Element content-based mother rock mineral composition prediction method, system and device and storage medium

Abstract

The invention discloses a mother rock mineral composition prediction method, system and device based on element content and a storage medium, and the method comprises the following steps: obtaining and preprocessing the element content and mineral composition data of a mother rock, and obtaining a preprocessed data set; selecting a classification algorithm to construct an initial classification model, and training the initial classification model through the preprocessing data set to obtain a classification model; and predicting mother rock mineral composition through the classification model. According to the method, element content analysis and an advanced machine learning technology are combined, so that rapid and accurate prediction of the mother rock mineral composition is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of mineral detection, and in particular relates to a method, system, device and storage medium for predicting the composition of parent rock minerals based on element content. Background Art

[0002] In the field of geological exploration and geotechnical engineering, accurate prediction of parent rock mineral composition is of great significance for understanding geological structures, assessing resource potential, and managing environmental risks. Traditional parent rock mineral composition analysis methods rely on laboratory tests, which are accurate but often time-consuming and costly. With the development of technology, more and more studies have begun to explore new methods that combine elemental content analysis with machine learning technology to predict parent rock mineral composition.

[0003] In the field of geotechnical engineering, the application of intelligent technologies is also developing. For example, the intelligent transformation of geotechnical engineering has been achieved by combining 3D geological modeling, the Internet of Things (IoT), artificial intelligence (AI), deep learning, and extended reality technology. The application of these technologies not only improves the efficiency and accuracy of geotechnical engineering investigation and testing, but also provides new means for monitoring and early warning of geotechnical engineering.

[0004] However, there is still a lack of a mineral composition prediction method for rapid prediction. Summary of the invention

[0005] In order to solve the above technical problems, the present invention proposes a method and system for predicting the composition of parent rock minerals based on element content to solve the problems existing in the above prior art.

[0006] To achieve the above object, the present invention provides a method for predicting the composition of parent rock minerals based on element content, comprising the following steps:

[0007] Obtaining the element content and mineral composition data of the parent rock and preprocessing them to obtain a preprocessed data set;

[0008] Selecting a classification algorithm to construct an initial classification model, and training the initial classification model using the preprocessed data set to obtain a classification model;

[0009] The parent rock mineral composition is predicted by the classification model.

[0010] Preferably, obtaining the element content and mineral composition data of the parent rock includes:

[0011] The element content of the parent rock is obtained by X-ray fluorescence spectroscopy testing, where the parent rock element content includes Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co and Ni;

[0012] X-ray diffraction test is used to obtain the mineral composition data of the matrix, wherein the mineral composition data of the matrix include the mineral types and contents in each group of matrix rocks; wherein the mineral types include matrix quartz, plagioclase, orthoclase, hornblende, tremolite, pyroxene, talc, olivine, dolomite, calcite, chlorite, epidote and mica.

[0013] Preferably, preprocessing the acquired data includes:

[0014] Through data cleaning, outliers in the element content and mineral composition data of the parent rock are removed to obtain cleaned data;

[0015] The cleaned data are classified according to the mineral types and content ranges in each group of parent rock samples to obtain a total data set, and the total data set is divided into a training set, a validation set, and a test set in proportion.

[0016] Preferably, when the initial classification model is trained, the accuracy is evaluated through a confusion matrix.

[0017] The present invention also provides a system for predicting the composition of parent rock minerals based on element content, comprising:

[0018] A data acquisition module is used to acquire the element content and mineral composition data of the parent rock and perform preprocessing to obtain a preprocessed data set;

[0019] A model building module is used to select a classification algorithm to build an initial classification model, and train the initial classification model through the preprocessed data set to obtain a classification model;

[0020] A prediction module is used to predict the mineral composition of the parent rock through the classification model.

[0021] Preferably, the data acquisition module includes:

[0022] A data acquisition unit, used to obtain element content and mineral composition data of the parent rock;

[0023] The preprocessing unit is used to preprocess the element content and mineral composition data of the parent rock to obtain a preprocessed data set.

