Regional Geochemical Exploration Extension Method and Device

Abstract

The present invention provides a method and device for regional geochemical exploration extension. The exploration extension method includes: providing a first geochemical database and a second geochemical database, where the element types and contents in the first element group and the second element group in the first geochemical database are known quantities, and the content in the second element group in the second geochemical database is an unknown quantity; constructing a deep neural network with a fully connected layer as the main architecture and configuring the parameter information of the deep neural network; using the first geochemical database to train the deep neural network to generate a prediction model; based on the prediction model, using the first element group in the second geochemical database as the input quantity to determine the element content in the second element group in the second geochemical database, so as to generate a geochemical anomaly map at the second sampling density. By adopting the above technical solution, the present invention can quickly and effectively estimate the content value of the second element group.

Description

Technical Field

[0001] The present invention belongs to the technical field of mineral resource exploration, and particularly relates to a method and device for extending regional geochemical exploration. Background Art

[0002] Eleven mineral species such as iron, aluminum, potassium salt, boron, beryllium, lithium, manganese, niobium, titanium, uranium, and zircon are strategic scarce minerals in China. Increasing their resource reserves through exploration is the key to solving the supply-demand contradiction. Among them, regional geochemical exploration is one of the most effective means. By measuring the contents of 11 elements or oxides such as Fe2O3, Al2O3 / SiO2, K2O, B, Be, Li, Mn, Nb, Ti, U, and Zr, direct prospecting indicators for the corresponding mineral species can be obtained.

[0003] However, in important metallogenic belts or small-scale regional geochemical anomalies, in the 1:50,000 geochemical exploration carried out with the goal of geological prospecting, most of the work only analyzes and tests 16 mandatory elements required by the "Geochemical Prospecting Specification (1:50,000)", namely Au, Ag, As, Bi, Cd, Cr, Co, Cu, Hg, Mo, Ni, Pb, Sb, Sn, W, and Zn. The above 11 elements or oxides are less involved, which brings great challenges to the exploration of the above strategic scarce minerals.

[0004] Although data can be obtained by re-analyzing and testing the duplicate samples, this will consume a large amount of manpower, material resources, financial resources, and time. Therefore, for the above strategic scarce minerals, there is an urgent need for the emergence of an efficient and low-cost method for extending 1:50,000 regional geochemical exploration.

[0005] At present, the research on estimating element contents using artificial intelligence methods is not rare and has achieved certain actual effects, but these methods also have obvious defects. Although various interpolation methods (such as inverse distance weighting, Kriging, spline function, trend surface, etc.) can estimate the contents of target elements or oxides at unknown positions by interpolating internal sparse data to achieve data encryption, they basically cannot generate new abnormal information with practical value; the disadvantages of various machine learning methods (such as linear regression, random forest, and shallow neural network, etc.) are that they cannot stimulate the advantages of big data in exploration geochemistry and cannot achieve more accurate estimation of the contents of target elements or oxides. In summary, it is often difficult to systematically and comprehensively discover the complex and hidden mapping relationships between massive element or oxide data using the existing methods above, which is insufficient for accurately predicting the content values of target elements or oxides.

[0006] Therefore, how to more quickly and effectively estimate the content values of target elements or oxides has become a difficult point in geochemical exploration. Summary of the Invention

[0007] In view of this, embodiments of the present invention provide a method and device for regional geochemical exploration extension, which can quickly and effectively estimate the content values of the second element group.

[0008] Embodiments of the present invention provide a method for regional geochemical exploration extension, including:

[0009] Providing a first geochemical database obtained based on a first sampling density and a second geochemical database obtained based on a second sampling density, wherein the element types and contents in the first element group and the second element group in the first geochemical database are known quantities, and the element contents in the second element group in the second geochemical database are unknown quantities;

[0010] Constructing a deep neural network with a fully connected layer as the main architecture and configuring the parameter information of the deep neural network;

[0011] Using the first geochemical database to train the deep neural network to generate a prediction model;

[0012] Based on the prediction model, using the first element group in the second geochemical database as the input quantity to determine the element contents in the second element group in the second geochemical database, so as to generate a geochemical anomaly map at the second sampling density.

[0013] Optionally, the constructing a deep neural network with a fully connected layer as the main architecture and configuring the parameter information of the deep neural network includes:

[0014] Determining a network architecture with the fully connected layer as the main body, including: determining the input layer of the network architecture, the input layer is composed of multiple first neurons, and the number of the first neurons is equal to the number of elements in the first element group; determining the output layer of the network architecture, the output layer is composed of multiple second neurons, and the number of the second neurons is equal to the number of elements in the second element group; determining multiple hidden layers for connecting the input layer and the output layer, and each hidden layer is interconnected;

[0015] Configuring the data normalization processing method for each hidden layer of the deep neural network, where the normalization processing formula is:

[0016]

[0017] Where represents the value after batch normalization, γ and β respectively represent the values for scaling and shifting the normalized data, x i represents the i-th input value in the hidden layer, x meanrepresents the average value of the input values in the hidden layer, σ represents the standard deviation of the input values in the hidden layer, and ξ is a very small constant value;

[0018] Define the activation functions in each hidden layer of the deep neural network, where the activation function is:

[0019] PReLU(x) = max(0, x) + ηmin(0, x)

[0020] where η represents the learning parameter;

[0021] and determine the number of neurons randomly deactivated in the hidden layer during each training process of the deep neural network.

