Chlorite growth prediction method based on microbial activity

Abstract

The invention discloses a chlorite growth prediction method based on microbial activity, and relates to the technical field of geological environment monitoring and prediction.The method comprises the steps that S1, dynamic quantitative monitoring of microbial activity and synchronous acquisition of environmental parameters are conducted based on a geological environment monitoring system, and a comprehensive data set is generated; s2, establishing a kinetic model of influence of microorganisms on chlorite nucleation growth in combination with the comprehensive data set and molecular dynamics; constructing a standardized training data set; s3, constructing a microbial activity space-time prediction model based on the standardized training data set; and S4, carrying out dynamic updating and error correction on the microbial activity space-time prediction model to obtain a final space-time prediction model. When the prediction result shows that the growth rate or morphology of chlorite is abnormally changed, the early warning mechanism of the geological environment monitoring system is triggered, and the monitoring frequency and the sampling strategy are automatically adjusted.

Description

Technical Field

[0001] This application relates to the field of geological environment monitoring and prediction technology, and in particular to a method for predicting chlorite growth based on microbial activity. Background Technology

[0002] Chlorite is a common iron-magnesium-aluminate layered silicate mineral found in sedimentary environments, and its formation is influenced by a variety of physical, chemical, and biological factors. For a long time, research on chlorite growth has mainly focused on physicochemical conditions such as temperature, pH, and solution composition, and our understanding of its genesis and distribution has largely been based on laboratory synthesis and field surveys. However, recent studies have revealed that microorganisms play an indispensable role in mineral formation. Some microorganisms can significantly influence the crystal structure, morphology, and formation rate of chlorite by secreting metabolites, regulating the local chemical environment, or mediating crystal nucleation and growth.

[0003] Existing monitoring methods mostly employ single technologies, which, while providing some information on bioactivity, suffer from limitations in applicability, spatial resolution, or dynamic response capabilities, making it difficult to comprehensively and continuously reflect the actual role of microorganisms in the chlorite formation process. Furthermore, most current chlorite growth models have not fully integrated microbial data, and prediction methods still rely primarily on static parameter fitting, lacking the ability to respond to dynamic changes in microorganisms within the actual geological environment.

[0004] In conclusion, how to achieve dynamic monitoring of microbial activity in real geological environments and effectively incorporate it into chlorite growth prediction models to improve the accuracy and adaptability of predictions has become an urgent technical problem to be solved. Summary of the Invention

[0005] The purpose of this application is to provide a method for predicting chlorite growth based on microbial activity. This method can accurately predict and dynamically warn of chlorite nucleation, growth kinetics, and crystal morphology by quantifying microbial activity in real time and integrating multi-source data such as Raman spectroscopy, electrochemical impedance spectroscopy, fluorescent probes, and environmental parameters, combined with molecular dynamics simulation and an adaptive deep learning architecture.

[0006] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a method for predicting chlorite growth based on microbial activity, the method comprising: S1. Based on the geological environment monitoring system, a comprehensive dataset is generated by dynamic quantitative monitoring of microbial activity and synchronous collection of environmental parameters, including: microbial activity indicators, environmental parameters and chlorite growth data; S2. A kinetic model of the influence of microorganisms on the nucleation and growth of chlorite is established by combining a comprehensive dataset and molecular dynamics; and a standardized training dataset is constructed based on the input and output data of the kinetic model. S3. Construct a spatiotemporal prediction model for microbial activity based on a standardized training dataset; S4. Dynamically update and correct the spatiotemporal prediction model of microbial activity to obtain the final spatiotemporal prediction model, and output the corrected chlorite growth prediction results based on the final spatiotemporal prediction model.

[0007] Optionally, the step S1, which involves dynamic quantitative monitoring of microbial activity and synchronous acquisition of environmental parameters based on a geological environment monitoring system to generate a comprehensive dataset, specifically includes: The geological environment monitoring system includes: a Raman spectroscopy detection module, an electrochemical impedance measurement module, a fluorescent probe labeling module, an environmental parameter monitoring module, and a data fusion analysis module; Raman spectroscopy detection module, used to identify characteristic peaks of microbial metabolites; An electrochemical impedance spectroscopy module is used to detect changes in the resistance of microbial membranes; The fluorescent probe labeling module is used to label the extracellular polymer secretion state; The environmental parameter monitoring module is used to acquire environmental factor data; The data fusion and analysis module is used to establish a multi-source data fusion framework that integrates real-time monitoring data, historical geological data, and laboratory analysis results. Through the multi-source data fusion framework, a comprehensive dataset containing microbial activity indicators, environmental parameters, and chlorite growth data is constructed.

[0008] Optionally, the multi-source data fusion framework specifically includes: The multi-source data fusion framework adopts a layered fusion architecture, including sensor-level fusion, feature-level fusion, and decision-level fusion; Among them, sensor-level fusion performs time registration, spatial registration and data association processing on raw data from sensors of the same type; Feature-level fusion is used to extract correlated features and complementary information between different data sources and to construct a multi-dimensional feature vector space. Decision-level fusion is used to combine the fusion results of sensor-level fusion and feature-level fusion to generate microbial activity assessment results.

[0009] Optionally, S2 specifically includes: The complexation reaction mechanism between microbial extracellular polymers and chlorite precursor ions was predicted by molecular dynamics simulation and verified by chlorite synthesis experiments under controlled laboratory conditions, thus establishing a kinetic model of the effect of microbial activity on chlorite nucleation and growth. A kinetic model was used to quantify the regulatory effects of microbial metabolites on the morphology and growth direction of chlorite crystals. A standardized training dataset was constructed based on a comprehensive dataset, which includes: temporal characteristics of microbial activity, spatial distribution patterns, and changes in environmental parameters; A standardized training dataset was used to capture the impact of environmental parameter changes on the succession of microbial community structure, as well as the feedback regulation mechanism of microbial activity fluctuations on chlorite growth patterns.

[0010] Optionally, constructing the standardized training dataset specifically includes: The spatiotemporal gridding method is used to organize the comprehensive dataset, which is divided into multiple time windows according to the time dimension. Each time window contains 6 to 24 hours of continuous monitoring data. A three-dimensional grid is established based on the location of the monitoring points, with a grid resolution of 1–10 m. The three-dimensional grid is adjusted according to the geological complexity of the monitoring area. Each spatiotemporal grid cell contains microbial activity indicators, environmental parameters, and chlorite growth data.

[0011] Optionally, the spatial distribution pattern specifically includes: Spatial interpolation, spatial clustering, and spatial autocorrelation analysis; Among them: spatial interpolation is based on microbial activity data from discrete sampling points to generate a continuous spatial distribution map; The spatial clustering is used to identify high-value areas, low-value areas and transition areas of microbial activity, revealing the spatial distribution pattern of microbial activity. The spatial autocorrelation analysis is used to quantitatively assess the spatial correlation and spatial heterogeneity of microbial activity.

[0012] Optionally, the establishment of a kinetic model of the effect of microbial activity on the nucleation and growth process of chlorite specifically includes: a nucleation kinetic model, a growth kinetic model, and a morphology control model; The nucleation kinetics model includes: based on classical and non-classical nucleation theories, considering the influence of microbial metabolites on the nucleation energy barrier, nucleation rate and critical nucleus size, and calculating the nucleation rate equation that includes biological factors; Growth kinetics model: Using spiral growth mechanism, two-dimensional nucleation mechanism and multi-step growth mechanism, the growth rate of microbial extracellular polymers on each crystal face of chlorite crystal was calculated; Morphology control model: Based on Wulff configuration theory of crystal morphology, considering the selective adsorption of organic molecules and surface energy modification effects, predict the morphology evolution of chlorite crystals under microbial activity conditions.

[0013] Optionally, the construction of the spatiotemporal prediction model for microbial activity based on the standardized training dataset specifically includes: Design an adaptive deep learning architecture that integrates convolutional neural networks, long short-term memory networks, and attention mechanisms; A spatiotemporal prediction model for microbial activity was constructed by combining a standardized training dataset with an adaptive deep learning architecture.

[0014] Optionally, the adaptive deep learning architecture specifically includes: A multi-layered structure consisting of an input layer, a feature extraction layer, a feature fusion layer, a temporal modeling layer, and an output layer; Among them: the input layer is used to standardize and encode the comprehensive dataset, converting different types of data into vector forms that the network can process; The feature extraction layer is used to process temporal and spatial data by combining one-dimensional and two-dimensional convolutional neural networks, respectively, to extract local and global features. The feature fusion layer is used to adaptively fuse different types of local and global features using attention and gating mechanisms to generate fused features that include information on microbial activity and environmental parameters. The temporal modeling layer is used to model the temporal dependencies between microbial activity and environmental parameters based on the fused features output by the feature fusion layer. The output layer is used to output chlorite growth prediction results by employing fully connected layers and activation functions, combined with temporal dependencies.

