An oil-based mud while-drilling electrical imaging logging apparatus and method

Abstract

An oil-based mud while-drilling electrical imaging logging device and method, the main part of the logging device is the button electrode at the center position of the electrical imaging probe device, and the shielding electrode and the return electrode are sequentially arranged outward, the numerical simulation method and the physical experiment model method can determine the instrument constant of the oil-based mud while-drilling resistivity imaging logging, and the conversion between the measured impedance amplitude and phase information and the formation resistivity and the formation dielectric constant is realized; the fusion splicing strategy of the formation resistivity is formulated, the measured data under the multi-frequency condition are used to realize the fusion splicing calculation of the full-range formation resistivity; the neural network fitting method is used to establish the minimum objective function, and the inversion or optimization algorithm is used to realize the calculation of the formation resistivity, the formation dielectric constant and the electrode plate gap; the neural network algorithm is used to fit the forward response function, and then the inversion or optimization algorithm is used to obtain the formation resistivity, the gap and the formation dielectric constant, so that the oil and gas formation interpretation and evaluation are better served.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geophysical well logging technology, and particularly relates to an oil-based mud while-drilling electrical imaging logging device and method. BACKGROUND

[0002] The oil exploration and development industry plays an important role in the process of national economy and energy development. Oil and gas reservoir evaluation can provide strong technical support for oil exploration and development, among which, logging technology and interpretation evaluation play an important role. Logging, also known as geophysical well logging, includes open hole logging and logging while drilling, and belongs to the field of applied geophysics. It is to measure the parameters such as formation resistivity, dielectric constant, density, radioactivity and their distribution in the underground borehole by using electromagnetic, acoustic, nuclear magnetic, radioactivity, electrochemical and other measuring instruments, and is widely used in the field of oil and gas exploration and development.

[0003] Open hole electrical imaging logging is an electrical logging method, also known as micro-resistivity scanning imaging logging. It uses a micro-resistivity scanning array composed of densely distributed array button electrodes on the instrument electrode plate, combined with supporting shielding electrodes, return electrodes, adjustment and acquisition circuits and other components, to simultaneously measure dozens to hundreds of resistivity curves, and obtain high-definition wellbore formation resistivity distribution images through data processing. Logging while drilling electrical imaging logging uses a single or multiple button electrodes embedded in the drill collar, combined with the rotation of the drill collar itself during drilling, to measure the wellbore formation in different directions, and obtain wellbore formation resistivity distribution images through data processing, which can be used for formation fracture, pore identification, thin layer analysis, geological structure interpretation, etc.

[0004] During drilling, the wellbore contains mud, which plays a role in lubrication, transmission, and maintaining the stability of the well pressure. The mud commonly used is low-resistivity water-based mud, which appeared earliest and is widely used. Another type of mud is a dispersed system formed by using oil (diesel oil or mineral oil) as the continuous phase, water or oil-wet solids (organic soil, oxidized pitch, etc.) as the dispersed phase, and adding appropriate amounts of treating agents, lime and weighting materials. The main advantages of oil-based mud are high temperature resistance, strong inhibition, and good lubricity, and can effectively reduce damage to oil and gas layers, and the resistivity value is very high. The original while-drilling electrical imaging logging instrument is suitable for low-resistivity water-based mud. However, in many cases, water-based mud cannot meet the requirements of field operations, such as highly deviated wells, horizontal wells, shale formations, deep-sea reservoirs, and other special environments. Oil-based mud has good lubricity, high temperature resistance, high pressure resistance, and can maintain wellbore stability and improve operation efficiency, and is widely used in these special environments. However, oil-based mud is a continuous phase of oil, with high resistivity, usually several hundred times or even tens of thousands times that of water-based mud, which limits the use of the original while-drilling electrical imaging logging instrument suitable for water-based mud.

[0005] Currently, measures taken to develop electrical imaging logging technology suitable for oil-based mud include developing conductive oil-based mud, four-terminal measurement method, and capacitive coupling method. Patent applications with application numbers 202010117447.2, 202110912504.0, and 202310569238.5 disclose a structure of a while-drilling electrical imaging logging instrument and an imaging method suitable for water-based mud. Patent application with application number 201910294886.8 discloses an oil-based mud micro-resistivity scanning imaging logging method based on a recessed electrode structure, which uses one frequency but does not consider the coupling between formation resistivity and formation dielectric constant, is suitable for relatively low-resistivity formations, only measures formation resistivity, and thus has single measurement data, which is inconvenient for subsequent formation interpretation and analysis and increases the difficulty of mechanical manufacturing of the electrode plate. The paper titled “Four-parameter calculation method for oil-based mud electrical imaging logging in low-resistivity formations based on a pair of recessed electrodes” published in Issue 8 of Acta Petrolei Sinica in 2020 discloses a four-parameter calculation method for oil-based mud micro-resistivity scanning imaging logging based on a pair of recessed electrodes, which uses one frequency and obtains four parameters including mud resistivity, mud dielectric constant, mud cake thickness, and formation resistivity through equivalent model calculation, thus enriching the measurement data. However, this method also does not consider the coupling between formation resistivity and formation dielectric constant and is only suitable for low-resistivity formations, like the patent application with application number 201910294886.8. Patent application with application number 202011019657.4 is based on the patent application with application number 201910294886.8 and identifies and judges wellbore fractures and holes, but also does not consider the coupling between formation resistivity and formation dielectric constant and is only suitable for low-resistivity formations. Patent application with application number 201910124532.9 discloses a multi-frequency correction method for oil-based mud environment micro-resistivity scanning imaging, which also takes obtaining formation resistivity as the main purpose and does not fully consider the influence of formation capacitive coupling. The paper titled “Response analysis and quantitative inversion of electrical imaging logging in oil-based drilling fluid environment” published in Issue 3 of Journal of China University of Petroleum (Edition of Natural Science) in 2018 analyzes the response characteristics of oil-based mud micro-resistivity scanning imaging logging and studies formation resistivity and mud cake thickness (interval) between the electrode plate and the wellbore wall using inversion methods, which needs to repeatedly search for the forward modeling database until the objective function is met, the process is tedious, cannot meet the real-time requirements of logging site interpretation, and lacks analysis and inversion processing methods of the influence of formation dielectric constant. Patent application with application number 202211164950.9 discloses an oil-based mud electrical imaging logging parameter determination method, which mainly focuses on open-hole electrical imaging logging and does not specifically involve oil-based mud while-drilling electrical imaging logging instrument structure and data processing methods.Through the above analysis, the existing method has the disadvantages of high development cost, limited applicable formation conditions, unable to effectively separate mud cake / mud signal and formation signal, affected by formation capacitance coupling, single data, complicated data processing process, etc., and mainly focuses on open hole well environment, and lacks an oil-based mud while drilling electrical imaging logging method and device. SUMMARY

[0006] In order to overcome the defects of the prior art, the purpose of the present application is to provide an oil-based mud while drilling electrical imaging logging device and method, which is suitable for measuring and calculating the resistivity and dielectric constant of oil-based mud while drilling electrical imaging logging, based on an electrical imaging probe device, using a neural network algorithm to fit the forward response function, and then using an inversion or optimization algorithm to obtain the formation resistivity, gap and formation dielectric constant, which has the characteristics of quickly and accurately obtaining multiple parameters based on while drilling electrical imaging logging in an oil-based mud environment, and is beneficial to reservoir geological evaluation.

