Machine learning based deep geothermal reservoir parameter prediction system

Abstract

The application relates to the technical field of geological data processing, in particular to a deep geothermal reservoir parameter prediction system based on machine learning. The system comprises a data acquisition module for acquiring gravity data, magnetic field intensity and seismic wave data of an observation point; a predicted temperature acquisition module for acquiring a first predicted temperature, a second predicted temperature and external environment stability of the observation point; a geological interference analysis module for judging whether the observation point is an abnormal observation point and then acquiring geological interference of the observation point; a stratum uniformity analysis module for calculating stratum uniformity based on a travel time curve of the observation point; a fusion predicted temperature acquisition module for training a CNN model and acquiring stratum uniformity and geological interference of a target observation point; and a fusion predicted temperature of the target observation point is obtained according to the stratum uniformity, the geological interference, the external environment stability, a first preset value and the second predicted temperature. The application can improve the accuracy of deep geothermal reservoir temperature prediction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geological data processing, in particular to a deep geothermal reservoir parameter prediction system based on machine learning. BACKGROUND

[0002] Deep geothermal energy, as a clean and sustainable energy form, plays an important role in global energy transformation. China is rich in geothermal resources, and efficient exploration and development of geothermal energy is the key to the development of new replacement energy. The reservoir temperature is the core parameter for calculating the reserves of geothermal resources. Through temperature prediction, the total amount of thermal energy of geothermal fluid can be estimated to determine the development potential of geothermal fields, avoid equipment damage or insufficient production capacity caused by temperature anomalies, and improve development efficiency to optimize well site selection and mining strategy to maximize energy output. Therefore, in the whole life cycle of geothermal exploration and development, reservoir temperature prediction is an indispensable key link.

[0003] Traditional reservoir temperature prediction for un-drilled areas mainly uses electromagnetic inversion method, which mainly uses the characteristics of the change of rock conductivity (or resistivity) with temperature to estimate the reservoir temperature. However, this relationship is not simply linear, but is affected by various factors such as rock type, porosity, fluid composition, etc., and the underground electrical structure obtained by electromagnetic inversion may have multiple solutions, i.e. different geological models may produce similar electromagnetic responses. At the same time, due to external interference and the complexity of geological conditions, the complexity and uncertainty of electromagnetic response may exist, so the temperature prediction result obtained by the electromagnetic inversion method alone may have errors. SUMMARY

[0004] In order to solve the above technical problems, the purpose of the present application is to provide a deep geothermal reservoir parameter prediction system based on machine learning, and the technical solution adopted is as follows:

[0005] One embodiment of the present application provides a deep geothermal reservoir parameter prediction system based on machine learning, which comprises:

[0006] A data acquisition module for acquiring gravity data, magnetic field strength and seismic wave data of observation points;

[0007] A predicted temperature acquisition module for acquiring first and second predicted temperatures of observation points; a sensor is arranged at the observation point, and the stability of the external environment of the observation point is calculated according to the data collected by the sensor;

[0008] A geological interference analysis module for determining whether the observation point is an abnormal observation point according to the gravity data and magnetic field strength of the observation point. If it is a normal observation point, the geological interference of the observation point is a preset value, and if it is an abnormal observation point, the geological interference of the observation point is calculated according to the adjacent observation points of the observation point.

[0009] a stratum uniformity analysis module configured to obtain a travel time curve according to seismic wave data of the observation point; and calculate stratum uniformity based on changes of data points on the travel time curve;

[0010] a fusion predicted temperature acquisition module configured to train a CNN model based on the travel time curve, gravity data, magnetic field strength, stratum uniformity and geological interference of each observation point; input the travel time curve, gravity data and magnetic field strength of the target observation point into the CNN model to obtain stratum uniformity and geological interference of the target observation point; and obtain the fusion predicted temperature of the target observation point according to the stratum uniformity, geological interference, external environment stability, first predicted temperature and second predicted temperature.

[0011] Preferably, the first predicted temperature and the second predicted temperature of the observation point are obtained by:

[0012] The stratum temperature at the same depth of an observation point is obtained by a heat conduction model method and an electromagnetic inversion method, and is recorded as the first predicted temperature and the second predicted temperature, respectively.

[0013] Preferably, the external environment stability of the observation point is calculated according to the data collected by the sensor, and includes:

[0014] The absolute value of the difference between a temperature data and the slope of the adjacent temperature data in the temperature data collected by a sensor is recorded as the slope change difference of the temperature data; the reciprocal of the sum of the slope change difference and the first preset value is obtained and negatively correlated to obtain the mutation rate of the temperature data; the temperature data with the mutation rate greater than the mutation rate threshold is recorded as the mutation data; the proportion of the number of the mutation data in the temperature data corresponding to the sensor is recorded as the mutation data occurrence probability of the sensor; and the difference between the first preset value and the mean of the mutation data occurrence probabilities of all sensors is recorded as the external environment stability of the observation point.

