, the steps of which are as follows: Step 1: Establish a two-dimensional axisymmetric forward model for transient electromagnetic logging through the casing, including the wellbore, casing, and formation; Step 2: Construct an n-transmitter single-receiver coil detector, n≥3, and calibrate the target observation time window that maximizes the distinguishability of the formation information based on the formation signal strength at the source distance of the main transmitting coil. Step 3: Under standard casing conditions, global optimization is performed using a genetic algorithm. The proportion of formation signals within the target observation time window is used as the fitness function to determine the number of turns of each transmitting coil and the reference transmission delay time in the n-transmitter single-receiver coil detector. Step 4: Change the different casing environment conditions, and perform global optimization again through genetic algorithm to determine the different transmission delay times corresponding to each casing environment condition; extract the double logarithmic slope of the induced electromotive force curve in the very early time window of 5-10ms under each casing environment condition as the environmental feature fingerprint, and deeply bind the environmental feature fingerprint with the corresponding transmission delay time to construct a multidimensional discrete mapping table. Step 5: Substitute the time series parameters corresponding to the multidimensional discrete mapping table into the forward model, perform forward modeling by traversing different formation resistivity, extract the absolute amplitude of the combined induced electromotive force within the target observation time window, and establish an apparent resistivity interpretation chart between the absolute amplitude of the induced electromotive force and the formation resistivity. Step 6: During actual downhole casing measurement, the double logarithmic slope of the induced electromotive force curve within the very early time window of 5-10ms is measured using an n-transmitter single-receiver coil detector. Step 7: The double logarithmic slope of the induced electromotive force curve measured in the actual well is used as an environmental feature fingerprint and entered into the multidimensional discrete mapping table for lookup matching to obtain the corresponding launch delay time command under the current casing environment, and the multi-source launch timing is reconfigured in real time. Step 8: Based on the reconfigured multi-source emission timing, measure the absolute amplitude of the combined induced electromotive force within the target observation time window, and substitute it into the apparent resistivity interpretation chart for inversion to obtain the actual resistivity of the formation.

[0006] In some embodiments, the method for calibrating the target observation time window in step two includes: An n-transmitter single-receiver coil detector is constructed. The coil detector includes a main transmitting coil, m pairs of auxiliary transmitting coils, and a receiving coil, where m ≥ 1. Each pair of auxiliary transmitting coils includes a symmetrically configured first auxiliary transmitting coil and a second auxiliary transmitting coil. The source distance of the main transmitting coil is fixed at 1.0m. The source distance of the first auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 0.7m, and the source distance of the second auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 1.3m. The source distance of the first auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 0.7m and less than 1.0m, and they are arranged in sequence. The source distance of the second auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 1.0m and less than 1.3m, and they are arranged in sequence. In a single-transmitter, single-receiver coil structure with a main transmitting coil source-to-source distance of 1.0m, simulations were performed for resistivity greater than 100... High resistivity strata and resistivity less than 10 The total field response of low-resistivity strata; Extract the total field response difference signal between the high-resistivity and low-resistivity formations; The interval in which the total field response difference signal reaches its peak is defined as the target observation time window.

[0007] In some embodiments, the method for determining the number of turns of each transmitting coil and the reference transmission delay time in step three includes: Set spatial turns constraints and time delay constraints, wherein the time delay constraint requires that the electromagnetic signal emitted by the auxiliary transmitting coil arrives at the receiving coil before the target observation time window opens; Using the maximization of the proportion of the formation signal at the receiving coil within the target observation time window as the fitness function, a genetic algorithm is used to perform global optimization iteration within the set spatial turns constraint and time delay constraint. When the iteration reaches the preset termination condition, the number of turns of each transmitting coil and the reference transmission delay time in the n-transmitter single-receiver coil detector are obtained.

[0008] In some embodiments, the spatial turns constraint is as follows: the number of turns of the main transmitting coil is limited to the range of [0, 100] turns, and the number of turns of the auxiliary transmitting coil is limited to the range of [-100, 0] turns.

[0009] In some embodiments, the time delay constraint is to delay the time within the interval [11, 34].

[0010] In some embodiments, the preset termination condition is: the number of iterations reaches a preset maximum number of iterations.

[0011] In some embodiments, the method for extracting the double logarithmic slope of the induced electromotive force curve in step four includes: The one-dimensional time series of induced electromotive force of the main wave field in the very early time window of 5-10ms under various bushing environmental conditions was extracted to obtain the induced electromotive force curve. The slope of the tangent line to the induced electromotive force curve in a double logarithmic coordinate system is expressed as:

[0012] In the formula, Let be the slope of the tangent line to the induced electromotive force curve. for Constantly sensing electromotive force. For time.

[0013] In some embodiments, the method for establishing an apparent resistivity interpretation chart in step five includes: plotting response curves for different casing environment conditions with the absolute amplitude of the induced electromotive force as the vertical axis and the formation resistivity as the horizontal axis, thereby forming an apparent resistivity interpretation chart.