[0022] Optionally, the number of the third neurons in each hidden layer is greater than the number of the first neurons;

[0023] During each training process of the deep neural network, the number of neurons randomly deactivated in the hidden layer is 20% of the total number of all neurons.

[0024] Optionally, training the deep neural network using the first geochemical database to generate a prediction model includes:

[0025] Divide the first element group and the second element group in the first geochemical database into a training data set and a validation data set according to a preset ratio;

[0026] Input the training data set into the deep neural network in sequence and perform forward propagation to determine the cost function of the prediction result of the deep neural network, where the cost function is:

[0027]

[0028] where m is the number of training samples; y gt i is the true value of the i-th training sample; y pre i is the predicted value of the i-th training sample; L(y gt i , y pre i ) is the loss function of training sample i; δ is a threshold; λ is a penalty coefficient; R(w) is a penalty term, where w is the weight of the network;

[0029]

[0030] where n is the number of network weights, w i is the i-th weight;

[0031] Backpropagation is performed using the gradient descent method to update the weight parameters in the cost function, obtaining an initial prediction model;

[0032] All the validation data sets are input into the initial prediction model and propagated forward, and one or more of the mean square error and absolute error of the prediction results are determined as the validation evaluation metrics;

[0033] Repeat the iterative training and validation steps until the cost function value and the validation evaluation metrics meet the requirements.

[0034] Optionally, in the process of determining the cost function of the prediction results of the deep neural network, it further includes:

[0035] Determine an optimization function adapted to the deep neural network and optimize the weight parameters in the cost function during training, where the optimization function includes:

[0036]

[0037] Where w is the weight parameter, := means that the value on the left is updated or replaced by the value on the right, α is the learning rate, β1 is the first momentum, β2 is the second momentum, dw is the derivative of the cost function with respect to the weight parameter, v dw is the momentum exponentially weighted average of dw, s dw is the momentum exponentially weighted average of dw 2 The momentum exponentially weighted average, ε is a constant value, and t represents the number of iterations;

[0038] And initialize the weight parameters.

[0039] Optionally, before dividing the first element group and the second element group in the first geochemical database into training data sets and validation data sets according to a preset ratio, it further includes:

[0040] Preprocess the first element group and the second element group in the first geochemical database, and the first element group in the second geochemical database. The preprocessing includes: replacing the missing elements in the first element group and the second element group with any value between zero and the element detection limit, where the detection limit is the lowest value at which the corresponding element is detected; and performing standardization processing on the data in the first element group and the second element group after missing value processing, and the standardization parameters in the standardization processing of the first geochemical database and the second geochemical database are the same.

[0041] Optionally, the first geochemical database and the second geochemical database further include geographical landscape nominal scale variables, and the geographical landscape nominal scale variables are used to characterize different types of geographical landscapes;

[0042] Using the first geochemical database to train the deep neural network to generate a prediction model, the training process further includes: using the nominal scale variables of the geographical landscape as a data set to train the deep neural network.

[0043] Optionally, based on the prediction model, using the first element group in the second geochemical database as an input quantity to determine the element contents in the second element group in the second geochemical database, so as to generate a geochemical anomaly map at the second sampling density, including:

[0044] Using the first element group in the second geochemical database as an input quantity and inputting it into the prediction model to generate geochemical standard values corresponding to each element in the second element group and restoring them to the original values of each element in the second element group;

[0045] Performing interpolation processing on the original values of each element in the second element group to generate a geochemical map corresponding to the second element group;

[0046] Using the mean value of the original values of each element in the second element group and the standard deviation of the original values of each element as the anomaly lower limit to process the geochemical map to generate a geochemical anomaly map.

[0047] Optionally, the regional geochemical exploration extension method further includes:

[0048] Based on the geochemical anomaly map and multi-source information data, determining the prospecting target area for the target ore species.

[0049] Correspondingly, an embodiment of the present invention further provides a regional geochemical exploration extension device, including:

[0050] A geochemical database providing unit, adapted to provide a first geochemical database obtained based on a first sampling density and a second geochemical database obtained based on a second sampling density, wherein the element types and contents in the first element group and the second element group in the first geochemical database are known quantities, and the element contents in the second element group in the second geochemical database are unknown quantities;

[0051] A neural network construction unit, adapted to construct a deep neural network with a fully connected layer as the main architecture and configure the parameter information of the deep neural network;

[0052] A processing unit, adapted to use the first geochemical database to train the deep neural network to generate a prediction model; and based on the prediction model, using the first element group in the second geochemical database as an input quantity, determine the element contents in the second element group in the second geochemical database to generate a geochemical anomaly map at the second sampling density.

[0053] Compared with the prior art, the technical solution of the embodiment of the present invention has the following advantages:

[0054] By using the regional geochemical exploration extension method provided by the embodiment of the present invention, the element types and contents in the first element group in the first geochemical database and the second element group are known quantities, indicating that the first geochemical database is reliable. Thus, using the first geochemical database to train the deep neural network enables the prediction model to better learn and more comprehensively and objectively reflect the relationship between the first element group and the second element group in the first geochemical database. In this way, based on the prediction model, the element contents in the second element group in the second geochemical database can be quickly and effectively identified, and then a geochemical anomaly map for guiding prospecting exploration work can be generated. In other words, the above process trains a deep neural network by using the first geochemical database and, through transfer learning, is used for predicting the element contents in the second geochemical database. The whole process is more convenient, and thus the content values of the second element group can be quickly and effectively estimated. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on the provided drawings.

