Modularized X-ray analysis device and method for oil-gas geology

Abstract

The invention relates to the technical field of X-ray analysis instrument software, in particular to a modular X-ray analysis device and method for oil and gas geology, a time-sharing control unit drives a multi-target X-ray source to alternately excite fluorescence and diffraction signals, and time-sharing or synchronous acquisition of XRF and XRD is realized in combination with a time schedule controller; the environment adaptation module identifies sample types based on a classification model and switches a vacuum or atmospheric environment, the anti-seismic unit inhibits external vibration and temperature drift interference, the dynamic strategy module triggers an energy filtering mode according to a mineral phase identification result, a monochromator improves diffraction resolution, and a gradient descent algorithm iteratively updates model parameters; and target material parameters and time sequence weights are synchronously adjusted. According to the invention, the problems of low detection efficiency, data splitting and insufficient allotrope identification capability caused by independent operation of XRF / XRD are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of X-ray analysis instrument software, and particularly to a modular X-ray analysis device and method for oil and gas geology. Background Art

[0002] X-ray fluorescence spectroscopy (XRF) is a non-destructive analysis technique that measures the characteristic fluorescence spectra released by a substance after being excited by high-energy X-rays, and analyzes the types and contents of elements in the sample. Its principle is based on the characteristic peaks of energy differences generated by the electron transitions of different elemental atoms after being excited. X-ray diffraction (XRD) utilizes the Bragg diffraction phenomenon generated by the interaction between X-rays and crystalline materials, and analyzes the crystal structure, phase composition, lattice parameters and other microstructural information of the material by analyzing the diffraction angle and intensity distribution. The two techniques respectively provide material characterization data from the dimensions of elemental composition and crystal structure. XRF can quickly locate the elemental composition, while XRD can accurately identify the compound morphology. The two form a complementary relationship in material analysis, and their combined application can effectively solve the complex analysis requirements of allotropes that cannot be distinguished by a single technique or samples with the same elements but different phase structures.

[0003] In the prior art, since XRF and XRD devices operate independently, the detection of elemental composition and phase structure needs to be completed step by step, resulting in a long detection cycle, poor data correlation, and the inability to effectively distinguish samples with the same elements but different phase isomers (such as hematite and magnetite). For example, in geological cuttings analysis, the existing methods need to use XRF to measure the elemental content and XRD to identify the mineral composition respectively. The process is cumbersome and it is difficult to synchronously verify the correlation between elemental distribution and phase genesis, affecting the efficiency and accuracy of comprehensive judgment. Summary of the Invention

[0004] Aiming at the deficiencies of the prior art, the present invention provides a modular X-ray analysis device and method for oil and gas geology, and the present invention solves the problems of low detection efficiency, fragmentation of elemental composition and phase structure data, and insufficient ability to identify allotropes or samples with the same elements but different phase structures caused by the independent operation of existing XRF and XRD technologies.

[0005] To solve the above technical problems, the specific technical solutions of the present invention are as follows:

[0006] In the first aspect, a modular X-ray analysis device for oil and gas geology provided by the present invention includes:

[0007] A collection module, configured to trigger a sample to generate an X-ray fluorescence signal and an X-ray diffraction signal through a preset multi-target X-ray source, synchronously collect the energy and intensity distribution curves of the fluorescence signal and the diffraction angle and intensity pattern of the diffraction signal, and transmit the collected signals to the data processing module;

[0008] An environment adaptation module, connected to the acquisition module, is configured to identify the sample type based on the energy distribution curve of the fluorescence signal and the diffraction angle pattern of the diffraction signal through a pre-trained classification model, obtain the sample type, and switch between a vacuum or atmospheric environment according to the sample type;

[0009] A data processing module, which receives the fluorescence signal and diffraction signal transmitted by the acquisition module, calibrates the energy channels of the fluorescence signal, and performs angle and intensity normalization processing on the diffraction signal to generate a standardized data set with time-series correlation;

[0010] An intelligent analysis module, which calls the element and phase mapping rules in the calibration database, inputs the standardized data set into a pre-trained random forest model, outputs the correlation between element content and mineral phase composition, and generates reservoir permeability and porosity prediction results in combination with core scan data;

[0011] A dynamic strategy module, which generates a three-dimensional visualization report based on the correlation and reservoir parameter prediction results output by the intelligent analysis module, and adjusts the target material selection or signal acquisition mode of the X-ray source in the acquisition module through mobile interaction instructions.

[0012] Furthermore, for the modular X-ray analysis device for oil and gas geology of the present invention, the acquisition module includes:

[0013] A time-sharing control unit, connected to the data processing module through a timing controller, is configured to alternately activate fluorescence signal acquisition and diffraction signal acquisition according to a preset timing instruction;

[0014] Among them, during the fluorescence signal acquisition stage, the time-sharing control unit drives the first target material to excite the sample to generate an energy distribution curve, and transmits the energy distribution curve to the data processing module through the timing controller;

[0015] During the diffraction signal acquisition stage, the time-sharing control unit switches to the second target material to generate a diffraction pattern, synchronously triggers the sample stage to rotate to eliminate crystal orientation deviation, and transmits the diffraction pattern to the data processing module through the timing controller.

[0016] Furthermore, for the modular X-ray analysis device for oil and gas geology of the present invention, the environment adaptation module includes:

[0017] A vacuum switching unit, connected to the time-sharing control unit in the acquisition module, is configured to start a vacuum environment according to the acquisition stage instruction of the time-sharing control unit during light element detection, switch to an atmospheric mode during bulk sample detection, and send an environment status signal to the data processing module to trigger timing calibration;

[0018] The anti-seismic unit is connected to the X-ray source and the detector in the acquisition module through a shock-absorbing base, monitors the external vibration intensity in real time, activates the dynamic damping when the vibration intensity exceeds a preset threshold, and controls the temperatures of the X-ray source and the detector within a preset threshold through a heat dissipation system. The temperature control signal of the anti-seismic unit is synchronously transmitted to the data processing module to adjust the noise filtering parameters.

[0019] Furthermore, for the modular X-ray analysis device for oil and gas geology of the present invention, the intelligent analysis module includes:

[0020] A data fusion unit, which receives the standardized data set generated by the data processing module, aligns the standardized data set with the core scan imaging data through a spatial coordinate registration algorithm, and generates a three-dimensional model including element distribution and mineral phases. The three-dimensional model serves as the input data for the pre-trained random forest model;

[0021] A verification unit, which is connected to the data fusion unit, receives the element quantitative data of a third-party detection device, corrects the mineral phase proportion error output by the random forest model based on the element distribution data and mineral phase proportion data in the three-dimensional model through an error comparison algorithm, and feeds back the corrected error value to the data fusion unit to update the spatial coordinate registration parameters.

[0022] Furthermore, for the modular X-ray analysis device for oil and gas geology of the present invention, the dynamic strategy module includes:

[0023] A mode switching unit, which is connected to the time-sharing control unit in the acquisition module and the intelligent analysis module, responds to a mobile terminal instruction to adjust the signal acquisition mode to time-sharing acquisition or energy-filtered synchronous acquisition, and sends a mode switching instruction to the intelligent analysis module to update the input parameters of the pre-trained random forest model. The input parameters include the timing correlation weight of the fluorescence signal energy distribution curve and the diffraction pattern collected in the time-sharing mode;

[0024] An algorithm iteration unit, which is connected to the verification unit in the intelligent analysis module, updates the training parameters of the pre-trained random forest model through a gradient descent algorithm according to the mineral phase data in the regional geological database and the corrected error value output by the verification unit, optimizes the non-linear mapping rule of the mineral phase proportion and the element content, and synchronizes the updated model parameters to the intelligent analysis module.

[0025] In a second aspect, the modular X-ray analysis method for oil and gas geology provided by the present invention is applied to the modular X-ray analysis device for oil and gas geology as described above, and includes:

[0026] Step 1, obtaining the preprocessed sample data, generating X-ray fluorescence signals and X-ray diffraction signals by time-sharing or synchronously exciting the sample through a preset multi-target X-ray source, and alternately collecting the energy spectrum distribution and diffraction pattern of the fluorescence signal based on a time-sharing control instruction;

[0027] Step 2, based on the energy distribution curve of the fluorescence signal and the diffraction angle spectrum of the diffraction signal, the sample type is identified by a pre-trained classification model to obtain the sample type, and the vacuum or atmospheric environment is switched according to the sample type;

[0028] Step 3, calibrate the energy channel of the collected fluorescence signal, and perform angle and intensity normalization on the diffraction signal to generate a time-series-related standardized data set;

[0029] Step 4, aligning the standardized data set with the core scanning imaging data through a spatial coordinate registration algorithm to generate a three-dimensional model including element distribution and mineral phases, and inputting the three-dimensional model into a pre-trained random forest model to output the correlation between element content and mineral phase composition and the prediction results of reservoir permeability and porosity;

[0030] Step 5: Generate a three-dimensional visualization report based on the association relationship, adjust the target material selection or signal acquisition mode of the X-ray source according to the real-time feedback, and synchronously update the input parameters and training parameters of the random forest model.

