A Multi-Physics Field Monitoring Method for Suppressing CT Metal Artifacts Based on Artificial Intelligence

Abstract

This invention provides an AI-based multiphysics monitoring method for suppressing CT metal artifacts, comprising: conducting indoor rock physical hydraulic fracturing experiments and performing multiphysics monitoring; constructing a training dataset: the dataset includes CT data without piezoelectric ceramic sensor metal artifacts, CT data containing piezoelectric ceramic sensor metal artifacts, and a piezoelectric ceramic sensor location map; constructing a dual-domain, three-channel artifact suppression module, which uses both image domain and chordogram domain information; training a detection network using the training dataset; inputting preprocessed CT data containing piezoelectric ceramic sensor metal artifacts into the artifact suppression module to obtain corresponding CT data with piezoelectric ceramic sensor metal artifacts removed and a piezoelectric ceramic sensor location map; and updating the dataset and the artifact suppression module. This AI-based multiphysics monitoring method for suppressing CT metal artifacts can suppress CT metal artifacts and improve CT image quality.

Description

Technical Field

[0001] This invention relates to the field of image data processing technology, specifically to an artificial intelligence-based multi-physics field monitoring method for suppressing CT metal artifacts. Background Technology

[0002] Hydraulic fracturing technology involves injecting high-pressure fluid into shale reservoirs to create complex artificial fractures, increasing reservoir connectivity and improving single-well production. Monitoring and evaluating different stages of hydraulic fracturing reservoir stimulation are prerequisites for efficient development and safe production. However, in hydraulic fracturing field monitoring, the complex conditions at the well site, the limited range of the observation system, and the inability to obtain accurate background data of the actual work area (velocity structure, rock mechanical parameters, etc.) make it difficult to verify the accuracy of fracture inversion results during hydraulic fracturing through field monitoring.

[0003] The laboratory's rock physics hydraulic fracturing experiments can provide stable and controllable stress loading conditions and signal measurement conditions, enabling the accurate acquisition of various property parameters of rock samples. This allows for detailed analysis and research on the changes in rocks during hydraulic fracturing, further providing guidance and assistance for actual hydraulic fracturing.

[0004] Indoor rock physics hydraulic fracturing experiments can perform multi-physics monitoring (ultrasound, acoustic emission, CT). In laboratory rock physics experiments, CT, ultrasound, and acoustic emission monitoring can be performed simultaneously. Computed Tomography (CT) is an imaging technique that acquires information about the object being tested in a non-destructive manner; CT imaging uses excited X-ray beams to perform tomographic scanning of rock samples, obtaining information about the internal structure of the rock sample. Active-source ultrasound uses piezoelectric ceramic sensors (PZTs) to excite seismic waves, which propagate within the rock sample. By inverting the received seismic waves, information about the internal medium, such as velocity, can be obtained. Passive-source acoustic emission uses piezoelectric ceramic sensors to continuously and passively acquire acoustic emission signals generated by internal rock fracturing. Analysis of these acoustic emission signals can obtain source parameters related to internal rock fracture. In these monitoring methods, piezoelectric ceramic sensors are installed on the rock surface and arranged in a specific array of detectors. These PZT sensors use a rapid automatic switching system to switch between receiving and transmitting functions, allowing for both passive acquisition of acoustic emission signals and active excitation of ultrasonic signals.

[0005] However, CT artifacts exist in multiphysics field monitoring. The imaging process of a CT scanner is as follows: the X-ray tube moves in a circle around the object (the object under test), exposing it at different angles. X-rays are emitted from the tube, attenuated after absorption by the object, and the remaining X-rays reach the detector. The detector converts the received light intensity signal into an electrical signal. The signals received by all units on the detector form projection data (projection image). The projection data is used to generate the final CT image through an image reconstruction algorithm. The most commonly used reconstruction algorithm in the industry is the filtered backprojection algorithm. This algorithm is characterized by convolution and filtering of the projection at each acquisition angle before backprojection, compensating for high-frequency components, thereby improving the blurriness problem of the reconstructed image after direct backprojection, resulting in a better reconstructed image quality.

[0006] When a high-density object (such as metal) is present within the CT imaging field of view, striped and banded black-and-white artifacts will appear in the reconstructed image. The main causes of metal artifacts are: X-ray beam hardening, scattering effects, complete photon attenuation, and the partial volume effect of the metal. Among these, beam hardening is the most significant cause. X-rays produced by an X-ray tube have a certain spectral width, containing X-rays of varying energies. When multi-energy X-rays pass through an object, low-energy rays are easily absorbed, while high-energy rays pass through more easily, resulting in a higher average energy of the received rays and a gradual hardening of the rays—a phenomenon known as beam hardening. When X-rays encounter a very dense substance (metal), beam hardening is exacerbated, causing drastic changes in the acquired projection data between metallic and non-metallic regions, resulting in metal artifacts in the reconstructed CT image.

[0007] Compared to rock samples, PZT sensors have a higher density, approaching that of metal, which results in strong "metal artifacts" during CT acquisition. These artifacts severely interfere with the interpretation of internal images of rock samples and reduce the research value of multiphysics monitoring.