[0015] Preferably, whether the observation point is an abnormal observation point is determined according to the gravity data and the magnetic field strength of the observation point, and includes:

[0016] The average value of the gravity data of an observation point after removing the maximum value and the minimum value is recorded as the gravity average value, the gravity data is arranged in ascending order, and the median and mode are obtained; the mean of the gravity average value, the median and the mode is recorded as the reference value; the reciprocal of the difference between a gravity data and the reference value is obtained by subtracting the first preset value to obtain the gravity data anomaly degree of the gravity data; the magnetic field strength anomaly degree of each magnetic field strength is obtained in the same way; if the gravity data anomaly degree of the gravity data or the magnetic field strength anomaly degree of the magnetic field strength of an observation point at any time is greater than or equal to the abnormal threshold, the observation point is an abnormal observation point.

[0017] Preferably, the geological interference of an observation point is calculated according to neighboring observation points of the observation point, including:

[0018] The distance between one observation point and other observation points is calculated, and a preset number of other observation points with the smallest distance are taken as neighboring observation points of the observation point; the difference between the mean value of the gravity data of each neighboring observation point and the mean value of the gravity data of the observation point is calculated respectively, and then the average value of the difference corresponding to each neighboring observation point is taken and the absolute value is taken to obtain the local difference of the gravity data of the observation point; the local difference of the magnetic field intensity of the observation point is obtained in the same way; the reciprocal of the local difference of the gravity data of the observation point and the reciprocal of the local difference of the magnetic field intensity are negatively correlated mapped by the first preset value respectively to obtain a first mapping result and a second mapping result; the sum of the first mapping result and the second mapping result is the local difference of the observation point; the ratio of the number of abnormal data of the observation point to the total number of data is calculated, and the abnormal proportion of the observation point is obtained by subtracting the ratio from the first preset value; the local difference and the abnormal proportion of the observation point are weighted and summed to obtain the geological interference of the observation point.

[0019] Preferably, the travel time curve includes a travel time curve of a direct wave and a travel time curve of a reflected wave.

[0020] Preferably, the stratum uniformity is calculated based on the change of the data points on the travel time curve, including:

[0021] The slope between every two adjacent data points on the travel time curve of the direct wave is obtained, and the slope sequence is formed in the order of the abscissa; a triangle is constructed by using three adjacent data points on the travel time curve of the reflected wave, and the curvature of the data point located in the center of the three adjacent data points is calculated by using the minimum circumscribed circle of the triangle; the average value of the absolute value of the difference between every two adjacent slopes in the slope sequence is calculated and added to the first preset value to obtain an addition result, and the reciprocal of the addition result is obtained to obtain the direct wave reflection characteristic; the reflected wave reflection characteristic is obtained by analyzing the curvature sequence in the same way; and the stratum uniformity of the observation point is obtained by averaging the direct wave reflection characteristic and the reflected wave reflection characteristic.

[0022] Preferably, the CNN model is trained, including:

[0023] The travel time curve, the gravity data and the magnetic field intensity of each observation point of one observation point are taken as a sample, and the stratum uniformity and the geological interference of the observation point are taken as the label of the sample, to obtain the sample corresponding to each observation point with the label, and the CNN model is trained.

[0024] Preferably, the fusion prediction temperature of the target observation point is obtained according to the stratum uniformity, the geological interference, the external environment stability, the first predicted temperature and the second predicted temperature, including:

[0025] According to the stratum uniformity, the geological interference, and the external environment stability of the target observation point, weights corresponding to the first predicted temperature and the second predicted temperature are obtained, and the first predicted temperature and the second predicted temperature are weighted and summed according to the weights corresponding to the first predicted temperature and the second predicted temperature to obtain the fusion predicted temperature of the target observation point.

[0026] Preferably, obtaining the weights corresponding to the first predicted temperature and the second predicted temperature comprises:

[0027] The stratum uniformity of the target observation point is taken as the credibility of the first predicted temperature of the target observation point, the external environment stability of the target observation point is multiplied by the difference between the first preset value and the geological interference of the target observation point to obtain the credibility of the second predicted temperature of the target observation point, the ratio of the credibility of the first predicted temperature to the sum of the credibility of the first predicted temperature and the credibility of the second predicted temperature is the weight corresponding to the first predicted temperature, and the ratio of the credibility of the second predicted temperature to the sum of the credibility of the first predicted temperature and the credibility of the second predicted temperature is the weight corresponding to the second predicted temperature.