[0056] Figure 1 Shows a flowchart of a regional geochemical exploration extension method in an embodiment of the present invention;

[0057] Figure 2 Shows a flowchart of determining a deep neural network in an embodiment of the present invention;

[0058] Figure 3 Shows a schematic structural diagram of a deep neural network in an embodiment of the present invention;

[0059] Figure 4 Shows a schematic diagram of the inactivation principle of a deep neural network in an embodiment of the present invention;

[0060] Figure 5The flowchart shows the generation process of a prediction model in an embodiment of the present invention;

[0061] Figure 6 The structural schematic diagram of a regional geochemical exploration extension device in an embodiment of the present invention is shown. Detailed implementation manners

[0062] As described in the background art, in the 1:50,000 regional geochemical exploration method, re-testing and analysis of elements or oxides such as Fe2O3, Al2O3 / SiO2, K2O, B, Be, Li, Mn, Nb, Ti, U, Zr, etc. will consume a large amount of manpower, material resources, financial resources and time.

[0063] The inventor found in the actual application process that the 1:200,000 regional geochemical exploration in China has collected a large amount of data on 39 elements or oxides (including 16 elements measured in the 1:50,000 regional geochemical exploration, namely: Au, Ag, As, Bi, Cd, Cr, Co, Cu, Hg, Mo, Ni, Pb, Sb, Sn, W, Zn), such as more than 1.47 million data points, and the total data volume is about 63.21 million. These elements or oxides are SiO2, Al2O3, K2O, Na2O, CaO, MgO, Fe2O3, Ag, As, Au, B, Ba, Be, Bi, Cd, Co, Cr, Cu, F, Hg, La, Li, Mn, Mo, Nb, Ni, P, Pb, Sb, Sn, Sr, Th, Ti, U, V, W, Y, Zn, Zr respectively. And for the 1:200,000 and 1:50,000 regional geochemical explorations, in similar geographical landscape areas, although the sampling density is significantly different, the sampling media and analysis methods are basically the same. Therefore, the correlation relationships between different elements reflected by the two should also be similar.

[0064] Based on this, an embodiment of the present invention provides a regional geochemical exploration extension method. The element types and contents in the first element group and the second element group in the first geochemical database are both known quantities, indicating that the first geochemical database is reliable. In this way, by using the first geochemical database to train the deep neural network, the prediction model can better learn, more comprehensively and objectively reflect the relationship between the first element group and the second element group in the first geochemical database. In this way, based on the prediction model, the element contents in the second element group in the second geochemical database can be quickly and effectively identified, and then a geochemical anomaly map for guiding prospecting exploration work can be generated. In other words, the above process trains the deep neural network by using the first geochemical database and, through transfer learning, is used for predicting the element contents in the second geochemical database. The whole process is more simple, so the content values of the second element group can be quickly and effectively estimated.

[0065] To enable those skilled in the art to better understand the inventive concept, working principle, and advantages of the embodiments of the present invention, the following provides a detailed description of the regional geochemical exploration extension method in the embodiments of the present invention.

[0066] Refer to Figure 1 the flowchart of a regional geochemical exploration extension method in the embodiments of the present invention shown in Figure 1 as shown, the following steps may be performed:

[0067] S11, Provide a first geochemical database obtained based on a first sampling density and a second geochemical database obtained based on a second sampling density.

[0068] Specifically, the first geochemical database and the second geochemical database can characterize the distribution information of elements under different geographical conditions, and the first geochemical database and the second geochemical database are obtained at different sampling densities.

[0069] In this embodiment, the first sampling density may refer to 1:200,000 exploration geochemistry, and the second sampling density may refer to 1:50,000 exploration geochemistry.

[0070] It should be noted that the numerical values listed in the above examples are only for illustrative purposes and are used to represent different sampling densities, and should not be construed as a limitation of the present invention. In some other embodiments, the first sampling density and the second sampling density may also be other values, as long as they are different.

[0071] In this embodiment, the element types and contents in the first element group and the second element group in the first geochemical database are all known quantities. Among them, as a non-limiting example, the first element group may include 16 elements (i.e., Au, Ag, As, Bi, Cd, Cr, Co, Cu, Hg, Mo, Ni, Pb, Sb, Sn, W, Zn), and the second element group may include 11 elements (i.e., Fe2O3, Al2O3 / SiO2, K2O, B, Be, Li, Mn, Nb, Ti, U, Zr elements or oxides).

[0072] The element contents in the second element group in the second geochemical database are unknown quantities, or rather, the second element group in the second geochemical database is a null value, which is the quantity to be predicted in the present invention.

[0073] S12, Construct a deep neural network with a fully connected layer as the main architecture and configure the parameter information of the deep neural network.

[0074] Specifically, a deep neural network with a fully connected layer as the main architecture can better learn, more comprehensively and objectively reflect the relationship between the first element group and the second element group in the first geochemical database, and by configuring the parameter information of the deep neural network, it can better process the first element group and the second element group in the first geochemical database.