[0008] Currently, the main methods for suppressing metal artifacts are as follows: (1) Projection domain interpolation algorithm. The projection domain linear interpolation method is based on the similarity of the linear attenuation coefficients of adjacent tissues in the human body (taking the human body as an example), that is, the value near the metal trajectory is linearly interpolated to fill the metal trajectory. The projection domain interpolation method can effectively correct most metal artifacts by filling the metal projection trajectory. However, due to the instability of metal segmentation accuracy and the abrupt change in data at the edge of the metal implant trajectory, it is easy to cause secondary artifacts, which also reduces the quality of the reconstructed image. (2) Iterative reconstruction algorithm. This method starts from an initial image and iteratively updates the reconstructed image by continuously reducing the error between the actual projection value and the theoretical projection value. The iterative reconstruction method mainly uses the projection data that is not missing in the sine curve to reconstruct the image, and reconstructs an artifact-free CT image after a sufficient number of iterations. Due to its complicated algorithm steps and huge computational load, the computation time of the iterative algorithm is usually very long. (3) Deep learning algorithm. The deep learning algorithm mainly includes key steps such as feature learning, mapping relationship establishment and model training. This method automatically learns latent feature patterns from a database and establishes a mapping relationship from artifact-containing images to artifact-free images through training, thereby removing artifacts. It can be categorized into supervised, unsupervised, and semi-supervised methods. Deep learning-based MAR (Metal Artifact Reduction) methods are mainly divided into three categories: chordmap enhancement, CT image enhancement, and joint dual enhancement of chordmap and CT images. Among these, joint dual enhancement methods based on chordmaps and CT images, such as artifact removal methods based on the unfolding concept, have achieved performance improvements, but due to the optimization algorithms and network structure design they employ, the artifact removal effect is still not ideal. Summary of the Invention

[0009] The purpose of this invention is to at least partially overcome the shortcomings of the prior art and provide a method for suppressing CT metal artifacts based on artificial intelligence and multi-physics field monitoring.

[0010] The present invention also aims to provide a method for suppressing CT metal artifacts in multi-physics field monitoring based on artificial intelligence, which effectively solves the problem of CT artifacts in indoor rock physical hydraulic fracturing experiments during multi-physics field monitoring.

[0011] The present invention also aims to provide a method for suppressing CT metal artifacts based on artificial intelligence and multi-physics field monitoring, thereby improving image quality.

[0012] To achieve the above-mentioned objectives or one of them, the technical solution of the present invention is as follows:

[0013] A multi-physics-based method for suppressing CT metal artifacts, the method comprising:

[0014] Indoor rock physical hydraulic fracturing experiments were conducted, and multi-physics field monitoring was performed.

[0015] The training dataset is constructed as follows: The dataset includes CT data without piezoelectric ceramic sensor metal artifacts, CT data with piezoelectric ceramic sensor metal artifacts, and piezoelectric ceramic sensor location maps. The three are matched and correspond to form a data volume, and all data exist in the form of a data volume. The dataset consists of real rock sample experimental data and simulated synthetic data. The data is augmented to improve the generality of the dataset. All data undergo the same preprocessing steps.

[0016] A dual-domain, three-channel artifact suppression module is constructed, utilizing information from both the image domain and the chordmap domain. The first channel is constructed as a UNet-type submodule based on the image domain, used to obtain the position information of the piezoelectric ceramic sensor and provide constraints for artifact suppression. The second channel is a UNet-type submodule jointly enhanced by the image and chordmap domains, used to suppress the influence of metal artifacts from the piezoelectric ceramic sensor. The third channel is a chordmap domain channel. The UNet-type submodule of the first channel uses a skip-layer connection, and the UNet-type submodule of the first channel and the UNet-type submodule of the second channel share piezoelectric ceramic sensor information through a cross-spatial attention mechanism to provide information constraints for the artifact suppression process.

[0017] Train the detection network using the training dataset;

[0018] All data obtained from actual monitoring are preprocessed using the same steps. The preprocessed CT data containing metal artifacts of the piezoelectric ceramic sensor is then input into the artifact suppression module to obtain the corresponding CT data with metal artifacts of the piezoelectric ceramic sensor removed and the piezoelectric ceramic sensor location map.

[0019] Update the dataset and artifact suppression module.

[0020] According to a preferred embodiment of the present invention, the "conducting indoor rock physical hydraulic fracturing experiments and performing multi-physics field monitoring" includes:

[0021] Rock samples of different sizes were collected according to the research objectives. The rock samples were covered with a prefabricated rubber sleeve. The rubber sleeve had multiple probe holes, but no piezoelectric ceramic sensors were attached to the holes.

[0022] A single CT data was collected from a rock sample without a piezoelectric ceramic sensor attached using an indoor hydraulic fracturing experimental setup. This CT data did not contain metal artifacts from the piezoelectric ceramic sensor and served as the first-stage CT data.

[0023] The piezoelectric ceramic sensor is placed in the probe hole and then bonded to the surface of the rock sample.

[0024] The indoor hydraulic fracturing test apparatus was used to collect CT data on a rock sample with a piezoelectric ceramic sensor attached. The spatial position of the piezoelectric ceramic sensor was then determined using the CT data. The CT data included metal artifacts of the piezoelectric ceramic sensor and was used as the second-stage CT data.

