A method for identifying water source of high-chlorine mine well water inrush in gold mine based on principal component analysis

Abstract

The present application relates to mine water source identification and data processing technical field, especially to a kind of gold mine high-chlorine mine gushing water source identification method based on principal component analysis.The content includes: determining candidate water source, collecting regional background reference sample and known water source sample, and constructing water source identification index sequence;Known water source sample is processed, the characteristic response value of high-chlorine water source is calculated, the characteristic matrix of high-chlorine water source is constructed and effective principal component is extracted, the principal component characteristic center and principal component dispersion range of each candidate water source are obtained, and identification margin threshold and maximum allowable water source deviation threshold are determined;Obtain the mine gushing water sample to be identified, calculate the characteristic response value of high-chlorine water source and project to principal component space, calculate its water source deviation degree and water source consistency with each candidate water source;Water source is identified in combination with maximum allowable water source deviation threshold and identification margin threshold.Solve the problems of insufficient utilization of traditional method coordination change, insufficient consideration of dimension and dispersion, lack of threshold constraint.

Description

Technical Field

[0001] This invention relates to the field of mine water source identification and data processing technology, and in particular to a method for identifying high-chlorine mine water sources in gold mines based on principal component analysis. Background Technology

[0002] As gold mining depths increase, the conditions and hydraulic connections of groundwater in mining areas become increasingly complex. Fault structures, different aquifers, and seawater from coastal areas can all contribute to the mine water inflow process. Particularly in high-chloride mining areas, groundwater from different sources develops differentiated hydrochemical compositions and stable isotopic characteristics under long-term runoff, surrounding rock effects, and water mixing, providing a crucial information foundation for analyzing mine water inflow sources. Conventional hydrochemical ions, ion ratios, and stable isotopes of hydrogen, oxygen, and chloride can reflect groundwater recharge sources, runoff evolution, and water body connections from different perspectives, while principal component analysis (PCA) can comprehensively extract and reduce the dimensionality of multidimensional water source characteristics. Therefore, combining multi-type hydrochemical and stable isotope detection data with PCA to identify the water sources of high-chloride gold mine water inflow is of great significance for understanding the characteristics of the groundwater system in mining areas, analyzing mine water inflow sources, and carrying out mine water hazard prevention and control.

[0003] However, current methods for identifying mine water sources often rely on single hydrochemical indicators, ion ratios, or simple statistical discrimination. They fail to adequately utilize the synergistic changes among multiple hydrochemical indicators and stable isotope indicators in high-chlorine mine water, making it difficult to effectively distinguish between similar water sources such as seawater, shallow weathered fissure water, and groundwater from different aquifers. Furthermore, they do not adequately consider the different dimensions of indicators, information redundancy, and the inherent dispersion of the water source itself, making them prone to misjudgment due to a single chloride ion anomaly. In addition, some methods lack constraints on identification thresholds and unknown water source conditions. When the mine water to be identified is close to or significantly deviates from all known water sources, a single result may still be forced, leading to insufficient stability and reliability in identification. Summary of the Invention

[0004] This invention provides a method for identifying high-chloride mine water sources in gold mines based on principal component analysis. This method addresses the shortcomings of existing mine water source identification methods that rely heavily on single hydrochemical indicators, ion ratios, or simple statistical discrimination. These methods often fail to adequately utilize the synergistic changes among multiple hydrochemical indicators and stable isotope indicators in high-chloride mine water, making it difficult to effectively distinguish between similar water sources such as seawater, shallow weathered fissure water, and groundwater from different aquifers. Furthermore, these methods often lack sufficient consideration for different indicator dimensions, information redundancy, and the inherent dispersion of the water source itself, making them prone to misjudgment due to single chloride ion anomalies. Additionally, the lack of constraints on identification thresholds and unknown water source conditions means that even when the mine water source to be identified is close to or significantly deviates from all known sources, a single result may still be forced, leading to insufficient identification stability and reliability.

