Geological hazard identification method and system based on multi-source data fusion

Abstract

The present invention provides a geological disaster identification method and system based on multi-source data fusion, which relates to the field of reservoir geological risk technology. By collecting reservoir data and performing correlation analysis on the data, iterative calculation is performed through multiple iterations and setting a learning rate, thereby generating a geological disaster probability occurrence value for reflecting the probability of reservoir geological disasters. The present invention collects the coordinate height, inclination, slope depth, and groundwater level depth of the reservoir slope, and performs correlation analysis on the coordinate height, inclination, slope depth, and groundwater level depth data of each day, and combines the collected data 360 times. Through data probability iteration, an iterative probability regression index is generated, and the minimum value of the iterative probability regression index is obtained, thereby generating a geological disaster probability occurrence value for reflecting the probability of reservoir geological disasters, automatically outputting the geological disaster probability, and helping staff better understand the reservoir geological disaster risk.

Description

Technical Field

[0001] The present invention relates to the field of reservoir geological risk technology, and in particular to a geological disaster identification method and system based on multi-source data fusion. Background Art

[0002] Reservoir geological risks refer to potential threats to the reservoir and its surrounding environment, infrastructure, and human activities caused by geological factors. These risks may affect the safe operation of the reservoir, leading to geological disasters such as dam failure, debris flow, and landslide, and even have serious impacts on downstream areas.

[0003] In general reservoir risk assessments, only data related to coordinate height, inclination, slope depth, and groundwater level depth are monitored. Emergency warnings are only issued when the relevant data exceeds the threshold. Since geological disasters are often not the result of a single factor, and this monitoring method does not comprehensively process the data, the errors in its emergency warnings are large and emergency warnings cannot be issued more accurately.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a geological disaster identification method and system based on multi-source data fusion to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A multi-source data fusion geological disaster identification method is used to analyze the probability of occurrence of geological disasters on reservoir slopes. The method is characterized by comprising the following specific steps:

[0008] S1. Collecting initial reservoir data, including initial slope coordinate height, initial slope inclination, initial slope depth, and initial groundwater level depth data;

[0009] S2. Reservoir data is collected once a day and analyzed with the initial reservoir data to obtain reservoir deviation data. The reservoir data is numbered according to the number of days. The reservoir deviation data includes the coordinate height deviation of the highest point of the slope, the slope inclination deviation, the slope depth deviation, and the slope groundwater level depth deviation. The collection days are 360 ​​days;

[0010] S3. performing correlation analysis on the deviation of the coordinate height of the highest point of the slope, the deviation of the slope inclination, the deviation of the slope depth, and the deviation of the groundwater level depth of the slope, respectively, to generate normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth, and integrating the normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth data of each day into classification groups;

[0011] S4. For all classification groups, the geological disaster probability is set respectively, and the disaster occurrence probability prediction value is set, and the correlation analysis of the geological disaster probability is performed to generate the initial probability;

[0012] S5. Establish an iterative model, perform K iterations on the initial probability, output the residual of the k-th iteration, and define the k-th iteration probability regression index;

[0013] S6. Set the learning rate, perform correlation analysis on the k-th iteration probability regression index, generate a geological disaster probability occurrence value, and use the geological disaster probability occurrence value to reflect the probability of geological disasters. Input the reservoir data collected in real time into the iteration model and output the geological disaster probability occurrence value.

[0014] Furthermore, the initial coordinate height A0 of the slope is determined by satellite GPS positioning, the initial inclination B0 of the slope is measured by an inclinometer, the initial depth C0 of the slope is measured by a deep displacement sensor, and the initial groundwater level depth D0 of the slope is measured by a water level sensor.

[0015] Furthermore, the height deviation of the highest point of the slope on day i is expressed as A i The slope deviation on the i-th day is expressed as B i The slope depth deviation on the i-th day is expressed as C i The groundwater level deviation of the slope on day i is expressed as D i Represents, where the subscript i is used as the index of day i.