[0024] Preferably, the data acquisition unit comprises:

[0025] The first acquisition unit is used to obtain the element content of the parent rock through X-ray fluorescence spectrometry, wherein the element content of the parent rock includes Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co and Ni;

[0026] The second acquisition unit is used to obtain the matrix mineral composition data by X-ray diffraction test, wherein the matrix mineral composition data includes the mineral types and contents in each group of matrix; wherein the mineral types include matrix quartz, plagioclase, orthoclase, hornblende, tremolite, pyroxene, talc, olivine, dolomite, calcite, chlorite, epidote and mica.

[0027] Preferably, the pre-processing unit comprises:

[0028] A cleaning unit is used to remove abnormal values ​​in the element content and mineral composition data of the parent rock through data cleaning to obtain cleaned data;

[0029] The classification unit is used to classify the cleaned data according to the mineral types and content ranges in each group of parent rock samples to obtain a total data set, and divide the total data set into a training set, a validation set, and a test set in proportion.

[0030] The present invention also provides a computer device, comprising a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method.

[0031] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the method are implemented.

[0032] Compared with the prior art, the present invention has the following advantages and technical effects:

[0033] The present invention discloses a method, system, device and storage medium for predicting the mineral composition of parent rocks based on element content. The method includes the following steps: obtaining the element content and mineral composition data of the parent rock and preprocessing it to obtain a preprocessed data set; selecting a classification algorithm to construct an initial classification model, training the initial classification model through the preprocessed data set to obtain a classification model; and predicting the mineral composition of the parent rock through the classification model. The present invention provides a fast, accurate and cost-effective method for predicting the mineral composition of parent rocks by combining element content analysis and machine learning technology, which can significantly improve prediction efficiency and accuracy, reduce costs in geological exploration and geotechnical engineering, and at the same time enhance data processing capabilities, support scientific decision-making, and be environmentally friendly, bringing innovative technical solutions to related fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0035] Figure 1 The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0036] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0037] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0038] Embodiment 1

[0039] like Figure 1 As shown, this embodiment provides a method for predicting the composition of parent rock minerals based on element content, comprising the following steps:

[0040] Obtaining the element content and mineral composition data of the parent rock and preprocessing them to obtain a preprocessed data set;

[0041] Select a classification algorithm to build an initial classification model, train the initial classification model by preprocessing the data set, and obtain a classification model;

[0042] Predict parent rock mineral composition through classification model.

[0043] The specific steps include:

[0044] (1) The element content and mineral composition of the parent rock were tested to form a complete parent rock element content-mineral composition data set. The element content of the parent rock was tested by X-ray fluorescence spectroscopy (XRF), and the data set included the content of multiple elements such as Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co, and Ni in the parent rock; the mineral composition of the parent rock was tested by X-ray diffraction (XRD), and the test results were qualitatively and quantitatively analyzed to obtain the mineral types and contents in each group of parent rocks. The data set included the mineral types and corresponding contents of parent rocks such as quartz, plagioclase, orthoclase, amphibole, tremolite, pyroxene, talc, olivine, dolomite, calcite, chlorite, epidote, and mica.

[0045] (2) Preprocessing: For the data used in the prediction model, outlier data samples are removed through data cleaning, and then classified according to the mineral types and content ranges in each group of parent rock samples to form the final data set, which is divided into training set, validation set and test set according to a certain ratio. Among them, the training set is the data source for model training, the validation set is used to evaluate the model performance during the training process and as a basis for adjustment to form a prediction model with good training performance; the test set is independent of the model training data and is used to test the accuracy and robustness of the trained model. In this model, the test set accounts for 20% of the total data, and the rest is used for training set and validation set. K-fold cross validation is used to deal with insufficient data. Specifically, the data set is divided into k equal sub-samples. In each iteration, one of the sub-samples is used as the validation set, and the remaining k-1 sub-samples are used as the training set for model training. Subsequently, the prediction performance of the trained model on the validation set is evaluated. This process is repeated k times to ensure that each sub-sample has the opportunity to be used as a validation set. Finally, the results of these k iterations (i.e., accuracy) are averaged. The advantage of cross-validation is that when there is less parent rock data in the data set, the limited data can be used more fully for training and validation, while reducing the possibility that the evaluation results are affected by specific data division methods.