[0015] Optionally, the step of dynamically updating and error-correcting the spatiotemporal prediction model of microbial activity to obtain the final spatiotemporal prediction model specifically includes: The final spatiotemporal prediction model for microbial activity is generated through dynamic updating and error correction using Kalman filtering and particle filtering algorithms, including: When the prediction results show abnormal changes in the growth rate or morphology of chlorite, the early warning mechanism of the geological environment monitoring system is triggered, and the monitoring frequency and sampling strategy are adjusted. If the prediction results are within the preset growth pattern range, the current monitoring strategy will be maintained and chlorite growth prediction will continue. If the prediction results exceed the preset range, the monitoring density will be increased and the prediction model parameters will be recalibrated until the prediction accuracy meets the requirements.

[0016] According to the specific embodiments provided in this application, this application has the following technical effects: This application provides a method for predicting chlorite growth based on microbial activity. The integrated monitoring system lays a data foundation for quantitative research on the coupling relationship between microbial activity and the environment, and significantly improves the spatiotemporal resolution and reliability of the prediction model.

[0017] The structured and comprehensive dataset formed by the multi-source data fusion framework provides data support for the construction of chlorite growth prediction models, breaking through the problems of single data dimension, poor timeliness, and lack of microbial participation variables in traditional chlorite research.

[0018] By employing molecular dynamics simulations and experimental verification, we investigated the complexation reaction mechanism between microbial extracellular polymers and chlorite precursor ions, and constructed a kinetic model of chlorite nucleation and growth under microbial regulation, which significantly advanced the quantitative understanding of the mineral-microbe interaction mechanism.

[0019] Based on the fused multi-source dataset, the influence of environmental changes on microbial community succession and its feedback regulation mechanism on chlorite growth were captured, realizing the whole-process modeling of the complex geochemical process of "environmental disturbance - microbial response - mineral precipitation".

[0020] A chlorite growth prediction model that can adaptively extract the spatiotemporal characteristics of microbial activity has successfully solved the modeling challenges of geological environment monitoring data with high noise, multiple scales, and strong nonlinearity.

[0021] In summary, the chlorite growth prediction method of this application systematically constructs an intelligent geological monitoring platform that can dynamically sense microbial activity, respond to environmental disturbances, and output prediction results in real time by integrating multiple advanced monitoring technologies, data fusion strategies, and intelligent modeling algorithms.

[0022] The steps form an organic closed loop, which improves the real-time performance, accuracy and adaptability of geological system prediction, and makes up for the shortcomings of traditional geological prediction methods, such as insufficient understanding of microbial regulation mechanisms and weak modeling capabilities. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of a chlorite growth prediction method based on microbial activity in one embodiment of this application; Figure 2 This is a schematic diagram of the sampling location in one embodiment of this application; Figure 3 This is a schematic diagram showing the correlation test results between sedimentary environment and clay minerals. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] like Figure 1 As shown in the embodiments of this application, a method for predicting chlorite growth based on microbial activity is provided, the method comprising: S1. Based on the geological environment monitoring system, a comprehensive dataset is generated by dynamic quantitative monitoring of microbial activity and synchronous collection of environmental parameters, including: microbial activity indicators, environmental parameters and chlorite growth data; S2. A kinetic model of the influence of microorganisms on the nucleation and growth of chlorite is established by combining a comprehensive dataset and molecular dynamics; and a standardized training dataset is constructed based on the input and output data of the kinetic model. S3. Construct a spatiotemporal prediction model for microbial activity based on a standardized training dataset; S4. Dynamically update and correct the spatiotemporal prediction model of microbial activity to obtain the final spatiotemporal prediction model, and output the corrected chlorite growth prediction results based on the final spatiotemporal prediction model.

[0028] Optionally, the step S1, which involves dynamic quantitative monitoring of microbial activity and synchronous acquisition of environmental parameters based on a geological environment monitoring system to generate a comprehensive dataset, specifically includes: The geological environment monitoring system includes: a Raman spectroscopy detection module, an electrochemical impedance measurement module, a fluorescent probe labeling module, an environmental parameter monitoring module, and a data fusion analysis module; Raman spectroscopy detection module, used to identify characteristic peaks of microbial metabolites; An electrochemical impedance spectroscopy module is used to detect changes in the resistance of microbial membranes; The fluorescent probe labeling module is used to label the extracellular polymer secretion state; The environmental parameter monitoring module is used to acquire environmental factor data; The data fusion and analysis module is used to establish a multi-source data fusion framework that integrates real-time monitoring data, historical geological data, and laboratory analysis results. Through the multi-source data fusion framework, a comprehensive dataset containing microbial activity indicators, environmental parameters, and chlorite growth data is constructed.

[0029] Specifically, through the synergistic integration of Raman spectroscopy detection module, electrochemical impedance measurement module, fluorescent probe labeling module, and environmental parameter monitoring module, multi-dimensional and dynamic monitoring of microbial activity status and geological environmental parameters has been achieved.

[0030] This combination configuration is mainly reflected in two aspects: first, it improves the accuracy of acquiring microbial activity signals; second, it realizes the synchronous recording and corresponding analysis of environmental conditions and microbial states.

[0031] Specifically, the Raman spectroscopy detection module can identify the Raman characteristic peaks of specific microbial metabolites in real time, helping to characterize changes in their metabolic activity; the electrochemical impedance spectroscopy module measures the resistance changes of biofilms by constructing a three-electrode system, reflecting the changing trends of microbial community structure and attachment activities; the fluorescent probe module can label and track the secretion behavior of extracellular polymers, thereby revealing the spatial distribution characteristics of their reaction potential with mineral precursors; and the environmental parameter monitoring module acquires real-time data on environmental factors such as temperature, pH, Eh, and conductivity, providing background support for biogeochemical reactions. This integrated monitoring system lays a data foundation for the quantitative study of the coupling relationship between microbial activity and the environment, significantly improving the spatiotemporal resolution and reliability of prediction models.

[0032] Furthermore, the Raman spectroscopy detection module is specifically configured as follows: A multi-wavelength excitation laser system, including a 532nm green laser, a 785nm near-infrared laser, and a 1064nm long-wavelength laser, is used to selectively detect different types of microbial metabolites through wavelength switching technology. Equipped with a highly sensitive charge-coupled device detector, it can accurately detect key metabolites secreted by microorganisms in the concentration range of 10⁻⁴ to 10⁻⁶ molar concentration. By utilizing a surface-enhanced Raman scattering substrate, the intensity of the detection signal can be increased by 1 to 2 orders of magnitude; It is also equipped with a temperature compensation system and a humidity control device, with an operating temperature range of -10℃ to 60℃ and a relative humidity range of 20% to 90%, in order to eliminate the influence of environmental factors on the accuracy of spectral measurement and achieve accurate reflection of the activity state of microorganisms.

[0033] In summary, this Raman spectroscopy detection module achieves selective detection of different microbial metabolites through a multi-wavelength excitation system. Combined with a high-sensitivity detector, it can accurately identify low-abundance metabolites within a concentration range of 10⁻⁴ to 10⁻⁸ molarities, significantly improving detection sensitivity and accuracy. Surface-enhanced Raman scattering technology further amplifies the signal intensity by 1 to 2 orders of magnitude, effectively improving the detectability of weak signals. The equipped temperature and humidity control system can operate stably in environments ranging from -10℃ to 60℃ and 20% to 90% humidity, suppressing the impact of external disturbances on spectral accuracy. Overall, it achieves efficient and stable monitoring of microbial activity under complex environments.

[0034] Furthermore, the electrochemical impedance measurement module adopts a three-electrode system configuration, with a platinum or gold electrode as the working electrode, a saturated calomel electrode or a silver / silver chloride electrode as the reference electrode, and a platinum wire electrode as the counter electrode. The measurement frequency range is set to 0.1 Hz to 10 kHz, the applied AC voltage amplitude is controlled within the range of 10 to 50 mV, and the measurement interval is set to perform a complete impedance spectrum scan every 2 to 6 hours. The solution resistance, charge transfer resistance, and double-layer capacitance are extracted by the equivalent circuit fitting method.