[0005] The present invention provides a method for identifying water sources in high-chloride gold mines based on principal component analysis, which specifically includes the following steps:

[0006] S1. Identify candidate water sources, collect regional background reference samples and known water source samples, obtain water chemical detection data and stable isotope detection data, and construct a water source identification index sequence; based on the regional background reference samples, process the known water source samples, calculate the characteristic response value of high chlorine water sources, and construct a high chlorine water source characteristic matrix.

[0007] S2. Perform principal component analysis on the feature matrix of high-chlorine water sources, extract effective principal components, construct the principal component feature centers and principal component discrete ranges for each candidate water source, and generate the identification margin threshold and the maximum allowable water source deviation threshold.

[0008] S3. Obtain the water source identification index of the mine water sample to be identified, calculate the characteristic response value of high chlorine water source and project it to the principal component space, calculate the water source deviation degree and water source consistency degree between it and each candidate water source; combine the maximum allowable water source deviation threshold and the identification margin threshold to identify the water source.

[0009] Preferably, S1 specifically includes:

[0010] The water chemistry detection data includes K + Na + Ca 2+ Mg 2+ Cl - SO4 2- HCO3 - CO3 2- and the mass concentration of total dissolved solids; the stable isotope detection data includes the parts per thousand deviation values ​​of stable isotopes of hydrogen, oxygen, and chlorine in water; based on water chemistry detection data, stable isotope detection data, and Na... + Cl - Mg 2+ and Ca 2+ Na mass concentration calculated + With Cl - mg equivalent concentration ratio, Mg 2+ With Ca 2+ The milligram equivalent concentration ratio and Ca 2+ with Na + The milligram equivalent concentration ratio was used to construct a sequence of water source identification indicators.

[0011] Preferably, S1 specifically includes:

[0012] Based on the regional background reference samples, calculate the regional background reference parameters; for all samples except Cl... -And Na + With Cl - In addition to the milligram equivalent concentration ratio, water source identification indicators are used. Combined with regional background reference parameters, the standardized shift of the water source identification indicators is calculated. Then, combined with the high-chlorine co-enhancing factor, the characteristic response values ​​of high-chlorine water sources for known water samples are calculated. For Cl... - And Na + With Cl - The milligram equivalent concentration ratio is used as the standardized offset of the water source identification index as the characteristic response value of high chlorine water source for known water source samples.

[0013] Preferably, S1 specifically includes:

[0014] The high-chlorine co-enhancing factor is constructed by calculating the positive high-chlorine anomaly degree of known water source samples and the anomaly degree of the water source identification index itself.

[0015] Preferably, S2 specifically includes:

[0016] Principal component analysis was performed on the characteristic matrix of high-chlorine water sources after centering, and the cumulative contribution rate was calculated. Based on the cumulative contribution rate, the effective principal components were determined, and the scores of the effective principal components were calculated.

[0017] Preferably, S2 specifically includes:

[0018] Known water source samples are grouped according to water source category. The mean and standard deviation of the effective principal component scores of each group of known water source samples are calculated and used as the principal component feature center and principal component dispersion range of the candidate water source, respectively. The identification margin threshold and the maximum allowable water source deviation threshold are determined by leave-one-out verification.

[0019] Preferably, S3 specifically includes:

[0020] Based on the various water source identification indicators of the mine water inflow sample to be identified, the characteristic response value and effective principal component score of the high chlorine water source are calculated according to the methods described in steps S1 and S2. Based on the difference between the effective principal component score and the principal component feature center of the candidate water source and the principal component dispersion range of the candidate water source, the degree of water source deviation of the mine water inflow sample to be identified relative to the candidate water source is calculated.

[0021] Preferably, S3 specifically includes:

[0022] Based on the degree of deviation of the mine water sample to be identified from the candidate water sources, the water source consistency is calculated, and the first and second candidate water sources are determined. Combining the maximum allowable water source deviation threshold and the identification margin threshold, the water source is identified, and the status of the main identified water source, the composite water source to be judged, or the unknown water source status is output.