[0025] Different loading strategies were employed to pressurize the rock samples. The pressurization phase included an isotropic loading phase, an axial pressure increase phase, a water injection phase to increase pore pressure, and a pressure unloading phase. During the water injection phase to increase pore pressure, several CT data points were acquired at set intervals. These CT data points, containing piezoelectric ceramic sensor metal artifacts, served as the third-stage CT data. During the pressure unloading phase, one CT data point was acquired, which also contained piezoelectric ceramic sensor metal artifacts, serving as the fourth-stage CT data. Finally, the piezoelectric ceramic sensor was removed from the rock sample, and another CT data point was acquired. This new CT data point, free of piezoelectric ceramic sensor metal artifacts, served as the fifth-stage CT data.

[0026] All acquired CT data were processed.

[0027] According to a preferred embodiment of the present invention, the rock sample has the following specifications: a diameter of 50 mm and a length of 125 mm, and is cylindrical.

[0028] Rock samples include two types: ordinary rock samples and disturbed rock samples;

[0029] The common rock sample types include sandstone and shale; they are further divided into five categories based on whether they contain bedding and the direction of bedding: bedding directions of 0±20°, 45±20°, 90±20°, 135±20°, and homogeneous media without bedding; and two categories based on axial compressive strength: 50±10MPa and 90±10MPa. Five rock samples were selected from each category, for a total of 100 rock samples.

[0030] The interfering rock samples include two types of rocks: sandstone and shale. They are homogeneous media without bedding and contain sediments. Five rock samples of each type are selected, for a total of 10 rock samples.

[0031] According to a preferred embodiment of the present invention, the generation of real rock sample experimental data includes:

[0032] Using ordinary rock samples and interfering rock samples, CT data from the first stage, the second stage, the fourth stage, and the fifth stage were collected to form the total dataset.

[0033] According to a preferred embodiment of the present invention, in the step of "training the detection network using the training dataset", the dataset is divided into a training set and a test set in a ratio of 8:2; the artifact suppression module adopts the stochastic gradient descent optimization method, sets a dynamic learning rate, sets the initial value to 0.0001, reduces it by half every 50 times, sets the batch size to 40, and sets the number of iterations to 200; the training of the artifact suppression module is performed on the GPU image processing unit.

[0034] According to a preferred embodiment of the present invention, when any of the following conditions are triggered, CT data containing piezoelectric ceramic sensor metal artifacts, CT data without piezoelectric ceramic sensor metal artifacts, and the extracted piezoelectric ceramic sensor location map are used as data volumes to update the training dataset:

[0035] The current nth image containing artifacts Compared with the (n-1)th image containing artifacts Perform image similarity analysis to obtain the image similarity between two adjacent images containing artifacts. When this value is greater than 5%; or

[0036] Current image after removing artifacts for the nth time Compared with the (n-1)th artifact removal image Image similarity analysis was performed to obtain the image similarity between two adjacent images after artifact removal. When this value is greater than 5%; or

[0037] Current position diagram of the nth piezoelectric ceramic sensor Location diagram of the (n-1)th piezoelectric ceramic sensor Image similarity analysis was performed to obtain the image similarity between two adjacent images of the electro-ceramic sensor positions. When this value is greater than 3%.

[0038] According to a preferred embodiment of the present invention, the artifact suppression module is updated when the following conditions are met:

[0039] After the experiment begins, the cumulative number of updated data volumes is greater than 5; or

[0040] No. and The similarity difference between the images is greater than 10%.

[0041] According to a preferred embodiment of the present invention, the probe holes on the rubber sleeve are positioned as follows: four rows are arranged at 0°, 90°, 180° and 270° along the circumference of the rock sample, and multiple probe holes are evenly arranged on each row. The height of the piezoelectric ceramic sensor in every two rows of probe holes is different, so that the four rows of piezoelectric ceramic sensors do not appear at the same height at the same time.

[0042] According to a preferred embodiment of the present invention, the indoor hydraulic fracturing experimental apparatus includes a pressure vessel, a loading system, an acoustic emission counting and waveform acquisition system, and a CT monitoring system.

[0043] According to a preferred embodiment of the present invention, during the pressurization of rock samples using different loading strategies, active source ultrasonic data are collected at set times. During active source ultrasonic monitoring, some piezoelectric ceramic sensors are used as transmitting probes to excite ultrasonic signals, and the remaining piezoelectric ceramic sensors are used as receiving probes to receive ultrasonic signals.

[0044] During the pressurization of rock samples using different loading strategies, the piezoelectric ceramic sensor is used as both a transmitting and receiving probe during active source ultrasonic acquisition, and at other times it serves as a receiving probe to receive acoustic emission signals generated by changes in the rock sample.

[0045] According to a preferred embodiment of the present invention, the data processing of CT data includes: CT imaging, suppression of metal artifacts by piezoelectric ceramic sensors, and joint analysis with acoustic emission results.

[0046] According to a preferred embodiment of the present invention, the data processing of acoustic emission data includes: valid event picking, first arrival picking, source location, source mechanism analysis, magnitude calculation, and stress field analysis.

[0047] According to a preferred embodiment of the present invention, the data processing of ultrasound data includes: identification of ultrasound events, acquisition of ultrasound first arrival, and velocity inversion analysis.

[0048] The present invention provides a multi-physics field monitoring method for suppressing CT metal artifacts based on artificial intelligence. This method utilizes an adaptive AI network update mechanism to dynamically suppress PZT metal artifacts in CT images, thereby improving the accuracy of CT image analysis. Attached Figure Description

[0049] Figure 1 The arrangement of a rock sample and a piezoelectric ceramic sensor according to an embodiment of the present invention is shown, with the rock sample on the left and the PZT on the surface of the rock sample on the right.