[0035] In summary, this electrochemical impedance spectroscopy module, through a three-electrode system configuration, ensures the stability of the measurement process and the high accuracy of the data. Using platinum or gold as the working electrode material provides excellent conductivity and biocompatibility, enhancing the electrochemical response to microbial membranes. The use of a standard electrode system for the reference and counter electrodes ensures the stability of potential control. Setting a frequency range of 0.1 Hz to 10 kHz and an AC voltage amplitude of 10–50 mV helps cover the impedance characteristics of microbial membranes from diffusion to the interface. Scanning the impedance spectrum every 2–6 hours facilitates dynamic tracking of changes in microbial activity. Extracting key parameters through equivalent circuit fitting accurately reflects the structure and metabolic state of the microbial membrane, providing a reliable basis for the quantitative assessment of microbial activity.

[0036] Furthermore, the fluorescent probe labeling module uses fluorescein isothiocyanate with an excitation wavelength of 495 nm and an emission wavelength of 519 nm to label extracellular polysaccharides, rhodamine B with an excitation wavelength of 550 nm and an emission wavelength of 570 nm to label extracellular proteins, and anthocyanin dye with an excitation wavelength of 650 nm and an emission wavelength of 670 nm to label extracellular DNA. Fluorescence detection is performed using a fluorescence microscope with a spatial resolution of 1 μm and a temporal resolution of acquiring fluorescence images every 1–2 hours. The secretion intensity and spatial distribution patterns of various extracellular polymers are quantitatively analyzed using image processing algorithms. A standard curve between fluorescence intensity and extracellular polymer concentration is established, and a fluorescence intensity change threshold of 30% is set to monitor significant changes in microbial secretory activity.

[0037] In summary, this fluorescent probe labeling module employs a multi-channel fluorescent labeling strategy, using specific wavelength fluorescent dyes for extracellular polysaccharides, proteins, and DNA to achieve selective identification and high-resolution monitoring of different extracellular polymeric components. The combination of fluorescein isothiocyanate, rhodamine B, and anthocyanin dyes can cover the major components of microbial extracellular polymers, significantly improving the characterization accuracy of secretory activity. Using fluorescence microscopy with 1-micrometer spatial resolution and 1-2 hour time interval sampling, the distribution and intensity changes of extracellular polymers can be dynamically acquired. Image processing algorithms are used to analyze fluorescence intensity and construct a standard curve, quantifying its relationship with actual concentrations. A 30% fluorescence intensity change threshold is set to effectively identify significant fluctuations in microbial secretory activity, thus providing crucial biological evidence for predicting its impact on chlorite nucleation and growth.

[0038] Furthermore, the environmental parameter monitoring module includes a temperature sensor array, a pH detection unit, a dissolved oxygen concentration measuring device, an ion concentration detection unit, and hydrogeological parameter monitoring equipment. Specifically: the temperature sensor array employs a high-precision thermistor sensor with a measurement accuracy of ±0.1℃, a measurement range of -20℃ to 80℃, and a response time of less than 10 seconds; the pH detection unit uses a composite pH electrode composed of a glass electrode and a reference electrode, equipped with an automatic calibration function and a temperature compensation mechanism to ensure a measurement accuracy better than ±0.1℃ within the pH range of 2–12. The system comprises 1 pH unit; the ion concentration detection system integrates selective electrodes for potassium, sodium, calcium, magnesium, and silicate ions, with a measurement range of 10⁻⁵ to 10⁻¹ mol / L, enabling real-time monitoring of key ion concentration changes related to chlorite formation; the hydrogeological parameter monitoring equipment includes a groundwater level monitor with an accuracy of ±1 cm, an osmotic pressure measuring device with a measurement range of 0–1000 kPa, and a soil moisture sensor with a measurement range of 0%–100% volumetric water content, providing complete environmental parameter background data for microbial activity and chlorite growth.

[0039] In summary, this environmental parameter monitoring module integrates high-precision, multi-functional sensor equipment to achieve comprehensive and real-time monitoring of key physicochemical parameters in the chlorite growth environment. The temperature sensor array employs a thermistor with a fast response and a measurement accuracy of ±0.1℃, ensuring precise capture of temperature changes and covering a wide range from -20℃ to 80℃, adapting to complex geological conditions. The pH detection unit is equipped with a composite glass electrode and automatic calibration and temperature compensation functions, ensuring accurate measurements within a pH range of 2 to 12, meeting the dynamic acid-base monitoring needs during microbial metabolism and mineral precipitation processes. The ion concentration detection system uses selective electrodes for potassium, sodium, calcium, magnesium, and silicate ions, effectively monitoring changes in solution chemical composition closely related to chlorite nucleation and crystal growth. The hydrogeological parameter equipment covers three major indicators: groundwater level, osmotic pressure, and soil moisture content, accurately reflecting the hydrological environment at the centimeter, kilopascal, and percentage volumetric water content levels, respectively, forming a closed loop of environmental background data on microbial activity and mineral formation. The overall module enables the system to achieve high-resolution synchronous monitoring of environmental factors and biogeological processes, providing scientific and comprehensive environmental support for chlorite growth prediction.

[0040] Optionally, the multi-source data fusion framework specifically includes: The multi-source data fusion framework adopts a layered fusion architecture, including sensor-level fusion, feature-level fusion, and decision-level fusion; Among them, sensor-level fusion performs time registration, spatial registration and data association processing on raw data from sensors of the same type; Feature-level fusion is used to extract correlated features and complementary information between different data sources and to construct a multi-dimensional feature vector space. Decision-level fusion is used to combine the fusion results of sensor-level fusion and feature-level fusion to generate microbial activity assessment results.

[0041] Specifically, by establishing a multi-source data fusion framework that integrates real-time monitoring data, historical geological data, and laboratory analysis results, a structured comprehensive dataset was formed, providing data support for the construction of a chlorite growth prediction model. This breaks through the problems of single data dimension, poor timeliness, and lack of microbial participation variables in traditional chlorite research.

[0042] like Figure 2 As shown, samples will be collected from the Badain Jaran Desert in typical river, delta, and coastal environments. like Figure 3 As shown, based on the analysis of samples to be collected from the Badain Jaran Desert, the correlation between sedimentary environment and clay minerals is as follows: coarse-grained, high-energy, oxidizing environments are conducive to illite preservation, while low-energy, fine-grained, reducing environments promote chlorite formation. Illite has a stable crystal structure and slightly higher hydraulic equivalence, allowing it to co-deposit with coarse-grained clay. A reducing environment is conducive to Fe²⁺ transformation, providing the necessary conditions for the formation of chlorite.

[0043] This fusion framework effectively integrates dynamically collected microbial activity indicators, on-site environmental parameters, historical mineral formation data, and chlorite formation parameters under controlled experimental conditions, constructing a multi-scale, cross-domain data platform that enables continuous modeling from macroscopic environmental background to microscopic reaction mechanisms.

[0044] Specifically, it can perform attribution analysis on the relationship between microbial metabolic activities and mineral phase transitions under different geological scenarios, reveal their co-evolutionary paths, and support preprocessing operations such as feature selection, variable normalization, and data missing repair, providing highly cleaned training data for subsequent model training and parameter optimization.

[0045] Furthermore, the data fusion analysis module adopts a multi-layered data fusion architecture, including a data preprocessing layer, a feature extraction layer, an information fusion layer, and a decision output layer. Specifically: the data preprocessing layer performs time synchronization, noise filtering, missing value imputation, and outlier detection on the raw data from each detection module to ensure data quality and consistency; the feature extraction layer extracts key feature parameters reflecting microbial activity and environmental changes from multi-dimensional detection data; the information fusion layer weightedly fuses information from different detection modules to construct a comprehensive fusion index reflecting the microbial activity status; and the decision output layer generates a microbial activity assessment report based on the fusion results, including activity level classification, trend prediction, and anomaly identification, providing reliable microbial activity input parameters for chlorite growth prediction.

[0046] In summary, this data fusion and analysis module achieves efficient integration and intelligent analysis of multi-source heterogeneous data through a multi-layered architecture, significantly improving the accuracy and reliability of microbial activity monitoring. First, the data preprocessing layer ensures temporal alignment of data from each module through time synchronization. Combined with noise filtering, missing value completion, and outlier detection, it effectively improves the completeness and quality of input data, avoiding misjudgments caused by data anomalies. Second, the feature extraction layer extracts key indicators closely related to microbial activity and environmental changes from multi-dimensional monitoring data, such as electrochemical impedance spectroscopy, fluorescence intensity changes, and environmental factor fluctuation characteristics, achieving information refinement and dimensionality reduction. The information fusion layer integrates features from multiple modules based on a weighted algorithm, forming a fusion index reflecting the overall state of microbial activity, overcoming the limitations of single detection methods and enhancing sensitivity to complex environmental changes. Finally, the decision output layer classifies activity levels, predicts trends, and identifies abnormal states based on the fusion results, generating a structured evaluation report that provides accurate and dynamic input data support for subsequent chlorite growth prediction.