[0075] S13. Use the first geochemical database to train the deep neural network to generate a prediction model.

[0076] Specifically, the element types and contents in the first element group and the second element group in the first geochemical database are known quantities, indicating that the first geochemical database is reliable. When using the first geochemical database to train the deep neural network, the prediction model can better learn, more comprehensively and objectively reflect the relationship between the first element group and the second element group in the first geochemical database, making the prediction model have better generalization ability.

[0077] S14. Based on the prediction model, use the first element group in the second geochemical database as the input quantity to determine the element contents in the second element group in the second geochemical database, so as to generate a geochemical anomaly map at the second sampling density.

[0078] Specifically, the prediction model is based on the first geochemical database, enabling the prediction model to quickly and effectively identify the element contents in the second element group in the second geochemical database, and then a geochemical anomaly map for guiding prospecting and exploration work can be generated.

[0079] In other words, the above process trains a deep neural network by using the first geochemical database and, through transfer learning, is used for predicting the element contents in the second geochemical database. The whole process is more convenient, and thus can quickly and effectively estimate the content values of the second element group.

[0080] In this embodiment, in combination with Figure 2 and Figure 3 , where Figure 2 is a flowchart for determining a deep neural network in an embodiment of the present invention, Figure 3 is a schematic diagram of the architecture of a deep neural network in an embodiment of the present invention. As shown in Figure 2 and Figure 3 , the following steps can be executed:

[0081] S21. Determine a network architecture with the fully connected layer as the main body.

[0082] Specifically, the fully connected layer can integrate the extracted features, enabling the output of the deep neural network to more accurately reflect the relationship between the first element group (which can be used as an independent variable) and the second element group (which can be used as a dependent variable) in the first geochemical database.

[0083] In this embodiment, the following steps can be performed:

[0084] Determine the input layer of the network architecture.

[0085] Specifically, the input layer is used to process the first element group in the first geochemical database input into the network architecture, enabling the deep neural network to correctly identify the first element group.

[0086] In this embodiment, the input layer is composed of multiple first neurons, and the number of first neurons is equal to the number of elements in the first element group, so that the first neurons can fully correspond to the elements in the first element group.

[0087] As a specific embodiment, when the number of elements in the first element group is 17, the input layer can be composed of 17 first neurons. Thus, the first neurons numbered 1 to 16 in the input layer respectively correspond to Au, Ag, As, Bi, Cd, Cr, Co, Cu, Hg, Mo, Ni, Pb, Sb, Sn, W, Zn in the first element group, and the first neuron numbered 17 is the geographical landscape.

[0088] Determine the output layer of the network architecture.

[0089] Specifically, the output layer is used to process the second element group in the first geochemical database input into the network architecture, enabling the deep neural network to correctly identify the second element group.

[0090] In this embodiment, the output layer is composed of multiple second neurons, and the number of second neurons is equal to the number of elements in the second element group, so that the second neurons can fully correspond to the elements in the second element group.

[0091] As a specific embodiment, when the number of elements in the second element group is 11, the output layer can be composed of 11 second neurons. Thus, the second neurons numbered 1 to 11 in the output layer respectively correspond to Fe2O3, Al2O3 / SiO2, K2O, B, Be, Li, Mn, Nb, Ti, U, Zr in the second element group.

[0092] Determine the multiple hidden layers used to connect the input layer and the output layer, and each hidden layer is connected to each other.

[0093] In this embodiment, the setting of the hidden layer includes the setting of the number of hidden layers and the number of neurons in each hidden layer. Among them, the more the number of hidden layers and the number of third neurons, the more features can be extracted. However, at the same time, the number of parameters will be larger, and the risk of overfitting will also increase.

[0094] In this case, the number of third neurons in each hidden layer is greater than the number of first neurons, which can avoid missing important features.

[0095] Furthermore, the number of hidden layers should be deep enough to extract more advanced features.

[0096] In a specific embodiment, the number of hidden layers can be set to 20, and the number of third neurons in each hidden layer can be set to 32.

[0097] S22. Configure the data normalization processing method of the deep neural network for each hidden layer.

[0098] Specifically, by performing data normalization processing, the search for hyperparameters becomes simple, and the deep neural network becomes more robust and easier to train.

[0099] In this embodiment, the normalization processing formula is:

[0100]

[0101] Among them, represents the value after batch normalization, γ and β respectively represent the values for scaling and translation of the normalized data, and x i represents the i-th input value in the hidden layer, and x mean represents the average value of the input values in the hidden layer, σ represents the standard deviation of the input values in the hidden layer, and ξ is a very small fixed value to prevent the denominator from being 0.

[0102] S23. Define the activation function in each hidden layer of the deep neural network.

[0103] Specifically, by introducing the activation function, the deep neural network has the ability to solve non-linear problems. Currently, the ReLU activation function is used more in the hidden layer. It can solve the problem of gradient disappearance caused by the saturated activation function and greatly accelerate the learning speed. However, the disadvantage of this function is that when there is a very large gradient during the backpropagation process, the backpropagation update may cause the weight distribution center to be less than 0, so that the derivative at that place is always 0, and the backpropagation cannot update the weights, that is, it enters the inactivated state.

[0104] Therefore, different from the ReLU activation function in the existing solution, the activation function in the embodiment of the present invention is:

[0105] PReLU(x) = max(0, x) + ηmin(0, x) (2)

[0106] Among them, η represents the learning parameter.