[0016] Furthermore, the coordinate height deviation A of the highest point of the slope i , Slope inclination deviation B i , slope depth deviation C i , slope groundwater level depth deviation D i Perform correlation analysis separately to generate normalized coordinate height A′ i , normalized slope B′ i , normalized slope depth C′ i , normalized groundwater level depth D′ i , based on the formula:

[0017]

[0018] Among them, max{A i} and min{A i} respectively represent the maximum and minimum values ​​of the coordinate height deviation of the highest point of all slopes, max{B i} and min{B i} are used to represent the total slope inclination deviation B i The maximum and minimum values ​​of max{C i} and min{C i} are used to represent the total slope depth deviation C i The maximum and minimum values ​​of max{D i} and min{D i} are used to represent the groundwater level deviation D of the entire slope. i The maximum and minimum values ​​of the normalized coordinate height A′ i , normalized slope B′ i , normalized slope depth C′ i , normalized groundwater level depth D′ i Both are used to ensure data consistency, the coordinate height deviation of the highest point of the slope is A i Used to reflect the difference between the coordinate height of the highest point on the i-th day and the initial coordinate height of the slope, and the slope inclination deviation B i It is used to reflect the deviation value of the slope inclination on the i-th day from the initial slope inclination and the slope depth deviation C i It is used to reflect the deviation value between the slope depth on the i-th day and the initial slope depth, and the slope groundwater level depth deviation D i It is used to reflect the deviation between the depth of the slope groundwater level on the i-th day and the initial slope groundwater level.

[0019] Furthermore, in said S3, the classification group of normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth on the i-th day is set as T i , for all classification groups T i Set the geological disaster probability y i , the probability of geological disaster y i Perform correlation analysis to generate the initial probability F0 based on the following formula:

[0020]

[0021] Perform K iterations, where the output probability of the k-1th iteration is F k-1 , set the residual of the kth iteration to The formula is:

[0022]

[0023] Among them, the residual of the kth iteration Used to reflect the difference between the output probability and the geological disaster probability.

[0024] Furthermore, the kth iteration probability regression index is defined as h k , based on the formula:

[0025]

[0026] in, It is used to reflect the sum of squares of the differences between the residuals of all samples and the probability of geological disasters. Arg in is used to find the target formula The minimum value in h is obtained by computer software XGBoost k .

[0027] Furthermore, the learning rate μ is set to 0.5, and the correlation analysis is performed on the k-th iteration probability regression index to generate the geological disaster probability value s. The formula is:

[0028]

[0029] Among them, the value of K is not less than 10, and the geological disaster probability occurrence value s is used to reflect the predicted probability of geological disasters, and its value range is [0,1].

[0030] Furthermore, when the geological disaster probability occurrence value s is 0, it means that a geological disaster will not occur, and when the geological disaster probability occurrence value s is 1, it means that a geological disaster will occur.

[0031] The present invention also provides a multi-source data fusion geological hazard identification system for executing a multi-source data fusion geological hazard identification method, comprising:

[0032] Initial data acquisition module, used to collect initial reservoir data;

[0033] Reservoir data collection module, used to collect reservoir data once a day and number the reservoir data according to the number of days;

[0034] Normalization module, used to perform correlation analysis on the deviation of the coordinate height of the highest point of the slope, the deviation of the slope inclination, the deviation of the slope depth, and the deviation of the groundwater level depth of the slope, and generate normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth. The normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth data of each day are integrated into classification groups.

[0035] The evaluation module is used to set the geological disaster probability y for all classification groups. i, conduct correlation analysis on geological disaster probability and generate initial probability;

[0036] The iteration module is used to establish an iterative model, perform K iterations on the initial probability, output the residual of the kth iteration, and define the kth iteration probability regression index;

[0037] The output module performs correlation analysis on the k-th iteration probability regression index to generate a geological disaster probability occurrence value, inputs the real-time collected reservoir data into the iteration model, and outputs the geological disaster probability occurrence value.

[0038] Compared with the prior art, the present invention has the following beneficial effects:

[0039] The present invention is based on;

[0040] The present invention collects the coordinate height, inclination, slope depth and groundwater level of the reservoir slope, and performs correlation analysis on the daily coordinate height, inclination, slope depth and groundwater level data in combination with the collected data 360 times. Through data probability iteration, an iterative probability regression index is generated, and the minimum value of the iterative probability regression index is obtained, thereby generating a geological disaster probability occurrence value used to reflect the probability of geological disasters in the reservoir, automatically outputting the geological disaster probability, and helping staff to better understand the geological disaster risks of the reservoir. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0042] Figure 2 Schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0043] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0044] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0045] Example:

[0046] See also Figure 1 , the present invention provides a technical solution:

[0047] The multi-source data fusion geological disaster identification method is used to analyze the probability of geological disasters on reservoir slopes. With the help of new-generation information technologies such as the Internet of Things, cloud computing, big data, and artificial intelligence, the geological disaster-related data is edge-calculated. Reservoir data is collected through the data collection gateway of the Internet of Things, and the data is correlated and processed to output a geological disaster risk assessment value. A database is established through data collection, and the data in the database is analyzed to generate a geological disaster probability value and evaluate the probability of geological disasters in the reservoir. The specific steps include:

[0048] Step 1: Collect initial reservoir data, including the initial coordinate height, initial inclination, initial depth, and initial groundwater level data of the slope. The initial coordinate height A0 of the slope is determined by GPS positioning, the initial inclination B0 of the slope is measured by an inclinometer, the initial depth C0 of the slope is measured by a deep displacement sensor, and the initial groundwater level D0 of the slope is measured by a water level sensor.

[0049] Step 2: Collect reservoir data once a day, analyze and process it with the initial reservoir data to obtain reservoir deviation data, and number the reservoir data according to the number of days. The reservoir deviation data includes the coordinate height deviation of the highest point of the slope, the slope inclination deviation, the slope depth deviation, and the slope groundwater level depth deviation. The collection days are 360 ​​days; the coordinate height deviation of the highest point of the slope on the i-th day is represented by A i The slope deviation on the i-th day is expressed as B i The slope depth deviation on the i-th day is expressed as C iThe groundwater level deviation of the slope on day i is expressed as D i Represents, where the subscript i is used as the index of day i.

[0050] Step 3: Perform correlation analysis on the deviation of the coordinate height of the highest point of the slope, the deviation of the slope inclination, the deviation of the slope depth, and the deviation of the groundwater level depth of the slope to generate normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth. The normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth data of each day are integrated into classification groups.

[0051] Coordinate height deviation of the highest point of the slope A i , Slope inclination deviation B i , slope depth deviation C i , slope groundwater level depth deviation D i Perform correlation analysis separately to generate normalized coordinate height A′ i , normalized slope B′ i , normalized slope depth C′ i , normalized groundwater level depth D′ i , based on the formula:

[0052]

[0053] Among them, max{A i} and min{A i} respectively represent the maximum and minimum values ​​of the coordinate height deviation of the highest point of all slopes, max{B i} and min{B i} are used to represent the total slope inclination deviation B i The maximum and minimum values ​​of max{C i} and min{C i} are used to represent the total slope depth deviation C i The maximum and minimum values ​​of max{D i} and min{D i} are used to represent the groundwater level deviation D of the entire slope. i The maximum and minimum values ​​of the normalized coordinate height A′ i , normalized slope B′ i , normalized slope depth C′ i , normalized groundwater level depth D′ i Both are used to ensure data consistency, the coordinate height deviation of the highest point of the slope is A i Used to reflect the difference between the coordinate height of the highest point on the i-th day and the initial coordinate height of the slope, and the slope inclination deviation b iIt is used to reflect the deviation value of the slope inclination on the i-th day from the initial slope inclination and the slope depth deviation C i It is used to reflect the deviation value of the slope depth on the i-th day from the initial slope depth and the slope groundwater level depth deviation d i It is used to reflect the deviation between the depth of the slope groundwater level on the i-th day and the initial slope groundwater level.

[0054] Step 4: For all classification groups, the geological disaster probability is set accordingly, and the disaster probability prediction value is set. The correlation analysis of the geological disaster probability is performed to generate the initial probability;

[0055] The normalized coordinate height, normalized slope, normalized slope depth, and normalized groundwater level depth classification group of day i are set as T i , for all classification groups T i Set the geological disaster probability y i The method of obtaining the probability of geological disasters is as follows: geological experts calculate the slope height deviation A according to the coordinate height deviation of the highest point of the slope. i , Slope inclination deviation B i , slope depth deviation C i , slope groundwater level depth deviation D i The numerical value sets the probability of the corresponding disaster occurring, and the average value of the probability of the corresponding disaster occurring is taken as the geological disaster probability y i , the probability of geological disaster y i Perform correlation analysis to generate the initial probability F0 based on the following formula:

[0056]

[0057] Perform K iterations, where the output probability of the k-1th iteration is F k-1 , set the residual of the kth iteration to The formula is:

[0058]

[0059] Among them, the residual of the kth iteration Used to reflect the difference between the output probability and the geological disaster probability.