[0047] Optionally, S2 specifically includes: The complexation reaction mechanism between microbial extracellular polymers and chlorite precursor ions was predicted by molecular dynamics simulation and verified by chlorite synthesis experiments under controlled laboratory conditions, thus establishing a kinetic model of the effect of microbial activity on chlorite nucleation and growth. A kinetic model was used to quantify the regulatory effects of microbial metabolites on the morphology and growth direction of chlorite crystals. A standardized training dataset was constructed based on a comprehensive dataset, which includes: temporal characteristics of microbial activity, spatial distribution patterns, and changes in environmental parameters; A standardized training dataset was used to capture the impact of environmental parameter changes on the succession of microbial community structure, as well as the feedback regulation mechanism of microbial activity fluctuations on chlorite growth patterns.

[0048] Specifically, through molecular dynamics simulations and experimental verification, the complexation reaction mechanism between microbial extracellular polymers and chlorite precursor ions was studied, and a kinetic model of chlorite nucleation and growth under microbial regulation was constructed, significantly advancing the quantitative understanding of the mineral-microorganism interaction mechanism. Its advancements are mainly reflected in two aspects: First, molecular dynamics simulations can reveal, at the atomic scale, the binding modes and energy barrier changes between specific extracellular polymeric functional groups (such as carboxyl and hydroxyl groups) and precursor ions such as Fe²⁺ and Mg²⁺, clarifying their nucleation site selection and preferential crystallization direction; second, controlled experiments simulate the influence of microbial activity regulation conditions (such as carbon source concentration and redox state) on the morphology (flaky, fibrous, etc.) and growth rate of chlorite, achieving mutual verification between theoretical simulations and actual processes.

[0049] Based on a fused multi-source dataset, a standardized training dataset was constructed, incorporating temporal characteristics of microbial activity, spatial distribution patterns, and changes in environmental parameters. This dataset captures the impact of environmental changes on microbial community succession and its feedback regulation mechanism on chlorite growth. Its advancement lies in achieving full-process modeling of the complex geochemical process of "environmental disturbance—microbial response—mineral precipitation." By introducing spatial distribution characteristics and temporal evolution paths, the dynamic succession patterns of microbial activity under different environmental factors (such as drastic fluctuations in pH and temperature) can be identified. Furthermore, by combining information on changes in chlorite deposition rates, the temporal lag characteristics and nonlinear feedback of the existing biogeochemical coupling process can be deduced.

[0050] Furthermore, the molecular dynamics simulation employs the GROMACS or LAMMPS software platform, selects CHARMM or AMBER force field parameters, the simulation system contains 1000–5000 atoms, the simulation time is 5–50 nanoseconds, the time step is set to 1–2 femtoseconds, the temperature is controlled within the range of 25–50℃, and isothermal or isothermal-isobaric simulations are performed using NVT or NPT ensembles. The coordination structure of microbial extracellular polymers and metal ions is analyzed through radial distribution function, the average interaction energy and diffusion coefficient are calculated, the intermolecular interaction potential energy function is established, and the dynamic process and mechanism of complexation reaction are revealed through molecular dynamic trajectory analysis.

[0051] In summary, this molecular dynamics simulation, utilizing the GROMACS or LAMMPS platform and combined with CHARMM or AMBER force field parameters, accurately simulates the molecular interactions between extracellular polymers and metal ions in microorganisms. The simulation system scale covers 1000 to 5000 atoms, with a time span of 5 to 50 nanoseconds, ensuring sufficient spatial and temporal resolution. The time step is controlled within 1–2 femtoseconds to ensure numerical stability. Temperatures are controlled between 25 and 50 °C, employing NVT or NPT systems to reproduce the physical environment under isothermal or isothermal-isobaric conditions. Radial distribution function analysis accurately describes the coordination structure characteristics of extracellular polymers and metal ions, calculating the average interaction energy and diffusion coefficient, revealing the intermolecular binding strength and dynamic behavior. Based on the intermolecular interaction potential energy function constructed from the simulated trajectories, the dynamic evolution process and mechanism of the complexation reaction are analyzed in depth, providing microscopic theoretical support for microbial regulation of chlorite nucleation and growth.

[0052] Furthermore, the process for identifying characteristic peaks of microbial metabolites includes the following steps: First, the spectral data is automatically baseline corrected using a combination of polynomial fitting and asymmetric least squares method to effectively remove the influence of fluorescence background and baseline drift; then, the peak boundaries of the corrected spectral data are identified based on the second derivative method and continuous wavelet transform, and the peak shape is fitted using a Gaussian function; finally, the integrated area under the peak is calculated using a numerical integration method to achieve a quantitative reflection of the content of microbial metabolites; next, based on the established database containing standard Raman spectra of 200–300 common microbial metabolites, the calculated peak area and its characteristics are processed using a spectral matching algorithm to complete the rapid identification of metabolites.

[0053] In summary, high-precision spectral baseline correction is achieved by introducing polynomial fitting and asymmetric least squares method, peak position identification is performed by combining second derivative method and continuous wavelet transform, and the content of metabolites is quantified by using Gaussian fitting and numerical integration. Finally, efficient and accurate rapid identification of metabolites is achieved by spectral matching with standard database, thereby significantly improving the automation, quantitative accuracy and identification efficiency of microbial metabolite analysis process.

[0054] Furthermore, in the process of detecting changes in microbial membrane resistance, firstly, impedance spectroscopy is measured using a frequency scanning method with a sinusoidal excitation signal to obtain the impedance response characteristics of the microbial membrane at different frequencies. Subsequently, based on the measured impedance spectroscopy data, an equivalent circuit model including solution resistance, charge transfer resistance, double-layer capacitance, and Warburg diffusion impedance is used to fit the impedance data to accurately describe the electrochemical behavior of the microbial membrane-electrode interface. Finally, the fitting results are processed using a nonlinear least squares fitting algorithm to extract membrane resistance, membrane capacitance, ion transport impedance, and reaction kinetic parameters.

[0055] The equivalent circuit model processing includes the following steps: First, the electrochemical behavior of the microbial membrane-electrode interface is abstracted into a circuit network composed of solution resistance, charge transfer resistance, double-layer capacitance, and Warburg diffusion impedance. Second, by fitting impedance spectrum data, parameters are estimated according to the physical meaning of each circuit element: solution resistance reflects the conductivity of the electrolyte and is represented by the real part of the high-frequency impedance; charge transfer resistance corresponds to the impedance of the electron transfer reaction occurring on the electrode surface and mainly affects the mid-frequency impedance; double-layer capacitance simulates the capacitance characteristics of the charge double layer at the electrode interface and affects the change of the imaginary part of the impedance; Warburg diffusion impedance describes the diffusion process of ions near the electrode surface and is represented by the frequency dependence of the low-frequency impedance. Finally, nonlinear least squares fitting is used to fit the real and imaginary parts of the impedance spectrum data with the equivalent circuit model based on the above physical correspondences, determining the parameter values ​​of each element one by one, and realizing the quantitative analysis of the resistance change of the microbial membrane.

[0056] The reaction kinetic parameters include: complexation reaction rate constant, which reflects the rate of complex formation; coordination number and complexation ligand type, which affect the stability of the complex; diffusion coefficient, which affects the frequency of encounter between ions and polymer molecules; and surface adsorption constant, which describes the regulation of the crystal growth surface by the extracellular polymer.

[0057] In summary, by measuring the impedance spectrum of microbial membranes using frequency scanning, fitting the impedance data with an equivalent circuit model containing multiple electrochemical components, and extracting key electrochemical parameters using a nonlinear least squares algorithm, we have achieved accurate modeling of the microbial membrane-electrode interface behavior and efficient extraction of parameters such as membrane resistance. This improves the accuracy, resolution, and quantification capabilities of detecting the electrochemical properties of microbial membranes.

[0058] Furthermore, the standardized training dataset constructed from the multi-source data fusion results is organized using a spatiotemporal gridding method. The dataset is divided into different time windows according to the time dimension, with each window containing 6–24 hours of continuous monitoring data. Spatially, a three-dimensional grid is established based on the monitoring point locations, with a grid resolution of 1–10 m, adjusted according to the geological complexity of the monitoring area. Each spatiotemporal grid unit contains a complete record of microbial activity indicators, environmental parameters, and chlorite growth data. The data standardization process employs Z-score standardization or min-max standardization methods to ensure the comparability of data with different dimensions.

