System and method for multi-source heterogeneous data fusion analysis of oil and gas reservoirs

Abstract

The present invention relates to the technical field of oil and gas reservoir data analysis, and provides a system and method for multi-source heterogeneous data fusion analysis of oil and gas reservoirs. The system includes a data acquisition module, a data processing unit, and a comprehensive analysis module. The data acquisition module realizes multi—source data collection and calibration through a distributed sensor network; the data processing unit performs classification, feature extraction, and anomaly detection on heterogeneous data; and the comprehensive analysis module constructs a reservoir characteristic model and provides decision support. The system further contains dynamic optimization, environment adaptation, and resource management fiinctions to improve the stability and intelligence level of the system. The present application can significantly improve the comprehensiveness and accuracy of reservoir analysis, optimize data processing processes, and is suitable for exploration tasks of deep and unconventional oil and gas reservoirs under complex geological conditions. FIG. 1

Description

TECHNICAL FIELD The present invention relates to the technical eld of oil and gas reservoir data analysis, and in particular to a . BACKGROUND In the process of exploration and development of oil and gas reservoirs, data acquisition, processing, and analysis are key links to achieve efcient resource evaluation and optimized production decisionmaking. With the continuous advancement of exploration technologies, the data sources related to oil and gas reservoirs have become increasingly diversied, including seismic data, logging data, geological data, production dynamic data, and experimental analysis data. These data have the characteristics of multisource heterogeneity, meaning there are signicant differences in data format, structure, and semantics, along with large data volume and complex dimensions. However, existing data analysis methods and technical means often process a single data source or specic type of data independently, lacking the ability for in-depth fusion and comprehensive analysis of multisource heterogeneous data. This limitation leads to low data utilization, serious information silo phenomenon, and one-sidedness and inconsistency of analysis results, making it difcult to illy reect the true characteristics and dynamic change laws of reservoirs. In addition, traditional data analysis methods usually rely on manual experience or simple mathematical models, lacking intelligent and automated technical support, which makes it difcult to meet the highdimensional data processing needs under complex reservoir conditions. At the same time, existing technologies insufciently consider spatiotemporal correlation and multiscale features in the data ision process, irther limiting the accuracy and reliability of analysis results. Faced with the growing data scale and complex geological conditions in oil and gas reservoir development, how to effectively integrate multi-source heterogeneous data and tap its potential value has become an urgent technical problem to be solved. Therefore, developing a system and method capable of efcient fusion and intelligent analysis of multisource heterogeneous data is of great signicance for improving the scienticity of oil and gas reservoir research and the practicality of engineering applications. SUMMARY To solve the technical problems, the present invention provides a , which can realize highprecision analysis of reservoir characteristics under complex geological conditions and optimize the data processing process and resource allocation efciency. The basic solution provided by the present invention is that a system for multisource heterogeneous data fusion analysis of oil and gas reservoirs includes a data acquisition module, a data processing unit, and a comprehensive analysis module. The data acquisition module includes a distributed sensor network, a signal conversion module, and a data calibration module. The data processing unit includes a data classication module, a feature extraction module, and an anomaly detection module. The comprehensive analysis module includes a model construction module, a result evaluation module, and a decision support module. The distributed sensor network is composed of multiple nodes, and each node is connected to the signal conversion module through optical ber or wireless communication; The signal conversion module automatically selects the corresponding data format for conversion according to the type of input signals to adapt to subsequent processing needs. The data calibration module is arranged after the signal conversion module and is congured to dynamically calibrate the acquired data to eliminate errors caused by environmental changes. The data classication module in the data processing unit receives original data from the data acquisition module and divides heterogeneous data from different sources and types into independent channels for processing; The feature extraction module conducts multilevel analysis on the classied data and extracts key characteristic information of the reservoir, such as porosity, permeability, and uid distribution characteristics. identifying potential abnormal areas or data anomaly points by an anomaly detection module based on the extracted information to provide reference for subsequent analysis. The model construction module in the comprehensive analysis module generates a threedimensional visualization model of reservoir characteristics by integrating the results of the feature extraction module. The result evaluation module veries the accuracy of the model and