[0028] The embodiment of the present application has at least the following beneficial effects: the present application predicts the temperature of the deep geothermal reservoir of the observation point by two methods of the heat conduction model and the electromagnetic inversion method, obtains the first predicted temperature and the second predicted temperature, then arranges sensors at the observation point, calculates the external environment stability of the observation point according to the data collected by the sensors, analyzes the gravity data and the magnetic field strength of the observation point to obtain the geological interference of the observation point, thereby obtaining two factors affecting the predicted temperature by the electromagnetic inversion method; further, the travel time curve is obtained according to the seismic wave data of the observation point; the stratum uniformity is calculated based on the change of the data points on the travel time curve to obtain the factor affecting the predicted temperature by the heat conduction model; then the CNN model is trained based on the travel time curve, the gravity data, the magnetic field strength, the stratum uniformity and the geological interference of each observation point, the complex calculation in the subsequent actual use process is reduced, the efficiency of the actual use is improved, and the stratum uniformity, the geological interference and the external environment stability of the target observation point are further obtained; finally, the first predicted temperature and the second predicted temperature predicted by the two methods are fused according to the stratum uniformity, the geological interference and the external environment stability to obtain the fusion predicted temperature of the target observation point, the deficiencies of the predicted temperatures by the two methods are made up, the influence of each factor on the prediction of the temperature of the deep geothermal reservoir is reduced, and the accuracy of the prediction of the temperature of the deep geothermal reservoir is improved. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, and the advantages thereof, below will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

[0030] Figure 1 A system block diagram of a deep geothermal reservoir parameter prediction system based on machine learning provided by an embodiment of the present application is provided.

[0031] Figure 2 A travel time curve schematic diagram of a deep geothermal reservoir parameter prediction system based on machine learning provided by an embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined purpose, the specific embodiments, structure, features and effects of a deep geothermal reservoir parameter prediction system based on machine learning according to the present application are described in detail below in combination with the drawings and preferred embodiments. Different "one embodiment" or "another embodiment" in the following description do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0034] The specific scheme of a deep geothermal reservoir parameter prediction system based on machine learning provided by the present application is described in detail below in combination with the drawings.

[0035] Embodiment:

[0036] The main application scenario of the present application is to predict the deep geothermal reservoir temperature during exploration and development of unexplored mining areas. For unexplored mining areas to be temperature predicted, since the heat conduction model assumes that the stratum is uniform, the more uniform the stratum, the greater the weight of the heat conduction model, and the electromagnetic inversion method is affected by exploration geology and external environment, i.e. the weight of the electromagnetic inversion method is high in the case of small external environmental influence and small geological interference. The geological data of the terrain to be explored can be obtained by using the gravity and magnetic method and the seismic wave velocity, the stratum uniformity of the terrain is analyzed, the external environment stability and the geological interference are analyzed to obtain the weight of the predicted temperature data of the two methods, and the final predicted temperature is obtained by weighted fusion.

[0037] Please refer to Figure 1It shows a system block diagram of a deep geothermal reservoir parameter prediction system based on machine learning provided by the embodiment of the application, and the system comprises the following modules:

[0038] A data acquisition module is configured to acquire gravity data, magnetic field intensity and seismic wave data of the observation point.

[0039] Firstly, the collected data is acquired. The geological data used in the application is mainly acquired through the gravity and magnetic method and the seismic wave velocity method. The gravity and magnetic method needs to use a gravimeter (such as an absolute gravity measuring instrument or a relative gravity measuring instrument) to observe the gravity value of each measuring point in the to-be-measured area, and use a magnetometer (such as a proton precession magnetometer or an optical pump magnetometer) to observe the magnetic field intensity, magnetic inclination, magnetic declination and other magnetic parameters of each measuring point in the to-be-measured area. The seismic wave velocity is acquired by the exploration equipment (such as a seismic detector) to obtain the seismic wave signal on the ground, and the mode is single-channel excitation, that is, receiving on one side after the seismic source excitation.

[0040] Specifically, in the observation time, the gravimeter and the magnetometer are used to acquire the gravity data and the magnetic field intensity of each observation point, and the seismic detector is used to acquire the seismic wave data, wherein the acquisition frequency is set by the implementer according to the actual situation.

[0041] A predicted temperature acquisition module is configured to acquire a first predicted temperature and a second predicted temperature of the observation point; a sensor is arranged at the observation point, and the external environment stability of the observation point is calculated according to the data collected by the sensor.

[0042] The predicted temperature is mainly acquired by the electromagnetic inversion method and the heat conduction model. The heat conduction model needs to acquire the ground temperature data through the meteorological station observation, remote sensing technology and other methods, acquire the geothermal gradient data through drilling, temperature measuring well and other methods, and acquire other related parameters such as the thermal conductivity and density of the rock through laboratory testing or geological investigation data. The electromagnetic inversion method needs to use electromagnetic exploration equipment (such as wide-area electromagnetic inversion method detection equipment) to arrange an electrode array on the ground, emit electromagnetic signals and receive the reflected signals to obtain the potential difference or current distribution data between different positions.

[0043] The heat conduction model calculates the formation temperature according to the collected data, and the mathematical model is as follows:

[0044] ;

[0045] Wherein, represents the formation temperature at a depth of , represents the ground temperature, represents the geothermal gradient, that is, the change rate of the formation temperature per unit depth, represents the depth.