[0023] The beneficial effects of the technical solution of the present invention are:

[0024] This invention unifies the hydrochemical characteristics, ion ratio characteristics, and stable isotope characteristics of different candidate water sources within a gold mine research area, and constructs a feature expression that reflects the comprehensive differences between different water sources by combining high-chloride anomaly states. This reduces misjudgments caused by relying solely on chloride ion content for water source identification. Principal component analysis extracts representative comprehensive characteristics between different water sources, reducing information duplication and dimensional differences among multiple detection indicators. It also establishes principal component feature centers and principal component dispersion ranges for each candidate water source, enabling quantitative comparison of mine water inflows to be identified with different candidate water sources within a unified feature space. Simultaneously, leave-one-out validation determines the identification margin threshold and the maximum allowable water source deviation threshold, and sets unknown water source states and composite water source pending judgment states to avoid forcibly giving a unique identification result when the water source characteristic differences are insufficient or exceed the known water source range. This improves the accuracy, stability, and interpretability of identifying high-chloride mine water sources in gold mines, providing a reliable basis for mine water inflow source analysis and subsequent water hazard prevention and control. Attached Figure Description

[0025] Figure 1 This is a flowchart of a method for identifying water sources in high-chlorine gold mines based on principal component analysis, as described in this invention. Detailed Implementation

[0026] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0027] 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 this invention pertains.

[0028] The following description, in conjunction with the accompanying drawings, details a specific scheme for identifying the source of inrush water in high-chlorine gold mines based on principal component analysis, provided by the present invention.

[0029] See attached document Figure 1 The diagram illustrates a flowchart of a method for identifying water sources in high-chlorine gold mines based on principal component analysis, according to an embodiment of the present invention. The method includes the following steps:

[0030] S1. Identify candidate water sources, collect regional background reference samples and known water source samples, obtain water chemical detection data and stable isotope detection data, and construct a water source identification index sequence; based on the regional background reference samples, process the known water source samples, calculate the characteristic response value of high chlorine water sources, and construct a high chlorine water source characteristic matrix.

[0031] Candidate water sources that may form mine water inflow within the gold mine study area were identified. Regional background reference samples and known water source samples were collected to obtain corresponding hydrochemical detection data and stable isotope detection data, and a fixed water source identification index sequence was established. The background reference state of each water source identification index was established using the regional background reference samples. The water source identification indexes of known water source samples were converted into high-chlorine water source characteristic response values, and a high-chlorine water source characteristic matrix was constructed.

[0032] In practice, the first step is to determine the types of water sources that may cause mine water inflow based on existing drilling data, hydrogeological profiles, distribution of aquifers and impermeable layers, fault structures, and groundwater recharge, runoff, and discharge conditions in the gold mine research area. The determined types of water sources are then used as candidate water sources, including shallow weathered fissure water, groundwater from different water-bearing aquifers, and seawater.

[0033] Water samples with identified sources used for source identification are collectively referred to as known source samples. For each candidate source, at least 10 known source samples should be obtained, and the source type, sampling location, sampling time, and corresponding hydrological period should be recorded. Shallow weathered fissure water collected from areas unaffected by mining drainage and seawater replenishment is collectively referred to as regional background reference samples, with at least 10 regional background reference samples. Regional background reference samples are only used to establish a background reference state and are not used as known source samples for candidate shallow weathered fissure water sources in the corresponding mining area. Mine water samples whose sources actually need to be identified are collectively referred to as mine water samples to be identified.