[0014] In a second aspect, the present invention provides a through-sleeve resistivity measurement system based on multi-source spatiotemporal joint optimization, used to implement the through-sleeve resistivity measurement method based on multi-source spatiotemporal joint optimization described in the first aspect of the present invention, comprising: The model building module is configured to: establish a two-dimensional axisymmetric forward model of transient electromagnetic logging through the casing, including the wellbore, casing, and formation; The n-transmitter single-receiver coil detector is used to measure the formation signal intensity under the source distance of the main transmitting coil, and to measure the double logarithmic slope of the induced electromotive force curve within a very early time window of 5-10ms during actual downhole casing measurement; based on the reconfigured multi-source transmission timing, the absolute amplitude of the combined induced electromotive force is measured within the target observation time window. The calibration module is configured to: calibrate the target observation time window that maximizes the distinguishability of the formation information based on the formation signal strength at the source distance of the main transmitting coil; The parameter determination module is configured to: perform global optimization using a genetic algorithm under a standard casing environment, and determine the number of turns and reference transmission delay time of each transmitting coil in the n-transmitter single-receiver coil detector using the proportion of formation signals within the target observation time window as the fitness function; The mapping module is configured to: change different casing environment conditions, perform global optimization again through genetic algorithm to determine the different transmission delay times corresponding to each casing environment condition; extract the double logarithmic slope of the induced electromotive force curve in the very early time window of 5-10ms under each casing environment condition as the environmental feature fingerprint, and deeply bind the environmental feature fingerprint with the corresponding transmission delay time to construct a multidimensional discrete mapping table. The chart generation module is configured to: substitute each time series parameter corresponding to the multidimensional discrete mapping table into the forward model, perform forward modeling calculations by traversing different formation resistivity, extract the absolute amplitude of the combined induced electromotive force within the target observation time window, and establish an apparent resistivity interpretation chart between the absolute amplitude of the induced electromotive force and the formation resistivity. The configuration module is set to: use the double logarithmic slope of the induced electromotive force curve measured in the actual well as an environmental feature fingerprint to enter the multidimensional discrete mapping table for lookup matching, obtain the corresponding launch delay time command under the current casing environment, and reconfigure the multi-source launch timing in real time; The resistivity calculation module is configured to: substitute the absolute amplitude of the combined induced electromotive force into the apparent resistivity interpretation chart for inversion to obtain the actual resistivity of the formation.

[0015] In some embodiments, the n-transmitter single-receiver coil detector includes a main transmitting coil, m pairs of auxiliary transmitting coils (m≥1), and a receiving coil. Each pair of auxiliary transmitting coils includes a symmetrically configured first auxiliary transmitting coil and a second auxiliary transmitting coil. The source distance of the main transmitting coil is fixed at 1.0m. The source distance of the first auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 0.7m, and the source distance of the second auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 1.3m. The source distance of the first auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 0.7m and less than 1.0m, and they are arranged in sequence. The source distance of the second auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 1.0m and less than 1.3m, and they are arranged in sequence.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: The present invention provides a method and system for measuring resistivity through casing based on multi-source spatiotemporal joint optimization, comprising an offline numerical simulation stage for constructing a multidimensional discrete mapping table and a downhole actual logging stage. In the offline numerical simulation stage for constructing the multidimensional discrete mapping table, in an n-transmitter single-receiver coil detector, the target observation window most favorable for extracting formation information is first determined in the numerical simulation by the formation signal intensity at the source distance of the main transmitting coil. Subsequently, a genetic algorithm is used to optimize and determine the number of turns and the reference transmission time of each transmitting coil using the proportion of formation signal within this time window as the fitness function. Based on this, different casing environment conditions are simulated, and the genetic algorithm is used again to optimize and determine different transmission times under different casing environment conditions. Simultaneously, the double logarithmic slope of the induced electromotive force within the very early time window of 5-10 ms under each condition is extracted as an environmental feature fingerprint and bound to the corresponding transmission time depth to establish a multidimensional discrete mapping table. The corresponding time-series parameters in the multidimensional discrete mapping table are substituted into the constructed forward model for forward modeling, forming an apparent resistivity interpretation chart between the absolute amplitude of the combined induced electromotive force and the formation resistivity. During the actual downhole logging phase, the double logarithmic slope of the induced electromotive force (EMF) within the very early time window of 5-10 ms is measured in real time. This slope is used to directly query a multidimensional discrete mapping table to quickly obtain the multi-source emission time command corresponding to the current environment, and the emission timing is reconfigured. Combined induced EMFs are collected within the target observation time window, and by substituting them into the apparent resistivity interpretation chart, accurate formation apparent resistivity can be quickly obtained through inversion. This invention, on the one hand, breaks the rigid constraints of physical hardware and overcomes the strong dependence of traditional pure spatial differential arrays on fixed physical parameters. When facing casing magnetic anomalies or corrosion thinning, adaptive compensation can be achieved by reconfiguring the multi-source emission timing, avoiding interference window failure. On the other hand, it innovatively extracts the double logarithmic slope of the induced EMF curve within the very early time window of 5-10 ms as an environmental feature fingerprint, binding it with the optimized emission delay time to form a multidimensional discrete mapping table. This reduces the dimensionality of the massive forward iterative model, significantly reducing the real-time computational load in downhole, effectively adapting to the computing power constraints of downhole embedded hardware, and ensuring the real-time performance of logging signal processing. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of the first aspect of the present invention, which is a method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization. Figure 2 This is a schematic diagram of the forward modeling computational model described in an embodiment of the present invention; Figure 3 This is a schematic flowchart of the method for calibrating the target observation time window according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of the transmit-receive coil detector in Embodiment 3 of the present invention; Figure 5 A schematic flowchart illustrating the method for determining the number of turns of each transmitting coil and the reference transmission delay time in an embodiment of the present invention; Figure 6 This is a structural block diagram of the through-sleeve resistivity measurement system based on multi-source spatiotemporal joint optimization according to an embodiment of the present invention; Figure 7 This is a diagram illustrating the adaptive interferometric reconstruction effect under the casing corrosion and thinning condition according to an embodiment of the present invention. Figure 8 This is a diagram illustrating the calculation of apparent resistivity in an embodiment of the present invention.