[0046] The electromagnetic inversion method obtains the predicted temperature: the observed electromagnetic field data is converted into the resistivity distribution of the underground medium through inversion calculation, and the temperature distribution of the geothermal reservoir is inferred according to the known resistivity characteristics at different temperatures. Thus, the formation temperature at the same depth of an observation point can be obtained by the thermal conduction model method and the electromagnetic inversion method, which are respectively denoted as the first predicted temperature and the second predicted temperature. The thermal conduction model method and the electromagnetic inversion method are known technologies, and will not be described here.

[0047] The electromagnetic inversion method mainly uses the characteristics of the change of rock conductivity (or resistivity) with temperature to estimate the reservoir temperature, so it is easy to be covered or distorted by external electromagnetic interference, which leads to errors in the predicted temperature. At the same time, due to the geological structure (such as rock type, porosity, permeability and other factors), the complexity and uncertainty of the electromagnetic response may cause errors in the temperature estimation. Therefore, the stability of the external environment and the geological interference may affect the error of the predicted temperature by the electromagnetic inversion method.

[0048] Since the electromagnetic inversion method is used for geological exploration, the underground electromagnetic interference is often related to the topography and geology, mainly underground ore bodies, pipelines, geological structures, etc. Such interference is relatively weak and closely related to geological exploration, to some extent, reflecting the underground situation. Therefore, the main electronic interference that needs to be excluded by the electromagnetic inversion method comes from the power system, industrial equipment, and wireless communication system on the ground. Since the sensor is susceptible to electromagnetic interference, any number of sensors can be placed around each observation point to collect corresponding data and analyze whether the data has changed abruptly, thereby analyzing the stability of the environment. It should be noted that when arranging the sensors, the data collected by the sensors should be relatively stable when there is no obvious change in the external environment, that is, the probability of data mutation is low.

[0049] Preferably, the sensor provided in the present application is a temperature sensor that collects temperature data. The collection time is the length of the observation time, and the collection frequency is once every 1 second, which can be determined according to the actual situation. The temperature sensors are arranged in a circular array with the observation point as the center of the circle, and the preset length is the radius of the circle. The temperature sensors are uniformly distributed on the boundary of the circle, and the central angle between two adjacent temperature sensors and the observation point is a preset degree. The preset length and the preset degree are set by the implementer according to the actual situation. Preferably, the preset length and the preset degree in the embodiment of the present application are 1 meter and 60°, respectively.

[0050] Since the interference will cause the sensor data to mutate, the temperature data is subjected to mutation data judgment, specifically, the absolute value of the difference between a temperature data and the slopes of the adjacent temperature data in the temperature data collected by a sensor is obtained, denoted as the slope change difference of the temperature data; the reciprocal of the slope change difference plus the first preset value is obtained and subjected to negative correlation mapping to obtain the mutation rate of the temperature data. The specific calculation model is:

[0051] ;

[0052] Wherein, A represents the mutation rate of the temperature data of a temperature sensor at a moment, represents the slope of the temperature data at the moment and the previous temperature data, represents the slope of the temperature data at the moment and the next temperature data, 1 represents the first preset value, represents the slope change difference, the greater the value, the greater the value of the negative correlation mapping of the sum of the slope change difference and the first preset value by the first preset value, indicating that the mutation of the temperature data at the moment is more obvious, that is, the mutation rate A is greater.

[0053] The temperature data collected by all sensors of the observation point at each moment is calculated for the mutation rate, according to the experimental statistics, the mutation rate of most data is near 0.3, therefore, the mutation rate threshold is set to 0.7, the temperature data with a mutation rate greater than the mutation rate threshold is recorded as mutation data, further, the external environment stability of the observation point is obtained according to the mutation of the temperature data around the observation point, specifically: the proportion of the number of mutation data in the temperature data corresponding to a sensor is obtained, denoted as the mutation data occurrence probability of the sensor, the difference between the first preset value and the average of the mutation data occurrence probabilities of all sensors is obtained, denoted as the external environment stability of the observation point. The specific calculation model is:

[0054] ;

[0055] Wherein, a represents the external environment stability of an observation point, N represents the number of sensors (temperature sensors) placed around the observation point, represents the number of mutation data in the temperature data collected by the i th sensor, M represents the number of temperature data collected by each sensor, The probability of the occurrence of mutation data in the data collected by the i th sensor, that is, the mutation data occurrence probability corresponding to the i th sensor, if the probability of the occurrence of mutation data in the temperature data collected by the sensors placed around an observation point is small, it indicates that the electromagnetic interference around the observation point is small or not obvious, then the external stability of the observation point is large.

[0056] The geological interference analysis module is configured to determine whether the observation point is an abnormal observation point according to the gravity data and the magnetic field intensity of the observation point. If the observation point is a normal observation point, the geological interference of the observation point is a preset value. If the observation point is an abnormal observation point, the geological interference of the observation point is calculated according to the neighboring observation points of the observation point.