[0007] In order to achieve the above purpose, the technical scheme of the present application is as follows:

[0008] An oil-based mud while drilling electrical imaging logging device, comprising a logging instrument main body 10, an electrical imaging probe device 13 is embedded on the circumference of the logging instrument main body 10, the electrical imaging probe device 13 comprises a button electrode 14 located at the center position, and a shielding electrode 15 and a return electrode 16 outwardly in turn; an A insulating ring 17 is arranged between the button electrode 14 and the shielding electrode 15, a B insulating ring 18 is arranged between the shielding electrode 15 and the return electrode 16, an insulating belt 19 is arranged between the return electrode 16 and the logging instrument main body 10, and a stabilizer 20 is installed on the logging instrument main body 10 around the electrical imaging probe device 13.

[0009] The surface area of the return electrode 16 is greater than the surface area of the button electrode 14.

[0010] The surface shape of the button electrode 14 is rectangular, and the surface shapes of the shielding electrode 15, the return electrode 16, the A insulating ring 17, the B insulating ring 18 and the insulating belt 19 are all rectangular rings.

[0011] The number of the electrical imaging probe devices 13 distributed on the circumference of the logging instrument main body 10 is N, N≥1, which is an integer.

[0012] The materials of the A insulating ring 17, the B insulating ring 18 and the insulating belt 19 are all non-metallic materials.

[0013] An oil-based mud while drilling electrical imaging logging method, comprising the following steps:

[0014] Step A: The oil-based mud while drilling electrical imaging logging device measures the well wall formation 4 at every certain sampling interval.

[0015] Step B, the button electrode 14 emits A current 22 at a certain frequency, which passes through the mud / mud cake layer 21 into the formation 4, and again passes through the mud / mud cake layer 21 back to the return electrode 16; at the same time, the shield electrode 15 is kept at the same potential as the button electrode 14, and emits B current 23, which also passes through the mud / mud cake layer 21 into the formation 4, and again passes through the mud / mud cake layer 21 back to the return electrode 16, and the B current 23 shields the A current 22 to ensure that the A current 22 can enter the formation 4 more;

[0016] Alternatively, the button electrode 14 and the shield electrode 15 are kept at the same potential, and the outer return electrode 16 emits A current 22 and B current 23 at a certain frequency; the A current 22 passes through the mud / mud cake layer 21 into the formation 4, and again passes through the mud / mud cake layer 21 back to the central button electrode 14; at the same time, the B current 23 passes through the mud / mud cake layer 21 into the formation 4, and again passes through the mud / mud cake layer 21 back to the shield electrode 15;

[0017] Step C, the amplitude A i and phase of the A current 22 emitted or received by the button electrode 14 are measured , the amplitude A u and phase of the emission voltage of the button electrode 14 are measured , or the amplitude A u and phase of the emission voltage of the return electrode 16 are measured , as the drill bit 6 is continuously rotated and drilled, a series of current and voltage values are measured, and the measured impedance amplitude A z and phase of the button electrode 14 are calculated , that is

[0018]

[0019]

[0020] Step D, a mathematical model of the measured impedance of the button electrode 14 is established, that is

[0021]

[0022] wherein A z , is the amplitude and phase of the measured impedance of the button electrode 14, G represents the geometric response of the electrical imaging probe device 13, which remains unchanged when the structural size of the electrical imaging probe device 13 is determined; f represents the frequency combination of the emitted B current 23, which generally includes more than 2 frequencies, and the number and size of the frequencies in f have been determined when the logging-while-drilling operation is performed, R m is the oil-based mud resistivity in the wellbore 7, and ε mis the oil-based mud dielectric constant (relative) in the wellbore 7, sd is the spacing (gap) between the button electrode 14 and the formation 4, R t is the resistivity of the formation 4, ε f is the dielectric constant (relative) of the formation 4; F represents a non-linear function describing the measurement process;

[0023] Step E, a numerical simulation method is used to establish a homogeneous formation model without a wellbore, and the formation resistivity R t and the dielectric constant ε f are determined; z Step F, the homogeneous formation model is measured using the electrical imaging probe device 13 to obtain the measured impedance amplitude A z and phase Step G, according to the measured impedance amplitude A t , phase , and the determined formation resistivity R f and dielectric constant ε t , the instrument constant K is obtained by scaling;

[0024] Alternatively, in a laboratory environment, a physical model is built, i.e. the container 24 is filled with an aqueous solution, the salinity of the aqueous solution is changed, and an impedance grid analyzer is used to measure the resistivity and dielectric constant of the aqueous solution, which are equivalent to the formation resistivity R f and dielectric constant ε z ; the logging instrument main body 10 is placed in the aqueous solution, and the electrical imaging probe device 13 of the logging instrument main body 10 is used to measure the aqueous solution, and the measured impedance amplitude A z and phase are obtained by calculation; finally, according to the measured impedance amplitude A t , phase , and the known formation resistivity R f and dielectric constant ε thi , the instrument constant K is obtained by scaling;

[0025] Step F, n frequency values are used, and n+1 resistance threshold points R ri (i=1, 2, …, n+1) arranged in ascending order are introduced, and the measured impedance real part Z ri of the button electrode 14 at each frequency, the corresponding resistivity R a1 , and the impedance ratio amplitude ξ ri at the lowest frequency are calculated, where the measured impedance real part Z ri corresponds to the resistivity R

[0026]

[0027] The impedance ratio is calculated using the following formula:

[0028] ξa = KZ a (5)

[0029] The fusion processing of the formation resistivity is performed by the following strategy, that is

[0030] ① In the interval [R min , R th1 ], the resistivity R ri corresponding to the highest frequency is used to represent the change of the formation resistivity, wherein R min is the minimum value of the formation resistivity;

[0031] ② In the interval [R th1 , R th2 ], the resistivity R ri corresponding to the second highest frequency is used to represent the change of the formation resistivity;

[0032] ③ In the interval [R th2 , R th3 ], the resistivity R ri corresponding to the third highest frequency is used to represent the change of the formation resistivity;

[0033] ④ In this way, until in the interval [R th,n+1 , R max ], the amplitude of the impedance ξ a corresponding to the lowest frequency is used to represent the change of the formation resistivity, wherein R max is the maximum value of the formation resistivity;

[0034] The fusion processing of the full-range formation resistivity is realized by sequentially performing steps ①-④;

[0035] Alternatively, one frequency f s in the higher range is selected from the n frequency values, one resistivity threshold point R ths is set, the measured impedance real part Z rs corresponding to the selected frequency is calculated, the resistivity R rs corresponding to the measured impedance real part Z eqs is calculated, and the equivalent resistivity R rs is calculated, wherein the resistivity R rs corresponding to the measured impedance real part Z eqs is calculated by formula (5), that is

[0036]

[0037] The equivalent resistivity R eqs is calculated by the following formula, that is

[0038]

[0039] Then the fusion processing of the formation resistivity is performed by the following strategy, that is

[0040] ①In the interval [R min ,R ths ], the resistivity R rs of the formation is characterized by the real part of the measured impedance at the selected frequency, wherein R min is the minimum value of the resistivity of the formation;

[0041] ②In the interval [R ths ,R max ], the resistivity R eqs of the formation is characterized by the equivalent resistivity at the selected frequency, wherein R max is the maximum value of the resistivity of the formation;

[0042] The fusion processing steps ①-② are sequentially performed, and the fusion processing of the resistivity of the formation in the full range can also be achieved;