[0107] By using formula (2), the parameter η can be adaptively learned from the data, which can not only solve the above problems, but also has the characteristics of fast convergence speed and low error rate.

[0108] It should be noted that data normalization processing has been performed before defining the activation function.

[0109] S24. Determine the number of neurons randomly inactivated in the hidden layer during each training process of the deep neural network.

[0110] Specifically, the most likely problem encountered during the training process of a deep neural network is the overfitting problem. To alleviate overfitting and improve the generalization ability of the deep neural network, random inactivation is adopted in the model. Random inactivation means that during each training process, for the neurons in each hidden layer, a certain proportion of neurons are randomly made not to participate in the model training.

[0111] For example, referring to Figure 4 the schematic diagram of the inactivation principle of a deep neural network in the embodiment of the present disclosure shown in Figure 4 as shown, during each training process, the probability of random inactivation can be made 20%, that is: during each training process of the deep neural network, the number of neurons randomly inactivated in the hidden layer (such as Figure 4 shown by the dotted line in) is 20% of the total number of neurons in the hidden layer, which can effectively prevent overfitting.

[0112] Therefore, by adopting a network architecture mainly composed of fully connected layers and supplemented by data normalization processing, activation functions, and random inactivation methods, the prediction structure of the deep neural network is more accurate and reliable.

[0113] In this embodiment, after configuring the parameter information of the deep neural network, the deep neural network can be trained to generate a prediction model.

[0114] Referring to Figure 5 the flowchart of the generation process of a prediction model in the embodiment of the present invention shown in Figure 5 as shown, the following generation steps can be executed:

[0115] S31. Divide the first element group and the second element group in the first geochemical database into a training data set and a validation data set according to a preset ratio.

[0116] Specifically, the first element group and the second element group in the first geochemical database can be randomly shuffled to ensure that the relationships between the samples in the first geochemical database are random and there is no logic in the order, thereby preventing similar samples from being too concentrated in the training set or the validation set, resulting in a low generalization ability of the trained deep learning model. Then, according to a certain ratio, such as 7:3, 8:2, or 9:1, the first geochemical database with the randomly shuffled order is divided into corresponding training set and validation set folders to form a training data set and a validation data set for the training and validation processes of the deep neural network.

[0117] In this embodiment, considering that the contents of some elements in the first element group and the second element group are relatively low, such as below the detection limit, or the formats of some elements in the first element group and the second element group are inconsistent, before dividing the first element group and the second element group in the first geochemical database into a training data set and a validation data set according to a preset ratio, the regional geochemical exploration extension method may further include:

[0118] Preprocessing the first element group and the second element group in the first geochemical database, and the first element group in the second geochemical database.

[0119] By performing the preprocessing operation, the integrity of the elements in the first element group and the second element group in the first geochemical database, and the first element group in the second geochemical database can be improved, and the consistency of the data format can be improved, which is beneficial to reducing the training difficulty.

[0120] In this embodiment, the preprocessing may include:

[0121] Replacing the missing elements in the first element group and the second element group with any value between zero and the detection limit, where the detection limit is the lowest value at which the corresponding element is detected.

[0122] Specifically, to make up for the defect that some elements cannot be detected, this solution replaces the missing values with random values between zero and the detection limit, so that there are no missing elements in the first element group and the second element group, and it can prevent the data analysis results from being biased or misleading.

[0123] Further, as shown in formula (3), the random value follows a uniform distribution, and the reason for using the uniform distribution is that the probability of the random value taking any value in this interval is equal.

[0124] X~U(0, ρ)(3)

[0125] Where X represents the random value, U represents the uniform distribution, and ρ represents the detection limit of the element.

[0126] Next, the data in the first element group and the second element group after missing value processing is standardized, which can eliminate the influence of the dimension between different variables and make them comparable.

[0127] In addition, to improve the convergence speed and stability, the standardization method in this solution can be seen in formula (4):

[0128]

[0129] Among them, x i is the i-th value, x mean is the average value of the training set, σ is the standard deviation of the training set, ξ is a very small fixed value to prevent the denominator from being 0, and x std is the value after standardization.

[0130] Moreover, the parameters of the standardization in the first geochemical database and the second geochemical database are the same.

[0131] It should be particularly noted that the so-called same parameters here mean that the mean value and the standard deviation used for the standardization of the first element group and the second element group are both the mean value and the standard deviation of the training data set. The purpose is to ensure that the model evaluation and model prediction are consistent with the training set in terms of measurement and guarantee the reliability of the prediction effect.

[0132] S32: Input the training data set into the deep neural network in sequence and perform forward propagation to determine the cost function of the prediction result of the deep neural network.

[0133] Specifically, the cost function is an index representing the "badness" of the deep learning model and is used to measure the effect of the weight parameter w on the training set. When the cost function value converges to a suitable accuracy, it indicates that the deep neural network reaches the optimal state during the training phase.

[0134] By inputting the training data set into the deep neural network in sequence, the prediction result of the deep neural network for the training data set in the current state can be determined. Further, the cost function can be determined.