[0031] Furthermore, in the modular X-ray analysis method for oil and gas geology of the present invention, step 1 comprises:

[0032] In the time-sharing mode, the first target is activated first according to the preset timing instructions to generate fluorescence signals and record the full element energy spectrum. After the acquisition is completed, the timing controller switches to the second target to generate diffraction signals, and the angle of the goniometer is adjusted to the preset diffraction angle range and the sample stage is triggered to rotate to eliminate the crystal orientation deviation.

[0033] In the energy filtering mode, the monochromator is activated based on the interactive instructions of the mobile terminal to separate the fluorescence signal and the diffraction signal, and the intensity distribution of the two is recorded synchronously, and the timing information of the fluorescence signal and the diffraction signal is associated through the timestamp.

[0034] Furthermore, in the modular X-ray analysis method for oil and gas geology of the present invention, step 2 comprises:

[0035] Based on the fluorescence signal energy distribution curve collected in step 1, the peak position shift caused by the detector gain drift is eliminated by an energy channel calibration algorithm, and the calibrated energy spectrum is matched with the element characteristic peaks in the calibration database to determine the content range of the target element in the sample;

[0036] Based on the diffraction angle spectrum of the diffraction signal collected in Step 1, the diffraction signal is processed by an angle and intensity normalization algorithm, and the lattice spacing is calculated in combination with the preset Bragg diffraction equation, and compared with the diffraction peaks in the mineral phase standard sample database to identify the types and proportions of mineral phases in the sample.

[0037] Further, in the modular X-ray analysis method for oil and gas geology of the present invention, Step 4 includes:

[0038] Align the standardized data set generated in Step 3 with the core scan imaging data through a spatial coordinate registration algorithm to generate a three-dimensional model including element distribution and mineral phases, and input the three-dimensional model into a pre-trained random forest model to establish an association relationship between the element content threshold and the mineral phase proportion through a non-linear mapping rule;

[0039] Receive the element quantitative data of a third-party detection device, correct the mineral phase proportion error output by the random forest model through an error comparison algorithm based on the element and phase mapping rule in the calibration database, and feedback the corrected error value to the spatial coordinate registration algorithm to update the coordinate registration parameters of the three-dimensional model.

[0040] Further, in the modular X-ray analysis method for oil and gas geology of the present invention, Step 5 includes:

[0041] When samples of allotropes or elements with the same element but different phase structures are detected based on the mineral phase composition association relationship output in Step 4, switch to the energy filtering mode according to the activation instruction of the energy filtering mode, improve the resolution of the diffraction signal through a monochromator, and update the training parameters of the random forest model through a gradient descent algorithm based on the error value corrected in Step 9;

[0042] According to the reservoir permeability and porosity prediction results generated in Step 4, dynamically adjust the target tube voltage and beam current intensity in the time-sharing control instruction to adapt to the detection requirements of different geological lithologies, and synchronously optimize the timing correlation weight to match the real-time detection mode.

[0043] Advantages of the present invention;

[0044] The beneficial effects of the present invention are reflected in the collaborative optimization of the time-sharing control unit and the dynamic strategy module, achieving efficient acquisition and data fusion of XRF and XRD signals. The time-sharing control unit alternately drives the multi-target excitation fluorescence and diffraction signals based on timing instructions, combines the rotation of the sample stage to eliminate crystal orientation deviation, significantly shortens the detection cycle and improves data synchronization; the intelligent analysis module uses a spatial coordinate registration algorithm to align the standardized data set with the core scan imaging data, constructs a three-dimensional model integrating element distribution and mineral phase, and establishes an association rule between the element content threshold and the mineral phase ratio through a random forest model to solve the problem of the disconnection between element composition and phase structure data; the dynamic strategy module triggers the energy filtering mode according to the mineral phase identification result, the monochromator filters out non-target energy segments to improve the diffraction resolution, and combines the gradient descent algorithm to iteratively optimize the model parameters to enhance the discrimination ability of allotropes and phase isomerism samples. Each module dynamically adjusts the target parameters and acquisition strategies through a closed-loop feedback mechanism, realizing the integrated analysis of oil and gas reservoir characteristics while maintaining the detection efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the present invention, the following will briefly introduce the drawings required in the embodiments. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the drawings.

[0046] Figure 1 It is a flowchart of a modular X-ray analysis device and method for oil and gas geology provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with the specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. The following will describe in detail the technical solutions provided by each embodiment of the present invention in conjunction with the drawings. To better understand the objectives of the present invention, the present invention will be further described in detail below.

[0048] In a first aspect, a modular X-ray analysis device for oil and gas geology provided by the present invention includes:

[0049] An acquisition module for triggering a sample to generate X-ray fluorescence signals and X-ray diffraction signals through a preset multi-target X-ray source, synchronously collecting the energy and intensity distribution curves of the fluorescence signals and the diffraction angles and intensity spectra of the diffraction signals, and transmitting the collected signals to a data processing module;

[0050] An environmental adaptation module, connected to the acquisition module, for identifying the sample type based on the energy distribution curve of the fluorescence signal and the diffraction angle pattern of the diffraction signal through a pre-trained classification model to obtain the sample type, and switching between a vacuum or atmospheric environment according to the sample type;

[0051] A data processing module, which receives the fluorescence signal and diffraction signal transmitted by the acquisition module, calibrates the energy channels of the fluorescence signal, and normalizes the angle and intensity of the diffraction signal to generate a time-sequentially correlated standardized data set;

[0052] An intelligent analysis module, which calls the element-phase mapping rules in the calibration database, inputs the standardized data set into a pre-trained random forest model, outputs the correlation between element content and mineral phase composition, and generates reservoir permeability and porosity prediction results in combination with core scan data;

[0053] A dynamic strategy module, which generates a three-dimensional visualization report based on the correlation and reservoir parameter prediction results output by the intelligent analysis module, and adjusts the target material selection or signal acquisition mode of the X-ray source in the acquisition module through mobile interaction instructions.

[0054] A modular X-ray analysis device for oil and gas geology provided by the present invention realizes the functions and logical connection of each module through the following technical solutions:

[0055] The acquisition module triggers the sample to generate X-ray fluorescence signals and X-ray diffraction signals through a multi-target X-ray source, specifically including selecting a Cu / Mo dual target as the excitation source and dynamically adjusting the tube voltage and beam current according to the sample characteristics. The fluorescence signal is captured by a silicon drift detector for the full element energy spectrum (1 - 50 keV), and the diffraction signal is recorded by a two-dimensional array detector for the diffraction angle (5° - 80°) and intensity distribution. The acquisition module is built with a time-sharing control unit, which alternately activates the fluorescence and diffraction signal acquisition processes through a timing controller. During the fluorescence acquisition stage, the Mo target is preferentially driven to generate high-energy X-rays. After the signal recording is completed, it switches to the Cu target to excite the diffraction signal, and simultaneously triggers the sample stage to rotate to eliminate crystal orientation deviation. The collected original signals are preliminarily denoised by an independent arithmetic unit and then transmitted to the data processing module.

[0056] The environment adaptation module identifies the sample type through a pre-trained classification model, which is trained based on the XRF / XRD joint dataset of NIST reference materials and mineral standard samples. It extracts the peak position features of the fluorescence energy spectrum and the angular distribution features of the diffraction pattern through a convolutional neural network, and outputs the classification result of the sample type (such as powder, bulk or thin film). According to the classification result, the vacuum switching unit activates the vacuum environment during light element detection to reduce air scattering interference, and switches to the atmospheric mode during bulk sample detection to adapt to the rapid loading requirement. The environment adaptation module monitors the external vibration intensity in real time. When the vibration exceeds the preset threshold, it activates the dynamic damping system and controls the temperature of the X-ray source and the detector below 40°C through the heat dissipation unit. The temperature control signal is synchronously transmitted to the data processing module to adjust the noise filtering parameters.

[0057] The data processing module calibrates the energy channels of the received fluorescence signals, corrects the energy deviation caused by the detector gain drift through the characteristic peak positions in the calibration database, and ensures the accurate matching of the fluorescence energy spectrum and the element types. It performs angular and intensity normalization on the diffraction signals, calculates the lattice spacing based on the Bragg equation, and eliminates the influence of the sample surface roughness on the diffraction intensity. The calibrated fluorescence data and the normalized diffraction data are correlated by time stamps to generate a standardized dataset, providing temporally consistent inputs for subsequent analysis.

[0058] The intelligent analysis module calls the element and phase mapping rules in the calibration database, aligns the standardized dataset and the core scan imaging data through a spatial coordinate registration algorithm, and generates a three-dimensional model containing element distribution and mineral phases. This model is input into a pre-trained random forest model to establish the correlation between the element content (such as Fe = 15%) and the mineral phase ratio (such as hematite:pyrite = 3:1) through a non-linear mapping rule. The verification unit introduces third-party detection data (such as the SEM-EDS element quantitative results), corrects the mineral phase ratio error output by the model through an error comparison algorithm, and feeds back the correction result to the coordinate registration algorithm to optimize the accuracy of the three-dimensional model.