[0034] The same detection methods were used on regional background reference samples and known water source samples to obtain water chemical detection data; for example, atomic absorption spectrophotometry or inductively coupled plasma atomic emission spectrometry was used to obtain K. + Na + Ca 2+ Mg 2+ The mass concentration of Cl was obtained by ion chromatography. - SO4 2- The mass concentration of HCO3 was obtained by acid-base titration. - CO3 2-The mass concentration of total dissolved solids (TDS) is obtained by gravimetric method, referring to standard GB / T 5750.6, GB / T 5750.4, DZ / T 0064, and other conventional standards in this field; based on Na... + Cl - Mg 2+ and Ca 2+ Given the mass concentration and the equivalent mass of the corresponding ion, calculate the milliequivalent concentration of the ion. , , and And further calculate Na + With Cl - mg equivalent concentration ratio, Mg 2+ With Ca 2+ The milligram equivalent concentration ratio and Ca 2+ with Na + The milligram equivalent concentration ratios are denoted as follows: , and The milliequivalent concentration of the ion is obtained by dividing the ion mass concentration by the corresponding mass. To avoid an abnormal increase in the ratio due to the denominator ion concentration approaching zero under extreme operating conditions, all ions used as the denominator (i.e., Cl) are considered. - Ca 2+ Na + The milligram equivalent concentration is set with a minimum positive protection threshold. (For example That is, when the denominator is less than When, take the denominator as At the same time, an upper limit threshold is set for the milligram equivalent concentration ratio. (e.g., 100), if it exceeds, take .

[0035] Simultaneously acquire stable isotope detection data, including , and ,in, , , These represent the per-thousandth deviations of stable isotopes of hydrogen, oxygen, and chlorine in water, used to characterize the isotopic features of water bodies from different sources. The per-thousandth deviations of hydrogen and oxygen stable isotopes are based on the Vienna Standard Mean Ocean Water (VSMOW), while the per-thousandth deviations of chlorine stable isotopes are based on the Standard Mean Ocean Chloride (SMOC). These reference standards are all well-known in the field. The same reference standard is used for regional background reference samples, known water source samples, and samples of mine water to be identified. During testing... and Detection was performed using wavelength scanning-cavity ring-down spectroscopy. Detection was performed using thermal ionization mass spectrometry. K... + Na + Ca 2+ Mg 2+ Cl - SO4 2- HCO3 - CO3 2- TDS , , , , and A sequence of water source identification indicators is formed in a fixed order.

[0036] Let the total number of known water source samples be... The known water source sample number is , Let the serial number of the water source identification index be... , , For the total number of water source identification indicators, the first The first known water source sample The water source identification index value is Calculate the first using background reference samples from the entire region Average value of water source identification indicators and standard deviation Cl of all region background reference samples - The average and standard deviation of the mass concentration are denoted as . and , , , , This serves as the regional background reference parameter. Based on Z-score normalization, interaction effect characteristic construction, and hydrogeochemical ion ratio analysis, the characteristic response values ​​of high-chlorine water sources are calculated; for the first... Water source identification index, when the water source identification index is excluding Cl - and When identifying water sources other than those specified, the characteristic response value of high-chlorine water sources is calculated using the following formula:

[0037]

[0038] in, Indicates the first The first known water source sample The characteristic response values ​​of high-chlorine water sources corresponding to the water source identification indicators; Indicates the first Cl from a known water source sample - mass concentration; This represents a very small positive number that is much smaller than the background standard deviation of the corresponding water source identification index. It only serves to prevent the denominator from being zero when the background standard deviation approaches zero. In this embodiment, for water chemical indicators, Values For stable isotope indices, Values The three milligram equivalent concentrations were compared to the corresponding values. Based on water chemistry indicators, the denominator of the positive high chlorine anomaly term is related to... Added Belongs to Cl - The mass concentration value is always taken according to the water chemistry index and does not change with the [number of items]. The category of the water source identification indicator has changed; Indicates the first The first known water source sample The standardized offset of the water source identification index relative to the background reference state; Indicates the first The degree of positive high chlorine anomaly in a known water source sample, and only when At that time, this item participated in the combined enhancement of high chlorination. At that time, the abnormality level item takes a value of 0; Indicates the first The first known water source sample The degree of abnormality of the water source identification indicators themselves; Cl - Abnormalities and the first The degree of synergy in the simultaneous occurrence of abnormal water source identification indicators; This indicates a high-chlorine combined enhancement factor.