[0050] Figure 2 The modular architecture of a multiphysics monitoring CT metal artifact suppression method according to an embodiment of the present invention is shown, mainly comprising two parts: a CT image PZT artifact suppression part and a module update part; and

[0051] Figure 3 This is a schematic diagram of the artifact suppression module according to an embodiment of the present invention. Detailed Implementation

[0052] Exemplary embodiments of the present invention are described in detail below with reference to the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements. Furthermore, in the following detailed description, numerous specific details are set forth for ease of explanation to provide a thorough understanding of the embodiments disclosed herein. However, it will be apparent that one or more embodiments may be practiced without these specific details. In other instances, well-known structures and apparatuses are illustrated to simplify the drawings.

[0053] As mentioned earlier, in indoor rock physical hydraulic fracturing experiments, CT, ultrasound, and acoustic emission data were simultaneously acquired, enabling multi-physics field monitoring of the indoor rock physical hydraulic fracturing experiments. During CT acquisition, the PZT sensor used for ultrasound and acoustic emission signal acquisition, due to its high density, produced strong "metal artifacts." These artifacts severely interfered with the judgment and analysis of the internal images of the rock samples, reducing the research value of multi-physics field monitoring.

[0054] This application proposes an artificial intelligence network that simultaneously employs image domain and chordogram domain divisors to suppress "metal artifacts" caused by PZT probes in CT images during rock physics experiments. Specifically, a one-channel image domain-based UNet-type submodule is constructed to obtain PZT location information and provide constraints for artifact suppression. A two-channel image domain and chordogram domain jointly dual-enhanced UNet-type submodule is constructed to suppress the metal artifact effects of PZT probes. This invention uses different types of rock samples (rock type, different bedding orientations, different axial compressive strengths), and performs CT scans with PZT probes attached to the rock sample surfaces to construct a training set. During testing, an adaptive data update mechanism and a module update mechanism based on the difference analysis between adjacent images are designed. The method proposed in this invention can dynamically suppress PZT metal artifacts and improve the accuracy of CT image analysis.

[0055] Artificial Intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or computers-controlled machines to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. Machine Learning (ML) is a multidisciplinary field involving probability theory, statistics, approximation theory, convex analysis, and algorithm complexity theory. It specifically studies how computers can simulate or implement human learning behavior to acquire new knowledge or skills and reorganize existing knowledge structures to continuously improve their performance. Machine learning is the core of artificial intelligence and the fundamental way to endow computers with intelligence; its applications span all areas of artificial intelligence. Machine learning and deep learning typically include techniques such as artificial neural networks, belief networks, reinforcement learning, transfer learning, inductive learning, and learn-by-doing.

[0056] The following describes the specific process of the AI-based multiphysics field monitoring CT metal artifact suppression method according to an embodiment of the present invention.

[0057] 1. Basic situation of the rock physics experiment involved in this invention

[0058] 1.1 Preparation of experimental samples:

[0059] Experimental samples are generally derived from rocks from specific regions. The size of the rock samples can vary depending on the actual research purpose. The standard size used for conventional indoor hydraulic fracturing experiments is a cylinder with a diameter of 50 mm and a length of 125 mm.

[0060] Two types of rock samples were set up: ordinary rock samples and disturbed rock samples.

[0061] Common rock sample types include: rock type (sandstone, shale, etc., 2 types), bedding with different orientations (bedding directions are approximately 0±20°, 45±20°, 90±20°, 135±20°, etc., and homogeneous media without obvious bedding, etc., 5 types), and axial compressive strength (50±10MPa, 90±10MPa, etc., 2 types); 5 rock samples are selected from each type, for a total of 100 rock samples.

[0062] Interfering rock sample types include: rock types (sandstone, shale, etc.), homogeneous media without obvious bedding, and rock samples containing obvious sediments, such as heavy mineral impurities and clastic particles. These sediments are internal characteristics of the rock samples and will be visible in CT images regardless of whether a PZT probe is attached to the surface, thus increasing the generalizability of the rock samples. Five rock samples from each type were selected, for a total of 10 rock samples.

[0063] The rock sample is covered with a prefabricated rubber sleeve containing 24 PZT probe holes (the number is not limited to 24; it depends on the sleeve's design). The PZT probes are placed in the holes, and the bottom of the PZT probes is glued to the rock sample surface. The positions of the PZT probe holes on the rubber sleeve are shown below. Figure 1 As shown, four columns of probes are arranged along the rock sample at 0°, 90°, 180°, and 270°, with each column having a different probe height to ensure that the four columns of probes do not appear at the same height simultaneously. PZT is a metallic material, which can produce severe metal artifacts on the rock sample surface, affecting the CT imaging quality of the rock sample; this design will reduce the metal artifacts caused by multiple PZT probes at any height (e.g., ...). Figure 1 ).

[0064] 1.2 Experimental System Composition:

[0065] The experimental system is an indoor hydraulic fracturing experimental device, mainly consisting of a pressure vessel, a loading system, and an acoustic emission counting and waveform acquisition system.

[0066] (1) The pressure vessel is a high-pressure resistant metal container, and the loaded rock sample is placed inside the container.

[0067] (2) The loading system generally adopts a triaxial compressive stress loading system. This system can provide axial pressure, injection pressure and confining pressure to the inside of the pressure vessel to simulate the in-situ formation conditions of the rock sample.