[0059] The dynamic strategy module generates a visualization report superimposing the element heat map and the three-dimensional distribution of mineral phases based on the reservoir permeability and porosity prediction results output by the intelligent analysis module. It adjusts the acquisition mode in real time through mobile interaction instructions: optimizing the temporal correlation weight in the time-sharing mode to balance the XRF / XRD detection efficiency; activating the monochromator in the energy filtering mode to improve the diffraction signal resolution. The algorithm iteration unit updates the training parameters of the random forest model based on the regional geological database, optimizes the non-linear mapping rule between the mineral phase and the element content through the gradient descent algorithm, and forms a closed-loop optimized reservoir property analysis system.

[0060] Data is transmitted between modules through a high-speed serial bus (LVDS) and Gigabit Ethernet. The timing controller ensures the synchronization of time-division acquisition and signal processing. Sample identification, environmental adaptation, data processing, and intelligent analysis form a series link. The dynamic strategy module uses a feedback mechanism to reverse-optimize the acquisition parameters and analysis model, achieving efficient analysis of oil and gas geological samples and dynamic prediction of reservoir characteristics.

[0061] Specifically, for the modular X-ray analysis device for oil and gas geology of the present invention, the acquisition module includes:

[0062] A time-division control unit, connected to the data processing module through a timing controller, configured to alternately activate fluorescence signal acquisition and diffraction signal acquisition according to a preset timing instruction;

[0063] Among them, in the fluorescence signal acquisition stage, the time-division control unit drives the first target to excite the sample to generate an energy distribution curve, and transmits the energy distribution curve to the data processing module through the timing controller;

[0064] In the diffraction signal acquisition stage, the time-division control unit switches to the second target to generate a diffraction pattern, synchronously triggers the sample stage to rotate to eliminate crystal orientation deviation, and transmits the diffraction pattern to the data processing module through the timing controller.

[0065] The time-division control unit in the acquisition module of the present invention is connected to the data processing module through a timing controller to achieve alternate acquisition of fluorescence signals and diffraction signals. Based on a preset timing instruction, in the fluorescence signal acquisition stage, the time-division control unit drives the first target (such as a Mo target) to excite the sample to generate an energy distribution curve. The Mo target operates in a high-energy X-ray mode to excite the characteristic fluorescence spectrum in the sample, and the silicon drift detector captures the energy distribution signal with a microsecond-level response time. The energy distribution curve is synchronized with the subsequent data processing stage through the timestamp marking of the timing controller and transmitted to the data processing module for energy channel calibration.

[0066] In the diffraction signal acquisition stage, the time-division control unit switches to the second target (such as a Cu target) to excite the diffraction signal of the sample through low-energy X-rays. The switching process is completed by an electromagnetic drive mechanism, and the target switching time is controlled within milliseconds to avoid signal interference. The sample stage rotation device is synchronously triggered, and the sample stage is driven by a stepper motor to continuously rotate at a preset angular step (such as 0.1°) to eliminate the influence of crystal orientation deviation on the diffraction pattern. The two-dimensional array detector records the diffraction intensity distribution at different rotation angles to generate map data containing the relationship between diffraction angle and intensity. The diffraction pattern is associated with the timestamp of the fluorescence signal through the timing controller, and after ensuring timing consistency, it is transmitted to the data processing module.

[0067] The timing instructions of the time-sharing control unit are generated by the FPGA controller. The preset instructions include the acquisition duration of the fluorescence signal (such as 100 ms), the acquisition interval of the diffraction signal (such as 200 ms), and the sample stage rotation angle sequence. The FPGA controller receives the signal quality evaluation results feedback by the data processing module through Gigabit Ethernet, and dynamically adjusts the timing parameters to optimize the acquisition efficiency. For example, when the fluorescence signal intensity is lower than the threshold, the excitation time of the Mo target is extended or the tube current is increased to ensure the signal-to-noise ratio of the energy distribution curve.

[0068] As a data transfer node, the timing controller preliminarily encapsulates the energy distribution curve of the fluorescence signal and the diffraction pattern, and adds time stamps and environmental status tags (such as vacuum degree, temperature). The encapsulated data is transmitted to the data processing module through a high-speed serial bus, and the bus protocol adopts the LVDS standard to reduce the transmission delay. The synchronization between the sample stage rotation angle and the diffraction signal acquisition is calibrated in real time by an optical encoder, and the encoder signal is fed back to the time-sharing control unit to dynamically correct the rotation step deviation and ensure the accuracy of the diffraction angle measurement.

[0069] The interaction logic between the time-sharing control unit and the data processing module forms a closed-loop control link. After completing a single time-sharing acquisition, the data processing module sends a new round of timing instructions to the time-sharing control unit based on the signal integrity analysis results to achieve the adaptive acquisition mode switch. For example, when a high content of light elements is detected, the acquisition cycle of the fluorescence signal is shortened and the vacuum environment is preferentially activated to improve the recognition accuracy of the light element characteristic peaks. This time-sharing control mechanism maximizes the utilization rate of hardware resources while ensuring the data coordination of XRF and XRD, meeting the high-throughput detection requirements of oil and gas geological samples.

[0070] Specifically, for the modular X-ray analysis device for oil and gas geology described in the present invention, the environment adaptation module includes:

[0071] A vacuum switching unit, connected to the time-sharing control unit in the acquisition module, is used to start the vacuum environment according to the acquisition stage instructions of the time-sharing control unit during light element detection, switch to the atmospheric mode during bulk sample detection, and send an environmental status signal to the data processing module to trigger timing calibration;

[0072] An earthquake-resistant unit, connected to the X-ray source and the detector in the acquisition module through a shock-absorbing base, real-time monitors the external vibration intensity, activates dynamic damping when the vibration intensity exceeds the preset threshold, and controls the temperatures of the X-ray source and the detector within the preset threshold through a heat dissipation system. The temperature control signal of the earthquake-resistant unit is synchronously transmitted to the data processing module to adjust the noise filtering parameters.

[0073] The vacuum switching unit in the environmental adaptation module of the present invention is connected to the time-sharing control unit and dynamically adjusts the detection environment according to the acquisition stage instruction. When the time-sharing control unit activates the fluorescence signal acquisition mode, the vacuum switching unit determines whether it is a light element detection requirement based on the sample type recognition result: if it is recognized as a light element (such as C, O), the vacuum pump is started to reduce the air pressure in the detection chamber to below 10 -3 Pa to eliminate the scattering interference of air molecules on low-energy X-rays; if it is recognized as a bulk core sample, it switches to the atmospheric mode and opens the sample chamber door rapid loading mechanism, and the chamber sealing state is monitored in real time through a pressure sensor. The environmental state signals (including vacuum degree, air pressure value and chamber door state) are transmitted to the data processing module through the CAN bus, triggering a timing calibration program to adjust the signal acquisition interval and match the detector response time in different environments.

[0074] The anti-seismic unit is rigidly connected to the X-ray source and the detector through a shock-absorbing base, and a piezoelectric vibration sensor is built in the base to collect triaxial vibration data in real time. When the vibration intensity exceeds the preset threshold (such as 0.5g), the dynamic damping system adjusts the damping coefficient through magnetorheological fluid and attenuates the vibration amplitude to a safe range within 10 ms. The temperatures of the X-ray source and the detector are monitored in real time by thermocouples, and the heat dissipation system uses a semiconductor refrigeration chip and air cooling to work together to control the temperature below 40°C. The temperature control signal is converted into a digital quantity through the ADC module and transmitted to the data processing module, and the noise filtering algorithm dynamically adjusts the filter cut-off frequency according to the temperature data to suppress the influence of thermal noise on the fluorescence energy spectrum baseline.

[0075] The data of the vacuum switching unit and the anti-seismic unit are synchronously interacted with the time-sharing control unit through a high-speed serial interface. For example, during the start-up stage of the vacuum environment, when the vibration data exceeds the threshold, the operation of closing the chamber door is automatically delayed to avoid mechanical shock causing seal failure; under high-temperature working conditions, the heat dissipation system preferentially starts the semiconductor refrigeration mode, and at the same time the time-sharing control unit extends the signal acquisition interval to reduce the load of the X-ray source. The environmental state signals and the vibration temperature data are fused in the data processing module to generate an environmental quality assessment index, and this index reversely controls the timing instruction cycle of the time-sharing control unit, forming a balance mechanism between the stability of the detection environment and the data acquisition efficiency.