[0059] In summary, the constructed multi-source data fusion standardized training dataset achieves systematic organization through spatiotemporal gridding, significantly improving data structuring and analysis efficiency. In the temporal dimension, the data is divided into continuous time windows of 6–24 hours to ensure the capture of the dynamic evolution of microbial activity and environmental parameters. Spatially, a three-dimensional grid is established based on the monitoring point locations, with a resolution between 1 and 10 meters, which can be flexibly adjusted according to geological complexity to refine the spatial heterogeneity. Each spatiotemporal unit contains comprehensive data on microbial activity indicators, environmental factors, and chlorite growth, forming a multidimensional coupled feature set. By applying Z-score or min-max standardization methods, the dimensional differences between different variables are eliminated, enhancing data consistency and comparability.

[0060] Furthermore, the multi-source data fusion framework adopts a hierarchical fusion architecture, including three levels: sensor-level fusion, feature-level fusion, and decision-level fusion. Sensor-level fusion performs temporal registration, spatial registration, and data association processing on raw data from similar sensors to improve spatiotemporal consistency. Feature-level fusion extracts correlation features and complementary information between different data sources to construct a multi-dimensional feature vector space. Decision-level fusion integrates the fusion results from different levels to generate a unified microbial activity assessment result. Specifically, the integrated decision receives the multi-dimensional features extracted by feature-level fusion, weights each indicator based on weight allocation, calculates a comprehensive activity score using fuzzy comprehensive evaluation or machine learning algorithms, and then combines this with a threshold to determine the microbial activity status, ultimately outputting a unified assessment result and achieving unified decision-making based on multi-source information.

[0061] In summary, this multi-source data fusion framework effectively improves the spatiotemporal consistency and information representation capabilities of data through three-layer fusion: sensor-level, feature-level, and decision-level. Sensor-level fusion achieves temporal and spatial registration of similar data, ensuring accurate alignment of original data; feature-level fusion mines the correlation features and complementary information between different data sources, constructing a multi-dimensional feature space and enhancing data expressiveness; and decision-level fusion integrates the results from all levels to achieve unified and accurate assessment of microbial activity, significantly improving the monitoring accuracy and decision reliability of the system.

[0062] Furthermore, in the chlorite synthesis experiment, firstly, a reaction system was prepared using high-purity FeCl3·6H2O, MgCl2·6H2O, Al2(SO4)3·18H2O, and Na2SiO3·9H2O through precise stoichiometric control to simulate the ionic composition of the natural geological environment, and the pH value of the reaction system was controlled within the range of 5.5–8.5. Subsequently, based on a reaction vessel with temperature control and stirring functions, temperature regulation, atmospheric pressure reaction, pH adjustment, and reaction time control were implemented to precisely manage the reaction conditions. Next, the product was characterized by X-ray diffraction, scanning electron microscopy, transmission electron microscopy, and energy-dispersive X-ray spectroscopy to comprehensively analyze the crystal structure, morphological characteristics, and chemical composition of chlorite. Finally, based on timed sampling and XRD quantitative analysis data, a quantitative relationship between the chlorite growth rate and reaction conditions was established, covering conversion rate, crystallinity, and grain size parameters.

[0063] In summary, the chlorite synthesis experiment successfully simulated natural geological conditions by precisely preparing high-purity metal salt precursors and strictly controlling the ionic composition and pH value of the reaction system. The precise adjustment of reaction conditions, including temperature, atmospheric pressure, pH, and time, ensured the controllability of the chlorite synthesis process. The product was comprehensively characterized using various advanced techniques, revealing its crystal structure, morphology, and chemical composition, ensuring accurate analysis of the synthesized chlorite. Kinetic analysis, combined with timed sampling and quantitative X-ray diffraction, established a quantitative relationship between growth rate and reaction parameters, providing a scientific basis for understanding the chlorite formation mechanism and optimizing synthesis conditions.

[0064] Optionally, constructing the standardized training dataset specifically includes: The spatiotemporal gridding method is used to organize the comprehensive dataset, which is divided into multiple time windows according to the time dimension. Each time window contains 6 to 24 hours of continuous monitoring data. A three-dimensional grid is established based on the location of the monitoring points, with a grid resolution of 1–10 m. The three-dimensional grid is adjusted according to the geological complexity of the monitoring area. Each spatiotemporal grid cell contains microbial activity indicators, environmental parameters, and chlorite growth data.

[0065] Optionally, the geological complexity is quantified using a multi-factor comprehensive evaluation method, mainly considering geological elements closely related to the growth environment of chlorite, including tectonic complexity, lithological complexity, and hydrogeological complexity. The tectonic complexity is evaluated by fault density, fold development degree, and tectonic stress field change, with a value range of 0.1 - 1.0, reflecting the impact of geological structures on the microbial habitat environment and ion migration paths; the lithological complexity is calculated based on rock type distribution, permeability difference, and chemical composition variability, with a value range of 0.1 - 0.9, characterizing the control effect of different lithologies on microbial metabolic activities and the supply of chlorite precursor ions; the hydrogeological complexity is calculated by combining changes in groundwater flow direction, permeability distribution, and groundwater chemical zoning, with a value range of 0.1 - 0.8, quantifying the regulation degree of hydrogeological conditions on microbial activity and ion transport. The comprehensive complexity index GCI is calculated by weighted fusion of the above three components, with the weight distribution being 0.4 for tectonic complexity, 0.35 for lithological complexity, and 0.25 for hydrogeological complexity.

[0066] Optionally, the three-dimensional grid adjustment adopts a hierarchical refinement strategy based on the GCI value: when GCI ≤ 0.35, a coarse grid resolution of 8 - 10m is used; when 0.35 < GCI ≤ 0.65, it is adjusted to a medium resolution of 4 - 6m; when GCI > 0.65, a fine resolution of 1 - 3m is adopted; the specific adjustment process is as follows: first, calculate the average GCI value and spatial gradient of each initial grid cell. When GCI exceeds the hierarchical threshold or the gradient is greater than 0.2, octree segmentation is triggered, and the grid cell is decomposed into 8 sub-grids to double the resolution. A buffer transition zone with a width of 0.5 - 1.0 times the size of the coarse grid is established at the junction of different resolutions, and the grid density is smoothly connected through gradual interpolation; at the same time, local optimization is combined with the distribution density of monitoring points. The grid is densified in areas with dense monitoring points, and the coarse grid is maintained in sparse areas. The geological attributes of the subdivided sub-grids are inherited and refined through Kriging interpolation to ensure the consistency of grid attributes with the actual geological conditions and monitoring data.

[0067] Strict quality control inspection is carried out on the adjusted three-dimensional grid: in terms of grid geometric quality, it is required that the aspect ratio of grid cells is controlled within the range of 1:1 - 2.5:1, the ratio of adjacent grid sizes does not exceed 1.5:1, and the internal angles of the grid are maintained between 45° and 135°; in terms of grid coverage integrity, it is ensured that all monitoring points are included in grid cells, and the grid boundary is consistent with the monitoring area boundary; in terms of data consistency, the fitting accuracy of grid attribute values and measured data is verified, and it is required that the interpolation error is less than 15% of the measured value. At the same time, the continuity of attribute values of adjacent grid cells is checked to avoid unreasonable mutations. Grid cells that fail the quality inspection need to be re-adjusted in resolution or re-performed attribute interpolation until all quality index requirements are met.

[0068] Optionally, the spatial distribution pattern specifically includes: Spatial interpolation, spatial clustering, and spatial autocorrelation analysis; Among them: spatial interpolation is based on microbial activity data from discrete sampling points to generate a continuous spatial distribution map; The spatial clustering is used to identify high-value areas, low-value areas and transition areas of microbial activity, revealing the spatial distribution pattern of microbial activity. The spatial autocorrelation analysis is used to quantitatively assess the spatial correlation and spatial heterogeneity of microbial activity.

[0069] Optionally, the establishment of a kinetic model of the effect of microbial activity on the nucleation and growth process of chlorite specifically includes: a nucleation kinetic model, a growth kinetic model, and a morphology control model; The nucleation kinetics model includes: based on classical and non-classical nucleation theories, considering the influence of microbial metabolites on the nucleation energy barrier, nucleation rate and critical nucleus size, and calculating the nucleation rate equation that includes biological factors; Growth kinetics model: Using spiral growth mechanism, two-dimensional nucleation mechanism and multi-step growth mechanism, the growth rate of microbial extracellular polymers on each crystal face of chlorite crystal was calculated; Morphology control model: Based on Wulff configuration theory of crystal morphology, considering the selective adsorption of organic molecules and surface energy modification effects, predict the morphology evolution of chlorite crystals under microbial activity conditions.