[0068] (3) The acoustic emission counting and waveform acquisition system consists of a PZT, a preamplifier and a high-speed acquisition device.

[0069] (4) The CT monitoring system is composed of CT units, with a field of view greater than 125mm, and the scanning mode is plain scan, usually 256-slice CT.

[0070] 1.3 Experimental Procedure:

[0071] The experimental system can employ different loading strategies depending on the actual research objective. The experimental procedure of this patent is as follows:

[0072] (1) Place the rock sample in a rubber sleeve and load it into a pressure vessel; place it on a CT scanner and collect CT data once. The CT data collected in this process does not contain interference from PZT metal artifacts.

[0073] (2) Remove the rock sample from the container and attach the PZT (or PZT probe) to the PZT probe hole in the outer rubber sleeve of the rock sample.

[0074] (3) Place the rock sample with PZT (or PZT probe) attached inside a pressure vessel; place it on a CT scanner and acquire CT data once. The CT data acquired during this process includes interference from PZT metal artifacts. Calibrate the spatial position of all probes. (Because the PZT probe aperture is larger than the probe diameter, there will be spatial deviations when installing the PZT probe in each experiment. Therefore, it is necessary to accurately extract the position using CT images after installing the PZT probe.)

[0075] (4) Isotropic loading stage (containment pressure).

[0076] (5) Add axial compression stage (axial compression).

[0077] (6) Water injection to increase pore pressure (injection pressure) until the rock fracturing stage and unloading stage. During this process, the CT scanner is turned on and CT data is collected according to the set time. The CT data collected during this process includes interference from PZT metal artifacts.

[0078] (7) Pressure unloading stage (injection pressure, confining pressure and axial pressure unloaded to 0), collect CT data once. The CT data collected in this process includes interference from PZT metal artifacts (after the rock sample is fractured, there are internal cracks); take the rock sample out of the container, remove the PZT probe pasted in the PZT probe hole of the outer rubber sleeve of the rock sample, and place the rock sample in the pressure container; place it on the CT machine and collect CT data once. The CT data collected in this process does not include interference from PZT metal artifacts.

[0079] In the above (4)-(6) process, active source ultrasound data are collected according to the set time. When performing active source ultrasound monitoring, some PZT probes are used as transmitting probes (excitation probes) to excite ultrasound signals, and the remaining probes are used as receiving probes to receive ultrasound signals.

[0080] During the above (4)-(6) process, due to the loading of external forces, the process will be accompanied by acoustic emission events. Except when the PZT probe is used as the transmitting probe during active source ultrasonic acquisition, it is used as the receiving probe for the rest of the time to receive the acoustic emission signals generated by the changes in the rock sample.

[0081] 1.4 Main data analysis contents:

[0082] The CT data processing workflow includes: data acquisition, CT imaging, PZT metal artifact suppression, and joint analysis with acoustic emission results.

[0083] The acoustic emission data processing flow includes: valid event picking (picking data segments with obvious phases), first arrival picking (extracting the first arrival time of the waveform from the valid event data), source location (obtaining the source location of the event based on the picked first arrival time and the spatial position of the probe), source mechanism analysis (inverting the source mechanism of the event based on the waveform of the event to obtain the rupture characteristics of the event), magnitude calculation (inverting the magnitude of the event based on the waveform of the event), and stress field analysis (based on the waveform of the event or the source mechanism of the event), etc.

[0084] The ultrasonic data processing workflow includes: ultrasonic event identification (identifying signals received by other channels based on ultrasonic excitation time), ultrasonic first arrival pickup (obtaining the first arrival time of the received signal), and velocity inversion analysis (inverting the velocity model of the rock sample based on the first arrival time, waveform, and other information of the signal).

[0085] 2. Construct the training dataset:

[0086] 2.1 Overview of the dataset.

[0087] The dataset includes: CT data without PZT metal artifacts, CT data with PZT metal artifacts, and a PZT location map (a binary map with a value of 1 at the PZT probe location and a value of 0 at other locations). These three are matched and matched to form a single data volume, and all data exists in the form of a data volume. All the above data have the same size, and the data space acquired from a single CT scan is [data volume size missing]. In this single experiment, a total of [missing information] were contained. Yes, the total data volume is .

[0088] 2.2 Dataset generation.

[0089] The dataset consists of real rock sample experimental data and simulated synthetic data.

[0090] (1) Generation of real rock sample data:

[0091] Use the common rock samples prepared in 1.1 to complete the experimental procedure in 1.3. Repeat the procedure using different common rock samples from 1.1. This experiment constitutes the total dataset.

[0092] In the kth experiment, CT data were obtained once each through processes 1.3(1) and 1.3(3). PZT locations were extracted from the CT data containing PZT metal interference to obtain a PZT location map. The obtained CT images were calibrated to form a data volume. This data volume was acquired before fracturing and did not contain fracturing cracks.

[0093] In the kth experiment, two CT data from process 1.3 (7) were selected. The PZT location was extracted from the CT data containing PZT metal interference to obtain the PZT location map. The obtained CT images were image calibrated to form a data volume. This data volume was acquired after fracturing, and its rock damage contained fracturing cracks.

[0094] Use the interference rock sample prepared in 1.1 to complete the experimental procedure in 1.3. Repeat the procedure using different interference rock samples from 1.1. This experiment constitutes the total dataset.