[0076] The collaborative work process of each unit is as follows: After the time-sharing control unit sends the acquisition phase instruction, the vacuum switching unit switches the environment according to the sample type, and the seismic resistance unit simultaneously starts vibration suppression and temperature control; the environmental state signal triggers the data processing module to load the corresponding calibration parameter library (such as the energy spectrum gain correction coefficient in the vacuum mode), and the vibration temperature data optimizes the noise filtering algorithm in real time; after a single detection is completed, the environmental quality assessment index updates the timing strategy of the time-sharing control unit. For example, in a high-temperature and high-vibration scenario, the single acquisition duration is shortened and the heat dissipation cycle is increased. This design maintains the stability of the X-ray source and the detector under complex working conditions and ensures the accuracy of element and phase analysis data.

[0077] Specifically, for the modular X-ray analysis device for oil and gas geology of the present invention, the intelligent analysis module includes:

[0078] A data fusion unit, which receives the standardized data set generated by the data processing module, aligns the standardized data set with the core scan imaging data through a spatial coordinate registration algorithm, and generates a three-dimensional model containing element distribution and mineral phases. The three-dimensional model serves as the input data for the pre-trained random forest model;

[0079] A verification unit, which is connected to the data fusion unit, receives the element quantitative data of the third-party detection device, corrects the mineral phase proportion error output by the random forest model based on the element distribution data and mineral phase proportion data in the three-dimensional model through an error comparison algorithm, and feeds the corrected error value back to the data fusion unit to update the spatial coordinate registration parameters.

[0080] The data fusion unit in the intelligent analysis module of the present invention receives the standardized data set generated by the data processing module and aligns it with the core scan imaging data through a spatial coordinate registration algorithm. The standardized data set contains the calibration results of the energy channels of the fluorescence signal and the angle and intensity normalization data of the diffraction signal. The core scan imaging data obtains a high-resolution three-dimensional voxel model through a CT scanner. The spatial coordinate registration algorithm adopts the iterative closest point (ICP) method, uses the distribution of characteristic elements (such as Fe, Si) in the fluorescence signal as the reference points, maps the three-dimensional coordinates of the core scan data to the spatial coordinate system of the standardized data set, and generates a three-dimensional model that fuses element distribution and mineral phase information. This model labels the percentage of element content and the corresponding mineral phase categories (such as hematite, quartz) in units of voxels, serving as the input data structure of the pre-trained random forest model.

[0081] The verification unit receives the elemental quantitative data transmitted by a third-party detection device (such as SEM-EDS) through the RS-485 interface, and extracts the measured values of the elemental content in the target area. Based on the elemental distribution data and mineral phase proportion data in the corresponding area of the three-dimensional model, the error comparison algorithm calculates the mean square error between the model prediction value and the measured value, and identifies the areas where the prediction deviation of the mineral phase proportion exceeds the threshold (such as ±5%). The correction process uses the least squares method to optimize the weight parameters of the random forest model and adjusts the mineral phase classification decision boundary. The corrected error value is fed back to the data fusion unit through the backpropagation mechanism to update the registration weight coefficient in the spatial coordinate registration algorithm. For example, the coordinate matching accuracy weight of the high-error area is increased to optimize the data alignment effect during the generation of the next three-dimensional model.

[0082] The data fusion unit and the verification unit form a closed-loop optimization link. In a single analysis process, after the data fusion unit generates the initial three-dimensional model, the random forest model outputs the prediction result of the mineral phase composition; the verification unit introduces external measured data to perform error correction, and the correction result improves the accuracy of subsequent data fusion through the update of the registration parameters. For example, when there is a significant difference between the predicted proportion of the quartz phase in a certain area and the measured value of EDS, the adjustment of the registration weight coefficient will strengthen the spatial matching intensity between the core scan data and the fluorescence energy spectrum peak position in this area, and correct the coordinate offset caused by the uneven surface of the sample.

[0083] The data interaction between each unit is realized through a high-speed data bus, and the voxel data of the three-dimensional model and the error correction parameters adopt a hierarchical transmission protocol. The three-dimensional model output by the data fusion unit is stored in a tensor structure, including the elemental content matrix, the mineral phase classification matrix, and the coordinate transformation parameters, and is subjected to feature normalization processing before being input into the random forest model. The correction parameters of the verification unit are fed back to the data fusion unit through an independent channel to avoid conflicts with the main data stream. This design improves the accuracy of mineral phase analysis through a dynamic parameter optimization mechanism while maintaining the high-throughput data processing efficiency, meeting the analytical requirements for the complex mineral composition of oil and gas reservoirs.

[0084] Specifically, for the modular X-ray analysis device for oil and gas geology described in the present invention, the dynamic strategy module includes:

[0085] A mode switching unit, connected to the time-sharing control unit in the acquisition module and the intelligent analysis module, responds to a mobile terminal instruction to adjust the signal acquisition mode to time-sharing acquisition or energy-filtered synchronous acquisition, and sends a mode switching instruction to the intelligent analysis module to update the input parameters of the pre-trained random forest model. The input parameters include the temporal correlation weight between the fluorescence signal energy distribution curve and the diffraction pattern collected in the time-sharing mode;

[0086] The algorithm iteration unit is connected to the verification unit in the intelligent analysis module. According to the mineral phase data in the regional geological database and the corrected error value output by the verification unit, it updates the training parameters of the pre-trained random forest model through the gradient descent algorithm, optimizes the non-linear mapping rule between the mineral phase ratio and the element content, and synchronizes the updated model parameters to the intelligent analysis module.

[0087] The mode switching unit in the dynamic policy module of the present invention is connected to the time-sharing control unit and the intelligent analysis module to achieve dynamic adaptation of the acquisition mode and the model parameters. The mode switching unit receives the mobile terminal interaction instruction through the MQTT protocol and parses the mode switching request in the instruction (such as time-sharing acquisition or energy filtering synchronous acquisition). In the time-sharing acquisition mode, the mode switching instruction triggers the time-sharing control unit to adjust the timing correlation weight. For example, the proportion of the fluorescence signal acquisition period is increased from 60% to 80% to preferentially obtain high signal-to-noise ratio element distribution data. In the energy filtering synchronous acquisition mode, the monochromator is activated to filter out X-rays in non-target energy segments, and the intensity distributions of the fluorescence and diffraction signals are synchronously recorded. The timing correlation weight is adjusted to 1:1 to balance the data integrity. The updated input parameters (including the timing weight and the energy filtering threshold) are transmitted to the intelligent analysis module through the high-speed data bus. After the pre-trained random forest model loads the new parameters, it adjusts the weights of the input layer nodes to optimize the real-time response speed of the mineral phase classification.

[0088] The algorithm iteration unit accesses the regional geological database through the ODBC interface and extracts the typical mineral phase distribution data of the target area (such as the quartz / clay phase ratio range in shale). Combining the corrected error value output by the verification unit (such as the hematite prediction error of ±3%), the gradient descent algorithm aims to minimize the loss function and iteratively updates the connection weights of the hidden layer neurons of the random forest model. For example, when the prediction error of the clay phase in a certain area continues to be high, the algorithm increases the feature weights of the corresponding elements (such as Al and Si) and weakens the influence coefficient of the interfering elements (such as Fe) to optimize the non-linear mapping rule between the mineral phase ratio and the element content. The updated model parameters are synchronized to the intelligent analysis module after encryption verification, replacing the original parameter file. The synchronization process prevents data conflicts through the version control mechanism and maintains the continuity of the model iteration.

[0089] The mode switching unit and the algorithm iteration unit work together through the data bus. After the mode switching is completed, the algorithm iteration unit dynamically adjusts the learning rate and batch size of the gradient descent algorithm according to the new input parameter structure (such as the time-series correlation weight in the energy filtering mode) to adapt to the differences in data characteristics in different acquisition modes. For example, the high-frequency signal in the time-sharing acquisition mode requires a smaller learning rate to avoid overfitting, while the full spectrum data in the energy filtering mode uses a larger batch to improve the convergence efficiency. The mineral phase data of the regional geological database and the error data of the verification unit are weighted and fused during the iteration process to form a regional adaptive training set, which enhances the generalization ability of the model for specific oil and gas reservoirs.

[0090] The interaction between the dynamic strategy module and the intelligent analysis module forms a two-way optimization link. The reservoir parameter prediction results output by the intelligent analysis module are reversely input into the dynamic strategy module to drive the mode switching unit to optimize the acquisition strategy. For example, when the prediction results show the characteristics of a high-porosity reservoir, the mode switching unit automatically increases the acquisition frequency of the energy filtering mode to improve the identification accuracy of the clay mineral phase; the algorithm iteration unit synchronously calls the mineral phase data of the corresponding area to strengthen the association rules between porosity and mineral phase ratio. While maintaining the hardware acquisition efficiency, this closed-loop optimization mechanism improves the analysis accuracy of oil and gas geological samples through dynamic iteration of model parameters to meet the real-time analysis needs of complex reservoir characteristics.