[0070] Furthermore, the kinetic model of the effect of microbial activity on the nucleation and growth process of chlorite includes three sub-models: a nucleation kinetic model, a growth kinetic model, and a morphology control model, wherein: The nucleation kinetics model is based on classical and non-classical nucleation theories, taking into account the influence of microbial metabolites on the nucleation energy barrier, nucleation rate and critical nucleus size, and establishing a nucleation rate equation that includes biological factors. The growth kinetics model employs a spiral growth mechanism, a two-dimensional nucleation mechanism, and a multi-step growth mechanism to describe the regulatory effect of microbial extracellular polymers on the growth rate of various crystal faces of chlorite crystals. The morphology control model is based on the Wulff configuration theory of crystal morphology, taking into account the selective adsorption of organic molecules and the surface energy modification effect, and predicts the morphology evolution of chlorite crystals under microbial activity conditions.

[0071] The nucleation kinetics model is expressed by the following formula: ,in, The pre-index factor depends on the nucleation site density and diffusion rate; For nucleation energy barriers; Boltzmann's constant is 1; T is temperature. A correction function for microbial metabolites that promote or inhibit nucleation; Nucleation rate refers to the number of new crystal nuclei formed per unit volume or unit area per unit time, reflecting the rate at which a crystal transforms from a disordered state to an ordered state.

[0072] The growth kinetic model is expressed by the following formula: ,in, crystal plane The growth rate constant; This represents the concentration of precursor ions in the solution. To balance the concentration; Microbial extracellular polymers affect crystal planes Growth inhibition coefficient; Indicates the coverage or binding strength of extracellular polymeric substances; Indicates the first crystal The growth rate of a crystal facet, i.e., the increase in length of that crystal facet along the normal direction per unit time, reflects the growth rate of the crystal along different crystal facet directions.

[0073] The topography control model is expressed by the following formula: ,in, crystal plane The original surface free energy; This refers to the decrease in surface energy caused by the adsorption of microbial metabolites or extracellular polymers. This is the corrected surface free energy; crystal plane The area.

[0074] In summary, this kinetic model of microbial activity on chlorite nucleation and growth, through three sub-models—nucleation kinetics, growth kinetics, and morphology control—achieves a systematic description of the entire mineral formation process induced by microorganisms. The nucleation kinetics model, combining classical and non-classical theories, quantifies the influence of microbial metabolites on the nucleation energy barrier, rate, and critical nucleus size, clarifying the crucial role of biological factors in early mineral formation. The growth kinetics model introduces helical, two-dimensional, and multi-step mechanisms, revealing the regulatory effect of extracellular polymers on the growth rate of different crystal faces, reflecting the precise control of crystal structure by microorganisms. The morphology control model, based on the Wulff configuration and considering organic molecule adsorption and surface energy modification, predicts the morphological evolution of chlorite crystals under microbial activity conditions, providing theoretical support for understanding microbial regulation of mineral crystal morphology and comprehensively enhancing the quantitative analysis capability of chlorite growth mechanisms.

[0075] Optionally, the construction of the spatiotemporal prediction model for microbial activity based on the standardized training dataset specifically includes: Design an adaptive deep learning architecture that integrates convolutional neural networks, long short-term memory networks, and attention mechanisms; A spatiotemporal prediction model for microbial activity was constructed by combining a standardized training dataset with an adaptive deep learning architecture.

[0076] Specifically, a deep learning architecture integrating Convolutional Neural Networks (CNN), Long Short-Term Memory Networks (LSTM), and attention mechanisms was designed to construct a predictive model for chlorite growth that can adaptively extract spatiotemporal features of microbial activity. The advanced aspect of this technology lies in successfully solving the modeling challenges posed by high noise, multi-scale, and strong nonlinearity in geological environmental monitoring data. The CNN module is primarily used to identify the spatial distribution patterns of microbial activity and environmental parameters, the LSTM is used to capture the evolutionary trend of microbial activity over time, and the attention mechanism enhances the model's responsiveness to key variables and mutation points, improving its interpretability and prediction accuracy. Furthermore, data augmentation, transfer learning, and cross-validation techniques enhance the model's robustness and generalization ability under different geological conditions. This model can not only predict the growth rate and spatial distribution of chlorite but also further deduce its morphological evolution under different microbial community dominance mechanisms, thus providing technical support for geological disaster early warning and mineral resource assessment.

[0077] Optionally, the adaptive deep learning architecture specifically includes: A multi-layered structure consisting of an input layer, a feature extraction layer, a feature fusion layer, a temporal modeling layer, and an output layer; Among them: the input layer is used to standardize and encode the comprehensive dataset, converting different types of data into vector forms that the network can process; The feature extraction layer is used to process temporal and spatial data by combining one-dimensional and two-dimensional convolutional neural networks, respectively, to extract local and global features. The feature fusion layer is used to adaptively fuse different types of local and global features using attention and gating mechanisms to generate fused features that contain information on microbial activity and environmental parameters. The temporal modeling layer is used to model the temporal dependencies between microbial activity and environmental parameters based on the fused features output by the feature fusion layer. The output layer is used to output chlorite growth prediction results by employing fully connected layers and activation functions, combined with temporal dependencies.

[0078] Furthermore, the extraction of time-series features of microbial activity includes three steps: time series preprocessing, feature parameter calculation, and pattern recognition. Specifically: time series preprocessing is used to eliminate the influence of measurement noise and outliers; feature parameter calculation includes the extraction of statistical features, time-domain features, and frequency-domain features; pattern recognition is used to identify periodic, sudden, and trend-based changes in microbial activity; and the feature extraction process also includes multi-scale analysis, capturing the changing patterns of microbial activity at different time scales through feature extraction at different time windows.

[0079] In summary, the extraction of time-series features of microbial activity achieves a comprehensive capture of the dynamic changes in microbial activity through three steps: time series preprocessing, feature parameter calculation, and pattern recognition. The preprocessing stage effectively removes noise and outliers, ensuring data quality; feature calculation covers statistical, time-domain, and frequency-domain indicators, enriching feature expression; and pattern recognition reveals periodic, sudden, and trend-based changes, reflecting the complex dynamics of microbial activity. Simultaneously, multi-scale analysis extracts features through different time windows, accurately capturing the patterns of activity changes at different time scales, thus enhancing the analytical capability for the spatiotemporal evolution of microbial activity.

[0080] Furthermore, the spatial distribution pattern analysis comprises three components: spatial interpolation, spatial clustering, and spatial autocorrelation analysis. Specifically: spatial interpolation generates a continuous spatial distribution map based on microbial activity data from discrete sampling points; spatial clustering is used to identify high-value areas, low-value areas, and transition areas of microbial activity, revealing the spatial distribution pattern of microbial activity; and spatial autocorrelation analysis is used to quantitatively assess the spatial correlation and spatial heterogeneity of microbial activity.

[0081] In summary, spatial distribution pattern analysis utilizes three techniques—spatial interpolation, clustering, and autocorrelation analysis—to achieve a precise spatial characterization of microbial activity. Spatial interpolation transforms discrete sampling point data into a continuous distribution map, intuitively reflecting the trend of activity changes; spatial clustering identifies high-value areas, low-value areas, and transitional zones, revealing the spatial structural characteristics of microbial activity; and spatial autocorrelation analysis quantitatively assesses the spatial correlation and heterogeneity of activity, deepening the understanding of the spatial distribution patterns of microorganisms and providing a scientific basis for environmental regulation and chlorite growth prediction.

[0082] Furthermore, the capture of the feedback regulation mechanism of microbial activity fluctuations on chlorite growth patterns includes the following steps: based on the known chemical structures of the main components of microbial extracellular polymers and chlorite precursor ions, an initial molecular structure is constructed using Avogadro, and charge distribution and force field parameterization are performed on the microbial extracellular polymer fragments; TIP3P water molecules are added around the initial molecular structure to form a buffer layer of at least 1.5 nm, and appropriate amounts of sodium and chloride ions are added according to the experimental pH conditions to achieve electroneutrality; unreasonable atomic overlap is removed by minimizing energy, and the temperature is slowly increased to the target temperature in an isobaric isothermal system; at least 1 nm of the process is first carried out in an isobaric isothermal system. Pressure equilibrium was achieved to stabilize the system density, temperature, and energy. The system was then run for another 1 ns in an isothermal system of equal volume to eliminate the influence of temperature fluctuations on molecular configuration. A production simulation of at least 100 ns was performed with a time step of 2 fs, recording trajectory files every 2 ps. The trajectory files were analyzed to calculate coordination number, coordination bond length distribution, and radial distribution function, observing the dynamic binding between metal ions and microbial extracellular polymer functional groups, and describing the complexation reaction pathway from both geometric and energy perspectives. Binding free energy was calculated for selected keyframes in the production simulation to decompose the energy contribution, quantify the influence of different functional groups on complexation stability, and determine the dominant complexation sites.