optimizes model parameters in combination with historical data. The decision support module generates development suggestions according to the evaluation results and provides them to users as operation references through an interactive interface. The principle and advantages of the present invention are as follows: through the design of a distributed sensor network and combined with the fusion technology of multi-source heterogeneous data, the comprehensiveness and accuracy of reservoir analysis can be signicantly improved. The distributed structure enables a wider data acquisition range, and at the same time, the quality and availability of data are improved through intelligent data calibration and classication technologies. The fusion analysis of multi-source heterogeneous data integrates various data types such as seismic, logging, and geological data, which can not only reect the overall characteristics of the reservoir but also depict local details in a rened manner. In addition, the introduction of the anomaly detection module effectively reduces the risk of misjudgment and improves the reliability of analysis results. Compared with existing technologies, the distributed sensor network structure adopted in this solution solves the shortcomings of traditional single data sources in coverage range and data diversity. The combination of multisource heterogeneous data ision technology enhances the adaptability of the system under complex geological conditions, and it is especially suitable for exploration tasks of deep oil and gas reservoirs and unconventional reservoirs. In addition, the collaborative work of the data classication module and the feature extraction module simplies the data processing process and improves the realtime performance and intelligence level of the system. Further, the data acquisition module also includes a dynamic optimization module, and the dynamic optimization module includes a sampling frequency adjustment unit, a signal strength monitoring unit, and a noise suppression unit. The sampling frequency adjustment unit monitors a working status of each sensor node in real time through a built-in time synchronization mechanism and dynamically adjusts a sampling frequency according to a data change trend to balance data quality and acquisition efciency. The signal strength monitoring unit samples and compares the amplitude of the output signal of each node to ensure the consistency of signal transmission. The noise suppression unit uses ltering technology to suppress background noise generated during the acquisition process and improve the signaltonoise ratio of data. Through the collaborative work of the sampling frequency adjustment unit, the signal strength monitoring unit, and the noise suppression unit, the dynamic optimization module eliminates data uctuations caused by environmental interference or equipment aging and ensures the longterm stability of the system. Further, the feature extraction module includes a reservoir characteristic analysis unit, a spatial distribution calculation unit, and a data association unit; and Hierarchical processing is performed on the received original data by a reservoir characteristic analysis unit to extract basic physical characteristic information of the reservoirs, such as rock mechanical properties, and uid saturation. Planar and crosssectional distribution maps of reservoir attributes are generated by a spatial distribution calculation unit by analyzing spatial distribution characteristics of the data; and logical connections between data are established by a data association unit based on the correlation between different types of data and providing support for reservoir modeling. The feature extraction module realizes comprehensive depiction of reservoir characteristics through multi-level analysis of the reservoir characteristic analysis unit, the spatial distribution calculation unit, and the data association unit, providing a reliable basis for subsequent comprehensive analysis. Further, the comprehensive analysis module also includes an environment adaptation module. The environment adaptation module includes a geological condition acquisition unit, a data weight adjustment unit, and an analysis mode selection unit; and the geological condition acquisition unit monitors geological conditions of a reservoir area in real time through integrated ground stress sensors, thermometers, and pressure gauges. The data weight adjustment unit automatically adjusts the weight of different types of data according to the acquired geological parameters to highlight key information. The analysis mode selection unit dynamically selects the optimal analysis mode according to reservoir characteristics and data characteristics to improve the accuracy of analysis results. Through the intelligent regulation of the geological condition acquisition unit, the data weight adjustment unit, and the analysis mode selection unit, the environment adaptation module enhances the analysis ability of the system under different geological conditions. Further, the comprehensive analysis module also includes a resource management unit and a fault early-waming unit. The resource management unit dynamically adjusts the working status of each module by monitoring the real-time resource occupancy of the system, such as reducing the data acquisition frequency or lowering the model calculation complexity, to extend the operation time of the system. The fault early-waming unit quickly locates and reports abnormal situations by monitoring an operating status of each module in real time, which is convenient for maintenance personnel to deal with them in time. The introduction of the resource management unit and the fault earlywaming unit improves the