[0018] In the diagram, 1. Wellbore, 2. Casing, 3. Coil, 4. Formation, 5. Magnetic core, 6. Transient electromagnetic wave resistivity measurement system through casing, 61. Model building module, 62. n-transmitter single-receiver coil detector, 63. Calibration module, 64. Parameter determination module, 65. Mapping module, 66. Chart generation module, 67. Configuration module, 68. Resistivity calculation module. Detailed Implementation

[0019] The present invention will now be described in detail with reference to the accompanying drawings through exemplary embodiments. However, it should be understood that, without further description, elements, structures, and features in one embodiment may be advantageously incorporated into other embodiments.

[0020] See Figure 1 The first embodiment of the present invention provides a method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization, comprising: S1, Step 1: Establish a two-dimensional axisymmetric through-casing transient electromagnetic logging forward model including wellbore 1, casing 2, and formation 4 (see...). Figure 2 ).

[0021] S2, Step 2: Construct an n-transmitter single-receiver coil detector, n≥3, and calibrate the target observation time window that maximizes the differentiation of the formation information based on the formation signal strength at the source distance of the main transmitting coil.

[0022] Specifically, in one embodiment of the present invention, see [link to relevant documentation]. Figure 3 Methods for determining the target observation time window include: S21. Construct an n-transmitter single-receiver coil detector, wherein the coil detector includes a main transmitting coil, m pairs of auxiliary transmitting coils, and a receiving coil, where m ≥ 1. Each pair of auxiliary transmitting coils includes a symmetrically configured first auxiliary transmitting coil and a second auxiliary transmitting coil. The source distance of the main transmitting coil is fixed at 1.0m. The source distance of the first auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 0.7m, and the source distance of the second auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 1.3m. The source distance of the first auxiliary transmitting coil in other pairs of auxiliary coils is greater than 0.7m and less than 1.0m, and they are arranged in sequence. The source distance of the second auxiliary transmitting coil in other pairs of auxiliary coils is greater than 1.0m and less than 1.3m, and they are arranged in sequence. S22. In a single-transmitter, single-receiver coil structure with a main transmitting coil source-to-source distance of 1.0m, simulate resistivity greater than 100... High resistivity strata and resistivity less than 10 The total field response of low-resistivity strata; S23. Extract the total field response difference signal between the high-resistivity strata and the low-resistivity strata; S24. The interval in which the total field response difference signal reaches its peak is defined as the target observation time window.

[0023] In this embodiment of the invention, an n-transmitter single-receiver detection structure is formed by the main transmitting coil and multiple pairs of symmetrical auxiliary transmitting coils. This structure can perform gradient control of the electromagnetic field in the detection space, weaken the interference field of non-target formations such as surrounding rock and wellbore, and improve the spatial resolution of the formation electromagnetic response signal. The total field response of high and low resistivity formations is simulated and the difference signal is extracted. The peak interval of the difference signal is calibrated as the target observation time window. This window can maximize the highlighting of the electrical differences of the formation, increase the proportion and purity of the effective formation signal, suppress primary field and clutter interference, and improve the accuracy of formation identification and resistivity inversion. Moreover, the window calibration is completed in advance based on simulation, eliminating the need for repeated on-site debugging of time parameters and reducing the workload of on-site debugging.

[0024] For example, see Figure 4 The coil detector includes a main transmitting coil Tx1, a symmetrically arranged first auxiliary transmitting coil Tx2 and second auxiliary transmitting coil Tx3, and a receiving coil Rx. The source distance of the main transmitting coil Tx1 is fixed at 1.0m, the source distance of the first auxiliary transmitting coil Tx2 is fixed at 0.7m, and the source distance of the second auxiliary transmitting coil Tx3 is fixed at 1.3m.