[0083] In summary, by combining molecular modeling and molecular dynamics simulation, an interaction model of microbial extracellular polymers and chlorite precursor ions was constructed, accurately reproducing the structural evolution process of the system in aqueous solution and specific pH environments. Furthermore, through trajectory analysis and free energy calculation, the dynamic complexation behavior between metal ions and microbial functional groups and their stability contribution were revealed in depth, thereby achieving precise capture and quantitative analysis of the mechanism by which microbial activity fluctuations regulate chlorite growth.

[0084] Optionally, the step of dynamically updating and error-correcting the spatiotemporal prediction model of microbial activity to obtain the final spatiotemporal prediction model specifically includes: The final spatiotemporal prediction model for microbial activity is generated through dynamic updating and error correction using Kalman filtering and particle filtering algorithms, including: When the prediction results show abnormal changes in the growth rate or morphology of chlorite, the early warning mechanism of the geological environment monitoring system is triggered, and the monitoring frequency and sampling strategy are adjusted. If the prediction results are within the preset growth pattern range, the current monitoring strategy will be maintained and chlorite growth prediction will continue. If the prediction results exceed the preset range, the monitoring density will be increased and the prediction model parameters will be recalibrated until the prediction accuracy meets the requirements.

[0085] Specifically, by introducing Kalman filtering and particle filtering algorithms, dynamic parameter updates and adaptive error correction of the model under real-time monitoring conditions are achieved, thereby enabling "closed-loop control" of the prediction model and improving its real-time response capability and robustness. During model operation, sudden changes in environmental data and microbial activity indicators can cause prediction results to deviate from the actual trend. The filtering algorithm can re-evaluate the model state and optimize parameters based on the latest observation data, promptly correcting prediction biases and ensuring the continuity and accuracy of results. Especially in situations such as abnormal chlorite growth rates and drastic changes in crystal morphology, the dynamic correction mechanism can effectively prevent error propagation, enhance the model's online learning capability and stability, and thus support the realization of a high-precision, low-latency geological environment prediction system.

[0086] When the growth rate or crystal morphology of chlorite deviates from the expected trend, an early warning mechanism is automatically triggered, and the monitoring frequency and sampling strategy are adjusted to optimize resource allocation and improve response efficiency, thereby enhancing the ability to respond to sudden geological events. This allows for increased data acquisition frequency, with a focus on monitoring abnormal areas and key indicator variables. If the prediction results remain within a set safe range, the regular monitoring density is maintained (the monitoring density includes temporal and spatial monitoring densities; temporal monitoring density is the number of data acquisitions per unit time, and spatial monitoring density is the number of monitoring points per unit area. Based on the type of anomaly in the prediction results, the temporal and / or spatial monitoring densities are adaptively increased to enhance the accuracy and reliability of monitoring microbial activity and the dynamic process of chlorite growth), reducing energy consumption and data processing burden. Furthermore, if the prediction results are continuously determined to be inconsistent with actual occurrences, model retraining or parameter recalibration can be automatically triggered to achieve a closed-loop optimization of "model-monitoring-feedback," ensuring long-term operational stability and prediction accuracy.

[0087] In summary, this adaptive deep learning architecture efficiently processes and fuses comprehensive datasets through a multi-layered structure, achieving accurate prediction of chlorite growth. The input layer standardizes and encodes different data types to ensure consistency; the feature extraction layer combines one-dimensional and two-dimensional convolutional neural networks to capture temporal and spatial features, extracting local and global information; the feature fusion layer employs attention and gating mechanisms to adaptively integrate key features, enhancing important information and suppressing redundancy; the temporal modeling layer deeply explores the temporal dependencies between microbial activity and environmental parameters; and the output layer, through fully connected layers and activation functions, accurately predicts growth rate, orientation, and crystal morphology parameters, significantly improving prediction accuracy and reliability.

[0088] Furthermore, the convolutional neural network adopts a ResNet-18 or ResNet-34 architecture, containing 18 or 34 convolutional layers. Each convolutional layer uses a 3×3 or 5×5 convolutional kernel, and the activation function is ReLU or Leaky ReLU. Batch normalization is used to accelerate the training process, and residual connections are used to alleviate the gradient vanishing problem. The network input layer receives a 64×64 or 128×128 pixel feature image, extracts spatial features through multiple convolutional and pooling layers, and finally outputs a feature vector through a fully connected layer. The network training uses the Adam or SGD optimization algorithm, with a learning rate set to 0.001–0.01, a batch size set to 16–64, and a training epoch set to 30–100 epochs. Dropout is used to prevent overfitting, with a dropout ratio set to 0.2–0.5.

[0089] In summary, this convolutional neural network employs a ResNet-18 or ResNet-34 architecture with depths of 18 and 34 layers respectively, combining 3×3 or 5×5 convolutional kernels to effectively extract spatial features. ReLU or Leaky ReLU activation functions and batch normalization techniques are used to accelerate training and improve network stability; residual connections effectively alleviate the vanishing gradient problem, ensuring the training effect of deep networks. The input layer accepts 64×64 or 128×128 pixel feature images, extracting rich spatial information through multiple convolutional and pooling layers. Training uses the Adam or SGD optimizer, combined with appropriate learning rate and batch size settings, undergoing 30–100 training rounds. Dropout is used to prevent overfitting, improving the model's generalization ability and prediction accuracy.

[0090] Furthermore, the Long Short-Term Memory (LSTM) network comprises 1 to 3 LSTM layers, each containing 64 to 256 hidden units. The input sequence length is set to 24 to 72 time steps, corresponding to 24 to 72 hours of historical data. Information flow is controlled through a gating mechanism, including a forget gate, an input gate, and an output gate. The activation function is the tanh function, and gradient pruning is used to prevent gradient explosion. The pruning threshold is set to 0.5 to 2.0. The network training uses the backpropagation algorithm, with a learning rate set to 0.001 to 0.01 and a dropout ratio set to 0.1 to 0.3 to prevent overfitting. An early stopping strategy is employed, stopping training when the validation set loss does not decrease for 5 to 10 consecutive rounds.

[0091] In summary, this Long Short-Term Memory (LSTM) network comprises 1 to 3 layers of LSTM, with 64 to 256 hidden units per layer. The input sequence covers 24 to 72 hours of historical data, effectively capturing long-term dependencies in time series. Gating mechanisms such as forget gates, input gates, and output gates regulate information flow, combined with the tanh activation function, to achieve accurate modeling of temporal features. Gradient pruning is employed to prevent gradient explosion and ensure training stability. During training, backpropagation is used for optimization, with appropriate learning rates and dropout ratios set to prevent overfitting. Early stopping is also employed to effectively avoid overtraining and improve the model's generalization ability and prediction accuracy.

[0092] Furthermore, the attention mechanism employs a multi-head self-attention structure, containing 4 to 8 attention heads, each with a dimension of 64. Attention weights are calculated using three matrices: query, key-value, and value. A scaled dot product attention mechanism is used, with the scaling factor set to the square root of the dimension. Attention weights are normalized using the softmax function, and residual connections and layer normalization techniques are employed to improve model performance. Position encoding is generated using sine and cosine functions, with a maximum sequence length of 256. The multi-head attention mechanism captures the dependencies between different time steps and different features, and an attention weight visualization system is established to intuitively display the key information that the model focuses on.

[0093] In summary, this attention mechanism employs a multi-head self-attention structure, containing 4 to 8 attention heads, each with a dimension of 64. Attention weights are calculated through queries, key-value pairs, and numerical matrices, and a scaled dot product mechanism is used to enhance computational stability. Softmax normalization of weights, combined with residual connections and layer normalization techniques, effectively improves model training performance and generalization ability. Positional encoding is generated based on sine and cosine functions, supporting sequence lengths up to 256, enhancing the model's ability to capture dependencies between different time steps and features within the sequence. Furthermore, an attention weight visualization system is established to intuitively display the key information the model focuses on, improving understanding and tuning efficiency.

[0094] Furthermore, the automatic adjustment of monitoring frequency and sampling strategy includes the following steps: automatically retrieving historical data related to microbial activity fluctuations, abnormal changes in environmental parameters, and monitoring error records based on the anomaly type; evaluating the possible driving factors and impact range of the anomaly through a data fusion analysis module; adaptively increasing the data acquisition frequency of relevant modules for monitoring areas or points where anomalies occur, based on the rate of environmental change and the intensity of the anomaly; if the spatial distribution of the anomaly shows an expanding trend, automatically identifying the spatial weight and anomaly correlation of neighboring monitoring points, and adding temporary sampling points or activating backup sensor nodes in necessary areas to achieve spatially encrypted monitoring of the anomaly area; dynamically optimizing sampling parameters, including Raman laser power, electrochemical measurement frequency, and fluorescence exposure time, based on the current environmental state and microbial response speed, to improve data quality and reduce the impact of external interference on sampling stability.