operation efciency and reliability of the system and reduces maintenance costs. The present invention also provides a method for multi-source heterogeneous data fusion analysis of oil and gas reservoirs, including the following steps: Sl: activating a distributed sensor network, selecting an initial data format through the signal conversion module, and performing dynamic calibration through a data calibration module; SZ: classifying the received original data by a data classication module and dividing the same into independent channels; S3: processing the classied data by a feature extraction module to extract key characteristic information of the reservoirs; S4: identifying potential abnormal areas or data anomaly points by an anomaly detection module based on the extracted information; SS: generating a three-dimensional visualization model of reservoir characteristics by a model construction module according to feature extraction results; and S6: verifying and optimizing a model by a result evaluation module, and generating development suggestions by a decision support module. Further, S3 includes the steps of: S301: performing hierarchical processing on the received original data by a reservoir characteristic analysis unit to extract basic physical characteristic information of the reservoirs; S302: generating planar and cross-sectional distribution maps of reservoir attributes by a spatial distribution calculation unit by analyzing spatial distribution characteristics of the data; S303: establishing logical connections between data by a data association unit based on the correlation between different types of data; and S304: generating a feature description of the reservoir by integrating results of reservoir characteristics, spatial distribution, and data association. Further, the method includes the steps of: S7: automatically adjusting a data weight and analysis mode by an environment adaptation module according to data of a geological condition acquisition unit; S701: optimizing a sampling frequency and signal strength of the distributed sensor network in real time by a dynamic optimization module; S702: dynamically adjusting a working status of each module by a resource management unit according to realtime resource occupancy of the system; Further, the method includes the steps of: S8: monitoring an operating status of each module in real time by a fault early-warning unit to quickly locate and report abnormal situations; and S801: adjusting a working mode by the comprehensive analysis module to maintain normal operation of the system when an abnormality is detected. BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of an overall structure of the present invention; FIG. 2 is a block diagram of an internal structure of a data acquisition module of the present invention; FIG. 3 is a block diagram of an internal structure of a feature extraction module of the present invention; FIG. 4 is a block diagram of a functional composition of a comprehensive analysis module of the present invention; and FIG. 5 is a schematic ow chart of an analysis method of the present invention. DETAILED DESCRIPTION Technical solutions of the present invention will be described clearly and completely in the following with reference to the attached drawings of the present invention. Obviously, all the described examples are only some, rather than all examples of the present invention. Based on the examples in the present invention, all other examples obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention. The present invention provides a system and method for multi-source heterogeneous data ision analysis of oil and gas reservoirs, and its specic implementation is described in detail with reference to FIGS. 1 to 5. As shown in FIG. 1, an overall structure block diagram shows three core modules of the system: the data acquisition module, the data processing unit, and the comprehensive analysis module. These modules work collaboratively to achieve highprecision analysis of oil and gas reservoir characteristics and optimized resource allocation. First, the design of the data acquisition module is the foundation of the entire system. According to an internal structure block diagram in FIG. 2, the data acquisition module includes a distributed sensor network, a signal conversion module, a data calibration module, and a dynamic optimization module. The distributed sensor network is composed of multiple nodes, and each node is connected to the signal conversion module through optical ber or wireless communication. In practical applications, these nodes can be deployed at different locations of the reservoir to ensure wide coverage. For example, in deep oil and gas reservoir exploration tasks, the arrangement of nodes is adjusted according to geological conditions to maximize the acquisition of reservoir information. The core function of the signal conversion module is to automatically select the corresponding data format for conversion according to the type of input signals, to adapt to subsequent processing needs. For example, when receiving seismic wave signals, the signal conversion module converts them into a time-series data format suitable for processing by the feature extraction module. The data calibration module is congured to dynamically calibrate the acquired data to eliminate errors caused by environmental changes. For example, in hightemperature and highpressure environments, the sensor output may drift; and the data calibration module monitors environmental parameters in real time and adjusts calibration algorithms to ensure data accuracy. The dynamic optimization module irther enhances the stability of data acquisition, and it includes a sampling frequency adjustment unit, a signal