[0025] For example, the coil detector includes a main transmitting coil, two pairs of auxiliary transmitting coils, and a receiving coil. Each pair of auxiliary transmitting coils includes a first auxiliary transmitting coil and a second auxiliary transmitting coil arranged symmetrically. The source distance of the main transmitting coil is fixed at 1.0m. The source distance of the first auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 0.7m and the source distance of the second auxiliary transmitting coil is fixed at 1.3m. The source distance of the first auxiliary transmitting coil in the second pair of auxiliary transmitting coils is fixed at 0.8m and the source distance of the second auxiliary transmitting coil is fixed at 1.2m.

[0026] S3. Step 3: Under standard casing conditions, global optimization is performed using a genetic algorithm. The proportion of formation signals within the target observation time window is used as the fitness function to determine the number of turns of each transmitting coil and the reference transmission delay time in the n-transmitter single-receiver coil detector.

[0027] Specifically, see Figure 5 Methods for determining the number of turns of each transmitting coil and the reference transmission delay time include: S31. Set spatial turns constraint and time delay constraint, wherein the time delay constraint requires that the electromagnetic signal emitted by the auxiliary transmitting coil arrives at the receiving coil before the target observation time window opens; S32. Using the maximization of the proportion of the formation signal at the receiving coil within the target observation time window as the fitness function, a genetic algorithm is used to perform global optimization iteration within the set spatial turns constraint and time delay constraint. S33. When the iteration reaches the preset termination condition, the number of turns of each transmitting coil and the reference transmission delay time in the n-transmitter single-receiver coil detector are obtained.

[0028] In this embodiment of the invention, by setting spatial turns constraints and time delay constraints, the electromagnetic signal of the auxiliary transmitting coil arrives at the receiving coil before the target observation time window opens, avoiding interference with the formation signal within the observation window. Using the maximization of the formation signal proportion within the observation window as the fitness function, a genetic algorithm is employed to globally optimize within the constraints, iteratively solving for the turns number of each transmitting coil and the reference transmission delay time. This method can automatically obtain the optimal parameter combination for multi-transmitter single-receiver coil detectors, effectively eliminating interference from auxiliary transmitting signals, increasing the proportion of effective formation signals in the received signal, enhancing the formation electromagnetic response resolution capability, reducing the workload of manual calculation and debugging, and improving the accuracy of formation resistivity inversion and formation identification.

[0029] Specifically, the spatial turns constraint is as follows: the number of turns of the main transmitting coil is limited to the range of [0, 100] turns, and the number of turns of the auxiliary transmitting coil is limited to the range of [-100, 0] turns. The time delay constraint is as follows: the time delay is within the range of [11, 34]. The preset termination condition is as follows: the number of iterations reaches the preset maximum number of iterations. It should be noted that the preset maximum number of iterations can be set according to actual needs.

[0030] S4, Step 4: Change the different casing environment conditions, and perform global optimization again through genetic algorithm to determine the different transmission delay times corresponding to each casing environment condition; extract the double logarithmic slope of the induced electromotive force curve in the very early time window of 5-10ms under each casing environment condition as the environmental feature fingerprint, and deeply bind the environmental feature fingerprint with the corresponding transmission delay time to construct a multidimensional discrete mapping table.

[0031] This invention innovatively extracts the double logarithmic slope of the induced electromotive force (EMF) curve within a very early time window of 5-10ms (during which the induced EMF is highly sensitive to changes in casing conditions) as an environmental feature fingerprint, which is then bound to the optimized launch delay time to form a multidimensional discrete mapping table. On the one hand, this reduces the dimensionality of the massive forward modeling iteration, significantly decreasing the real-time computational load in downhole operations, effectively adapting to the computing power constraints of downhole embedded hardware, and ensuring the real-time performance of logging signal processing. On the other hand, by establishing a one-to-one mapping relationship between the environmental fingerprint and the optimal launch delay time, after acquiring the induced EMF curve downhole, the double logarithmic slope feature is extracted, and the corresponding optimal launch delay time can be quickly matched by directly searching the multidimensional discrete mapping table. This enables automatic adjustment of the launch timing according to changes in the casing environment, improving the quality of electromagnetic detection signals, weakening casing environment interference, and improving the formation signal extraction effect of high and low resistivity formation total field response.

[0032] Specifically, methods for extracting the double logarithmic slope of the induced electromotive force curve include: The one-dimensional time series of induced electromotive force of the main wave field in the very early time window of 5-10ms under various bushing environmental conditions was extracted to obtain the induced electromotive force curve. The slope of the tangent line to the induced electromotive force curve in a double logarithmic coordinate system is expressed as:

[0033] In the formula, Let be the slope of the tangent line to the induced electromotive force curve. for Constantly sensing electromotive force. For time.

[0034] S5. Step 5: Substitute the time series parameters corresponding to the multidimensional discrete mapping table into the forward modeling calculation model, perform forward modeling calculations by traversing different formation resistivity, extract the absolute amplitude of the combined induced electromotive force within the target observation time window, and establish an apparent resistivity interpretation chart between the absolute amplitude of the induced electromotive force and the formation resistivity.