[0091] Second, see Figure 1 The modular X-ray analysis method for oil and gas geology provided by the present invention is applied to the modular X-ray analysis device for oil and gas geology as described above, comprising:

[0092] Step 1, obtaining the preprocessed sample data, generating X-ray fluorescence signals and X-ray diffraction signals by time-sharing or synchronously exciting the sample through a preset multi-target X-ray source, and alternately collecting the energy spectrum distribution and diffraction pattern of the fluorescence signal based on a time-sharing control instruction;

[0093] Step 2, based on the energy distribution curve of the fluorescence signal and the diffraction angle spectrum of the diffraction signal, the sample type is identified by a pre-trained classification model to obtain the sample type, and the vacuum or atmospheric environment is switched according to the sample type;

[0094] Step 3, calibrate the energy channel of the collected fluorescence signal, and perform angle and intensity normalization on the diffraction signal to generate a time-series-related standardized data set;

[0095] Step 4, aligning the standardized data set with the core scanning imaging data through a spatial coordinate registration algorithm to generate a three-dimensional model including element distribution and mineral phases, and inputting the three-dimensional model into a pre-trained random forest model to output the correlation between element content and mineral phase composition and the prediction results of reservoir permeability and porosity;

[0096] Step 5: Generate a 3D visualization report based on the association relationship, adjust the target material selection or signal acquisition mode of the X-ray source according to real-time feedback, and synchronously update the input parameters and training parameters of the random forest model.

[0097] The modular X-ray analysis method for oil and gas geology provided by the present invention realizes the logical connection and functional coordination of each step through the following technical solutions:

[0098] In Step 1, the multi-target X-ray source is driven by a time-sharing control unit. In the time-sharing mode, the Mo target is preferentially activated to generate high-energy X-rays to excite the fluorescence signal of the sample, and the silicon drift detector collects the full-element energy spectrum at a microsecond-level response speed. After the fluorescence signal is collected, the time-sharing control unit switches to the Cu target to excite low-energy X-rays, and simultaneously triggers the sample stage stepping motor to rotate at a preset angular step to eliminate the crystal orientation deviation. The two-dimensional array detector records the diffraction intensity distribution at different rotation angles. In the energy filtering synchronization mode, the monochromator filters out the X-rays in the non-target energy band, synchronously collects the intensity distributions of the fluorescence and diffraction signals, and associates the timing information through timestamps.

[0099] In Step 2, the pre-trained classification model is constructed based on the XRF / XRD joint dataset of NIST reference materials and mineral standard samples. The fluorescence energy spectrum characteristic peak positions and the diffraction pattern angle distribution characteristics are extracted through a convolutional neural network, and the sample type classification result (such as powder, bulk or thin film) is output. The classification result triggers the action of the vacuum switching unit: when detecting light elements, the vacuum pump is started to reduce the chamber air pressure to below 10 -3 Pa to reduce air scattering interference; when detecting bulk samples, switch to the atmospheric mode and turn on the fast loading mechanism. The environmental status signal is transmitted to the data processing module through the CAN bus, triggering the timing calibration program to adjust the signal acquisition interval.

[0100] In Step 3, the energy channel calibration of the fluorescence signal corrects the detector gain drift through the characteristic peak positions in the calibration database. For example, the Fe-Kα peak position is corrected to 6.40 keV to match the characteristic energy range of the element. The angle and intensity normalization processing of the diffraction signal calculates the lattice spacing based on Bragg's equation to eliminate the influence of the sample surface roughness on the intensity distribution, generating a standardized dataset containing timestamps. The timing association is realized through a unified clock source, and the time deviation between the fluorescence and diffraction data is controlled within milliseconds.

[0101] In step 4, the spatial coordinate registration algorithm uses the distribution of characteristic elements (such as Fe, Si) in the fluorescence signal as a reference point to map the three-dimensional coordinates of the core scan CT data to the coordinate system of the standardized data set. The registered three-dimensional model annotates the element content percentage and mineral phase category (such as hematite accounting for 30%) in units of voxels and inputs them into the pre-trained random forest model. The model establishes the correlation between the element content threshold (such as Ca>5%) and the mineral phase ratio (calcite phase ≥15%) through non-linear decision tree rules, and outputs the predicted values of reservoir permeability and porosity in combination with the core pore structure data.

[0102] In step 5, the three-dimensional visualization report overlays the element distribution heat map and the three-dimensional model of the mineral phase, and marks the permeability anomaly area and the pore development zone. The real-time feedback mechanism dynamically adjusts the target material selection and acquisition mode according to the user interaction command or the model prediction result: optimizing the excitation duration ratio of the Mo / Cu target in the time-sharing mode; adjusting the monochromator filtering range in the energy filtering mode. The input parameters (such as the time series correlation weight) and training parameters (such as the number of hidden layer nodes) of the random forest model are iteratively updated through the gradient descent algorithm, and the updated parameters are synchronized to the intelligent analysis module through the version control mechanism to maintain the timeliness and regional adaptability of the model prediction.

[0103] Signals and parameters are transmitted between steps through a high-speed data bus, and the interaction between the time-sharing control unit and the data processing module ensures the synchronization of the acquisition timing and signal processing. Sample type identification, environment adaptation, data calibration, and model analysis form a series link, and the dynamic strategy module reversely optimizes the acquisition parameters and model structure based on the feedback data to achieve the integrated and efficient analysis of the element, phase, and reservoir parameters of oil and gas geological samples.

[0104] Specifically, for the modular X-ray analysis method for oil and gas geology described in the present invention, step 1 includes:

[0105] In the time-sharing mode, according to the preset timing instruction, the first target material is preferentially activated to generate a fluorescence signal and the full-element energy spectrum is recorded. After the acquisition is completed, the second target material is switched through the timing controller to generate a diffraction signal, and the angle of the goniometer is synchronously adjusted to the preset diffraction angle range and the sample stage is triggered to rotate to eliminate the crystal orientation deviation;

[0106] In the energy filtering mode, based on the mobile terminal interaction instruction, the monochromator is activated to separate the fluorescence signal and the diffraction signal, and the intensity distributions of both are synchronously recorded and the timing information of the fluorescence signal and the diffraction signal is associated through a timestamp.

[0107] In the time-sharing mode of Step 1 of the present invention, a timing controller is used to achieve time-sharing acquisition of fluorescence and diffraction signals. The timing controller generates preset timing instructions based on FPGA. During the fluorescence signal acquisition stage, it drives the first target (such as Mo target) to excite the sample with high-energy X-rays, and a silicon drift detector captures the full-element energy spectrum with a microsecond-level response speed. After preliminary noise reduction, the energy spectrum data is transmitted to the data processing module through a high-speed serial bus. After completing the fluorescence acquisition, the timing controller sends a switching instruction to the electromagnetic drive mechanism to switch to the second target (such as Cu target) to excite low-energy X-rays, and synchronously triggers the goniometer stepper motor to adjust the angle to the preset diffraction angle range (such as 20° - 80°). The sample stage is driven by a servo motor to rotate in steps of 0.1° to eliminate the influence of crystal orientation deviation on the diffraction pattern.

[0108] In the energy filtering mode, the mobile terminal interaction instruction is transmitted to the monochromator control unit through the MQTT protocol to activate the monochromator to filter out X-rays in non-target energy segments. The monochromator separates the fluorescence signal (5 - 20 keV) and the diffraction signal (Cu-Kα 8.04 keV) based on the multi-layer film reflection principle. In the synchronous acquisition mode, a two-dimensional array detector records the intensity distributions of both. The timing controller generates a timestamp through a GPS synchronous clock, controlling the acquisition time deviation between the fluorescence signal energy spectrum and the diffraction angle-intensity spectrum within the millisecond level to ensure the consistency of the associated timing information.

[0109] The switching logic between the time-sharing mode and the energy filtering mode is driven by the sample type recognition result. When the classification model identifies a light element or a complex mineral phase sample, it automatically switches to the energy filtering mode; for homogeneous bulk samples, the time-sharing mode is preferentially adopted to improve the detection efficiency. The timing controller dynamically adjusts the signal acquisition period during mode switching: the acquisition interval between fluorescence and diffraction in the time-sharing mode is 200 ms, and the synchronous acquisition period in the energy filtering mode is shortened to 50 ms. The coordination between the goniometer angle adjustment and the sample stage rotation is calibrated in real time through an optical encoder. The encoder signal is converted by the ADC module and then input into the timing controller to correct the diffraction angle deviation caused by mechanical transmission errors.

[0110] The data acquisition processes of both modes interact with the data processing module through a high-speed data bus. The timing correlation weights of the time-sharing mode (such as fluorescence:diffraction = 3:1) and the timestamp parameters of the energy filtering mode are stored in a non-volatile memory for subsequent standardized processing calls. This design maintains the hardware control accuracy while adapting to the analysis requirements of different oil and gas geological samples through a dynamic mode switching mechanism, improving the collaborative acquisition efficiency of element and phase data.