[0046] (3) Select the model algorithm and use the classification algorithm as the core algorithm of the prediction model according to the type of prediction result. In the classification algorithm, the predicted value is the category or interval to which it belongs. The confusion matrix is ​​used for accuracy evaluation, which shows the relationship between the prediction result of the model on the test data set and the actual category in the form of a matrix. The rows of the matrix represent the actual category, the columns represent the category predicted by the model, and each cell contains the number of samples matching the corresponding category. The typical composition includes true positives (TP), false positives (FP), true negatives (TN), and false negatives (FN). The prediction accuracy can be calculated using Formula 1, which is the ratio of the number of correct predictions to the total number of samples.

[0047]

[0048] (4) Model training and evaluation: During the model training process, a variety of machine learning models are selected to train the data set, including one or more of the decision tree, discriminant analysis, logistic regression, naive Bayes, support vector machine, nearest neighbor classification, kernel approximation classification, neural network, etc. Afterwards, the classification of the model and model parameters need to be adjusted.

[0049] Furthermore, when using a classification algorithm, if the data volume differs too much, the model may tend to predict the category with a larger data volume, which will lead to model deviation and have no practical significance. If there are too many classifications, the complexity of the model will increase, which is not conducive to model training. However, a small number of classifications has no guiding significance for practical applications. Therefore, in the preprocessing stage, it is necessary to adjust according to the statistical distribution of the content of parent rock minerals in each group to ensure that the amount of data in each category is close. Taking carbonate minerals as an example, the main minerals are calcite, dolomite, a small amount of quartz and metamorphic minerals. The samples are classified according to whether the calcite content in different groups of samples is greater than 80%, whether the dolomite content is greater than 80%, and whether the quartz content is greater than 20%. According to the model training results, the content range is appropriately adjusted to adjust the classification situation and thus improve the model accuracy.

[0050] Furthermore, the model parameters are appropriately adjusted according to the different parameters required by different models. For example, in the decision tree model, the maximum number of splits, split criteria, and whether to perform alternative decision splits can be adjusted; in the neural network model, the number of hidden layers, the number of neurons in each layer, and the activation function can be adjusted. By training different machine learning models and adjusting different model parameters under the same machine learning model, the model with the highest accuracy is selected as the final prediction model. The model accuracy is calculated using a confusion matrix, and whether the accuracy is greater than or equal to 80% is used as the basis for model training judgment.

[0051] This embodiment also provides a parent rock mineral composition prediction system based on element content, which is characterized by comprising:

[0052] A data acquisition module is used to acquire the element content and mineral composition data of the parent rock and perform preprocessing to obtain a preprocessed data set;

[0053] The model building module is used to select a classification algorithm to build an initial classification model, and train the initial classification model by preprocessing the data set to obtain the classification model;

[0054] The prediction module is used to predict the mineral composition of the parent rock through the classification model.

[0055] Furthermore, the data acquisition module includes:

[0056] A data acquisition unit, used to obtain element content and mineral composition data of the parent rock;

[0057] The preprocessing unit is used to preprocess the element content and mineral composition data of the parent rock to obtain a preprocessed data set.

[0058] Furthermore, the data acquisition unit comprises:

[0059] The first acquisition unit is used to obtain the element content of the parent rock through X-ray fluorescence spectrometry, wherein the element content of the parent rock includes Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co and Ni;

[0060] The second acquisition unit is used to obtain the matrix mineral composition data by X-ray diffraction test, wherein the matrix mineral composition data includes the mineral types and contents in each group of matrix; wherein the mineral types include matrix quartz, plagioclase, orthoclase, hornblende, tremolite, pyroxene, talc, olivine, dolomite, calcite, chlorite, epidote and mica.