[0135] The cost function is:

[0136]

[0137] Among them, m is the number of training samples; y gt i is the true value of the i-th training sample; y pre i is the predicted value of the i-th training sample; L(y gt i ,y prei ) is the loss function of training sample i; δ is a threshold (e.g., 1); λ is a penalty coefficient (e.g., 0.01); R(w) is the penalty term, where w is the weight of the network;

[0138]

[0139] where n is the number of network weights, and w i is the i-th weight.

[0140] In this embodiment, in the process of determining the cost function of the prediction result of the deep neural network, it further includes: determining an optimization function adapted to the deep neural network, and optimizing the weight parameters in the cost function.

[0141] The optimization function is used to optimize the weight parameters. Among them, Adam combines the advantages of the Momentum and RMSprop methods, realizes efficient search in the parameter space, and thus is widely used. This deep learning model optimization adopts this method.

[0142] In this embodiment, the optimization function is:

[0143]

[0144] where w is the weight parameter, := means that the value on the left side is updated or replaced by the value on the right side, α is the learning rate, β1 is the first momentum, β2 is the second momentum, dw is the derivative of the cost function with respect to the weight parameter, and v dw is the exponentially weighted average of the momentum of dw, and s dw is the exponentially weighted average of the momentum of dw 2 momentum, ε is a fixed value, and t represents the number of iterations.

[0145] Next, initialize the weight parameters.

[0146] Specifically, the initialization of the weight parameters is to prevent the so-called "symmetry" problem, that is, to prevent all neurons in the same hidden layer from having the same function, and it is not allowed to initialize all weight parameters to 0. Therefore, random initialization is required.

[0147] Specifically, in order to meet the two conditions that the initialized parameters are relatively small and the changes are relatively stable, this deep learning method uses random values that conform to one percent of the standard normal distribution as the initialized parameters, that is, the formula:

[0148] W ∼ N(0,1) / 100(8)

[0149] where W is the weight parameter, and N(0,1) is the standard normal distribution.

[0150] S33. Perform backpropagation using the gradient descent method to update the weight parameters in the cost function, obtaining an initial prediction model.

[0151] Specifically, through steps S31 and S32, the cost function of the prediction result of the deep neural network can be determined. Furthermore, through the Adam gradient descent method, the weight parameters in the cost function can be further optimized, making the prediction structure of the deep neural network closer to the true result, thereby obtaining the corresponding initial prediction model.

[0152] S34. Input all the validation data sets into the initial prediction model and perform forward propagation to determine one or more of the mean squared error and absolute error of the prediction result as the validation evaluation index.

[0153] Specifically, by executing step S34, by inputting all the validation data sets into the initial prediction model and performing forward propagation, the validation evaluation index of the prediction result can be determined. For example, one or more of the mean squared error and absolute error are used as the validation evaluation index, so as to evaluate the quality of the initial prediction model.

[0154] Specifically, the mean squared error MES and absolute error MAE can be determined through the following formula:

[0155]

[0156] where m is the number of training samples; y gt i is the true value of the i-th training sample; y pre i is the predicted value of the i-th training sample.

[0157] S35. Repeat the training and validation steps iteratively until the cost function value and the validation evaluation index meet the requirements.

[0158] For example, repeat the above steps iteratively multiple times (e.g., 1000 times) until the cost function and the validation evaluation index reach the required accuracy, mainly manifested as the cost function value no longer decreasing significantly and the validation evaluation index no longer increasing significantly, that is, obtaining the optimal deep learning model (prediction model).

[0159] Thus, by using the above training method, on the one hand, by dividing the training and validation data sets, a sample basis is laid for the deep learning model based on big data; on the other hand, the loss function is improved, and by introducing a penalty coefficient and penalty terms, the problem of sample imbalance is alleviated to a certain extent, thereby improving the generalization ability of the prediction model.

[0160] In addition, for the loss function, the mean square error and the absolute error loss are fused, which can prevent the negative impact of outliers in exploration geochemical data on model training and improve the generalization ability of the model.

[0161] In this embodiment, the first geochemical database and the second geochemical database further include nominal scale variables of geographical landscapes, which are used to characterize different types of geographical landscapes. By adding such nominal scale independent variables of geographical landscapes, the trained model can adapt to different geographical landscapes.

[0162] In a specific embodiment, different identifiers can be used to represent different types of geographical landscapes. For example, 0 represents middle and low mountains, hills; 1 represents forest swamps; 2 represents arid and semi-arid deserts; 3 represents alpine mountainous areas; 4 represents alpine lake-marsh deserts; 5 represents karsts; 6 represents tropical rainforests; 7 represents the Loess Plateau; 8 represents alluvial-proluvial plains.

[0163] It should be noted that, first, the nominal scale variables listed in the above examples, namely 0 to 8, are a type of integers, but there is neither a size nor an order among them. The difference in numerical values only represents different categories of geographical landscapes; second, the geographical landscapes listed in the above examples are also for illustrative purposes and are used to represent different types of geographical landscapes.

[0164] In this case, using the first geochemical database, the deep neural network is trained to generate a prediction model. The training process further includes: using the nominal scale variables of geographical landscapes as a data set to train the deep neural network.

[0165] As a specific embodiment, then refer to Figure 3 , the first neuron numbered 17 in the input layer corresponds to the nominal scale variables of geographical landscapes.

[0166] In summary, by using the first geochemical database to train the deep neural network to generate a prediction model, then the prediction model can be used to predict the element contents in the second element group of the second geochemical database.