[0043] Step G, a well logging response forward flowchart is established, from left to right, the frequency value gradually increases, and at each frequency value, a mud surface is established, the horizontal and vertical coordinates of the mud surface are mud resistivity R m and mud dielectric constant ε m , the number of values is m1 and m2 respectively, so that m1×m2 mud points are obtained, at each mud point, a three-dimensional coordinate system is established, the coordinate axes are gap sd, formation resistivity R t and formation dielectric constant ε f , the number of values of the three is m3, m4 and m5 respectively, so that a three-dimensional grid body containing m3×m4×m5 data points is established at each mud point, for each three-dimensional grid body, the numerical simulation method of step E is used to carry out batch well logging response forward calculation, and the corresponding three-dimensional impedance amplitude data body and three-dimensional impedance phase data body are obtained;

[0044] Step H, a determination method of mud parameters is established, when the logging instrument main body 10 passes through the casing 8 during the operation of the logging while drilling, the electric imaging probe device 13 emits current signals at each frequency point, and the amplitude and phase signals of the voltage and current are collected at the same time, the amplitude A z and phase of the measured impedance at this time are calculated, and the mud resistivity and mud dielectric constant are calculated by using the measured impedance, that is,

[0045]

[0046] wherein ω is the angular frequency of the emitted current, and ω=2πf is satisfied;

[0047] Alternatively, during logging-while-drilling operations, after drilling a certain distance or time, the shielding electrode 15 on the electrical imaging probe device 13 no longer emits the shielding B current 23, but instead acts as a return electrode. That is, the emitting A current 22 of the button electrode 14 no longer flows back to the return electrode 16, but instead flows back to the shielding electrode 15. Since the emitting A current 22 is no longer shielded and the current loop path is shorter, the measurement impedance of the button electrode 14 comes from the mud, and the mud resistivity and dielectric constant are determined using formulas (7) and (8).

[0048] Step I: After determining the mud resistivity and dielectric constant at each frequency, select the corresponding measured impedance amplitude and phase forward modeling data volume, as well as the corresponding formation resistivity R. t , gap sd and formation dielectric constant ε f The data, using machine learning algorithms, including single-layer or multi-layer neural networks, is used to determine the formation resistivity R corresponding to the forward modeling data volume. t , gap sd and formation dielectric constant ε f As input, the impedance amplitude A is measured. z and phase The optimal neural network model is obtained by iteratively optimizing the input layer to the hidden layer, the input layer to the hidden layer, the hidden layer to the output layer, and the activation functions between hidden layers, either individually or together, by setting the number of neurons and layers in the hidden layer, and by adjusting the activation functions between the input and hidden layers. From this optimal model, the weights and thresholds of the connections between layers, as well as the neuron activation functions, are extracted to form the explicit fitting function expression h(R). t ,sd,ε f ) and g(R t ,sd,ε f ), or k(R) t ,sd,ε f ),Right now

[0049] A z =h(R) t ,sd,ε f (9)

[0050]

[0051] or

[0052]

[0053] Step J involves obtaining the measured impedance amplitude A at various frequencies. z and phase After fitting the function, in order to obtain the formation resistivity R... t , gap sd and formation dielectric constant ε f Construct the objective function C(x) to minimize:

[0054]

[0055] Among them, the vector x consists of the gap, the formation resistivity R at each frequency t and the formation dielectric constant ε f which are the values to be determined; S(x) is a combined form of the fitting functions h, g or k obtained in step G, and the vector y * consists of the impedance amplitude and phase of the electrical imaging probe device 13 at multiple frequencies in the actual measurement data, which is the actual reference value, and the vector x p is the value of the vector x after the previous iteration in the optimization iteration process, which is the reference value; W d and W p are preset weight matrices used to adjust the influence magnitude of each difference part in C(x), and ζ is a preset regularization coefficient; the optimization iteration process of the function C(x) is implemented by using the Gauss-Newton iteration algorithm, the particle swarm optimization algorithm or a combined form of the two; when C(x) < tol (tol is a preset error threshold) or the maximum number of iterations is reached, the vector x at this time is the optimal solution, and the calculation of the formation resistivity R t the gap sd and the formation dielectric constant ε f is completed;

[0056] Step K: Based on the above steps, calculate the formation resistivity, gap and formation dielectric constant corresponding to the measured impedance amplitude and phase of the electrical imaging probe device 13 one by one, and perform image generation and display processing on the calculation data according to the drilling speed and rotational angular velocity of the logging tool body 10, then the formation resistivity image, gap image and formation dielectric constant image of the wellbore formation can be obtained.

[0057] Compared with the prior art, the advantages of the present invention are as follows:

[0058] (1) The oil-based mud logging-while-drilling electrical imaging logging device of the present invention adopts button electrodes, shield electrodes, return electrodes and insulation design, can work in the oil-based mud environment, can accurately obtain multiple parameters, provides a dedicated device for logging-while-drilling electrical imaging, and the obtained parameters use the neural network algorithm to fit the forward response function, and then use the inversion or optimization algorithm to obtain the formation resistivity, gap and formation dielectric constant.

[0059] (2) Step E can determine the instrument constants of the oil-based mud logging-while-drilling resistivity imaging logging by using the numerical simulation method and the physical experiment model method, and realize the conversion between the measured impedance amplitude and phase information and the formation resistivity and formation dielectric constant.

[0060] (3) Step F establishes a fusion and splicing strategy for formation resistivity, and uses measurement data under multi-frequency conditions to realize the fusion and splicing calculation of formation resistivity across the entire range.

[0061] (4) Steps G and H established a forward modeling database calculation strategy for oil-based mud logging while drilling and two methods for determining mud resistivity and mud dielectric constant applicable to oil-based mud logging while drilling, laying the foundation for subsequent inversion optimization calculations of formation resistivity, formation dielectric constant and electrode gap.

[0062] (5) Steps I and J use neural network fitting to establish the response function of the measured impedance amplitude and phase, and establish the minimization objective function. Inversion or optimization algorithms are used to calculate the formation resistivity, formation dielectric constant and electrode gap.

[0063] In summary, this invention has developed a forward modeling data calculation strategy for oil-based mud logging while drilling (EPD), and specifically proposed a method for measuring and calculating the resistivity and dielectric constant of oil-based mud suitable for EPD logging while drilling. It also proposes using a neural network algorithm to fit the forward modeling response function, and then using inversion or optimization algorithms to obtain formation resistivity, interstitial space, and formation dielectric constant, which can better serve the interpretation and evaluation of oil and gas exploration and development. Attached Figure Description

[0064] Figure 1 This is an overall schematic diagram of the implementation of the oil-based mud logging-while-drilling electrical imaging method of the present invention.

[0065] Figure 2 This is a schematic diagram of the logging instrument body 10 and the electrical imaging probe device 13 in this invention, wherein... Figure 2 (a) in the diagram is a schematic diagram of the main body 10 of the logging instrument; Figure 2 (b) is a schematic diagram of the circular electrical imaging probe device 13. Figure 2 (c) is a schematic diagram of the circular electrical imaging probe device 13. Figure 2 (d) is a schematic diagram of the electrical imaging probe device 13 distributed upward around the main body 10 of the logging instrument.

[0066] Figure 3 A schematic diagram illustrating the working principle of oil-based mud logging while drilling (EPR) with electrical imaging. Figure 3 (a) in the figure represents the emission current of button electrode 14; Figure 3 (b) in the figure represents the emission current of the return electrode 16.

[0067] Figure 4 It is a numerical simulation model of a homogeneous layer.

[0068] Figure 5 It is a laboratory mean stratigraphic model.

[0069] Figure 6This is a flowchart of the forward modeling of well logging response, in which... Figure 6 (a) in the diagram is the forward modeling flowchart of the well logging response. Figure 6 (b) in the figure represents a three-dimensional coordinate system.