[0039] When the The water source identification index is Cl - or At that time, only the standardized offset is calculated. As a characteristic response value of high chlorine water source Do not execute Cl again - The combined effect of high chlorine levels on mass concentration and other indicators is enhanced. For , and The combined enhancement of high chlorine only indicates the numerical characteristic relationship between the stable isotope and the high chlorine state and their deviation from the background reference state; it does not indicate the relationship between Cl and the high chlorine state. - There is a direct causal relationship between changes in mass concentration and stable isotopes.

[0040] Characteristic response values ​​of high-chlorine water sources from all known water source samples A high-chlorine water source feature matrix is ​​constructed by arranging known water source samples as rows and water source identification indicators as columns.

[0041] S2. Perform principal component analysis on the feature matrix of high-chlorine water sources, extract effective principal components, construct the principal component feature centers and principal component discrete ranges for each candidate water source, and generate the identification margin threshold and the maximum allowable water source deviation threshold.

[0042] The high-chlorine water source feature matrix obtained in step S1 is centered and subjected to principal component analysis. Known water source samples are mapped to a unified principal component space. The principal component feature center and principal component discrete range of each candidate water source are established, and the threshold required for water source identification is obtained through leave-one-out verification.

[0043] In practical implementation, the characteristic matrix of high-chlorine water sources is calculated. The average value of the characteristic response values ​​of high-chlorine water sources from all known water source samples. The characteristic response values ​​of each high-chlorine water source Subtract the column average of the characteristic response values ​​of high-chlorine water sources The characteristic response values ​​of the centralized high-chlorine water source were obtained and uniformly denoted as... Based on the characteristic response values ​​of all centralized high-chlorine water sources A centralized high-chlorine water source feature matrix is ​​constructed, and the covariance matrix is ​​calculated. The obtained covariance matrix is ​​then decomposed into eigenvalues ​​to obtain eigenvalues ​​and corresponding unit eigenvectors. The eigenvalues ​​are arranged in descending order, and the eigenvalues ​​in the order listed are... The eigenvalue of a bit is denoted as The corresponding unit eigenvector of the first Each component is denoted as , The cumulative contribution rate is calculated according to the above arrangement order. When the cumulative contribution rate first reaches 85%, the current and previous principal components are determined as effective principal components. Principal components with eigenvalues ​​equal to 0 are not included in the effective principal components. The number of effective principal components is denoted as [missing information]. , No. One effective principal component is The main components, , Right now time , .

[0044] Furthermore, calculate the first... The known water source sample at the ... Scores on each effective principal component:

[0045]

[0046] in, Indicates the first The known water source sample at the ... Principal component scores on each effective principal component; Indicates the first The th unit eigenvector corresponding to the th effective principal component One component; This represents the total number of water source identification indicators.

[0047] The column average of the characteristic response values ​​of high-chlorine water sources determined from all known water source samples. Number of effective principal components , No. Eigenvalues ​​corresponding to each effective principal component and its eigenvector components They are used together as parameters in the principal component model.

[0048] Based on the pre-recorded water source category for each known water source sample, known water source samples belonging to the same candidate water source are grouped together. For the first... Candidate water sources of category 1, selecting all water source categories as the first category. Given known water source samples from candidate water sources, calculate the values ​​of these known water source samples in the first... Principal component scores on each effective principal component The average value is calculated and the resulting average value is denoted as the first... Candidate water sources in the first category Principal component feature centers on each effective principal component ; Calculate the first known water source sample in the same group on the [number]th [day]. Principal component scores on each effective principal component The standard deviation of is calculated, and the obtained standard deviation is denoted as the first standard deviation. Candidate water sources in the first category The discrete range of principal components on each effective principal component When the first Candidate water sources in the first category Principal component scores on each effective principal component When the standard deviation is 0, calculate the number of known water source samples in the th case. Principal component scores on each effective principal component The standard deviation of the principal component is used, and the larger of this standard deviation and the preset minimum positive number is taken as the discrete range of the principal component. Thus ensuring The preset minimum positive number is set according to the order of magnitude of the principal component scores, and in this embodiment, the value is 0.01.