[0167] More specifically, the following steps can be executed:

[0168] Take the first element group in the second geochemical database as an input quantity and input it into the prediction model to generate the geochemical standard values corresponding to the elements in the second element group and restore them to the original values of the elements in the second element group.

[0169] Specifically, 17 elements (i.e., the first element group) can be loaded into the trained optimal deep learning model to carry out prospecting prediction, estimate the geochemical standardized values ​​of 11 elements (i.e., the second element group), and then obtain the original value through formula (11):

[0170] x i =x mean +x std ×σ (11)

[0171] Among them, x i is the original value, x mean is the mean of the training set, x std is the estimated geochemical normalized value mentioned above, and σ is the standard deviation of the training set.

[0172] Next, the original values ​​of the elements in the second element group are interpolated to generate a geochemical map corresponding to the second element group.

[0173] Finally, the geochemical map is processed by taking the mean of the original values ​​of each element in the second element group and the standard deviation of the original values ​​of each element as the anomaly lower limit to generate a geochemical anomaly map.

[0174] For example, the mean of the original geochemical value + 2 times the standard deviation is taken as the lower limit of the anomaly.

[0175] Furthermore, based on the generated geochemical anomaly map, the prospecting target areas of scarce minerals such as iron, aluminum, potassium salt, boron, beryllium, lithium, manganese, niobium, titanium, uranium and zirconium can be determined based on the geochemical anomaly map and multi-information data (such as geology, geophysical exploration and remote sensing).

[0176] Finally, field verification is carried out on the prospecting target area of ​​the target mineral species delineated on the surface. Based on its exposure, degree of mineralization, surface occurrence and other information, a certain degree of geological mapping, trenching and drilling (including chemical sample analysis) is used to determine the shape, scale, spatial distribution, grade, co-existing elements and resource quantity of the ore body in the target area.

[0177] The embodiment of the present invention also provides a device corresponding to the regional geochemical exploration extension method, such as Figure 6 The structural diagram of a regional geochemical exploration extension device in an embodiment of the present invention is shown in FIG. Figure 6 As shown, the regional geochemical exploration extension device 100 may include:

[0178] The geochemical database providing unit 110 is adapted to provide a first geochemical database obtained based on a first sampling density and a second geochemical database obtained based on a second sampling density. Among them, the element types and contents in the first element group and the second element group in the first geochemical database are known quantities, and the element contents in the second element group in the second geochemical database are unknown quantities;

[0179] The neural network construction unit 120 is adapted to construct a deep neural network with a fully connected layer as the main architecture and configure the parameter information of the deep neural network;

[0180] The processing unit 130 is adapted to use the first geochemical database to train the deep neural network to generate a prediction model; and based on the prediction model, using the first element group in the second geochemical database as the input quantity, determine the element contents in the second element group in the second geochemical database to generate a geochemical anomaly map under the second sampling density.

[0181] Among them, for the specific working processes and principles of the geochemical database providing unit 110, the neural network construction unit 120, and the processing unit 130, reference can be made to the relevant descriptions in the foregoing examples.

[0182] It can be understood that the above division of each unit is only a division of logical functions. In actual implementation, they can be fully or partially integrated into a physical entity, or physically separated. In addition, the above units can be implemented in the form of a processor calling software.

[0183] Although the embodiments of the present invention are disclosed as above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the protection scope of the present invention should be subject to the scope defined by the claims.

Claims

1. A regional geochemical exploration extension method, characterized in that: include: Providing a first geochemical database obtained based on a first sampling density and a second geochemical database obtained based on a second sampling density, wherein the types and contents of elements in the first element group and the second element group in the first geochemical database are known, and the contents of elements in the second element group in the second geochemical database are unknown; Construct a deep neural network with a fully connected layer as the main architecture, and configure parameter information of the deep neural network; The first geochemical database is used to train the deep neural network to generate a prediction model, including: dividing the first element group and the second element group in the first geochemical database into a training data set and a verification data set according to a preset ratio; inputting the training data set into the deep neural network in sequence, and forward propagating to determine the cost function of the prediction result of the deep neural network; using the gradient descent method for back propagation to update the weight parameters in the cost function to obtain an initial prediction model; inputting all the verification data sets into the initial prediction model, and forward propagating to determine one or more of the mean square error and absolute error of the prediction result as a verification evaluation index; repeating the iterative training and verification steps until the cost function value and the verification evaluation index meet the requirements; Based on the prediction model, the first element group in the second geochemical database is used as input, and the element content in the second element group in the second geochemical database is determined to generate a geochemical anomaly map under the second sampling density, including: taking the first element group in the second geochemical database as input, and inputting it into the prediction model, generating geochemical standard values ​​corresponding to each element in the second element group, and restoring them to the original values ​​of each element in the second element group; interpolating the original values ​​of each element in the second element group to generate a geochemical map corresponding to the second element group; taking the mean of the original values ​​of each element in the second element group and the standard deviation of the original values ​​of each element as the lower limit of the anomaly, processing the geochemical map to generate a geochemical anomaly map.