[0111] Specifically, for the modular X-ray analysis method for oil and gas geology described in the present invention, Step 2 includes:

[0112] Based on the fluorescence signal energy distribution curve collected in Step 1, the peak position shift caused by the detector gain drift is eliminated through the energy channel calibration algorithm, and the calibrated energy spectrum is matched with the element characteristic peaks in the calibration database to determine the content range of the target elements in the sample;

[0113] Based on the diffraction angle pattern of the diffraction signal collected in Step 1, the diffraction signal is processed through the angle and intensity normalization algorithm, the lattice spacing is calculated in combination with the preset Bragg diffraction equation, and compared with the diffraction peaks in the mineral phase standard sample database to identify the types and proportions of the mineral phases in the sample.

[0114] In Step 2 of the present invention, the accuracy of element and mineral phase analysis is improved through energy channel calibration and diffraction signal normalization processing. The specific implementation process is as follows:

[0115] Based on the fluorescence signal energy distribution curve collected in Step 1, the energy channel calibration algorithm eliminates the peak position shift caused by the detector gain drift through the reference peak positions in the calibration database (such as Fe-Kα = 6.40 keV, Cu-Kα = 8.04 keV). The calibration database stores the nominal energy values and allowable deviation ranges (±0.05 keV) of the element characteristic peaks. The calibration algorithm uses the least squares method to fit the linear relationship between the actual peak position and the nominal value to generate a gain correction coefficient matrix. The calibrated energy spectrum is compared with the characteristic peaks in the calibration database through the peak matching algorithm. During the matching process, a signal-to-noise ratio threshold (such as SNR>10) is introduced to filter out the noise interference peaks, determine the content range of the target elements (such as Fe, Si, Ca) in the sample (such as Fe = 12 - 18%), and calculate the confidence interval through the probability distribution model.

[0116] For the diffraction angle pattern of the diffraction signal collected in Step 1, the angle and intensity normalization algorithm first eliminates the intensity deviation caused by the surface roughness of the sample and the non-uniformity of the detector response. The normalization process is based on the Bragg diffraction equation, with the wavelength of the Cu-Kα ray (λ = 0.154 nm) as the reference, calculates the lattice spacing (d = λ / (2sinθ)) corresponding to each diffraction angle, and generates a standardized d-value - intensity distribution curve. The mineral phase standard sample database stores the diffraction peak positions and full-width at half-maximum parameters of typical minerals (such as quartz d = 3.34d, calcite d = 3.03d). The measured d values are matched with the standard peak positions in the database through the multi-peak fitting algorithm, and the mineral phase proportion (such as quartz accounting for 35%, clay minerals accounting for 50%) is calculated by combining the peak area integration method.

[0117] The data of energy channel calibration and diffraction signal normalization are correlated by timestamps. The calibrated element content range is input into the multi-peak fitting algorithm as a constraint condition for mineral phase analysis. For example, when the Fe content exceeds 15%, the diffraction peak positions of hematite are preferentially matched instead of those of magnetite. The mineral phase proportion data output by the normalization algorithm reversely optimizes the characteristic peak matching weights of the calibration database. For example, the matching tolerance of high-proportion mineral phases is tightened from ±0.5% to ±0.2%, improving the detection sensitivity of low-content mineral phases.

[0118] The collaborative processing of the two parts forms a cross-validation mechanism: the element content range after fluorescence signal calibration constrains the mineral phase identification results of the diffraction signal, while the mineral phase proportion data reversely corrects the weight coefficients of the element quantitative model. For example, when a high proportion of pyrite is identified by diffraction analysis, the Fe content threshold in the fluorescence signal is automatically increased to exclude the interference of iron oxide. This design improves the analysis accuracy of complex mineral compositions through data cross-verification, meeting the technical requirements for the coupled analysis of elements and phases in oil and gas reservoirs.

[0119] Specifically, for the modular X-ray analysis method for oil and gas geology described in the present invention, step 4 includes:

[0120] Align the standardized data set generated in step 3 with the core scan imaging data through a spatial coordinate registration algorithm to generate a three-dimensional model containing element distribution and mineral phases, and input the three-dimensional model into a pre-trained random forest model to establish the correlation between the element content threshold and the mineral phase proportion through a non-linear mapping rule;

[0121] Receive the element quantitative data of a third-party detection device, and based on the element-phase mapping rule in the calibration database, correct the mineral phase proportion error output by the random forest model through an error comparison algorithm, and feedback the corrected error value to the spatial coordinate registration algorithm to update the coordinate registration parameters of the three-dimensional model.

[0122] Step 4 of the present invention realizes the multi-dimensional correlation analysis of element-mineral phase-reservoir parameters through spatial coordinate registration and model iterative optimization. The specific implementation process is as follows:

[0123] The standardized data set contains time - series - related fluorescence energy spectrum calibration data and diffraction signal normalization results. The spatial coordinate registration algorithm uses the distribution of characteristic elements (such as Fe, Si) in the fluorescence signal as the spatial reference points, and maps the voxel coordinates of the core - scanned CT imaging data to the three - dimensional coordinate system of the standardized data set through the Iterative Closest Point (ICP) algorithm. During the registration process, a feature point cloud matching strategy is introduced, and the region with a gradient change in element content (such as the Fe content changing suddenly from 10% to 20%) is used as the key registration point to generate a three - dimensional model that integrates element distribution and mineral phase information. This model labels the percentage of element content (such as Fe = 15% ± 0.5%) and mineral phase classification labels (such as hematite, quartz) in units of voxels, and performs feature normalization processing before inputting it into the pre - trained random forest model to eliminate the influence of different dimensions on classification decisions.

[0124] The random forest model establishes the correlation between the element content threshold and the mineral phase ratio through non - linear decision tree rules. For example, when the Fe content in the voxel exceeds 12% and the diffraction d - value matches 3.67 Å, the decision tree classifies the mineral phase as hematite; if the Si content is detected to be higher than 20% at the same time, the quartz phase weight coefficient is superimposed. The prediction of reservoir permeability and porosity is based on the regression model of the mineral phase ratio and the core pore structure data. For example, when the clay mineral ratio exceeds 40%, a low - permeability warning is triggered.

[0125] The verification unit receives the quantitative element data transmitted by a third - party detection device (such as a laser - induced breakdown spectrometer) through the RS - 485 interface, and extracts the measured element content values of the target area. The error comparison algorithm calculates the mean square error between the predicted mineral phase ratio of the random forest model and the measured value, and identifies the voxel areas where the deviation exceeds the preset threshold (such as ±5%). The correction process uses the weighted least - squares method to optimize the model parameters and adjusts the feature weights of the decision tree splitting nodes. The corrected error value is fed back to the spatial coordinate registration algorithm through the back - propagation mechanism to dynamically update the registration weight matrix. For example, the feature point cloud density in the high - error area is increased (from 10 points per cubic millimeter to 20 points) to optimize the spatial alignment accuracy of the subsequent three - dimensional model.

[0126] Data fusion and model correction form a closed - loop iterative link. In a single analysis process, after the initial three - dimensional model is generated, the random forest model outputs the mineral phase prediction results; the verification unit introduces external measured data to correct the model error, and the correction parameters improve the data fusion accuracy of the next round through registration weight adjustment. For example, when the predicted ratio of the quartz phase in a certain area is significantly different from the measured value, the registration algorithm strengthens the matching weight between the CT imaging pore characteristics and the fluorescence energy spectrum in this area, and corrects the coordinate offset caused by the uneven surface of the core. The three - dimensional model version management mechanism records the parameter changes of each iteration, supports historical data analysis and model back - tracking verification, and maintains the stability of reservoir parameter prediction.

[0127] Data interaction between steps is achieved through a hierarchical transmission protocol. The voxel data of the three-dimensional model is stored in a tensor structure, including an element matrix, a mineral phase classification matrix, and coordinate transformation parameters, and is subjected to principal component analysis for dimensionality reduction before being input into the random forest model. The error correction parameters are fed back to the registration algorithm through an independent data channel to avoid conflicts with the main data stream. This design maintains the high-throughput data processing ability while improving the mineral phase analysis accuracy of complex oil and gas reservoirs through dynamic parameter optimization.

[0128] Specifically, for the modular X-ray analysis method for oil and gas geology described in the present invention, step 5 includes:

[0129] When detecting samples of allotropes or samples with the same element but different phase structures based on the mineral phase composition correlation relationship output in step 4, switch to the energy filtering mode according to the activation instruction of the energy filtering mode, improve the diffraction signal resolution through a monochromator, and update the training parameters of the random forest model through the gradient descent algorithm based on the error value corrected in step 9;

[0130] According to the reservoir permeability and porosity prediction results generated in step 4, dynamically adjust the target tube voltage and beam current intensity in the time-sharing control instruction to adapt to the detection requirements of different geological lithologies, and synchronously optimize the timing correlation weight to match the real-time detection mode.