[0035] In this embodiment of the invention, the established apparent resistivity interpretation chart uses optimal time series parameters that match the casing operating conditions. The forward modeling conditions are consistent with the downhole measured conditions, eliminating interpretation errors introduced by time series parameter mismatch. This effectively suppresses casing environment interference, amplifies the differences in induced electromotive force response between high and low resistivity formations, and improves the accuracy of formation electrical property identification. The chart construction is completed offline on the surface, and quantitative interpretation can be achieved downhole only by looking up tables, balancing the real-time performance of signal processing and interpretation accuracy.

[0036] Specifically, the method for establishing an apparent resistivity interpretation chart includes: plotting response curves for different casing environment conditions with the absolute amplitude of the induced electromotive force as the vertical axis and the formation resistivity as the horizontal axis, thus forming an apparent resistivity interpretation chart.

[0037] S6, Step Six: During actual downhole casing measurement, the double logarithmic slope of the induced electromotive force curve within the very early time window of 5-10ms is measured using an n-transmitter single-receiver coil detector.

[0038] S7. Step 7: Use the double logarithmic slope of the induced electromotive force curve measured in the actual well as an environmental feature fingerprint to enter the multidimensional discrete mapping table for lookup matching, obtain the corresponding launch delay time command under the current casing environment, and reconfigure the multi-source launch timing in real time.

[0039] S8. Step 8: Based on the reconfigured multi-source emission timing, measure the absolute amplitude of the combined induced electromotive force within the target observation time window, and substitute it into the apparent resistivity interpretation chart for inversion to obtain the actual resistivity of the formation.

[0040] See Figure 6 According to a second aspect of the present invention, a through-sleeve resistivity measurement system based on multi-source spatiotemporal joint optimization is provided to implement the through-sleeve resistivity measurement method based on multi-source spatiotemporal joint optimization described in the first aspect of the present invention.

[0041] The transient electromagnetic wave resistivity measurement system 6 through the bushing includes: Model building module 61 is configured to: establish a two-dimensional axisymmetric forward model of transient electromagnetic logging through the casing, including the wellbore, casing and formation; The n-transmitter single-receiver coil detector 62 is used to measure the formation signal intensity under the source distance of the main transmitting coil, and to measure the double logarithmic slope of the induced electromotive force curve within a very early time window of 5-10ms during actual downhole casing measurement; based on the reconfigured multi-source transmission timing, the absolute amplitude of the combined induced electromotive force is measured within the target observation time window. The calibration module 63 is configured to: calibrate the target observation time window that maximizes the distinguishability of the formation information based on the formation signal strength at the source distance of the main transmitting coil; The parameter determination module 64 is configured to: perform global optimization using a genetic algorithm under a standard casing environment, and determine the number of turns and reference transmission delay time of each transmitting coil in the n-transmitter single-receiver coil detector using the proportion of formation signals within the target observation time window as the fitness function; The mapping module 65 is configured to: change different casing environment conditions, perform global optimization again through genetic algorithm, determine the different transmission delay times corresponding to each casing environment condition; extract the double logarithmic slope of the induced electromotive force curve in the 5-10ms very early time window of each casing environment condition as the environmental feature fingerprint, and deeply bind the environmental feature fingerprint with the corresponding transmission delay time to construct a multi-dimensional discrete mapping table. The plot generation module 66 is configured to: substitute each time series parameter corresponding to the multidimensional discrete mapping table into the forward model, perform forward modeling calculations by traversing different formation resistivity, extract the absolute amplitude of the combined induced electromotive force within the target observation time window, and establish an apparent resistivity interpretation plot between the absolute amplitude of the induced electromotive force and the formation resistivity. Configuration module 67 is configured to: retrieve the double logarithmic slope of the induced electromotive force curve measured in actual downhole as an environmental feature fingerprint into the multidimensional discrete mapping table for lookup matching, obtain the corresponding launch delay time command under the current casing environment, and reconfigure the multi-source launch timing in real time; The resistivity calculation module 68 is configured to: substitute the absolute amplitude of the combined induced electromotive force into the apparent resistivity interpretation chart for inversion to obtain the actual resistivity of the formation.

[0042] Specifically, the n-transmitter single-receiver coil detector includes a main transmitting coil, m pairs of auxiliary transmitting coils (m≥1), and a receiving coil. Each pair of auxiliary transmitting coils includes a symmetrically configured first auxiliary transmitting coil and a second auxiliary transmitting coil. The source distance of the main transmitting coil is fixed at 1.0m. The source distance of the first auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 0.7m, and the source distance of the second auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 1.3m. The source distance of the first auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 0.7m and less than 1.0m, and they are arranged in sequence. The source distance of the second auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 1.0m and less than 1.3m, and they are arranged in sequence.

[0043] The effectiveness of the above-mentioned multi-source spatiotemporal joint optimization method for measuring resistivity through a bushing, as described in the present invention, will be explained below with specific examples.