[0070] Figure 7 This is a schematic diagram of a neural network for obtaining the fitting function of the measured impedance amplitude and phase, where Figure 7 (a) in the diagram is a schematic diagram of impedance amplitude fitting. Figure 7 (b) in the diagram is a schematic diagram of impedance phase fitting. Figure 7 (c) Schematic diagram of simultaneous fitting of impedance amplitude and phase.

[0071] Figure 8 This is the result of a fusion processing method characterizing formation resistivity changes, among which... Figure 8 (a) in the figure shows the results of the impedance ratio measurement and splicing. Figure 8 (b) in the figure is the result of splicing the resistivity calculated from the real part of the impedance; Figure 8 (c) in the figure represents the result of the equivalent resistivity splicing. Figure 8 In the diagram, (d) represents the splicing result of the low-frequency impedance. Figure 8 (e) in the figure represents the splicing result of the high-frequency impedance.

[0072] Figure 9 The results of mud parameter calculations are obtained using in-casing measurement methods, among which... Figure 9 (a) in the diagram is a schematic diagram of mud resistivity. Figure 9 (b) in the diagram is a schematic diagram of the dielectric constant results.

[0073] Figure 10 This is a schematic diagram illustrating the results of calculating the resistivity and dielectric constant of mud by changing the current emission method. Figure 10 (a) in the diagram is a schematic diagram of mud resistivity. Figure 10 (b) in the diagram is a schematic diagram of the dielectric constant results.

[0074] Figure 11 Curves showing the calculated results of formation resistivity, interstitial spacing, and high- and low-frequency dielectric constants in layered formations. Figure 11 (a) in the figure represents the calculation result for a fixed gap. Figure 11 (b) in the figure is the calculation result of the variable gap.

[0075] Figure 12 It is the imaging result of formation resistivity, gaps, and high and low dielectric constants in inclined strata. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0077] In combination with the aboveFigure 1 A schematic diagram of the overall implementation of the drilling operation is shown. This invention proposes an oil-based mud logging-while-drilling electrical imaging method and its logging device. The logging device proposed in this invention includes a logging instrument body 10. (The remaining text appears to be unrelated and possibly machine-generated.) Figure 2 As shown in (a), an electrical imaging probe device 13 is embedded circumferentially in the main body 10 of the logging tool, characterized in that, as Figure 2 As shown in (b), the electro-imaging probe device 13 includes a button electrode 14 located at its center, followed by a shielding electrode 15 and a return electrode 16. An insulating ring A 17 is provided between the button electrode 14 and the shielding electrode 15, an insulating ring B 18 is provided between the shielding electrode 15 and the return electrode 16, and an insulating strip 19 is provided between the return electrode 16 and the logging tool body 10. A stabilizer 20 is installed on the logging tool body 10 around the electro-imaging probe device 13 to keep the logging tool body 10 as close as possible to the wellbore axis position to ensure the measurement signal quality of the probe device 13.

[0078] The surface area of ​​the return electrode 16 is larger than that of the button electrode 14 to ensure that the contact resistance of the return electrode 16 with the mud is sufficiently small during signal acquisition.

[0079] like Figure 2 As shown in (c), the surface shape of the button electrode 14 is rectangular, and the surface shapes of the shielding electrode 15, return electrode 16, A insulating ring 17, B insulating ring 18, and insulating tape 19 are all rectangular rings. In fact, the surface shapes of these electrodes and insulating rings / tapes can be any combination of shapes.

[0080] The number of electrical imaging probe devices 13 distributed around the main body 10 of the logging tool is N (N≥1, an integer). For example... Figure 2 As shown in (d), N can be 1, 2 (180° symmetrical distribution), 3 (120° equidistant distribution), 4 (90° equidistant distribution), etc. In fact, the optimal N value can be selected based on the drilling speed of the drill bit 6 and the rotation speed of the logging tool body 10, according to the actual situation. The number of stabilizers 20 is generally set to 3.

[0081] The materials used for the insulating ring 17 (A), insulating ring 18 (B), and insulating strip 19 are all non-metallic, such as wear-resistant ceramics and rubber. The insulating strip 19 has a sufficiently large area and embedding depth to ensure good insulation between each electrode and the main body of the logging tool 10.

[0082] This invention proposes a method for logging while drilling using oil-based mud electrical imaging, comprising the following steps:

[0083] Step A: Driven by the drilling of drill bit 6, the oil-based mud continuously rotates and drills in with the drilling electrical imaging logging device. The diameter of the main body 10 of the drilling electrical imaging logging instrument is slightly larger than the diameter of drill pipe 5. This can reduce the influence of mud in the well on the measurement signal and better detect formation information. At certain sampling intervals, the oil-based mud drilling electrical imaging logging device measures the formation 4 on the well wall.

[0084] The entire oil-based mud logging-while-drilling electrical imaging system, such as Figure 1 As shown, a derrick 2 and a drilling operation platform 3 are installed at an appropriate location on the ground (sea surface) 1. The derrick 2 primarily serves to support, raise and lower equipment, and provide operating space. The drilling operation platform 3 primarily performs drilling, maintenance, and safety protection functions. Additionally, a mud pit and its associated equipment (not shown) are provided on the ground. By operating the drilling operation platform 3, the drill bit 6, powered by the rotation of the drill pipe 5 or a downhole motor (not shown), continuously drills through the formation 4 on the ground 4. By controlling the drilling direction of the drill bit 6, a wellbore 7 is formed in the formation 4, its trajectory being vertical, inclined, horizontal, or a combination thereof. The wellbore 7 is filled with mud, specifically oil-based mud. A casing 8 may be installed at the wellbore port or any other location, primarily to maintain wellbore stability. Due to the limited drilling depth, a connector 9 is also installed on the drill pipe 5, serving functions such as connection, sealing, and wear resistance. At a certain position at the rear end of the drill bit 6, a logging-while-drilling (LMD) or logging-while-drilling (LWD) tool or logging instrument body 10 is installed. This primarily provides information support for the drilling status of the drill bit 6 by measuring the fluid and formation data within the well, or by transmitting and receiving acoustic, electromagnetic, and other signals. Here, the logging instrument body 10 is a logging-while-drilling electrical imaging (EMI) instrument. Additionally, on the surface 1, there is a microcomputer control system 11 that transmits information to the drilling operation platform 3 and the logging instrument body 10 via wired or wireless means 12, and has functions such as information storage, display, and processing.

[0085] Step B, the specific application process of oil-based mud logging while drilling is as follows: Figure 3 As shown. Due to factors such as the different radii of curvature between the logging tool body 10 and the wellbore 7, the presence of the stabilizer 20, and the irregularity of the wellbore 10, mud / mud cake 21 exists between the probe device 13 and the formation 4. Figure 3 As shown in (a), the button electrode 14 emits current A 22 at a certain frequency. This current passes through the mud / cake layer 21 into the formation 4, and then flows back through the mud / cake layer 21 to the return electrode 16. Simultaneously, the shielding electrode 15, maintaining the same potential as the button electrode 14, emits current B 23. This current also passes through the mud / cake layer 21 into the formation 4, and then flows back through the mud / cake layer 21 to the return electrode 16. Current B 23 acts as a shield for current A 22, ensuring that more of current A 22 can enter the formation 4.

[0086] Alternatively, such as Figure 3 As shown in (b), the button electrode 14 and the shielding electrode 15 are kept at the same potential, and the outer return electrode 16 emits current A 22 and current B 23 at a certain frequency. Current A 22 passes through the mud / cake layer 21 into the formation 4, and then flows back through the mud / cake layer 21 to the central button electrode 14; at the same time, current B 23 passes through the mud / cake layer 21 into the formation 4, and then flows back through the mud / cake layer 21 to the shielding electrode 15. Since the resistivity of the mud / cake layer 21 is very high, the frequency range of current A 22 is generally from several megahertz to several hundred megahertz, and the number of frequencies is n, where n≥2.