[0057] The gravity data obtained by the gravity and magnetic method and the data obtained by the electromagnetic inversion method are different. The gravity data mainly reflects the density distribution of underground matter, and the distribution of gravity can be used to determine whether there is a high-density or low-density geological body, such as a mineral body, a rock layer, a fault, and the like. The magnetic force is mainly affected by the type of rock and the metal ore body. Therefore, the geological interference can be obtained by combining the differences between the data.

[0058] Because the conditions affecting the gravity data and the magnetic force data are slightly different, if the gravity data of a certain observation point is abnormal and the magnetic field intensity is not necessarily abnormal, it may be caused by the terrain. If both types of data are abnormal, and the anomalies around them are consistent, it may be caused by geological characteristics. If there is an anomaly that is inconsistent with the surrounding anomalies, it may be caused by an influencing factor. That is, if the data are consistent and consistent with the surrounding anomalies, the geological interference is small. The degree of abnormality of each observation point data in the type of data and the consistency of the abnormal position can be obtained to obtain the geological interference.

[0059] Further, the average value of the gravity data of an observation point after removing the maximum value and the minimum value is obtained, which is denoted as gravity mean value. The gravity data is arranged in ascending order, and the median and mode are obtained. The average value of the gravity mean value, the median and the mode is calculated, which is denoted as reference value. The reciprocal of the difference between the first preset value and the gravity data and the reference value is obtained to obtain the gravity data abnormality degree of the gravity data. Similarly, the magnetic field intensity abnormality degree of each magnetic field intensity is obtained. According to experimental statistics, the abnormal threshold is 0.6. If the gravity data abnormality degree of the gravity data of an observation point at any time or the magnetic field intensity abnormality degree of the magnetic field intensity is greater than or equal to the abnormal threshold, the observation point is an abnormal observation point.

[0060] Further, the abnormal observation point is analyzed, the geological interference of the observation point is calculated according to the adjacent observation points of the observation point, specifically, the distance between an observation point and other observation points is calculated, and a preset number of other observation points with the smallest distance are taken as the adjacent observation points of the observation point; the difference between the mean value of the gravity data of each adjacent observation point and the mean value of the gravity data of the observation point is calculated, then the average value of the difference corresponding to each adjacent observation point is calculated and the absolute value is taken, to obtain the local difference of the gravity data of the observation point; the local difference of the magnetic field intensity of the observation point is obtained in the same way; the reciprocal of the local difference of the gravity data of the observation point and the reciprocal of the local difference of the magnetic field intensity are respectively negatively correlated mapped by using the first preset value, to obtain the first mapping result and the second mapping result; the average of the sum of the first mapping result and the second mapping result is the local difference of the observation point; the ratio of the number of the data of the observation point with abnormality to the total number of data is subtracted from the first preset value, to obtain the abnormality proportion of the observation point; the local difference and the abnormality proportion of the observation point are weighted and summed, to obtain the geological interference of the observation point.

[0061] Specifically, the calculation model is:

[0062] ;

[0063] Wherein, γ represents the geological interference of an observation point (an abnormal observation point), C represents the number of adjacent observation points of the observation point, in order to prevent the geological change probability from being too far, 3 is set in the application; represents the mean value of the gravity data of the i-th adjacent observation point, represents the mean value of the gravity data of the observation point, represents the mean value of the magnetic field intensity of the i-th adjacent observation point, represents the mean value of the magnetic field intensity of the i-th adjacent observation point, h represents the number of the data of the observation point with abnormality, if only one of the gravity data abnormality degree of the gravity data or the magnetic field intensity abnormality degree of the magnetic field intensity is greater than or equal to the abnormal threshold, h takes 1, if both are greater than or equal to the abnormal threshold, h takes 2, H represents the number of data, two kinds of gravity data and magnetic field intensity. represents the weight, because whether the data abnormality performance is consistent is more important, therefore , .

[0064] is the local difference of the gravity data, represents the abnormality degree of the gravity data of the observation point compared with the surrounding observation points, the greater the difference is, the more abnormal the surrounding observation points are, and the more possible the geological interference is; For the local difference of magnetic field intensity, the magnetic data of the selected observation point is more abnormal than that of the surrounding observation points, the greater the difference is, the more abnormal the surrounding observation points are, and the greater the geological disturbance is likely to be. The first mapping result and the second mapping result are and The first mapping result and the second mapping result are respectively.

[0065] For normal observation points, the geological disturbance cannot be set to 0, which is due to the limitations of experimental conditions, instrument accuracy and environment and the inevitability of geological disturbance. The error cannot be completely eliminated, so according to the experimental calculation statistics, for the observation points without anomalies, the geological disturbance should be set to the error value 0.1.

[0066] The stratum uniformity analysis module is configured to obtain a travel time curve based on the seismic wave data of the observation point, and calculate the stratum uniformity based on the change of the data points on the travel time curve.