[0044] Example: A method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization, the steps of which are as follows: S1, see below. Figure 2A two-dimensional axisymmetric forward model for through-casing transient electromagnetic logging was established, including wellbore 1, casing 2, and formation 4. The casing model used was the P110TXP BTC casing model. The casing outer diameter was 139.7 mm, the casing thickness was 9.17 mm, and the casing conductivity was 10. 7 S / m, relative permeability 100, mud resistivity 1Ω·m.

[0045] S2. Construct a three-transmitter, single-receiver coil detector. See below. Figure 4 The coil detector includes a main transmitting coil Tx1, symmetrically arranged first auxiliary transmitting coils Tx2 and second auxiliary transmitting coils Tx3, and a receiving coil Rx. Magnetic core material is incorporated into both the transmitting and receiving coils. The source-to-source distance of the main transmitting coil Tx1 is fixed at 1.0m, the source-to-source distance of the first auxiliary transmitting coil Tx2 at 0.7m, and the source-to-source distance of the second auxiliary transmitting coil Tx3 at 1.3m. The radius of the transmitting and receiving coils is 0.03m, and the current intensity is 1A. Under a single-transmitter, single-receiver coil structure with a source-to-source distance of 1.0m for the main transmitting coil Tx1, simulations were performed with a resistivity of 300... High resistivity strata and 1 The total field response of the low-resistivity strata was analyzed. By extracting the difference signal between the total field response of the high-resistivity and low-resistivity strata, it was found that the difference reached a peak at 39.8ms. In order to maximize the preservation of useful information of the strata and take into account the signal-to-noise ratio, 35-50ms was defined as the target observation time window.

[0046] S3, in a standard bushing environment (relative permeability) Casing wall thickness Under the given conditions, a genetic algorithm is run. The fitness function of the genetic algorithm is to maximize the proportion of the formation signal at the receiving coil within the target observation time window of 35-50ms. Global optimization iterations are performed under preset spatial turns constraints (the number of turns of the main transmitting coil is limited to the range of [0,100] turns, and the number of turns of the auxiliary transmitting coil is limited to the range of [-100,0] turns) and time delay constraints (the time delay is within the range of [11,34]). At the same time, hardware digital quantization constraints are actively applied to lock the time search step size (i.e., resolution) to 0.01ms. When the iteration reaches the preset termination condition, the number of turns of the main transmitting coil is 96, the number of turns of the first auxiliary coil is -56, the number of turns of the second auxiliary coil is -76, and the reference transmission delay time of the main transmitting coil is obtained. The reference transmission delay time of the first auxiliary transmitting coil The reference transmission delay time of the second auxiliary transmitting coil .

[0047] S4. Maintaining the physical structure of the hardware coil turns established in step S3, modify the casing environment parameters to simulate actual downhole anomalies (including enhanced casing magnetism, casing magnetic degradation, moderate internal wall corrosion, and severe extreme corrosion). For each anomaly, release the temporal degrees of freedom and use a genetic algorithm to re-scan and optimize within the constraint space to obtain the new launch delay time combination that best cancels out the interference of the target observation time window under each condition. ).like Figure 7 As shown, the ultimate corrosion of the casing (casing wall thickness) (mm) This can cause the unadapted reference interferometric valley to shift severely to the right, thus completely deviating from the target observation time window. After time-series reconfiguration using a genetic algorithm (thick red and blue solid lines), the lagging interferometric valleys are accurately pulled back and strictly anchored within the target observation time window. Simultaneously, a one-dimensional time series of the induced electromotive force (EMF) within the very early 5-10 ms time window of the main transmitting wavefield (i.e., only the main transmitting coil transmits the signal) under various casing environmental conditions is extracted. Based on this one-dimensional time series of the induced EMF, the double logarithmic slope of the induced EMF curve for this segment is calculated in a double logarithmic coordinate system. Since this early slope is absolutely dominated by casing parameters, it can be used as an environmental feature fingerprint of the pole drift trend during mapping. The slopes under various typical casing environmental conditions are then analyzed. The optimal launch delay time obtained by each of the two quadratic genetic algorithms is deeply bound to the multidimensional discrete mapping table shown in Table 1.

[0048] Table 1 Multidimensional Discrete Mapping Table

[0049] S5. Substitute the time-series parameters corresponding to the multidimensional discrete mapping table into the forward model. Under the corresponding casing environment conditions, simulate different real formation resistivities (e.g., 1). -300 The absolute amplitude characteristics of the combined induced electromotive force (EMF) within the 35-50ms target observation time window under the corresponding reconfigured launch sequence were extracted. The absolute amplitude of the combined induced EMF was plotted on the vertical axis, and the formation resistivity was used as the horizontal axis. Using the horizontal axis as the plot, response curves corresponding to different bushing operating conditions are drawn, ultimately forming an apparent resistivity interpretation chart, such as... Figure 8 As shown.