[0087] Step C, measure the amplitude A of the current 22 emitted or received by the button electrode 14. i and phase Simultaneously measure the emission voltage amplitude A of button electrode 14. u and phase Or the emission voltage amplitude A of return electrode 16 u and phase As the drill bit 6 continuously rotates and drills, a series of current and voltage values ​​can be measured. The measurement impedance amplitude A of the button electrode 14 can then be calculated. z and phase Right now

[0088]

[0089]

[0090] Step D, according to Figure 3 The measurement process shown indicates that the measurement impedance of the button electrode 14 is a combination of contributions from the instrument structure, wellbore, mud / mud cake, and formation. Therefore, a mathematical model for the measurement impedance of the button electrode 14 is established, namely...

[0091]

[0092] Among them, A z , Button electrode 14 measures impedance amplitude and phase; G represents the geometric response of the electrical imaging probe device 13, which remains constant when the structural dimensions of the electrical imaging probe device 13 are determined; f represents the frequency combination of the emitted current B 23, generally including two or more frequency groups. During logging-while-drilling operations, the number and magnitude of the frequencies in f are determined; R... m The resistivity of the oil-based mud in wellbore 7, ε m is the relative dielectric constant of the oil-based mud in wellbore 7, sd is the distance (or gap) between button electrode 14 and formation 4, and R is... t The resistivity of formation 4, εf R is the dielectric constant (relative) of stratum 4; F represents the nonlinear function describing the measurement process, which cannot be directly obtained. m ε m sd, R t ε f These are unknown parameters; R is used to explain the characteristics of stratigraphic changes. t These are essential parameters, namely, obtaining formation resistivity images. To provide more accurate and comprehensive geological information in oil-based mud environments, sd and ε... f These are also the parameters that need to be obtained.

[0093] Step F, as follows Figure 4 As shown, a homogeneous formation model without a wellbore was established using numerical simulation, and the formation resistivity R was determined. t and dielectric constant ε f The homogeneous geological model was measured using an electrical imaging probe device 13, and the measured impedance amplitude A was obtained. z and phase Based on the measured impedance amplitude A z Phase and the determined formation resistivity R t and dielectric constant ε f The instrument constant K is obtained through calibration.

[0094] Alternatively, physical models can be built in a laboratory environment, such as... Figure 5 As shown, a suitably sized container 24, such as a PVC cylinder, is filled with an aqueous solution. By gradually adding NaCl salt from distilled water, the salinity of the aqueous solution is changed, thereby altering its resistivity and dielectric constant. Simultaneously, an impedance grid analyzer is used to measure the resistivity and dielectric constant of the aqueous solution, which are then used to equivalently measure the formation resistivity R. t and dielectric constant ε f The logging instrument body 10 is placed in an aqueous solution, and the electrical imaging probe device 13 of the logging instrument body 10 is used to measure the aqueous solution. The measured impedance amplitude A is then calculated. z and phase Finally, based on the measured impedance amplitude A... z and phase and the known formation resistivity R t and dielectric constant ε f The instrument constant K is obtained through calibration. Using the instrument constant K, the measured impedance data can be converted into resistivity and dielectric constant.

[0095] Step F, in order to describe the formation resistivity R t The variation is calculated using n frequency values, and n+1 resistivity threshold points R are introduced in ascending order. thi(i = 1, 2, ..., n+1). Calculate the real part Z of the measured impedance of the button electrode 14 at each frequency. ri The corresponding resistivity R ri and the impedance magnitude ξ at the lowest frequency a1 The real part of the measured impedance Z ri The corresponding resistivity R ri The following formula is used for calculation, namely

[0096]

[0097] The impedance ratio is calculated using the following formula:

[0098] ξ a =KZ a 5)

[0099] The following strategy was adopted for the fusion processing of formation resistivity, namely:

[0100] ① In the interval [R] min ,R th1 In the [section], the resistivity R corresponding to the highest frequency is used. ri Characterizing the variation in formation resistivity, where R min It is the minimum value of the formation resistivity, which can be set to a value such as 0.2Ωm.

[0101] ② In the interval [R] th1 ,R th2 In the [section], the resistivity R corresponding to the second highest frequency is used. ri Characterizing changes in formation resistivity;

[0102] ③ In the interval [R] th2 ,R th3 In the [section], the resistivity R corresponding to the third highest frequency is used. ri Characterizing changes in formation resistivity;

[0103] ④ Continue in this manner until the interval [R] is reached. th,n+1 ,R max In the [section], the impedance ratio ξ corresponding to the lowest frequency is used. a The amplitude characterizes the change in formation resistivity, where R max This is the maximum value of the formation resistivity, which can be set to a value such as 20000Ωm.

[0104] By executing steps ① through ④ in sequence, the fusion processing of the entire range of formation resistivity can be achieved.

[0105] Alternatively, in order to describe the formation resistivity R t The change is such that one frequency f is selected from the n frequency values ​​within the higher range. s Set a resistivity threshold point R thsCalculate the real part Z of the measured impedance at the selected frequency. rs The corresponding resistivity R rs and equivalent resistivity R eqs The real part of the measured impedance Z rs The corresponding resistivity R rs Using the same calculation method as formula (4), that is

[0106]

[0107] Equivalent resistivity R eqs The following formula is used for calculation, namely

[0108]

[0109] Then, the following strategy is used for formation resistivity fusion processing, namely...

[0110] ① In the interval [R] min ,R ths In the [reference], the resistivity R corresponding to the real part of the measured impedance at the selected frequency is used. rs Characterizing the variation in formation resistivity, where R min It is the minimum value of the formation resistivity, which can be set to a value such as 0.2Ωm.

[0111] ② In the interval [R] ths ,R max In the [section], the equivalent resistivity R of the selected frequency is used. eqs Characterizing the variation in formation resistivity, where R max This is the maximum value of the formation resistivity, which can be set to a value such as 20000Ωm.

[0112] By executing steps ①-② in sequence, the fusion processing of the entire range of formation resistivity can also be achieved.

[0113] To better characterize the changes in formation resistivity, and to obtain the interstitial space and formation dielectric constant, which will facilitate a better interpretation of geological phenomena, the following steps are taken.

[0114] Step G, establish as follows Figure 6 (a) shows the forward modeling flowchart of the logging response. From left to right, the frequency values ​​used gradually increase (three frequency values ​​are given in the figure; in actual operation, the number of frequencies used is n, satisfying n≥2). A mud surface is established at each frequency value, with the horizontal and vertical axes representing the mud resistivity R. m The dielectric constant of mud ε m The number of values ​​is m1 and m2, resulting in m1×m2 mud points. At each mud point, a system is established as follows: Figure 6 (b) shows a three-dimensional coordinate system with the axes being the gap sd and the formation resistivity R, respectively. tε, the dielectric constant of the formation f The values ​​for the three parameters are m3, m4, and m5, respectively, thus creating a three-dimensional mesh containing m3×m4×m5 data points at each mud point. For each three-dimensional mesh, forward modeling calculations of the logging response are performed in batches using the numerical simulation method described earlier, yielding the corresponding impedance amplitude data volume (three-dimensional) and impedance phase data volume (three-dimensional). It should be noted that the above parameters can be divided using linear or logarithmic equal intervals, ultimately resulting in hundreds of thousands to tens of millions of forward modeling values, the specific number depending on the value range of each parameter and the division interval.