[0095] In the kth experiment, CT data were obtained once each through processes 1.3(1) and 1.3(3). PZT locations were extracted from the CT data containing PZT metal interference to obtain a PZT location map. The obtained CT images were calibrated to form a data volume. This data volume was acquired before fracturing and did not contain fracturing cracks.

[0096] In the kth experiment, two CT data from process 1.3 (7) were selected. The PZT location was extracted from the CT data containing PZT metal interference to obtain the PZT location map. The obtained CT images were image calibrated to form a data volume. This data volume was acquired after fracturing, and its rock damage contained fracturing cracks.

[0097] (2) Generation of synthetic rock sample data:

[0098] Based on the same size as the real rock sample, and referring to the image features of the real rock sample (shale, sandstone) in (1), CT images (CT data) of the synthetic rock sample are produced; referring to the real rock sample experiment in (1), the fracture features (fracture width, extension morphology, etc.) generated in the fracturing process 1.3 (6) are added to the above-mentioned synthetic CT images (CT data) to form a synthetic rock sample with fractures.

[0099] Adding CT image features of a rubber sleeve to a synthetic rock sample creates a synthetic rock sample without a PZT probe; further adding CT image features of a PZT probe creates a synthetic rock sample with a PZT probe.

[0100] Based on the above-mentioned synthetic rock samples, CT images without PZT metal artifacts and CT images with PZT metal artifacts were obtained respectively. PZT locations were extracted from the CT data containing PZT metal interference to obtain a PZT location map. The above images were combined into a data volume.

[0101] Repeat this process to obtain multiple data volumes.

[0102] 2.3 Data augmentation.

[0103] During data generation, data augmentation is performed to improve the generalizability of the dataset. Data augmentation methods include:

[0104] Take the CT data obtained in step 1.3(1) during the interference rock sample experiment in 2.2 that does not contain PZT metal artifacts; extract the parts of the data that contain obvious interference, such as heavy mineral impurities, detrital particles, etc., to form a noise image; randomly add the above-extracted noise image to the data volume (do not add it to the PZT location map), and keep the addition position (including the relative position with the PZT probe), quantity and amplitude consistent in the data volume; in a single data volume, the number (0 to 5, randomly distributed) and amplitude (-2 to 2, randomly distributed) of the added interference objects are different.

[0105] 2.4 Data preprocessing.

[0106] All data undergo the same preprocessing steps.

[0107] 3. Constructing a PZT artifact suppression method:

[0108] 3.1 Module architecture.

[0109] Multiphysics monitoring methods for suppressing CT metal artifacts, such as Figure 2 As shown, it includes experimental setup, multiphysics data acquisition module (CT data preprocessing module), CT image PZT artifact suppression module, parameter update module, and PZT artifact suppression CT image.

[0110] The experimental setup and multiphysics data acquisition module mainly includes the experimental section in section 1.3, which corresponds to this section.

[0111] The CT image PZT artifact suppression module contains the main structure of the PZT artifact suppression method, mainly consisting of part 3.2.

[0112] The parameter update module mainly includes fine-tuning and updating the PZT artifact suppression module for CT images based on the updated training dataset.

[0113] 3.2 CT Image PZT Artifact Suppression Module

[0114] A dual-domain, three-channel artifact suppression module was constructed, utilizing information from both the image domain and the chordogram domain. A one-channel image-domain-based UNet submodule was built to obtain the PZT's location information and provide constraints for artifact suppression. A two-channel image-domain and chordogram-domain jointly dual-enhanced UNet submodule was constructed to suppress the metal artifacts introduced by the PZT probe.

[0115] In the UNet-type submodule for extracting PZT location information in one channel:

[0116] The image domain channels utilize 10 convolutional layer (Conv) units. The parameters for convolutional layers 1-10 are as follows: (256×3×3×3), (256×3×3×3), (128×3×3×3), (128×3×3×3), (64×3×3×3), (64×3×3×3), (128×3×3×3), (128×3×3×3), (256×3×3×3), (256×3×3×3). The first dimension represents the number of convolutions, and the latter three dimensions represent the convolutional size.

[0117] In the two-channel PZT artifact suppression UNet-type submodule:

[0118] The image domain channels utilize 10 convolutional layers. The parameters for convolutional layers 1-10 are as follows: (256×3×3×3), (256×3×3×3), (128×3×3×3), (128×3×3×3), (64×3×3×3), (64×3×3×3), (128×3×3×3), (128×3×3×3), (256×3×3×3), (256×3×3×3). The first dimension represents the number of convolutions, and the latter three dimensions represent the convolutional dimensions.

[0119] The chord map domain uses 10 convolutional layers. The parameters of convolutional layers 1-10 are as follows: (256×3×3×3), (256×3×3×3), (128×3×3×3), (128×3×3×3), (64×3×3×3), (64×3×3×3), (128×3×3×3), (128×3×3×3), (256×3×3×3), (256×3×3×3). The first dimension represents the number of convolutions, and the latter three dimensions represent the convolutional dimensions.

[0120] In the UNet-type submodule for PZT location information extraction, a skip connection is used to make PZT location information extraction more accurate.

[0121] The PZT location information extraction UNet-type submodule and the PZT artifact suppression UNet-type submodule share PZT information through a cross spatial attention mechanism to provide information constraints for the artifact suppression process.