[0131] Step 5 of the present invention improves the analysis ability of complex samples through a dynamic mode switching and parameter optimization mechanism, and the specific implementation process is as follows:

[0132] When the mineral phase composition correlation relationship output in step 4 detects samples of allotropes (such as hematite and magnetite) or samples with the same element but different phase structures, the intelligent analysis module generates an energy filtering mode activation instruction. This instruction is transmitted to the monochromator drive unit through the control bus to start the multilayer film reflective monochromator to filter out X-rays in non-target energy bands and improve the resolution of the target diffraction signal (such as the Fe oxide characteristic peak). The filtering range of the monochromator is dynamically adjusted according to the mineral phase type identified in step 4. For example, for hematite, the energy band of 8-9 keV is preferentially retained to suppress background noise. The fluorescence and diffraction signals collected synchronously in the energy filtering mode are correlated by time stamps to generate a high-precision timing correlation data set, which is input into the random forest model for secondary analysis.

[0133] The training parameters of the random forest model are updated based on the error value corrected in Step 9 (such as the hematite ratio error of ±3%). The gradient descent algorithm aims to minimize the loss function and iteratively adjusts the feature weights of the decision tree splitting nodes. For example, when the prediction error in the allotrope region remains high, the algorithm increases the weight coefficient of the diffraction signal d value and weakens the influence of the Fe content in the fluorescence signal to optimize the mineral phase classification rule. The updated model parameters are synchronously sent to the intelligent analysis module after encrypted verification to replace the historical parameter file. The parameter version control mechanism prevents data conflicts and maintains the continuity of model iteration.

[0134] Based on the reservoir permeability and porosity prediction results generated in Step 4, the dynamic strategy module adjusts the target tube voltage and beam current intensity in the time-sharing control instruction through the PID control algorithm. For example, the prediction of a high-porosity reservoir triggers the increase of the Cu target tube voltage from 40 kV to 50 kV to enhance the penetration of low-energy X-rays; for a low-permeability reservoir, the Mo target beam current intensity is increased to 20 mA to improve the signal-to-noise ratio of the fluorescence signal. The time-series correlation weight is dynamically optimized according to the real-time detection mode: the weight is set to 1:1 in the energy filtering mode to balance signal integrity; it is adjusted to 3:1 in the time-sharing mode to preferentially obtain element distribution data.

[0135] The dynamic strategy module and the time-sharing control unit interact in real time through the data bus, and the parameter adjustment instruction is embedded in the FPGA logic unit of the time-sequence controller. For example, when the change in the X-ray source load is caused by the adjustment of the tube voltage, the time-sharing control unit automatically extends the signal acquisition interval (from 100 ms to 150 ms) to match the thermal balance period of the heat dissipation system. The optimization result of the time-series correlation weight is stored in the non-volatile memory for subsequent generation of the standardized data set, forming a historical benchmark for parameter iteration.

[0136] The technical logic closed-loop of Step 5 is reflected in: the mineral phase identification result drives the switching of the acquisition mode, and the optimization of the model parameters improves the classification accuracy; the reservoir parameter prediction is fed back to the hardware control unit to dynamically adjust the target material parameters and acquisition strategy. For example, the prediction of a high clay mineral ratio triggers the high-frequency activation of the energy filtering mode and synchronously optimizes the recognition sensitivity of the model to phyllosilicates; for a low-porosity result, the time-sharing mode is preferentially used to extend the diffraction acquisition period to improve the measurement accuracy of the lattice parameters. This design realizes the dynamic balance of the detection efficiency and analysis accuracy of oil and gas geological samples through the collaborative optimization of software and hardware.

[0137] The specific implementation mode of the present invention combines the oil and gas geological sample analysis scenario to realize the collaborative detection and data fusion of XRF and XRD through the following technical solutions:

[0138] The acquisition module uses a Cu / Mo dual-target X-ray source. The timing controller is driven by the time-sharing control unit to alternately activate the acquisition of fluorescence and diffraction signals in the time-sharing mode. During the fluorescence signal acquisition stage, high-energy X-rays are generated by the Mo target (tube voltage 50 kV, beam current 20 mA). The silicon drift detector captures the full-element energy spectrum (1 - 50 keV) at a cycle of 100 ms. After the acquisition is completed, it switches to the Cu target (tube voltage 30 kV, beam current 15 mA) to excite the diffraction signal, and simultaneously triggers the sample stage stepping motor to rotate at a step size of 0.1°. The two-dimensional array detector records the diffraction angle (5° - 80°) and intensity distribution. The timing controller encapsulates the acquired data into a standardized signal packet with a timestamp through the LVDS bus and transmits it to the data processing module. The environment adaptation module identifies the sample type based on the classification model: when detecting light elements (such as C, O), the vacuum pump is started to reduce the chamber air pressure to below 10- 3 Pa. When detecting bulk samples, it switches to the atmospheric mode and activates the fast loading mechanism. At the same time, external vibrations are suppressed through the shock-absorbing base and dynamic damping (threshold 0.5 g). The heat dissipation system controls the temperature of the X-ray source below 40 °C, and the temperature signal is fed back to the noise filtering algorithm in real time to optimize the signal-to-noise ratio.

[0139] The data processing module calibrates the energy channels of the fluorescence signal, corrects the detector gain drift by calibrating the characteristic peak positions in the calibration database (such as Fe-Kα = 6.40 keV), and generates the element content distribution curve; normalizes the angle-intensity of the diffraction signal, and calculates the lattice spacing based on the Bragg equation (d = λ / (2sinθ), λ = 0.154 nm) to eliminate the influence of surface roughness. The calibrated data is aligned with the core scan CT imaging through the spatial coordinate registration algorithm. Taking the Fe and Si element distributions as reference points, the iterative closest point (ICP) algorithm is used to generate a three-dimensional model, and the voxel unit labels the element content (such as Fe = 15% ± 0.5%) and the mineral phase category (such as quartz, hematite). The random forest model establishes the correlation between the element threshold (Ca > 5%) and the mineral phase ratio (calcite ≥ 15%) based on the non-linear decision tree rule, and corrects the prediction error through the error comparison algorithm in combination with the third-party detection data (such as SEM-EDS). The error value is fed back to the coordinate registration parameters to update the accuracy of the three-dimensional model.

[0140] The dynamic strategy module triggers the energy filtering mode based on the mineral phase identification result. The monochromator filters out non-target energy bands (such as retaining the 8-9 keV energy band to enhance the hematite diffraction peak), synchronously collects fluorescence and diffraction signals, and correlates the timing information through timestamps. The gradient descent algorithm iteratively updates the model parameters based on the regional geological database (such as the quartz / clay phase ratio in shale) and the calibration error (±3%), optimizing the classification weights of mineral phases. The prediction results of reservoir parameters inversely control the time-sharing instructions: for high-porosity reservoirs, increase the Cu target tube voltage to 50 kV to enhance penetration; for low-permeability reservoirs, increase the Mo target beam current to 25 mA to optimize the fluorescence signal-to-noise ratio, and the timing correlation weight is dynamically adjusted to the time-sharing mode of 3:1 or the energy filtering mode of 1:1. This solution is verified through hardware collaborative control and data closed-loop.

[0141] The present invention solves the problems caused by the independent operation of existing XRF and XRD technologies through the following technical solutions: First, for the problem of low detection efficiency, the device works in cooperation with the time-sharing control unit and the dynamic strategy module to achieve time-sharing or synchronous acquisition of XRF and XRD signals. The time-sharing control unit generates timing instructions based on FPGA, alternately driving the multi-target X-ray source to activate the fluorescence and diffraction signal acquisition processes. The timing controller compresses the fluorescence signal acquisition period to the millisecond level and eliminates the crystal orientation deviation through synchronous adjustment of the sample stage rotation and the goniometer angle. The dynamic strategy module adjusts the target tube voltage and beam current intensity according to the real-time feedback, optimizes the timing correlation weight in the time-sharing mode, and shortens the detection cycle while ensuring data integrity.

[0142] Second, to solve the problem of the fragmentation of elemental composition and phase structure data, the intelligent analysis module uses the spatial coordinate registration algorithm to align the standardized data set with the core scan imaging data, generating a three-dimensional model that integrates elemental distribution and mineral phase information. The data fusion unit takes the distribution of characteristic elements (such as Fe and Si) as the reference points and realizes the spatial matching of cross-modal data through the iterative closest point algorithm. The random forest model establishes the correlation between the elemental content threshold and the mineral phase ratio based on the non-linear mapping rule, and combines the third-party detection data introduced by the verification unit to correct the error, forming a multi-dimensional data closed-loop verification mechanism for element-phase-reservoir parameters.

[0143] Finally, to improve the discrimination ability of allotropes or phase isomers, the dynamic strategy module triggers the energy filtering mode according to the correlation relationship of the mineral phase composition. The monochromator filters out non-target energy band X-rays, enhances the resolution of the target diffraction signal, and iteratively updates the training parameters of the random forest model through the gradient descent algorithm. For example, for the discrimination of hematite and magnetite, the algorithm increases the weight coefficient of the diffraction signal d value, weakens the interference of the Fe content in the fluorescence signal, and optimizes the classification decision boundary of mineral phases. This design realizes the accurate identification of complex mineral phases and the dynamic prediction of reservoir characteristics through the collaborative optimization of software and hardware.