[0060] Step 5: Establish an iterative model, perform K iterations on the initial probability, output the residual of the kth iteration, and define the kth iteration probability regression index;

[0061] Define the kth iteration probability regression index as h k , based on the formula:

[0062]

[0063] in, It is used to reflect the sum of squares of the differences between the residuals of all samples and the probability of geological disasters. Arg in is used to find the target formula The minimum value in h is obtained by computer software XGBoost k .

[0064] Step 6: Set the learning rate, perform correlation analysis on the kth iteration probability regression index, generate a geological disaster probability occurrence value, which is used to reflect the probability of geological disasters. Input the reservoir data collected in real time into the iterative model and output the geological disaster probability occurrence value.

[0065] Set the learning rate μ to 0.5, perform correlation analysis on the k-th iteration probability regression index, and generate the geological disaster probability occurrence value s based on the following formula:

[0066]

[0067] Among them, the value of K is not less than 10, and the geological disaster probability occurrence value s is used to reflect the predicted probability of geological disasters, and its value range is [0,1].

[0068] Input the latest acquired slope highest point coordinate height deviation, slope inclination deviation, slope depth deviation, slope groundwater level depth deviation, refer to the geology experts according to the slope highest point coordinate height deviation A i , Slope inclination deviation B i , slope depth deviation C i , slope groundwater level depth deviation D i The numerical values ​​are set for the probabilities of the corresponding disasters, and the probabilities of the four parameters causing the disasters are obtained. The probabilities of the four parameters causing the disasters are input into the iterative model, and it is iterated to output the geological disaster probability value s.

[0069] When the geological disaster probability value s is 0, it means that a geological disaster will not occur. When the geological disaster probability value s is 1, it means that a geological disaster will occur. The larger the value of s, the higher the probability of a geological disaster. The geological disaster probability value s represents the predicted probability of a geological disaster.

[0070] Reference Figure 2 The present invention also provides a geological disaster identification system based on multi-source data fusion, which is used to perform a geological disaster identification method based on multi-source data fusion, including:

[0071] Initial data acquisition module, used to collect initial reservoir data;

[0072] Reservoir data collection module, used to collect reservoir data once a day and number the reservoir data according to the number of days;

[0073] Normalization module, used to perform correlation analysis on the deviation of the coordinate height of the highest point of the slope, the deviation of the slope inclination, the deviation of the slope depth, and the deviation of the groundwater level depth of the slope, and generate normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth. The normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth data of each day are integrated into classification groups.

[0074] The evaluation module is used to set the geological disaster probability y for all classification groups. i , conduct correlation analysis on geological disaster probability and generate initial probability;

[0075] The iteration module is used to establish an iterative model, perform K iterations on the initial probability, output the residual of the kth iteration, and define the kth iteration probability regression index;

[0076] The output module performs correlation analysis on the k-th iteration probability regression index to generate a geological disaster probability occurrence value, inputs the real-time collected reservoir data into the iteration model, and outputs the geological disaster probability occurrence value.

[0077] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0078] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0079] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the objectives of this embodiment based on actual needs.