[0061] Furthermore, the preprocessing unit comprises:

[0062] A cleaning unit is used to remove abnormal values ​​in the element content and mineral composition data of the parent rock through data cleaning to obtain cleaned data;

[0063] The classification unit is used to classify the cleaned data according to the mineral types and content ranges in each group of parent rock samples, obtain the total data set, and divide the total data set into a training set, a validation set, and a test set in proportion.

[0064] This embodiment also provides a computer device, including a memory, a processor, and a computer program stored in the memory, and the processor executes the computer program to implement the steps of the method.

[0065] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method are implemented.

[0066] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for predicting parent rock mineral composition based on element content, characterized in that: The following steps are involved: Obtaining the element content and mineral composition data of the parent rock and preprocessing them to obtain a preprocessed data set; Selecting a classification algorithm to construct an initial classification model, and training the initial classification model using the preprocessed data set to obtain a classification model; The parent rock mineral composition is predicted by the classification model.

2. The method according to claim 1, characterized in that Obtaining element content and mineral composition data of the parent rock includes: The element content of the parent rock is obtained by X-ray fluorescence spectroscopy testing, where the parent rock element content includes Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co and Ni; X-ray diffraction test is used to obtain the mineral composition data of the matrix, wherein the mineral composition data of the matrix include the mineral types and contents in each group of matrix rocks; wherein the mineral types include matrix quartz, plagioclase, orthoclase, hornblende, tremolite, pyroxene, talc, olivine, dolomite, calcite, chlorite, epidote and mica.

3. The method according to claim 1, characterized in that The preprocessing of the acquired data includes: Through data cleaning, outliers in the element content and mineral composition data of the parent rock are removed to obtain cleaned data; The cleaned data are classified according to the mineral types and content ranges in each group of parent rock samples to obtain a total data set, and the total data set is divided into a training set, a validation set, and a test set in proportion.

4. The method according to claim 1, characterized in that: When training the initial classification model, the accuracy is evaluated through the confusion matrix.

5. A prediction system for parent rock mineral composition based on element content, characterized in that: include: A data acquisition module is used to acquire the element content and mineral composition data of the parent rock and perform preprocessing to obtain a preprocessed data set; A model building module is used to select a classification algorithm to build an initial classification model, and train the initial classification model through the preprocessed data set to obtain a classification model; A prediction module is used to predict the mineral composition of the parent rock through the classification model.

6. The system according to claim 5, characterized in that The data acquisition module comprises: A data acquisition unit, used to obtain element content and mineral composition data of the parent rock; The preprocessing unit is used to preprocess the element content and mineral composition data of the parent rock to obtain a preprocessed data set.

7. The system according to claim 6, characterized in that The data acquisition unit comprises: The first acquisition unit is used to obtain the element content of the parent rock through X-ray fluorescence spectrometry, wherein the element content of the parent rock includes Na, Mg, Al, Si, P, S, Cl, K, Ca, Sc, Ti, V, Cr, Mn, Fe, Co and Ni; The second acquisition unit is used to obtain the matrix mineral composition data by X-ray diffraction test, wherein the matrix mineral composition data includes the mineral types and contents in each group of matrix; wherein the mineral types include matrix quartz, plagioclase, orthoclase, hornblende, tremolite, pyroxene, talc, olivine, dolomite, calcite, chlorite, epidote and mica.

8. The system according to claim 6, characterized in that The pre-processing unit comprises: A cleaning unit is used to remove abnormal values ​​in the element content and mineral composition data of the parent rock through data cleaning to obtain cleaned data; The classification unit is used to classify the cleaned data according to the mineral types and content ranges in each group of parent rock samples to obtain a total data set, and divide the total data set into a training set, a validation set, and a test set in proportion.

9. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 4.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images