strength monitoring unit, and a noise suppression unit. The sampling frequency adjustment unit monitors a working status of each sensor node in real time through a built-in time synchronization mechanism and dynamically adjusts the sampling frequency according to data change trends. For example, in areas where the reservoir uid distribution changes rapidly, the sampling frequency is increased accordingly to capture more detailed information. The signal strength monitoring unit samples and compares the amplitude of the output signal of each node to ensure the consistency of signal transmission and avoid data loss caused by signal attenuation. The noise suppression unit uses ltering technology to suppress background noise generated during the acquisition process and improve a signal-to-noise ratio of the data. For example, an adaptive ltering algorithm is used for effectively suppressing lowfrequency noise while retaining useful information in high-frequency signals. Next, the data processing unit is responsible for classifying, extracting features from, and detecting anomalies in the received original data. The data classication module receives data from the data acquisition module and divides heterogeneous data from different sources and types into independent channels for processing. For example, seismic data, logging data, and geological data are classied separately to facilitate targeted processing by subsequent modules. The feature extraction module is the core of the data processing unit, and its internal structure is shown in FIG. 3. This module includes a reservoir characteristic analysis unit, a spatial distribution calculation unit, and a data association unit. The reservoir characteristic analysis unit performs hierarchical processing on the received original data and extracts basic physical characteristic information of the reservoir, such as rock mechanical properties and uid saturation. For example, by analyzing the reection characteristics of seismic waves, the porosity and permeability of the reservoir can be derived. The spatial distribution calculation unit analyzes the spatial distribution characteristics of the data to generate planar and crosssectional distribution maps of reservoir attributes. For example, based on the spatial distribution calculation of logging data, a threedimensional model of reservoir uid distribution can be generated. The data association unit establishes logical connections between data based on the correlation between different types of data, providing support for reservoir modeling. For example, by analyzing the correlation between seismic data and logging data, the local details of the reservoir can be depicted more accurately. The anomaly detection module identies potential abnormal areas or data anomaly points based on the extracted information. For example, by comparing historical data with current data, signicant changes in uid distribution in certain areas can be found to determine whether there is an abnormal situation. The comprehensive analysis module is a key part of the entire system, and its functional composition is shown in FIG. 4. The model construction module integrates results of the feature extraction module to generate a three-dimensional visualization model of reservoir characteristics. For example, by combining seismic data, logging data, and geological data, an overall three-dimensional model of the reservoir can be generated, intuitively showing the physical characteristics and uid distribution characteristics of the reservoir. The result evaluation module veries the accuracy of the model and optimizes model parameters in combination with historical data. For example, by comparing the model's prediction results with actual drilling data, key parameters in the model can be adjusted to improve prediction accuracy. The decision support module generates development suggestions based on the evaluation results and provides them to users as operation references through an interactive interface. For example, in areas where the reservoir uid distribution is relatively complex, the decision support module will recommend the use of horizontal well development to improve oil recovery. In addition, the comprehensive analysis module also includes an environment adaptation module, a resource management unit, and a fault early-warning unit. The environment adaptation module enhances the system's analysis capability under different geological conditions through intelligent regulation of the geological condition acquisition unit, the data weight adjustment unit, and the analysis mode selection unit. For example, in areas with high reservoir temperatures, the data weight adjustment unit automatically increases the weight of temperature data to highlight key information. The resource management unit dynamically adjusts the working status of each module by monitoring the real-time resource occupancy of the system. For example, when system resources are tight, the resource management unit reduces the data acquisition frequency or lowers the model calculation complexity to extend the system's operation time. The fault earlywarning unit quickly locates and reports abnormal situations by monitoring an operating status of each module in real time, which is convenient for maintenance personnel to deal with them in time. For example, when a sensor node fails, the fault early-warning unit immediately issues an alarm and indicates possible causes of the fault. The present invention also provides a method for multi-source heterogeneous data fusion analysis of oil and gas reservoirs, as shown in FIG. 5, with specic steps as follows. First, a distributed sensor network is activated, an initial data format is selected through the signal conversion module, and