[0080] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A multi-source data fusion method for identifying geological hazards is used to analyze the probability of occurrence of geological hazards on reservoir slopes. The method is characterized by: The specific steps include: S1. Collecting initial reservoir data, including initial slope coordinate height, initial slope inclination, initial slope depth, and initial groundwater level depth data; S2. Reservoir data is collected once a day and analyzed with the initial reservoir data to obtain reservoir deviation data. The reservoir data is numbered according to the number of days. The reservoir deviation data includes the coordinate height deviation of the highest point of the slope, the slope inclination deviation, the slope depth deviation, and the slope groundwater level depth deviation. The collection days are 360 ​​days; S3. performing correlation analysis on the deviation of the coordinate height of the highest point of the slope, the deviation of the slope inclination, the deviation of the slope depth, and the deviation of the groundwater level depth of the slope, respectively, to generate normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth, and integrating the normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth data of each day into classification groups; S4. For all classification groups, the geological disaster probability is set respectively, and the disaster occurrence probability prediction value is set, and the correlation analysis of the geological disaster probability is performed to generate the initial probability; S5. Establish an iterative model, perform K iterations on the initial probability, output the residual of the k-th iteration, and define the k-th iteration probability regression index; S6. Setting a learning rate, performing a correlation analysis on the k-th iteration probability regression index, generating a geological disaster probability occurrence value, which is used to reflect the probability of a geological disaster, inputting the real-time collected reservoir data into the iteration model, and outputting the geological disaster probability occurrence value; The initial coordinate height A0 of the slope is determined by satellite GPS positioning, the initial inclination B0 of the slope is measured by an inclinometer, the initial depth C0 of the slope is measured by a deep displacement sensor, and the initial groundwater level D0 of the slope is measured by a water level sensor; The coordinate height deviation of the highest point of the slope on the i-th day is expressed as A i The slope deviation on the i-th day is expressed as B i The slope depth deviation on the i-th day is expressed as C i The groundwater level deviation of the slope on day i is expressed as D i Indicates, where subscript i is used as the index of day i; Coordinate height deviation of the highest point of the slope A i , Slope inclination deviation B i , slope depth deviation C i , slope groundwater level depth deviation D i Perform correlation analysis separately to generate normalized coordinate height A′ i , normalized slope B′ i , normalized slope depth C′ i , normalized groundwater level depth D′ i , based on the formula: Among them, max{A i } and min{A i } respectively represent the maximum and minimum values ​​of the coordinate height deviation of the highest point of all slopes, max{B i } and min{B i } are used to represent the total slope inclination deviation B i The maximum and minimum values ​​of max{C i } and min{C i } are used to represent the total slope depth deviation C i The maximum and minimum values ​​of max{D i } and min{D i } are used to represent the groundwater level deviation D of the entire slope. i The maximum and minimum values ​​of the normalized coordinate height A′ i , normalized slope B′ i , normalized slope depth C′ i , normalized groundwater level depth D′ i Both are used to ensure data consistency, the coordinate height deviation of the highest point of the slope is A i It is used to reflect the difference between the coordinate height of the highest point of the slope on the i-th day and the initial coordinate height of the slope, and the slope inclination deviation B i It is used to reflect the deviation value of the slope inclination on the i-th day from the initial slope inclination and the slope depth deviation C i It is used to reflect the deviation value between the slope depth on the i-th day and the initial slope depth, and the slope groundwater level depth deviation D i It is used to reflect the deviation between the depth of the slope groundwater level on the i-th day and the initial slope groundwater level; In S3, the normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth classification group of the i-th day are set as T i , for all classification groups T i Set the geological disaster probability y i , the probability of geological disaster y i Perform correlation analysis to generate the initial probability F0 based on the following formula: Perform K iterations, where the output probability of the k-1th iteration is F k-1 , set the residual of the kth iteration to r i (k) , based on the formula: r i (k) =y i -F k-1 Among them, the residual r of the kth iteration i (k) Used to reflect the difference between the output probability and the geological disaster probability; Define the kth iteration probability regression index as h k , based on the formula: in, It is used to reflect the sum of squares of the differences between the residuals of all samples and the probability of geological disasters. Argmin is used to find the target formula The minimum value in h is obtained by computer software XGBoost k ; Set the learning rate μ to 0.5, perform correlation analysis on the k-th iteration probability regression index, and generate the geological disaster probability occurrence value s based on the following formula: Among them, the value of K is not less than 10, and the geological disaster probability occurrence value s is used to reflect the predicted probability of geological disasters, and its value range is [0,1].

2. The geological hazard identification method based on multi-source data fusion according to claim 1, characterized in that: When the probability value s of a geological disaster is 0, it means that no geological disaster will occur. When the probability value s of a geological disaster is 1, it means that a geological disaster will occur.

3. A multi-source data fusion geological hazard identification system for executing the multi-source data fusion geological hazard identification method according to claim 1, characterized in that: include: Initial data acquisition module, used to collect initial reservoir data; Reservoir data collection module, used to collect reservoir data once a day and number the reservoir data according to the number of days; Normalization module, used to perform correlation analysis on the deviation of the coordinate height of the highest point of the slope, the deviation of the slope inclination, the deviation of the slope depth, and the deviation of the groundwater level depth of the slope, and generate normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth. The normalized coordinate height, normalized inclination, normalized slope depth, and normalized groundwater level depth data of each day are integrated into classification groups. The evaluation module is used to set the geological disaster probability y for all classification groups. i , conduct correlation analysis on geological disaster probability and generate initial probability; The iteration module is used to establish an iterative model, perform K iterations on the initial probability, output the residual of the kth iteration, and define the kth iteration probability regression index; The output module performs correlation analysis on the k-th iteration probability regression index to generate a geological disaster probability occurrence value, inputs the real-time collected reservoir data into the iteration model, and outputs the geological disaster probability occurrence value.

Images