[0067] If the stratum is uniform, the propagation speed of the seismic wave will not change, and no medium boundary or interface will be encountered in the propagation process, so the propagation path of the seismic wave is a straight line. At the same time, the seismic wave contains P wave and S wave, and the propagation speed of P wave is usually faster than that of S wave. The travel time refers to the propagation time of the seismic wave from the source to the observation point. Based on the above characteristics, the travel time curve of the seismic wave mainly observes and records the direct wave and the reflected wave. The direct wave is the seismic wave that propagates directly from the excitation point to the receiving point without encountering the reflecting surface, so under the premise of stratum uniformity, it will appear as a straight line passing through the shot point, and the slope is the reciprocal of the speed. The reflected wave is assumed to have a horizontal reflecting interface underground, and there is a difference in wave impedance above and below the interface. The stratum above the interface is homogeneous and isotropic. The travel time curve will exhibit nonlinear characteristics of curves, and will exhibit smooth curves when the stratum is uniform. Thus, the travel time curve when the stratum is uniform can be obtained, and the travel time curve includes the travel time curve of the direct wave and the travel time curve of the reflected wave, as shown in Figure 2 .

[0068] Further, the stratum uniformity is calculated based on the change of the data points on the travel time curve. Specifically, the slope between every two adjacent data points on the travel time curve of the direct wave is obtained, and the slope sequence is formed in the order of the abscissa. A triangle is constructed using the three adjacent data points on the travel time curve of the reflected wave, and the curvature of the data point located at the center of the three adjacent data points is calculated using the minimum circumscribed circle of the triangle. The curvature sequence is formed in the order of the abscissa. The average of the absolute values of the difference between every two adjacent slopes in the slope sequence is calculated and added to the first preset value to obtain an addition result. The reciprocal of the addition result is obtained to obtain the reflection characteristics of the direct wave. Similarly, the curvature sequence is analyzed in the same way to obtain the reflection characteristics of the reflected wave. The reflection characteristics of the direct wave and the reflection characteristics of the reflected wave are averaged to obtain the stratum uniformity of the observation point.

[0069] The specific calculation model is:

[0070] ;

[0071] wherein β represents the formation uniformity of the observation point, n represents the number of the slope in the slope sequence corresponding to the travel time curve of the direct wave, m represents the number of the curvature in the curvature sequence corresponding to the travel time curve of the reflected wave, represents the i+1th slope in the slope sequence, represents the ith slope in the slope sequence, represents the slope difference mean value, if the slope difference mean value is closer to 0, it indicates that the seismic wave propagation is less affected, and the formation uniformity is stronger, represents the direct wave reflection characteristic, and the larger the value is, the better the formation uniformity is. represents the j+1th curvature in the curvature sequence, represents the jth curvature in the curvature sequence, represents the curvature difference mean value, if the curvature difference mean value is smaller, it indicates that the curve is smoother, the seismic wave propagation is less affected, and the formation uniformity is stronger, represents the reflected wave reflection characteristic.

[0072] The fusion predicted temperature acquisition module is configured to train a CNN model based on the travel time curve, the gravity data, the magnetic field intensity, the formation uniformity and the geological interference of each observation point; input the travel time curve, the gravity data and the magnetic field intensity of the target observation point into the CNN model to obtain the formation uniformity and the geological interference of the target observation point; and obtain the fusion predicted temperature of the target observation point according to the formation uniformity, the geological interference, the external environment stability, the first predicted temperature and the second predicted temperature.

[0073] In the above process, the formation uniformity and geological interference of the observation point are obtained. Subsequently, in order to avoid complex calculation during use, a machine learning model is used to learn the relationship between the original data of the observation point and the formation uniformity and geological interference of the observation point. Subsequently, only the original data needs to be input to obtain the formation uniformity and geological interference of the observation point. Preferably, the machine learning model selected by the present application is a CNN model, and the CNN model is trained based on the travel time curve, gravity data, magnetic field strength, formation uniformity and geological interference of each observation point. Specifically, the travel time curve, gravity data and magnetic field strength of each observation point of an observation point are taken as a sample, and the formation uniformity and geological interference of the observation point are taken as the label of the sample. The sample corresponding to each observation point with the label is obtained, the CNN model is trained, and the training of the CNN model is completed. It should be noted that when the number of samples is insufficient, the travel time curve, gravity data, magnetic field strength, formation uniformity and geological interference of other historical detection points can be obtained based on the past detection data to construct samples for training.

[0074] Finally, the observation point to be analyzed is denoted as a target observation point, the external environment stability of the target observation point is obtained, and the travel time curve, gravity data and magnetic field strength of the target observation point are input into the trained CNN model to obtain the formation uniformity and geological interference of the target observation point.

[0075] Finally, the fusion prediction temperature of the target observation point is obtained according to the formation uniformity, geological interference, external environment stability, first prediction temperature and second prediction temperature. Specifically, the weight corresponding to the first prediction temperature and the weight corresponding to the second prediction temperature are obtained according to the formation uniformity, geological interference and external environment stability of the target observation point, respectively. The first prediction temperature and the second prediction temperature are weighted and summed according to the weight corresponding to the first prediction temperature and the weight corresponding to the second prediction temperature to obtain the fusion prediction temperature of the target observation point.