dynamic calibration is performed through the data calibration module. For example, in an early stage of reservoir exploration, the signal conversion module selects an appropriate initial data format according to the type of sensor nodes, and the data calibration module performs preliminary calibration on the acquired data to ensure data quality. Next, the data classication module classies the received original data and divides it into independent channels. For example, seismic data, logging ll data, and geological data are classied separately to facilitate targeted processing by subsequent modules. Then, the feature extraction module processes the classied data to extract key characteristic information of the reservoir. The specic steps include: the reservoir characteristic analysis unit performs hierarchical processing on the received original data to extract basic physical characteristic information of the reservoir; the spatial distribution calculation unit analyzes the spatial distribution characteristics of the data to generate planar and cross-sectional distribution maps of reservoir attributes; the data association unit establishes logical connections between data based on the correlation between different types of data; and integrates the results of reservoir characteristics, spatial distribution, and data association to generate a feature description of the reservoir. For example, by analyzing the reection characteristics of seismic waves and the spatial distribution of logging data, a threedimensional distribution map of reservoir porosity and permeability can be generated. Subsequently, the anomaly detection module identies potential abnormal areas or data anomaly points based on the extracted information. For example, by comparing historical data with current data, signicant changes in uid distribution in certain areas can be found to determine whether there is an abnormal situation. A three-dimensional visualization model of reservoir characteristics is generated by a model construction module according to feature extraction results. For example, by combining seismic data, logging data, and geological data, an overall three-dimensional model of the reservoir can be generated, intuitively showing the physical characteristics and uid distribution characteristics of the reservoir. A model is veried and optimized by a result evaluation module, and development suggestions are generated by a decision support module. For example, by comparing the model's prediction results with actual drilling data, key parameters in the model can be adjusted to improve prediction accuracy. Finally, the environment adaptation module automatically adjusts the data weight and analysis mode according to the data from the geological condition acquisition unit; the dynamic optimization module optimizes the sampling frequency and signal strength of the distributed sensor network in real time; and the resource management unit dynamically adjusts the working status of each module according to the real-time resource occupancy of the system. For example, in areas with high reservoir temperatures, the data weight adjustment unit automatically increases the weight of temperature data to highlight key information. The fault early-warning unit monitors the operating status of each module in real time; and when an abnormality is detected, the comprehensive analysis module adjusts the working mode to maintain the normal operation of the system. To irther illustrate the technical effects of the present invention, the following explanation is provided in conjunction with specic algorithm formulas. In the feature extraction module, the reservoir characteristic analysis unit uses the following formulas to extract the porosity and permeability of the reservoir: porositF(Vp-Vm) / Vt, where Vp represents a total volume of a reservoir, Vm represents a matrix volume of the reservoir, and Vt represents a total pore volume of the reservoir. Permeability is calculated using Darcy's Law: K=QuL / (AAP), where Q represents the uid ow rate, 1.1 represents a uid viscosity, L represents a length of the reservoir, A represents a cross-sectional area of the reservoir, and AP represents a pressure difference. These formulas can accurately reect the physical characteristics of the reservoir and provide a reliable basis for subsequent modeling. In the anomaly detection module, a statisticsbased method is used for identifying anomaly points, and its core formula is Z=(X-u) / o, where X represents a current data value, 1.1 represents a mean value of historical data, and o represents a standard deviation of historical data. When a Z value exceeds a set threshold, a data point is determined to be an anomaly point. This method can effectively reduce the risk of misjudgment and improve the reliability of analysis results. In summary, through the design of the distributed sensor network, the ision technology of multi-source heterogeneous data, and the intelligent data processing process, the present invention signicantly improves the comprehensiveness and accuracy of reservoir analysis. In practical applications, this system has been successfully applied to exploration tasks of multiple deep oil and gas reservoirs and unconventional reservoirs, achieving good results. For example, in a deep oil and gas reservoir exploration project, this system integrated seismic, logging, and geological data to generate a threedimensional visualization model of the reservoir, accurately predicted the uid distribution characteristics of the reservoir, and provided an important basis for the formulation of development plans. Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, the scope of which is dened by the appended claims and equivalents thereof.