[0122] During training, the left input terminal receives a CT image without PZT artifact interference, while the right output terminal receives a CT image with PZT artifact interference and a PZT location map. The loss function is MSE (mean squared error). The total loss function is MSE = MSE1 + MSE2; where MSE1 is the error between the artifact-suppressed CT image calculated using the CT image with PZT metal interference and the CT image without PZT artifact interference; and MSE2 is the error between the PZT location map (piezoelectric ceramic sensor location map) calculated using the CT image with PZT metal interference and the actual PZT location map.

[0123] 4. Train the detection network using the training dataset.

[0124] 4.1 Divide the dataset into a training set and a test set in a ratio of 8:2.

[0125] 4.2 This PZT artifact suppression module uses a stochastic gradient descent optimization method; a dynamic learning rate is set, initially set to 0.0001, and halved every 50 iterations. The batch size is set to 40; the number of iterations is 200.

[0126] 4.3 The training of the PZT artifact suppression module is performed on the GPU image processing unit.

[0127] 5. Analysis and processing of actual monitoring data:

[0128] 5.1. Actual data preprocessing.

[0129] All data obtained from actual monitoring undergo the same preprocessing steps.

[0130] 5.2 Actual data analysis.

[0131] The preprocessed CT image containing PZT metal artifact interference is input into the PZT artifact suppression module to obtain the corresponding CT image with PZT metal artifact removed and the PZT location map.

[0132] 5.3 Dataset Update Mechanism

[0133] When any of the following conditions are triggered, the input image containing PZT metal artifacts, the image after artifact removal, and the extracted PZT locations are used as data volumes to update the training dataset.

[0134] (1) The current nth image containing artifacts Compared with the (n-1)th image containing artifacts Perform image similarity analysis to obtain the image similarity between two adjacent images containing artifacts (the nth and (n-1th) images). When this value is greater than 5%.

[0135] (2) The current image after removing artifacts for the nth time Compared with the (n-1)th artifact removal image Image similarity analysis was performed to obtain the image similarity between two adjacent artifact-removed images (the nth and (n-1th) images). When this value is greater than 5%.

[0136] (3) Current position of the nth PZT Location map of PZT for the (n-1)th time Image similarity analysis was performed to obtain the image similarity between two adjacent PZT location maps (the nth and (n-1th) times). When this value is greater than 3%.

[0137] The formula for calculating image similarity is: , , The dimensions of the 3D image. , , For indexing the images:

[0138]

[0139] 6. Parameter update module and parameter update:

[0140] 6.1 Module update mechanism.

[0141] The module triggers an update when the following conditions are met.

[0142] (1) After the experiment started, the number of accumulated update data volumes was greater than 5.

[0143] (2) In 5.3 arbitrary triggering conditions, two adjacent images (i.e., the first) and The similarity difference value of the two figures is greater than 10%.

[0144] 6.2 Parameter update method.

[0145] Based on the parameters of the aforementioned PZT artifact suppression module, this experiment uses newly added data, specifically the data updated in section 5.3 (both sets of data remain the same). To maintain consistent training volume, the PZT artifact suppression module is fine-tuned to improve training speed and maintain update efficiency.

[0146] 6.3 Update the parameters of the PZT artifact suppression module.

[0147] The present invention provides a multi-physics field monitoring method for suppressing CT metal artifacts based on artificial intelligence. This method utilizes an adaptive AI network update mechanism to dynamically suppress PZT metal artifacts in CT images, thereby improving the accuracy of CT image analysis.

[0148] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that variations may be made to these embodiments without departing from the principles and spirit of the invention. The scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence, characterized in that, The method includes: Indoor rock physical hydraulic fracturing experiments were conducted, and multi-physics field monitoring was performed. The training dataset is constructed as follows: The dataset includes CT data without piezoelectric ceramic sensor metal artifacts, CT data with piezoelectric ceramic sensor metal artifacts, and piezoelectric ceramic sensor location maps. The three are matched and correspond to form a data volume, and all data exist in the form of a data volume. The dataset consists of real rock sample experimental data and simulated synthetic data. The data is augmented to improve the generality of the dataset. All data undergo the same preprocessing steps. A dual-domain, three-channel artifact suppression module is constructed, utilizing information from both the image domain and the chordmap domain. The first channel is constructed as a UNet-type submodule based on the image domain, used to obtain the position information of the piezoelectric ceramic sensor and provide constraints for artifact suppression. The second channel is a UNet-type submodule jointly enhanced by the image and chordmap domains, used to suppress the influence of metal artifacts from the piezoelectric ceramic sensor. The third channel is a chordmap domain channel. The UNet-type submodule of the first channel uses a skip-layer connection, and the UNet-type submodule of the first channel and the UNet-type submodule of the second channel share piezoelectric ceramic sensor information through a cross-spatial attention mechanism to provide information constraints for the artifact suppression process. Train the detection network using the training dataset; All data obtained from actual monitoring are preprocessed using the same steps. The preprocessed CT data containing metal artifacts of the piezoelectric ceramic sensor is then input into the artifact suppression module to obtain the corresponding CT data with metal artifacts of the piezoelectric ceramic sensor removed and the piezoelectric ceramic sensor location map. Update the dataset and artifact suppression module.

2. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 1, characterized in that, The phrase "conducting indoor rock physical hydraulic fracturing experiments and conducting multi-physics field monitoring" includes: Rock samples of different sizes were collected according to the research objectives. The rock samples were covered with a prefabricated rubber sleeve. The rubber sleeve had multiple probe holes, but no piezoelectric ceramic sensors were attached to the holes. A single CT data was collected from a rock sample without a piezoelectric ceramic sensor attached using an indoor hydraulic fracturing experimental setup. This CT data did not contain metal artifacts from the piezoelectric ceramic sensor and served as the first-stage CT data. The piezoelectric ceramic sensor is placed in the probe hole and then bonded to the surface of the rock sample. The indoor hydraulic fracturing test apparatus was used to collect CT data on a rock sample with a piezoelectric ceramic sensor attached. The spatial position of the piezoelectric ceramic sensor was then determined using the CT data. The CT data included metal artifacts of the piezoelectric ceramic sensor and was used as the second-stage CT data. Different loading strategies were employed to pressurize the rock samples. The pressurization phase included an isotropic loading phase, an axial pressure increase phase, a water injection phase to increase pore pressure, and a pressure unloading phase. During the water injection phase to increase pore pressure, several CT data points were acquired at set intervals. These CT data points included metal artifacts from the piezoelectric ceramic sensor and were used as the third-stage CT data. During the pressure unloading phase, one CT data point was acquired, which also included metal artifacts from the piezoelectric ceramic sensor and was used as the fourth-stage CT data. Finally, the piezoelectric ceramic sensor was removed from the rock sample, and another CT data point was acquired. This new CT data point did not contain metal artifacts from the piezoelectric ceramic sensor and was used as the fifth-stage CT data. All acquired CT data were processed.

3. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 2, characterized in that: The rock sample has the following dimensions: 50 mm in diameter and 125 mm in length; it is cylindrical. Rock samples include two types: ordinary rock samples and disturbed rock samples; The common rock sample types include sandstone and shale; they are further divided into five categories based on whether they contain bedding and the direction of bedding: bedding directions of 0±20°, 45±20°, 90±20°, 135±20°, and homogeneous media without bedding; and two categories based on axial compressive strength: 50±10MPa and 90±10MPa. Five rock samples were selected from each category, for a total of 100 rock samples. The interfering rock samples include two types of rocks: sandstone and shale. They are homogeneous media without bedding and contain sediments. Five rock samples of each type are selected, for a total of 10 rock samples.

4. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 3, characterized in that, The generation of real rock sample experimental data includes: Using ordinary rock samples and interfering rock samples, CT data from the first stage, the second stage, the fourth stage, and the fifth stage were collected to form the total dataset.

5. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 1, characterized in that: In the "Training the Detection Network Using the Training Dataset" step, the dataset is divided into a training set and a test set in a ratio of 8:

2. The artifact suppression module adopts the stochastic gradient descent optimization method, sets a dynamic learning rate, initially set to 0.0001, and halved every 50 iterations. The batch size is set to 40, and the number of iterations is 200. The training of the artifact suppression module is performed on the GPU image processing unit.

6. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 1, characterized in that, When any of the following conditions are triggered, the CT data containing piezoelectric ceramic sensor metal artifacts, the CT data without piezoelectric ceramic sensor metal artifacts, and the extracted piezoelectric ceramic sensor location map will be used as the data volume to update the training dataset: The current nth image containing artifacts Compared with the (n-1)th image containing artifacts Perform image similarity analysis to obtain the image similarity between two adjacent images containing artifacts. When this value is greater than 5%; or Current image after removing artifacts for the nth time Compared with the (n-1)th artifact removal image Image similarity analysis was performed to obtain the image similarity between two adjacent images after artifact removal. When this value is greater than 5%; or Current position diagram of the nth piezoelectric ceramic sensor Location diagram of the (n-1)th piezoelectric ceramic sensor Image similarity analysis was performed to obtain the image similarity between two adjacent images of the electro-ceramic sensor positions. When this value is greater than 3%.

7. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 6, characterized in that, The artifact suppression module is updated when the following conditions are met: After the experiment begins, the cumulative number of updated data volumes is greater than 5; or No. and The similarity difference between the images is greater than 10%.

8. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 2, characterized in that: The probe holes on the rubber sleeve are arranged in four rows at 0°, 90°, 180° and 270° along the circumference of the rock sample. Multiple probe holes are evenly arranged in each row. The height of the piezoelectric ceramic sensor in every two rows of probe holes is different, so that the four rows of piezoelectric ceramic sensors do not appear at the same height at the same time.

9. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 2, characterized in that: The indoor hydraulic fracturing experimental setup includes a pressure vessel, a loading system, an acoustic emission counting and waveform acquisition system, and a CT monitoring system.

10. The method for suppressing CT metal artifacts based on multiphysics field monitoring using artificial intelligence according to claim 2, characterized in that: During the pressurization of rock samples using different loading strategies, active source ultrasonic data were collected at set times. During active source ultrasonic monitoring, some piezoelectric ceramic sensors acted as transmitting probes to excite ultrasonic signals, while the remaining piezoelectric ceramic sensors acted as receiving probes to receive ultrasonic signals. During the pressurization of rock samples using different loading strategies, the piezoelectric ceramic sensor is used as both a transmitting and receiving probe during active source ultrasonic acquisition, and at other times it serves as a receiving probe to receive acoustic emission signals generated by changes in the rock sample.

Images