[0095] Furthermore, the evaluation of the prediction accuracy adopts a diversified performance index system, including three dimensions: point prediction accuracy, interval prediction accuracy, and trend prediction accuracy. Point prediction accuracy is measured by root mean square error, mean absolute error, and mean absolute percentage error, requiring the root mean square error to be less than 10% of the actual value; interval prediction accuracy is measured by prediction interval coverage and average interval width, requiring the coverage of the 95% confidence interval to be no less than 80%; trend prediction accuracy is measured by directional accuracy and trend correlation coefficient, requiring the directional accuracy to be no less than 75%.

[0096] In summary, the prediction accuracy evaluation employs a multi-dimensional performance index system to comprehensively measure the model's capabilities in point prediction, interval prediction, and trend prediction. Point prediction is evaluated using root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAS%), ensuring that the MSE is less than 10% of the actual value to guarantee high accuracy in numerical prediction. Interval prediction is assessed using prediction interval coverage and average interval width, requiring a 95% confidence interval coverage of no less than 80%, reflecting the reliability and confidence level of the prediction results. Trend prediction is evaluated using directional accuracy and trend correlation coefficient, requiring a directional accuracy of over 75% to ensure the model accurately captures changing trends, effectively enhancing the practical value and decision support capabilities of chlorite growth prediction.

[0097] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0098] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0099] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0100] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting chlorite growth based on microbial activity, characterized in that, The method includes: S1. Based on the geological environment monitoring system, a comprehensive dataset is generated by dynamic quantitative monitoring of microbial activity and synchronous collection of environmental parameters, including: microbial activity indicators, environmental parameters and chlorite growth data; S2. A kinetic model of the influence of microorganisms on the nucleation and growth of chlorite is established by combining a comprehensive dataset and molecular dynamics; and a standardized training dataset is constructed based on the input and output data of the kinetic model. S3. Construct a spatiotemporal prediction model for microbial activity based on a standardized training dataset; S4. Dynamically update and correct the spatiotemporal prediction model of microbial activity to obtain the final spatiotemporal prediction model, and output the corrected chlorite growth prediction results based on the final spatiotemporal prediction model.

2. The method for predicting chlorite growth based on microbial activity according to claim 1, characterized in that, The S1 section describes the generation of a comprehensive dataset based on the dynamic quantitative monitoring of microbial activity and the synchronous acquisition of environmental parameters using a geological environment monitoring system. Specifically, this includes: The geological environment monitoring system includes: a Raman spectroscopy detection module, an electrochemical impedance measurement module, a fluorescent probe labeling module, an environmental parameter monitoring module, and a data fusion analysis module; Raman spectroscopy detection module, used to identify characteristic peaks of microbial metabolites; An electrochemical impedance spectroscopy module is used to detect changes in the resistance of microbial membranes; The fluorescent probe labeling module is used to label the extracellular polymer secretion state; The environmental parameter monitoring module is used to acquire environmental factor data; The data fusion and analysis module is used to establish a multi-source data fusion framework that integrates real-time monitoring data, historical geological data, and laboratory analysis results. Through the multi-source data fusion framework, a comprehensive dataset containing microbial activity indicators, environmental parameters, and chlorite growth data is constructed.

3. The method for predicting chlorite growth based on microbial activity according to claim 2, characterized in that, The multi-source data fusion framework specifically includes: The multi-source data fusion framework adopts a layered fusion architecture, including sensor-level fusion, feature-level fusion, and decision-level fusion; Among them, sensor-level fusion performs time registration, spatial registration and data association processing on raw data from sensors of the same type; Feature-level fusion is used to extract correlated features and complementary information between different data sources and to construct a multi-dimensional feature vector space. Decision-level fusion is used to combine the fusion results of sensor-level fusion and feature-level fusion to generate microbial activity assessment results.

4. The method for predicting chlorite growth based on microbial activity according to claim 1, characterized in that, S2 specifically includes: The complexation reaction mechanism between microbial extracellular polymers and chlorite precursor ions was predicted by molecular dynamics simulation and verified by chlorite synthesis experiments under controlled laboratory conditions, thus establishing a kinetic model of the effect of microbial activity on chlorite nucleation and growth. A kinetic model was used to quantify the regulatory effects of microbial metabolites on the morphology and growth direction of chlorite crystals. A standardized training dataset was constructed based on a comprehensive dataset, which includes: temporal characteristics of microbial activity, spatial distribution patterns, and changes in environmental parameters; A standardized training dataset was used to capture the impact of environmental parameter changes on the succession of microbial community structure, as well as the feedback regulation mechanism of microbial activity fluctuations on chlorite growth patterns.

5. The method for predicting chlorite growth based on microbial activity according to claim 4, characterized in that, The construction of the standardized training dataset specifically includes: The spatiotemporal gridding method is used to organize the comprehensive dataset, dividing the comprehensive dataset into multiple time windows according to the time dimension; A three-dimensional grid is established based on the location of the monitoring points, and the three-dimensional grid is adjusted according to the geological complexity of the monitoring area. Each spatiotemporal grid cell contains microbial activity indicators, environmental parameters, and chlorite growth data.

6. The method for predicting chlorite growth based on microbial activity according to claim 4, characterized in that, The spatial distribution pattern specifically includes: Spatial interpolation, spatial clustering, and spatial autocorrelation analysis; Spatial interpolation is based on microbial activity data from discrete sampling points to generate a continuous spatial distribution map. Spatial clustering is used to identify high-value areas, low-value areas, and transitional areas of microbial activity, revealing the spatial distribution patterns of microbial activity. Spatial autocorrelation analysis is used to quantitatively assess the spatial correlation and spatial heterogeneity of microbial activity.

7. The method for predicting chlorite growth based on microbial activity according to claim 4, characterized in that, The establishment of a kinetic model for the effect of microbial activity on the nucleation and growth process of chlorite specifically includes: a nucleation kinetic model, a growth kinetic model, and a morphology control model; The nucleation kinetics model includes: based on classical and non-classical nucleation theories, considering the influence of microbial metabolites on the nucleation energy barrier, nucleation rate and critical nucleus size, and calculating the nucleation rate equation that includes biological factors; Growth kinetics model: Using spiral growth mechanism, two-dimensional nucleation mechanism and multi-step growth mechanism, the growth rate of microbial extracellular polymers on each crystal face of chlorite crystal was calculated; Morphology control model: Based on Wulff configuration theory of crystal morphology, considering the selective adsorption of organic molecules and surface energy modification effects, predict the morphology evolution of chlorite crystals under microbial activity conditions.

8. The method for predicting chlorite growth based on microbial activity according to claim 1, characterized in that, The construction of the spatiotemporal prediction model for microbial activity based on a standardized training dataset specifically includes: Design an adaptive deep learning architecture that integrates convolutional neural networks, long short-term memory networks, and attention mechanisms; A spatiotemporal prediction model for microbial activity was constructed by combining a standardized training dataset with an adaptive deep learning architecture.

9. The method for predicting chlorite growth based on microbial activity according to claim 8, characterized in that, The adaptive deep learning architecture specifically includes: A multi-layered structure consisting of an input layer, a feature extraction layer, a feature fusion layer, a temporal modeling layer, and an output layer; Among them: the input layer is used to standardize and encode the comprehensive dataset, converting different types of data into vector forms that the network can process; The feature extraction layer is used to process temporal and spatial data by combining one-dimensional and two-dimensional convolutional neural networks, respectively, to extract local and global features. The feature fusion layer is used to adaptively fuse different types of local and global features using attention and gating mechanisms to generate fused features that include information on microbial activity and environmental parameters. The temporal modeling layer is used to model the temporal dependencies between microbial activity and environmental parameters based on the fused features output by the feature fusion layer. The output layer is used to output chlorite growth prediction results by employing fully connected layers and activation functions, combined with temporal dependencies.

10. The method for predicting chlorite growth based on microbial activity according to claim 1, characterized in that, The process of dynamically updating and error-correcting the spatiotemporal prediction model of microbial activity to obtain the final spatiotemporal prediction model specifically includes: The final spatiotemporal prediction model for microbial activity is generated through dynamic updating and error correction using Kalman filtering and particle filtering algorithms, including: When the prediction results show abnormal changes in the growth rate or morphology of chlorite, the early warning mechanism of the geological environment monitoring system is triggered, and the monitoring frequency and sampling strategy are adjusted. If the prediction results are within the preset growth pattern range, the current monitoring strategy will be maintained and chlorite growth prediction will continue. If the prediction results exceed the preset range, the monitoring density will be increased and the prediction model parameters will be recalibrated until the prediction accuracy meets the requirements.

Images