Claims

1. System for multi-source heterogeneous data fusion analysis of oil and gas reservoirs, consisting of a data acquisition module, a data processing unit and an extensive analysis module, whereby The data acquisition module consists of a distributed sensor network, a signal conversion module and a data calibration module; The data processing unit consists of a data classification module, a feature- extraction module and an anomaly detection module; the extensive analysis module, a model construction module, a result evaluation module and includes a decision support module; The distributed sensor network is composed of multiple nodes and each The node is connected to the via fiber optic or wireless communication. signal conversion module; the signal conversion module automatically a corresponding selects data format for conversion based on the type of input signals; and the The data calibration module is configured to dynamically calibrate; The data classification module receives original data from the data acquisition module and distributes heterogeneous data from various sources and of various types via independent processing channels; the The feature extraction module performs a multi-level analysis on the classified data to extract important characteristic information about the reservoirs; and the anomaly detection module identifies potential abnormal areas or data anomalies based on the extracted information; and The model building module generates a three-dimensional visualization model of reservoir characteristics by integrating the results of the characteristic extraction module; the Result assessment module verifies the accuracy of a model and optimizes the model parameters in combination with historical data; and the module for decision support generates development suggestions based on the assessment results.

2. System for multi-source heterogeneous data fusion analysis of oil and gas reservoirs according to claim l, where the data acquisition module is also a dynamic optimization module includes, and the dynamic optimization module a sampling rate adjustment unit, a signal strength monitoring unit and a noise reduction unit includes; and the sampling frequency adjustment unit the operating status of each sensor node in monitored in real time via a built-in time synchronization mechanism and the dynamically adjusts sampling frequency based on a data change trend; the signal strength monitoring unit samples and compares an amplitude of a output signal of each node to the consistency of the signal transmission safeguard; and the noise reduction unit suppresses background noise that during the The collection process is generated using filter technology.

3. System for multi-source heterogeneous data fusion analysis of oil and gas reservoirs according to claim l; where the feature extraction module consists of a reservoir feature analysis unit, a spatial distribution calculation unit and a data association unity; and the reservoir feature analysis unit performs hierarchical processing on the received original data to basic information about the physical characteristics of the reservoirs extract; the spatial distribution calculation unit planar and cross-section distribution maps of reservoir characteristics generated by the spatial to analyze distribution characteristics of the data; and the data association unit establishes logical connections between data based on the correlation between different types of data.

4. System for multi-source heterogeneous data fusion analysis of oil and gas reservoirs according to claim ] , where the extensive analysis module also a environmental adaptation module comprises; and the environmental adaptation module a unit for obtaining geological conditions, a unit for adjusting includes the data weight and a unit for selecting the analysis mode; and the unit for obtaining geological conditions the geological Conditions of a reservoir area monitored in real time via integrated soil tension sensors, thermometers and pressure measurements; the data weight adjustment unit automatically the weight of different types adjusts data based on the collected geological parameters; and the analysis mode selection unit dynamically selects an optimal analysis mode based on the reservoir characteristics and data characteristics.

5. System for multisource heterogeneous data analysis of oil and gas reservoirs according to claim 1, where the extended analysis module is also a resource management unit and includes an early warning unit; and the resource management unit dynamically adjusts the work status of each module by the to monitor real-time resource utilization of the system; and the error-early warning unit quickly locates and reports abnormal situations by the to monitor the business status of each module in real time.

6. Method for multisource heterogeneous data fusion analysis of oil and gas reservoirs, consisting of the following steps: Sl: activating a distributed sensometwork, selecting an initial data format via the signal conversion module and performing dynamic calibration via a data calibration module; SZ: classicization of the received original data by a data classification module and this divide into independent channels; S3: processing the classified data by an attribute extraction module to to extract important characteristic information from the reservoirs; S4: Identifying potential anomalous areas or data deviation points by a anomaly detection module based on the extracted information; S5: generating a three-dimensional visualization model of reservoir characteristics by a model construction module based on the results of the feature extraction; and S6: Verifying and optimizing a model through a result evaluation module and generating development suggestions by a decision support module.

7. Method for multisource heterogeneous data fusion analysis of oil and gas reservoirs according to claim 6, where step S3 comprises the following steps: S301: performing hierarchical processing on the received original data by a reservoir characteristic analysis unit for basic information about the physical characteristics to extract from the reservoirs; S302: generating planar and cross-sectional distribution maps of reservoir characteristics by a spatial distribution calculation unit by the spatial to analyze distribution characteristics of the data; S303: establishing logical connections between data by a data association unit based on the correlation between different species data; and S304: generating a characteristic description of the reservoir from the results of to integrate reservoir characteristics, spatial distribution, and data association.

8. Method for multi-source heterogeneous data fusion analysis of oil and gas reservoirs according to claim 6, further comprising the following steps: S7: automatic adjustment of a data weight and analysis mode by a environmental adaptation module based on geological condition acquisition data unit; S701: optimizing a sampling frequency and signal strength of the distributed sensor network in real-time through a dynamic optimization module; S702: dynamically adjusting a working status of each module by a resource management unit according to the real-time resource utilization of the system; S8: real-time monitoring of the operational status of each module by a fault early warning unit to quickly locate abnormal situations and to report; and S801: adjusting a working mode using the extensive analysis module to the to maintain normal operation of the system when a deviation is detected. 1 / 4Fig.1