[0076] The specific calculation model of the fusion prediction temperature of the target observation point is: wherein, represents the fusion prediction temperature of the target observation point, and respectively represent the weight corresponding to the first prediction temperature and the weight corresponding to the second prediction temperature, and respectively represent the first prediction temperature and the second prediction temperature.

[0077] The weight corresponding to the first predicted temperature and the weight corresponding to the second predicted temperature are specifically: taking the formation uniformity of the target observation point as the reliability of the first predicted temperature of the target observation point; multiplying the external environment stability of the target observation point by the difference between the first preset value and the geological interference of the target observation point to obtain the reliability of the second predicted temperature of the target observation point; the ratio of the reliability of the first predicted temperature to the sum of the reliability of the first predicted temperature and the reliability of the second predicted temperature is the weight corresponding to the first predicted temperature; and the ratio of the reliability of the second predicted temperature to the sum of the reliability of the first predicted temperature and the reliability of the second predicted temperature is the weight corresponding to the second predicted temperature.

[0078] Since the influencing factor of the heat conduction model is mainly the formation uniformity, and the influencing factor of the electromagnetic inversion method is mainly the external environment stability and the geological interference, the reliabilities of the two predicted temperatures of the target observation point can be obtained based on the above properties, and the calculation model is:

[0079] The reliability of the first predicted temperature, that is, the reliability of selecting the heat conduction model to predict the temperature of the target observation point, The reliability of the second predicted temperature, that is, the reliability of selecting the electromagnetic inversion method to predict the temperature of the target observation point, The external environment stability of the target observation point, The geological interference of the target observation point, The formation uniformity of the target observation point. The weight corresponding to the first predicted temperature is , and the weight corresponding to the second predicted temperature is .

[0080] In summary, by analyzing the formation uniformity, the geological interference, and the external environment stability of the observation point, the weights of the temperatures predicted by the two methods are obtained, which can improve the accuracy of the predicted temperature. At the same time, for the calculation of the formation uniformity and the geological interference which are relatively complex, in order to improve the boundary line in the use process, the model is trained, and the formation uniformity and the geological interference of the observation point are obtained by using the machine learning model, thereby improving the efficiency in the actual use process.

[0081] It should be noted that the above-mentioned order of the embodiments of the application is only for description, and does not represent the advantages and disadvantages of the embodiments. The above describes a specific embodiment of the present application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0082] ​​​The various embodiments in the specification are described in progressive manner, and the same or similar parts between the various embodiments can be mutually referred to, and each embodiment focuses on the difference from other embodiments.

[0083] The above description is merely preferred embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A deep geothermal reservoir parameter prediction system based on machine learning, characterized in that: The system includes: Data acquisition module, used to collect gravity data, magnetic field intensity and seismic wave data at the observation point; A predicted temperature acquisition module is used to obtain a first predicted temperature and a second predicted temperature of an observation point; a sensor is arranged at the observation point, and the stability of the external environment of the observation point is calculated based on the data collected by the sensor; The geological interference analysis module is used to determine whether the observation point is an abnormal observation point based on the gravity data and magnetic field strength of the observation point. If it is a normal observation point, the geological interference of the observation point is the preset value. If it is an abnormal observation point, the geological interference of the observation point is calculated based on the neighboring observation points of the observation point, including: Calculate the distance between an observation point and other observation points, and take a preset number of other observation points with the smallest distance as the neighboring observation points of the observation point; calculate the difference between the mean value of the gravity data of each neighboring observation point and the mean value of the gravity data of the observation point, and then average the differences corresponding to each neighboring observation point and take the absolute value to obtain the local difference of the gravity data of the observation point; similarly obtain the local difference of the magnetic field intensity of the observation point; use the first preset value to perform negative correlation mapping on the inverse of the local difference of the gravity data of the observation point and the inverse of the local difference of the magnetic field intensity to obtain the first mapping result and the second mapping result; the sum of the first mapping result and the second mapping result is the local difference of the observation point; calculate the ratio of the number of types of abnormal data to the total number of data types at the observation point, and use the first preset value to subtract the ratio to obtain the abnormal proportion of the observation point; perform weighted summation of the local difference and the abnormal proportion of the observation point to obtain the geological interference of the observation point; The formation uniformity analysis module is used to obtain travel time curves based on seismic wave data at the observation point and calculate the formation uniformity based on the changes in data points on the travel time curve, including: Obtain the slope between every two adjacent data points on the travel time curve of the direct wave, and form a slope sequence in the order of the horizontal coordinates; construct a triangle using three adjacent data points on the travel time curve of the reflected wave, calculate the curvature of the data point located at the center of the three adjacent data points using the minimum circumscribed circle of the triangle, and form a curvature sequence with the curvature corresponding to each data point in the order of the horizontal coordinates; calculate the average of the absolute values ​​of the differences between every two adjacent slopes in the slope sequence and add the average to a first preset value to obtain an addition result, invert the addition result to obtain the direct wave reflection characteristic, and similarly analyze the curvature sequence according to the same method to obtain the reflected wave reflection characteristic; average the direct wave reflection characteristic and the reflected wave reflection characteristic to obtain the formation uniformity of the observation point; The fused predicted temperature acquisition module is used to train the CNN model based on the travel time curves, gravity data, magnetic field strength, stratigraphic uniformity and geological interference of each observation point; the travel time curves, gravity data and magnetic field strength of the target observation point are input into the CNN model to obtain the stratigraphic uniformity and geological interference of the target observation point; and the fused predicted temperature of the target observation point is obtained based on the stratigraphic uniformity, geological interference, external environmental stability, the first predicted temperature and the second predicted temperature.