[0115] Step H: In order to calculate formation resistivity, interstitial space, and formation dielectric constant based on the aforementioned logging response data when processing actual logging data, it is first necessary to establish a method for determining mud parameters at multiple frequencies. To achieve this, during logging-while-drilling operations, when the logging instrument body 10 passes through the casing 8, the electrical imaging probe device 13 emits current signals at various frequency points, and simultaneously acquires the amplitude and phase signals of voltage and current, calculating the amplitude A of the measured impedance at this time. z and phase Because the casing is made of metal, its resistivity is much lower than that of the drilling mud, effectively isolating current signals from entering the formation. The measured impedance primarily comes from the contribution of the drilling mud. Therefore, the resistivity and dielectric constant of the drilling mud can be calculated using the measured impedance.

[0116]

[0117] Where ω is the angular frequency of the emitted current, satisfying ω=2πf.

[0118] Alternatively, during logging-while-drilling operations, after each drilling distance or time interval, the shielded electrode 15 on the electrical imaging probe device 13 no longer emits the shielded B current 23, but instead acts as a return electrode. That is, the emitted A current 22 of the button electrode 14 no longer flows back to the return electrode 16, but instead flows back to the shielded electrode 15. Since the emitted A current 22 is no longer shielded and the current loop is shorter, the measurement impedance of the button electrode 14 mainly comes from the mud, so the mud resistivity and dielectric constant can be determined using formulas (7) and (8).

[0119] Step I, after determining the resistivity and dielectric constant of the mud at various frequencies, in Figure 6 In the process, select the corresponding measured impedance amplitude and phase forward modeling data volume, as well as the corresponding formation resistivity R. t , gap sd and formation dielectric constant ε f Data. For example... Figure 7As shown, machine learning algorithms, such as single-layer or multi-layer neural networks, are used to analyze the formation resistivity R corresponding to the forward modeling data. t , gap sd and formation dielectric constant ε f As input, the impedance amplitude A is measured. z and phase Each as an output Figure 7 (a) Figure 7 [b] or both together as output [ Figure 7 In (c)], by setting the number of neurons and layers in the hidden layer, and the activation functions between the input layer and the hidden layer, the hidden layer and the output layer, and between the hidden layers, the optimal neural network model can be obtained through iterative optimization. From the optimal neural network model, the weights and thresholds of the connections between each layer are extracted, and combined with the neuron activation functions to form the explicit fitting function expression h(R). t ,sd,ε f ) and g(R t ,sd,ε f )[ Figure 7 (a) Figure 7 (b)], or k(R) t ,sd,ε f )[ Figure 7 (c)], that is

[0120] A z =h(R) t ,sd,ε f (9)

[0121]

[0122] or

[0123]

[0124] Step J involves obtaining the measured impedance amplitude A at various frequencies. z and phase After fitting the function, in order to obtain the formation resistivity R... t , gap sd and formation dielectric constant ε f Construct the objective function C(x) to minimize:

[0125]

[0126] Wherein, vector x is composed of the gap and the formation resistivity R at various frequencies. t and the dielectric constant ε of the formation f The composition is the value to be determined. S(x) is a combination of the fitting functions h, g, or k obtained in step H, and the vector y is... *It consists of the impedance amplitude and phase of the electric imaging probe device 13 at multiple frequencies in the actual measurement data, which is the actual reference value, vector x p is the value of vector x after the previous iteration during the optimization iteration process, which is the reference value. W d and W p are preset weight matrices used to adjust the influence of each difference part in C(x). ζ is a preset regularization coefficient. The optimization iteration process of function C(x) can be implemented using the Gauss-Newton iteration algorithm, the particle swarm optimization algorithm, or a combination of the two. When C(x) < tol (tol is a preset error threshold) or the maximum number of iterations is reached, the vector x at this time is the optimal solution, and the calculation of the gap sd, formation resistivity R t and formation dielectric constant ε f is completed.

[0127] Step L, based on the above steps, calculate the formation resistivity, gap, and formation dielectric constant corresponding to the measured impedance amplitude and phase of the electric imaging probe device 13 one by one. According to the drilling speed and rotational angular velocity of the logging tool body 10, perform image generation and display processing on the calculated data to obtain the formation resistivity image, gap image, and formation dielectric constant image of the wellbore formation.

[0128] To illustrate and verify the effects and benefits of the content of the present invention, the following several embodiments are studied.

[0129] Embodiment 1

[0130] To illustrate the technical content of the invention, the working process shown in (a) of Figure 3 is adopted. The formation resistivity gradually increases from 0.2 Ωm to 20000 Ωm, and the frequencies are 0.1 MHz, 1 MHz, 10 MHz, and 100 MHz respectively. Combining with the instrument constant, the measured impedance rate is calculated. The resistivity is calculated from the real part of the impedance, the equivalent resistivity, the splicing results of the multi-frequency impedance real part and the low-frequency impedance rate, and the splicing results of the high-frequency impedance real part and the equivalent resistivity are shown in (a)-(e) of Figure 8 respectively. It can be found that Figure 8 in (a) of Figure 8In (b) of the data, after considering the phase change, the real part resistivity of the impedance at the four frequencies shows good sensitivity in different formation resistivity ranges. In particular, the results at 100 MHz can well characterize the formation resistivity changes below 10 Ωm. It should be noted that when the formation resistivity increases, the real part resistivity of the impedance reverses, that is, the trend of change is not monotonic and can no longer characterize the formation resistivity change. Figure 8 In (c), in contrast to the change in the real part resistivity of the impedance, the equivalent resistivity at the four frequencies can well characterize the change in the resistivity of the formation in the high-resistivity region. In particular, the results at 100MHz show a monotonically increasing trend in the region above 10Ωm. Figure 8 Figure (d) shows the fusion result using the real part resistivity of the impedance at 1MHz, 10MHz, and 100MHz and the impedance at 0.1MHz. Figure 8 (e) in the paper presents the fusion processing results using the real part resistivity of the impedance and the equivalent resistivity at 100 MHz. It can be found that, although it cannot be completely quantitative, both fusion processing methods can characterize the monotonically increasing change of the formation resistivity from small to large, which verifies the effectiveness of the multi-frequency fusion processing proposed in this patent.

[0131] Example 2

[0132] To illustrate the invention, when the logging tool body 10 passes through the metal casing, the resistivity and dielectric constant of the drilling mud are measured using the electro-imaging probe device 13. The results are as follows: Figure 9 As shown, the B current 23 emitted by the button electrode 14 is highly guided by the metal casing. After passing through the high-resistivity mud / cake, it flows directly along the metal casing and does not enter the formation. At this point, the measured impedance has no formation contribution, and compared to the mud / cake, the contribution of the metal casing is very low and can be ignored. Therefore, Figure 9 In (a), the calculated mud resistivity and the actual value of mud resistivity form a straight line 'y=x', and the calculated mud dielectric constant and the actual value of mud dielectric constant form a straight line 'y=kx+b', indicating that the measurement method in the casing can well characterize the changes in mud parameters.

[0133] Example 3

[0134] To illustrate the technical content of this invention, the resistivity and dielectric constant of mud were measured by changing the current emission method, and the results are as follows: Figure 10 As shown. After the logging tool body 10 moves a certain distance or time in the well, the current emission mode is changed. That is, the emission current B 23 of the button electrode 14 no longer flows back to the return electrode 16, but flows back to the closer electrode 15. At this time, there is no longer a shielding effect of the shielding current A 22, and the radial detection depth of the current B 23 is shallower, with almost no contribution from the formation in the measured impedance. AsFigure 10 As shown in (a), the calculated mud resistivity and the actual mud resistivity also show a straight line 'y=x', and the calculated mud dielectric constant and the actual mud dielectric constant also show a straight line 'y=kx+b', indicating that changing the current emission mode at intervals of a certain distance or time can well characterize the changes in mud parameters.