2. The regional geochemical exploration extension method according to claim 1, characterized in that: The step of constructing a deep neural network with a fully connected layer as the main structure and configuring parameter information of the deep neural network includes: Determining a network architecture with the fully connected layer as the main body, including: determining an input layer of the network architecture, the input layer is composed of a plurality of first neurons, and the number of the first neurons is equal to the number of elements in the first element group; determining an output layer of the network architecture, the output layer is composed of a plurality of second neurons, and the number of the second neurons is equal to the number of elements in the second element group; determining a plurality of hidden layers for connecting the input layer and the output layer, and each hidden layer is connected to each other; The deep neural network is configured to perform data standardization processing on each hidden layer, wherein the standardization processing formula is: in, represents the value after batch normalization, γ and β represent the values ​​of scaling and translating the normalized data, respectively, and x i represents the i-th input value in the hidden layer, x mean represents the mean value of the input value in the hidden layer, and σ represents the standard deviation of the input value in the hidden layer; Define the activation function in each hidden layer of the deep neural network, where the activation function is: PReLU(x)=max(0,x)+ηmin(0,x) Where η represents the learning parameter; And determine the number of neurons that are randomly inactivated in the hidden layer during each training process of the deep neural network.

3. The regional geochemical exploration extension method according to claim 2, characterized in that: The number of third neurons in each hidden layer is greater than the number of the first neurons; During each training process of the deep neural network, the number of randomly inactivated neurons in the hidden layer is 20% of the number of all neurons.

4. The regional geochemical exploration extension method according to claim 1, characterized in that: The cost function is: Among them, m is the number of training samples; y gt i is the true value of the i-th training sample; y pre i is the predicted value of the i-th training sample; L(y gt i ,y pre i ) is the loss function of training sample i; δ is a threshold; λ is the penalty coefficient; R(w) is the penalty term, where w is the weight of the network; Among them, n is the number of network weights, w i is the ith weight.

5. The regional geochemical exploration extension method according to claim 4, characterized in that: In the process of determining the cost function of the prediction result of the deep neural network, it also includes: Determine an optimization function that is compatible with the deep neural network, and optimize the weight parameters in the cost function during the training process, wherein the optimization function includes: Where w is the weight parameter, := means that the value on the left is updated or replaced by the value on the right, α is the learning rate, β1 is the first momentum, β2 is the second momentum, dw is the derivative of the cost function with respect to the weight parameter, v dw is the momentum exponential weighted average of dw, s dw for dw 2 Momentum exponential weighted average, ε is a constant, and t represents the number of iterations; And initialize the weight parameters.

6. The regional geochemical exploration extension method according to claim 4, characterized in that: Before dividing the first element group and the second element group in the first geochemical database into a training data set and a verification data set according to a preset ratio, the method further includes: The first element group and the second element group in the first geochemical database, and the first element group in the second geochemical database are preprocessed, and the preprocessing includes: replacing the missing elements in the first element group and the second element group with any value between zero and the detection limit, wherein the detection limit is the lowest value of the corresponding element detected; and, standardizing the data in the first element group and the second element group after the missing processing, and the parameters of the standardization processing in the first geochemical database and the second geochemical database are the same.

7. The regional geochemical exploration extension method according to claim 1, characterized in that: The first geochemical database and the second geochemical database also include geographic landscape nominal scale variables, The geographical landscape nominal scale variables are used to characterize different types of geographical landscapes; The deep neural network is trained by using the first geochemical database to generate a prediction model. The training process also includes: using the geographical landscape nominal scale variables as a data set to train the deep neural network.

8. The regional geochemical exploration extension method according to claim 1, characterized in that: Also includes: Based on the geochemical anomaly map and multivariate information data, the prospecting target area for the target mineral species is determined.

9. A regional geochemical exploration extension device, characterized in that: include: a geochemical database providing unit, adapted to provide a first geochemical database obtained based on a first sampling density and a second geochemical database obtained based on a second sampling density, wherein the types and contents of elements in the first element group and the second element group in the first geochemical database are known quantities, and the contents of elements in the second element group in the second geochemical database are unknown quantities; A neural network construction unit, adapted to construct a deep neural network with a fully connected layer as the main architecture, and configure parameter information of the deep neural network; A processing unit is adapted to use the first geochemical database to train the deep neural network and generate a prediction model, comprising: dividing the first element group and the second element group in the first geochemical database into a training data set and a verification data set according to a preset ratio; inputting the training data set into the deep neural network in sequence, and forward propagating to determine the cost function of the prediction result of the deep neural network; using the gradient descent method for back propagation to update the weight parameters in the cost function to obtain an initial prediction model; inputting all the verification data sets into the initial prediction model, and forward propagating to determine one or more of the mean square error and the absolute error of the prediction result as a verification evaluation index; repeating the iterative training and verification steps until the cost function value and the verification evaluation index meet the requirements; and Based on the prediction model, the first element group in the second geochemical database is used as input, and the element content in the second element group in the second geochemical database is determined to generate a geochemical anomaly map under the second sampling density, including: taking the first element group in the second geochemical database as input, and inputting it into the prediction model, generating geochemical standard values ​​corresponding to each element in the second element group, and restoring them to the original values ​​of each element in the second element group; interpolating the original values ​​of each element in the second element group to generate a geochemical map corresponding to the second element group; taking the mean of the original values ​​of each element in the second element group and the standard deviation of the original values ​​of each element as the lower limit of the anomaly, processing the geochemical map to generate a geochemical anomaly map.

Images