2. A deep geothermal reservoir parameter prediction system based on machine learning according to claim 1, characterized in that: The obtaining of the first predicted temperature and the second predicted temperature of the observation point includes: The formation temperature at the same depth of an observation point is obtained by the heat conduction model method and the electromagnetic inversion method, and is recorded as the first predicted temperature and the second predicted temperature respectively.

3. The deep geothermal reservoir parameter prediction system based on machine learning according to claim 1, characterized in that: The calculation of the external environment stability of the observation point based on the data collected by the sensor includes: The absolute value of the difference between the slope of a temperature data collected by a sensor and the slope of the adjacent temperature data before and after it is obtained, and recorded as the slope change difference of the temperature data; the inverse of the slope change difference plus a first preset value is obtained and negative correlation mapping is performed to obtain the mutation rate of the temperature data; the temperature data with a mutation rate greater than the mutation rate threshold is recorded as mutation data; the proportion of mutation data in the temperature data corresponding to a sensor is obtained, and recorded as the probability of occurrence of mutation data of the sensor; the difference between the first preset value and the mean of the probability of occurrence of mutation data of all sensors is calculated, and recorded as the external environment stability of the observation point.

4. The deep geothermal reservoir parameter prediction system based on machine learning according to claim 1, characterized in that: The step of determining whether an observation point is an abnormal observation point based on the gravity data and magnetic field strength of the observation point includes: The method comprises the following steps: obtaining the average value of the gravity data of an observation point after removing the maximum and minimum values, which is recorded as the gravity mean value; arranging the gravity data in ascending order, and obtaining the median and mode therein; calculating the average of the gravity mean value, the median and the mode, and recording it as the reference value; subtracting the inverse of the difference between a gravity data and the reference value from a first preset value to obtain the degree of anomaly of the gravity data; similarly obtaining the degree of anomaly of each magnetic field strength; and if the degree of anomaly of the gravity data or the degree of anomaly of the magnetic field strength of an observation point at any moment is greater than or equal to the anomaly threshold value, the observation point is an abnormal observation point.

5. The deep geothermal reservoir parameter prediction system based on machine learning according to claim 1, characterized in that: The travel time curve includes a travel time curve of a direct wave and a travel time curve of a reflected wave.

6. The deep geothermal reservoir parameter prediction system based on machine learning according to claim 1, characterized in that: The training of the CNN model includes: The travel time curves, gravity data and magnetic field intensity of each observation point at an observation point are taken as a sample, and the stratigraphic uniformity and geological interference of the observation point are used as the label of the sample. The labeled samples corresponding to each observation point are obtained to train the CNN model.

7. The deep geothermal reservoir parameter prediction system based on machine learning according to claim 1, characterized in that: The method of obtaining the fused predicted temperature of the target observation point based on the formation uniformity, geological interference, external environment stability, the first predicted temperature and the second predicted temperature includes: According to the stratigraphic uniformity, geological interference, and external environmental stability of the target observation point, the weight corresponding to the first predicted temperature and the weight corresponding to the second predicted temperature are respectively obtained. According to the weight corresponding to the first predicted temperature and the weight corresponding to the second predicted temperature, the first predicted temperature and the second predicted temperature are weightedly summed to obtain the fused predicted temperature of the target observation point.

8. The deep geothermal reservoir parameter prediction system based on machine learning according to claim 7, characterized in that: The obtaining of the weight corresponding to the first predicted temperature and the weight corresponding to the second predicted temperature includes: The stratigraphic uniformity of the target observation point is used as the credibility of the first predicted temperature of the target observation point; the external environmental stability of the target observation point is multiplied by the difference between the first preset value and the geological interference of the target observation point to obtain the credibility of the second predicted temperature of the target observation point; the ratio of the credibility of the first predicted temperature to the sum of the credibility of the first predicted temperature and the credibility of the second predicted temperature is the weight corresponding to the first predicted temperature; the ratio of the credibility of the second predicted temperature to the sum of the credibility of the first predicted temperature and the credibility of the second predicted temperature is the weight corresponding to the second predicted temperature.

Images