[0050] S6. The three-transmitter, single-receiver coil detector is lowered into the well along with the logging cable to conduct continuous measurements at the target formation. Within a very early time window of 5-10 ms after the main transmitting wave field is turned off, the original sequence of induced electromotive force at the receiving coil is acquired in real time, and immediately linearly fitted and differentiated in a double logarithmic coordinate system to extract the double logarithmic slope of the actual induced electromotive force curve at the current measurement point. .

[0051] S7, the downhole embedded CPU will measure the slope. As an environmental feature fingerprint, it is retrieved into a multidimensional discrete mapping table for lookup and matching. Based on the matching result, the optimal auxiliary transmitting coil transmission delay time command under the current bushing environment condition is directly obtained. The trigger timing of the three transmitting coils is reconfigured in real time by a digital timing generator. Through this timing reconfiguration, under the current unknown variation of the casing environment, the interference of the three-source coherent wave field is forcibly generated, and the millisecond-level offset of the interference cancellation valley caused by environmental mismatch is accurately "pulled back" and strictly anchored within the 35-50m target observation window.

[0052] S8. Under the reconfigured time series, full-window excitation acquisition is performed to accurately extract the combined induced electromotive force amplitude characteristics within the 35-50ms target observation window after perfect physical background suppression. Subsequently, the measured combined induced electromotive force amplitude characteristics are substituted into the apparent resistivity interpretation chart. The interpretation curve matching the current working condition is selected for single-value fast inversion lookup table, and the actual formation apparent resistivity of the current layer with high fidelity and high signal-to-noise ratio is directly calculated.

[0053] The above embodiments are used to explain the present invention, but not to limit the present invention. Any modifications and changes made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization, characterized in that, include: Step 1: Establish a two-dimensional axisymmetric forward model for transient electromagnetic logging through the casing, including the wellbore, casing, and formation; Step 2: Construct an n-transmitter single-receiver coil detector, n≥3. Based on the formation signal strength under the source distance of the main transmitting coil, calibrate the target observation time window that maximizes the differentiation of formation information. Step 3: Under standard casing conditions, global optimization is performed using a genetic algorithm. The proportion of formation signals within the target observation time window is used as the fitness function to determine the number of turns of each transmitting coil and the reference transmission delay time in the n-transmitter single-receiver coil detector. Step 4: Change the different casing environment conditions, and perform global optimization again through genetic algorithm to determine the different transmission delay times corresponding to each casing environment condition; extract the double logarithmic slope of the induced electromotive force curve in the very early time window of 5-10ms under each casing environment condition as the environmental feature fingerprint, and deeply bind the environmental feature fingerprint with the corresponding transmission delay time to construct a multidimensional discrete mapping table. Step 5: Substitute the time series parameters corresponding to the multidimensional discrete mapping table into the forward model, perform forward modeling by traversing different formation resistivity, extract the absolute amplitude of the combined induced electromotive force within the target observation time window, and establish an apparent resistivity interpretation chart between the absolute amplitude of the induced electromotive force and the formation resistivity. Step 6: During actual downhole casing measurement, the double logarithmic slope of the induced electromotive force curve within the very early time window of 5-10ms is measured using an n-transmitter single-receiver coil detector. Step 7: The double logarithmic slope of the induced electromotive force curve measured in the actual well is used as an environmental feature fingerprint and entered into the multidimensional discrete mapping table for lookup matching to obtain the corresponding launch delay time command under the current casing environment, and the multi-source launch timing is reconfigured in real time. Step 8: Based on the reconfigured multi-source emission timing, measure the absolute amplitude of the combined induced electromotive force within the target observation time window, and substitute it into the apparent resistivity interpretation chart for inversion to obtain the actual resistivity of the formation.

2. The method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization as described in claim 1, characterized in that, In step two, the methods for determining the target observation time window include: An n-transmitter single-receiver coil detector is constructed. The coil detector includes a main transmitting coil, m pairs of auxiliary transmitting coils, and a receiving coil, where m ≥ 1. Each pair of auxiliary transmitting coils includes a symmetrically configured first auxiliary transmitting coil and a second auxiliary transmitting coil. The source distance of the main transmitting coil is fixed at 1.0m. The source distance of the first auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 0.7m, and the source distance of the second auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 1.3m. The source distance of the first auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 0.7m and less than 1.0m, and they are arranged in sequence. The source distance of the second auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 1.0m and less than 1.3m, and they are arranged in sequence. In a single-transmitter, single-receiver coil structure with a main transmitting coil source-to-source distance of 1.0m, simulations were performed for resistivity greater than 100 Ω·m. High resistivity strata and resistivity less than 10 The total field response of low-resistivity strata; Extract the total field response difference signal between the high-resistivity and low-resistivity formations; The interval in which the total field response difference signal reaches its peak is defined as the target observation time window.