Claims

1. A modular X-ray analysis device for oil and gas geology, characterized in that Including: A collection module, configured to trigger a sample to generate X-ray fluorescence signals and X-ray diffraction signals through a preset multi-target X-ray source, synchronously collect the energy and intensity distribution curves of the fluorescence signals and the diffraction angle and intensity patterns of the diffraction signals, and transmit the collected signals to a data processing module; An environment adaptation module, connected to the collection module, configured to identify the sample type based on the energy distribution curve of the fluorescence signal and the diffraction angle pattern of the diffraction signal through a pre-trained classification model to obtain the sample type, and switch the vacuum or atmospheric environment according to the sample type; A data processing module, receiving the fluorescence signals and diffraction signals transmitted by the collection module, calibrating the energy channels of the fluorescence signals, and performing angle and intensity normalization processing on the diffraction signals to generate a time-sequentially correlated standardized data set; An intelligent analysis module, invoking the element and phase mapping rules in the calibration database, inputting the standardized data set into a pre-trained random forest model, outputting the correlation between the element content and the mineral phase composition, and generating a prediction result of reservoir permeability and porosity in combination with core scan data; A dynamic strategy module, generating a three-dimensional visualization report according to the correlation and reservoir parameter prediction results output by the intelligent analysis module, and adjusting the target material selection or signal collection mode of the X-ray source in the collection module through a mobile terminal interaction instruction.

2. The modular X-ray analysis device for oil and gas geology according to claim 1, characterized in that The collection module includes: A time-sharing control unit, connected to the data processing module through a timing controller, configured to alternately activate the collection of fluorescence signals and diffraction signals according to a preset timing instruction; Among them, in the fluorescence signal collection stage, the time-sharing control unit drives the first target to excite the sample to generate an energy distribution curve, and transmits the energy distribution curve to the data processing module through the timing controller; In the diffraction signal collection stage, the time-sharing control unit switches to the second target to generate a diffraction pattern, synchronously triggers the sample stage to rotate to eliminate crystal orientation deviation, and transmits the diffraction pattern to the data processing module through the timing controller.

3. The modular X-ray analysis device for oil and gas geology according to claim 2, wherein, The environment adaptation module includes: A vacuum switching unit, connected to the time-sharing control unit in the collection module, configured to start a vacuum environment according to the collection stage instruction of the time-sharing control unit during light element detection, switch to an atmospheric mode during bulk sample detection, and send an environment status signal to the data processing module to trigger timing calibration; An earthquake resistance unit, connected to the X-ray source and detector in the collection module through a shock-absorbing base, real-time monitoring the external vibration intensity, activating dynamic damping when the vibration intensity exceeds a preset threshold, and controlling the temperature of the X-ray source and detector within a preset threshold through a heat dissipation system. The temperature control signal of the earthquake resistance unit is synchronously transmitted to the data processing module to adjust the noise filtering parameters.

4. The modular X-ray analysis device for oil and gas geology according to claim 1, characterized in that, The intelligent analysis module includes: A data fusion unit receives the standardized data set generated by the data processing module, and aligns the standardized data set with the core scanning imaging data through a spatial coordinate registration algorithm to generate a three-dimensional model including element distribution and mineral phases, wherein the three-dimensional model serves as input data for the pre-trained random forest model; A verification unit is connected to the data fusion unit, receives element quantitative data from a third-party detection device, corrects the mineral phase ratio error output by the random forest model through an error comparison algorithm based on the element distribution data and mineral phase ratio data in the three-dimensional model, and feeds the corrected error value back to the data fusion unit to update the spatial coordinate alignment parameters.

5. The modular X-ray analysis device for oil and gas geology according to claim 1, characterized in that, The dynamic strategy module includes: A mode switching unit, connected to the time-sharing control unit in the acquisition module and the intelligent analysis module, adjusts the signal acquisition mode to time-sharing acquisition or energy filtering synchronous acquisition in response to a mobile terminal instruction, and sends a mode switching instruction to the intelligent analysis module to update the input parameters of the pre-trained random forest model, wherein the input parameters include a time-series correlation weight between the energy distribution curve of the fluorescence signal collected in the time-sharing mode and the diffraction spectrum; The algorithm iteration unit is connected to the verification unit in the intelligent analysis module, and updates the training parameters of the pre-trained random forest model through a gradient descent algorithm according to the mineral phase data in the regional geological database and the corrected error value output by the verification unit, optimizes the nonlinear mapping rule between the mineral phase ratio and the element content, and synchronizes the updated model parameters to the intelligent analysis module.

6. A modular X-ray analysis method for oil and gas geology, applied to the modular X-ray analysis device for oil and gas geology according to any one of claims 1 to 5, characterized in that, include: Step 1, obtaining the preprocessed sample data, generating X-ray fluorescence signals and X-ray diffraction signals by time-sharing or synchronously exciting the sample through a preset multi-target X-ray source, and alternately collecting the energy spectrum distribution and diffraction pattern of the fluorescence signal based on a time-sharing control instruction; Step 2, based on the energy distribution curve of the fluorescence signal and the diffraction angle spectrum of the diffraction signal, the sample type is identified by a pre-trained classification model to obtain the sample type, and the vacuum or atmospheric environment is switched according to the sample type; Step 3, calibrate the energy channel of the collected fluorescence signal, and perform angle and intensity normalization on the diffraction signal to generate a time-series-related standardized data set; Step 4, aligning the standardized data set with the core scanning imaging data through a spatial coordinate registration algorithm to generate a three-dimensional model including element distribution and mineral phases, and inputting the three-dimensional model into a pre-trained random forest model to output the correlation between element content and mineral phase composition and the prediction results of reservoir permeability and porosity; Step 5: Generate a three-dimensional visualization report based on the association relationship, adjust the target material selection or signal acquisition mode of the X-ray source according to the real-time feedback, and synchronously update the input parameters and training parameters of the random forest model.

7. The modular X-ray analysis method for oil and gas geology according to claim 6, characterized in that, The step 1 comprises: In the time-sharing mode, according to the preset timing instructions, the first target is preferentially activated to generate fluorescence signals and the full-element energy spectrum is recorded. After the acquisition is completed, the second target is switched to generate diffraction signals through the timing controller, the angle of the goniometer is synchronously adjusted to the preset diffraction angle range, and the sample stage is triggered to rotate to eliminate the crystal orientation deviation. In the energy filtering mode, based on the mobile terminal interaction instructions, the monochromator is activated to separate the fluorescence signal and the diffraction signal, the intensity distributions of both are synchronously recorded, and the timing information of the fluorescence signal and the diffraction signal is associated through the time stamp.

8. The modular X-ray analysis method for oil and gas geology according to claim 7, characterized in that, Step 2 includes: Based on the fluorescence signal energy distribution curve collected in Step 1, the peak position shift caused by the detector gain drift is eliminated through the energy channel calibration algorithm, and the calibrated energy spectrum is matched with the element characteristic peaks in the calibration database to determine the content range of the target elements in the sample. Based on the diffraction angle pattern of the diffraction signal collected in Step 1, the diffraction signal is processed through the angle and intensity normalization algorithm, the lattice spacing is calculated in combination with the preset Bragg diffraction equation, and it is compared with the diffraction peaks in the mineral phase standard sample database to identify the types and proportions of the mineral phases in the sample.

9. The modular X-ray analysis method for oil and gas geology according to claim 6, wherein Step 4 includes: The standardized data set generated in Step 3 and the core scanning imaging data are aligned through the spatial coordinate registration algorithm to generate a three-dimensional model including element distribution and mineral phases, and the three-dimensional model is input into the pre-trained random forest model, and the correlation between the element content threshold and the mineral phase proportion is established through the non-linear mapping rule. Receive the element quantitative data of the third-party detection device, and based on the element and phase mapping rules in the calibration database, correct the mineral phase proportion error output by the random forest model through the error comparison algorithm, and feedback the corrected error value to the spatial coordinate registration algorithm to update the coordinate registration parameters of the three-dimensional model.

10. The modular X-ray analysis method for oil and gas geology according to claim 6, characterized in that Step 5 includes: When a sample with allotropes or the same elements but different phase structures is detected based on the mineral phase composition correlation relationship output in Step 4, switch to the energy filtering mode according to the activation instruction of the energy filtering mode, improve the diffraction signal resolution through the monochromator, and update the training parameters of the random forest model through the gradient descent algorithm based on the error value corrected in Step 9. According to the reservoir permeability and porosity prediction results generated in Step 4, dynamically adjust the target tube voltage and beam current intensity in the time-sharing control instructions to adapt to the detection requirements of different geological lithologies, and synchronously optimize the timing correlation weight to match the real-time detection mode.

Images