[0135] Example 4

[0136] To illustrate the technical content of this invention, a layered formation testing model was established. In this model, the formation resistivity was set to change interactively between low and high values, while the gap between the instrument and the formation was fixed. Rotary drilling was performed at two frequencies (low and high) using an instrument containing two electrical imaging probes 13 to scan the formation within the wellbore. The mud resistivity and dielectric constant were determined using an in-casing measurement method. Then, a neural network was used to determine the measurement response fitting function, and finally, the formation parameters were calculated using a Gauss-Newton inversion algorithm. Figure 11 Figure (a) presents the calculation results. Tracks 1 to 4 represent formation resistivity, interstitial space, low-frequency dielectric constant, and high-frequency dielectric constant, respectively. The two curves in each track correspond to the measurement results of two probes. It can be observed that the formation resistivity curve and the formation dielectric constant curves at the two frequencies can reflect the variation characteristics of the layered strata, and the two interstitial space curves almost overlap and remain unchanged.

[0137] To further illustrate the technical advantages of this invention, in the aforementioned test model, the gap between the instrument and the formation is varied, exhibiting high and low values ​​with depth, and the range of formation resistivity is expanded. Calculations are performed using the same method described above. Figure 11 (b) in the figure gives the calculation results, and the meaning of each curve is the same as... Figure 11 (a) Consistent, it can be found that even with changes in the gap, the contrast between adjacent layers increases, and the formation resistivity curve and formation dielectric constant curve can still reflect the formation change characteristics. At the same time, the gap curve can also reflect the change characteristics of the distance between the instrument and the formation on the wellbore, which well verifies the effectiveness of the technical content of the present invention.

[0138] Example 5

[0139] To illustrate the technical content of this invention, an inclined test model of a layered formation was established. In this formation model, the logging tool body 10 rotates and drills at a 60° inclination angle. During drilling, two measurement frequencies (low and high) are used, and two electrical imaging probes rotate to scan the formation around the wellbore. The mud resistivity and dielectric constant are determined by changing the current emission mode. Then, a neural network is used to determine the measurement response fitting function. Finally, the formation parameters are calculated using the Gauss-Newton inversion algorithm. Based on the drilling speed and rotational angular velocity, the data from the two probes are arranged into an image pixel matrix according to the longitudinal depth and circumferential angle, and finally, formation imaging is performed. Figure 12 The imaging results are presented, with channels 1 through 4 representing formation resistivity imaging, gap imaging, low-frequency formation dielectric constant imaging, and high-frequency formation dielectric constant imaging, respectively. Bright colors indicate high values, and dark colors indicate low values. It can be observed that the imaging results effectively reflect the changes in formation resistivity and dielectric constant. Due to the 60° inclination of the drilling direction, the formation interfaces in the images also exhibit an inclined characteristic, which better reflects the on-site conditions of logging while drilling, demonstrating the advantages of the technical content of this invention.

[0140] In summary, this invention provides a method and apparatus for logging while drilling using electrical imaging (EMI) in oil-based mud. First, it proposes an EPI probe device suitable for oil-based mud environments and details its working principle and impedance calculation method. Then, it characterizes formation resistivity changes by fusing the real part of the measured impedance and the equivalent impedance. Next, it proposes a method for establishing forward modeling data packages for oil-based mud EPI logging responses, as well as methods for measuring and calculating oil-based mud parameters. Finally, it employs a neural network algorithm to establish a logging response fitting function, and uses inversion or optimization algorithms to calculate and image formation resistivity, interstitial space, and formation dielectric constant. This invention can provide support for interpretation and evaluation services in oil and gas exploration and development.

[0141] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the protection scope of the present invention.