3. The method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization as described in claim 1, characterized in that, Step three involves determining the number of turns for each transmitting coil and the reference transmission delay time, including: Set spatial turns constraints and time delay constraints, wherein the time delay constraint requires that the electromagnetic signal emitted by the auxiliary transmitting coil arrives at the receiving coil before the target observation time window opens; Using the maximization of the proportion of the formation signal at the receiving coil within the target observation time window as the fitness function, a genetic algorithm is used to perform global optimization iteration within the set spatial turns constraint and time delay constraint. When the preset termination condition is met, the number of turns of each transmitting coil and the reference transmission delay time in the n-transmitter single-receiver coil detector are obtained.

4. The method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization as described in claim 3, characterized in that, The spatial turns constraint is as follows: the number of turns of the main transmitting coil is limited to the range of [0, 100] turns, and the number of turns of the auxiliary transmitting coil is limited to the range of [-100, 0] turns.

5. The method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization as described in claim 3, characterized in that, The time delay constraint is: the time delay is within the interval [11, 34].

6. The method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization as described in claim 3, characterized in that, The preset termination condition is: the number of iterations reaches the preset maximum number of iterations.

7. The method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization as described in claim 1, characterized in that, In step four, the method for extracting the double logarithmic slope of the induced electromotive force curve includes: The one-dimensional time series of induced electromotive force of the main wave field in the very early time window of 5-10ms under various bushing environmental conditions was extracted to obtain the induced electromotive force curve. The slope of the tangent line to the induced electromotive force curve in a double logarithmic coordinate system is expressed as: In the formula, Let be the slope of the tangent line to the induced electromotive force curve. for Constantly sensing electromotive force. For time.

8. The method for measuring resistivity through a bushing based on multi-source spatiotemporal joint optimization as described in claim 1, characterized in that, In step five, the method for establishing the apparent resistivity interpretation chart includes: plotting response curves for different casing environment conditions with the absolute amplitude of the induced electromotive force as the vertical axis and the formation resistivity as the horizontal axis, thus forming the apparent resistivity interpretation chart.

9. A through-sleeve resistivity measurement system based on multi-source spatiotemporal joint optimization, used to implement the through-sleeve resistivity measurement method based on multi-source spatiotemporal joint optimization as described in any one of claims 1 to 8, characterized in that, include: The model building module is configured to: establish a two-dimensional axisymmetric forward model of transient electromagnetic logging through the casing, including the wellbore, casing, and formation; The n-transmitter single-receiver coil detector is used to measure the formation signal intensity under the source distance of the main transmitting coil, and to measure the double logarithmic slope of the induced electromotive force curve within a very early time window of 5-10ms during actual downhole casing measurement; based on the reconfigured multi-source transmission timing, the absolute amplitude of the combined induced electromotive force is measured within the target observation time window. The calibration module is configured to: calibrate the target observation time window that maximizes the distinguishability of the formation information based on the formation signal strength at the source distance of the main transmitting coil; The parameter determination module is configured to: perform global optimization using a genetic algorithm under a standard casing environment, and determine the number of turns and reference transmission delay time of each transmitting coil in the n-transmitter single-receiver coil detector using the proportion of formation signals within the target observation time window as the fitness function; The mapping module is configured to: change different casing environment conditions, perform global optimization again through genetic algorithm to determine the different transmission delay times corresponding to each casing environment condition; extract the double logarithmic slope of the induced electromotive force curve in the very early time window of 5-10ms under each casing environment condition as the environmental feature fingerprint, and deeply bind the environmental feature fingerprint with the corresponding transmission delay time to construct a multidimensional discrete mapping table. The chart generation module is configured to: substitute each time series parameter corresponding to the multidimensional discrete mapping table into the forward model, perform forward modeling calculations by traversing different formation resistivity, extract the absolute amplitude of the combined induced electromotive force within the target observation time window, and establish an apparent resistivity interpretation chart between the absolute amplitude of the induced electromotive force and the formation resistivity. The configuration module is set to: use the double logarithmic slope of the induced electromotive force curve measured in the actual well as an environmental feature fingerprint to enter the multidimensional discrete mapping table for lookup matching, obtain the corresponding launch delay time command under the current casing environment, and reconfigure the multi-source launch timing in real time; The resistivity calculation module is configured to: substitute the absolute amplitude of the combined induced electromotive force into the apparent resistivity interpretation chart for inversion to obtain the actual resistivity of the formation.

10. The through-sleeve resistivity measurement system based on multi-source spatiotemporal joint optimization as described in claim 9, characterized in that, The n-transmitter single-receiver coil detector includes a main transmitting coil, m pairs of auxiliary transmitting coils, and a receiving coil, where m ≥ 1. Each pair of auxiliary transmitting coils includes a symmetrically configured first auxiliary transmitting coil and a second auxiliary transmitting coil. The source distance of the main transmitting coil is fixed at 1.0m. The source distance of the first auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 0.7m, and the source distance of the second auxiliary transmitting coil in the first pair of auxiliary transmitting coils is fixed at 1.3m. The source distance of the first auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 0.7m and less than 1.0m, and they are arranged in sequence. The source distance of the second auxiliary transmitting coil in the other pairs of auxiliary coils is greater than 1.0m and less than 1.3m, and they are arranged in sequence.

Images