Claims

1. A method for logging while drilling using oil-based mud electrical imaging, the logging device on which is based includes a logging instrument body (10), and an electrical imaging probe device (13) is embedded in the circumference of the logging instrument body (10), characterized in that, The electrical imaging probe device (13) includes a button electrode (14) located at its center, a shielding electrode (15) and a return electrode (16) arranged outwards, an A insulating ring (17) is provided between the button electrode (14) and the shielding electrode (15), a B insulating ring (18) is provided between the shielding electrode (15) and the return electrode (16), an insulating strip (19) is provided between the return electrode (16) and the main body of the logging tool (10), and a stabilizer (20) is installed on the main body of the logging tool (10) around the electrical imaging probe device (13). Its characteristic is that it includes the following steps: Step A: At certain sampling intervals, the oil-based mud logging-while-drilling electrical imaging device measures the formation (4) on the wellbore. In step B, the button electrode (14) emits current A (22) at a certain frequency. This current passes through the mud / cake layer (21) and enters the formation (4), then passes through the mud / cake layer (21) again and flows back to the return electrode (16). At the same time, the shielding electrode (15) maintains the same potential as the button electrode (14) and emits current B (23). Similarly, this current passes through the mud / cake layer (21) and enters the formation (4), then passes through the mud / cake layer (21) again and flows back to the return electrode (16). Current B (23) acts as a shield for current A (22) to ensure that current A (22) can enter the formation (4) more. Step C, measure the amplitude of the current (22) emitted or received by the button electrode (14). A i and phase φ i Simultaneously, the amplitude of the emission voltage of the button electrode (14) was measured. A u and phase φ u Or the amplitude of the emission voltage of the return electrode (16) A u and phase φ u As the drill bit (6) rotates and drills continuously, a series of current and voltage values ​​can be measured, and the measurement impedance amplitude of the button electrode (14) can be calculated. A z and phase φ z ,Right now （1） （2） Step D: Establish a mathematical model for measuring the impedance of the button electrode (14), i.e. （3） in, A z φ z The button electrode (14) measures the impedance amplitude and phase. G The geometric response of the electrical imaging probe device (13) is represented when the structural dimensions of the electrical imaging probe device (13) are determined. G Remain unchanged; f The frequency combination representing the emitted B current (23) generally includes more than 2 groups of frequencies, which have been determined during logging-while-drilling operations. f The number and magnitude of intermediate frequencies, R m It is the resistivity of the oil-based mud in the wellbore (7). It is the dielectric constant of the oil-based mud in the wellbore (7). sd It is the distance between the button electrode (14) and the formation (4). R t It is the resistivity of stratum (4). It is the dielectric constant of stratum (4); F This represents a nonlinear function describing the measurement process; Step E: Using numerical simulation, a homogeneous formation model without a wellbore is established, and the formation resistivity is determined. R t and dielectric constant The homogeneous geological model was measured using an electro-imaging probe device (13) to obtain the measured impedance amplitude. A z and phase φ z According to the measured impedance amplitude A z Phase φ z and the determined formation resistivity R t and dielectric constant The instrument constant is obtained through calibration. K ; Step F, using n Each frequency value is introduced. n +1 resistivity threshold points arranged in ascending order R thi (i=1,2,..., n +1), calculate the real part of the measured impedance of the button electrode (14) at each frequency. Z ri corresponding resistivity R ri and the impedance magnitude at the lowest frequency ξ a1 The real part of the measured impedance Z ri corresponding resistivity R ri The following formula is used for calculation, namely （4） The impedance ratio is calculated using the following formula: （5） The following strategy was adopted for the fusion processing of formation resistivity, namely: ① In the interval [ R min , R th1 In the [section], the resistivity corresponding to the highest frequency is used. R ri Characterizing changes in formation resistivity, among which, R min It is the minimum value of the formation resistivity; ② In the interval [ R th1 , R th2 In the [section], the resistivity corresponding to the second highest frequency is used. R ri Characterizing changes in formation resistivity; ③ In the interval [ R th2 , R th3 In the [section], the resistivity corresponding to the third highest frequency is used. R ri Characterizing changes in formation resistivity; ④ And so on, until in the interval [ R th,n+1 ,R max In the [section], the impedance corresponding to the lowest frequency is used. The amplitude characterizes the change in formation resistivity, where, R max It is the maximum value of the formation resistivity; By sequentially executing steps ①-④ of the fusion processing, the fusion processing of the entire range of formation resistivity can be achieved; Step G: Establish a forward modeling flowchart for the logging response. From left to right, the frequency values ​​used gradually increase. At each frequency value, a mud surface is established, with the horizontal and vertical axes representing the mud resistivity, respectively. R m Dielectric constant of mud The number of values ​​is m1 and m2, resulting in m1×m2 mud points. At each mud point, a three-dimensional coordinate system is established, with the coordinate axes representing the gap. sd Formation resistivity R t Formation dielectric constant The number of values ​​for the three are m3, m4, and m5, respectively. Thus, a three-dimensional grid containing m3×m4×m5 data points is established at each mud point. For each three-dimensional grid, the forward modeling calculation of logging response is carried out in batches using the numerical simulation method in step E to obtain the corresponding three-dimensional impedance amplitude data volume and three-dimensional impedance phase data volume. Step H: Establish a method for determining mud parameters. During logging-while-drilling operations, when the logging instrument body (10) passes through the casing (8), the electrical imaging probe device (13) emits current signals at various frequency points, and simultaneously collects the amplitude and phase signals of voltage and current, calculating the amplitude of the measured impedance at this time. A z and phase φ z The resistivity and dielectric constant of the mud are calculated by measuring impedance. （7） （8） in, ω It is the angular frequency of the transmitting current, which satisfies ω =2π f ; Step I: After determining the mud resistivity and dielectric constant at each frequency, select the corresponding measured impedance amplitude and phase forward modeling data volume, as well as the corresponding formation resistivity division. R t ,gap sd and the dielectric constant of the formation ε f The data, using machine learning algorithms, including single-layer or multi-layer neural networks, is used to determine the formation resistivity corresponding to the forward modeling data volume. R t ,gap sd and the dielectric constant of the formation ε f As input, measure the impedance amplitude. A z and phase φ z By setting the number of hidden layer neurons, the number of layers, and the activation functions between the input layer and hidden layers, the output layer, and between hidden layers, the optimal neural network model is obtained through iterative optimization. From this optimal model, the weights and thresholds of the connections between layers, as well as the neuron activation functions, are extracted to form the explicit fitting function expression. and ,or ,Right now （9） （10） or （11） Step J involves obtaining the measured impedance amplitude at various frequencies. A z and phase φ z After fitting the function, in order to obtain the formation resistivity R t ,gap sd and the dielectric constant of the formation ε f Construct the minimum objective function : （12） Where, vector x Based on the gaps and formation resistivity at various frequencies R t and the dielectric constant of the formation ε f The composition is the value to be determined; It is the fitting function obtained in step G. h , g or k Combination forms, vectors Composed of the impedance amplitude and phase of the electrical imaging probe device (13) at multiple frequencies in the actual measurement data, it serves as the actual reference value and vector. x p In the optimization iteration process, the vector after the previous iteration is... x The value is for reference only. W d , W p The preset weight matrix is ​​used for adjustment. The magnitude of the influence of each difference component. These are preset regularization coefficients; the function is implemented using a Gauss-Newton iterative algorithm, a particle swarm optimization algorithm, or a combination of both. The optimization and iterative process; when < tol, tol For the preset error threshold, or when The vector reaches the maximum number of iterations. x This is the optimal solution, which completes the formation resistivity calculation. R t ,gap sd and the dielectric constant of the formation ε f Calculation; Step K: Based on the above steps, calculate the formation resistivity, gap and formation dielectric constant corresponding to the measurement impedance amplitude and phase of the electrical imaging probe device (13). Based on the drilling speed and rotational angular velocity of the logging instrument body (10), perform image generation and display processing on the calculated data to obtain the formation resistivity image, gap image and formation dielectric constant image of the well wall formation.

2. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The alternative to step B is that the button electrode (14) and the shield electrode (15) are kept at the same potential, and the outer return electrode (16) emits current A (22) and current B (23) at a certain frequency; current A (22) passes through the mud / cake layer (21) into the formation (4), and then flows back through the mud / cake layer (21) to the central button electrode (14); at the same time, current B (23) passes through the mud / cake layer (21) into the formation (4), and then flows back through the mud / cake layer (21) to the shield electrode (15).

3. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The alternative to step E is to build a physical model in a laboratory environment, i.e., the container (24) is filled with an aqueous solution, the salinity of the aqueous solution is changed, and the resistivity and dielectric constant of the aqueous solution are measured using an impedance grid analyzer to represent the formation resistivity. R t and dielectric constant The logging instrument body (10) is placed in an aqueous solution, and the electrical imaging probe device (13) of the logging instrument body (10) is used to measure the aqueous solution. The measurement impedance amplitude is obtained by calculation. A z and phase φ z Finally, based on the measured impedance amplitude A z and phase φ z and known formation resistivity R t and dielectric constant The instrument constant is obtained through calibration. K .

4. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The alternative to step F is, from n Choose one frequency from the higher range of the given frequency values. f s Set a resistivity threshold point R ths Calculate the real part of the measured impedance at the selected frequency. Z rs corresponding resistivity R rs and equivalent resistivity R eqs The real part of the measured impedance Z rs corresponding resistivity R rs Using formula (5), that is （5） Equivalent resistivity R eqs The following formula is used for calculation, namely （6） Then, the following strategy is used for formation resistivity fusion processing, namely... ① In the interval [ R min , R ths In the [reference], the resistivity corresponds to the real part of the measured impedance at the selected frequency. R rs Characterizing changes in formation resistivity, among which, R min It is the minimum value of the formation resistivity; ② In the interval [ R ths , R max In the [section], the equivalent resistivity of the selected frequency is used. R eqs Characterizing changes in formation resistivity, among which, R max It is the maximum value of the formation resistivity; By executing steps ①-② in sequence, the fusion processing of the entire range of formation resistivity can also be achieved.

5. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The alternative to step H is that during logging-while-drilling, after drilling a certain distance or time, the shielding electrode (15) on the electrical imaging probe device (13) no longer emits the shielding B current (23), but instead acts as the return electrode. That is, the emitting A current (22) of the button electrode (14) no longer flows back to the return electrode (16), but instead flows back to the shielding electrode (15). Since the emitting A current (22) is no longer shielded, and the current loop path is shorter, the measurement impedance of the button electrode (14) comes from the mud, and the mud resistivity and dielectric constant are determined using formulas (7) and (8).

6. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The surface area of ​​the return electrode (16) is greater than that of the button electrode (14).

7. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The surface shape of the button electrode (14) is rectangular, and the surface shapes of the shielding electrode (15), return electrode (16), A insulating ring (17), B insulating ring (18), and insulating tape (19) are all rectangular rings.

8. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The number of the electrical imaging probe devices (13) distributed around the main body (10) of the logging tool is as follows: N, N ≥1, an integer.

9. The oil-based mud logging-while-drilling electrical imaging method according to claim 1, characterized in that, The materials of the insulating ring A (17), insulating ring B (18), and insulating tape (19) are all non-metallic materials.

Images