[0049] After completing the principal component analysis and establishing the principal component features of candidate water sources using the above method, leave-one-out validation is performed to obtain the principal component model. In each round, one sample is selected from all known water source samples as a validation sample, and the principal component model for the current round is rebuilt using the remaining known water source samples in the same way as described above. Using the principal component model for the current round of validation, water source identification is performed on the selected validation sample. The water source deviation and water source consistency of the validation sample relative to each candidate water source are calculated according to the method described in step S3, and the first candidate water source and the second candidate water source are determined. If the first candidate water source is consistent with the actual water source category of the validation sample, the validation sample is determined to be correctly identified.

[0050] After completing all leave-one-out validations, the number of correctly identified samples and the correct identification rate are counted. During the leave-one-out validation process, the candidate water source category set remains consistent with the actual water source identification in each round of validation. Only the current validation sample is removed from the candidate water source to which it belongs, and the remaining samples are used to build the principal component model for that round without deleting any candidate water source categories. When the number of correctly identified validation samples is not less than 10 and the correct identification rate is not less than 80%, for each correctly identified validation sample, the water source consistency difference between its first and second candidate water sources is calculated, and the 5th percentile of the water source consistency difference of all correctly identified validation samples is taken as the identification margin threshold. Simultaneously, the deviation of correctly identified verification samples from their actual water sources is recorded, and the 95th percentile of all deviations is taken as the maximum permissible water source deviation threshold. The degree of deviation of the verification sample from its actual water source refers to the deviation of the verification sample from its actual water source category, calculated according to the method described in step S3. Value; when the correctly identified verification sample does not meet the above conditions, it is determined that the current known water source samples are insufficient to stably identify the water source. It is necessary to supplement the known water source samples of the candidate water source to which the incorrectly identified verification sample belongs and then re-perform principal component analysis and leave-one-out verification.

[0051] S3. Obtain the water source identification index of the mine water sample to be identified, calculate the characteristic response value of high chlorine water source and project it to the principal component space, calculate the water source deviation degree and water source consistency degree between it and each candidate water source; combine the maximum allowable water source deviation threshold and the identification margin threshold to identify the water source.

[0052] The various water source identification indicators of the mine water sample to be identified were obtained. The hydrochemical and stable isotope detection data were obtained using the same detection methods as those for known water source samples. The three milligram equivalent concentration ratios were calculated using the same methods and the same minimum positive protection thresholds as those for known water source samples. and upper limit threshold Calculate and give the first The water source identification index value is recorded as follows: Cl - The mass concentration is denoted as ,in, The identifier for the mine water sample to be identified is compared with the known water source sample number. They are independent of each other; when there are multiple mine water inflow samples to be identified... , This represents the total number of mine water samples to be identified. The calculation formula for the characteristic response value of high-chlorine water sources in step S1 is used. , Replace each , Regional background reference parameters and Using the values ​​obtained in step S1, the first value of the mine water inflow sample to be identified is calculated. High-chlorine water source characteristic response values ​​corresponding to the water source identification indicators , among which, the The water source identification index is Cl - or hour, Only the standardized offset is taken; according to the calculation formula of the effective principal component score in step S2, the column average value in the principal component model parameters is used. and eigenvector components ,Will minus Later and Multiplication and pair Summing, we obtain the mine water inflow sample to be identified on the 1st... Principal component scores on each effective principal component .

[0053] Furthermore, the calculation of the mine water inflow sample to be identified relative to the first... Degree of deviation of candidate water sources:

[0054]

[0055] in, This indicates that the mine water sample to be identified is relative to the first... The degree of deviation of candidate water sources; For the first Candidate water sources in the first category Principal component feature centers on each effective principal component For the first Candidate water sources in the first category The discrete range of principal components on each effective principal component; This indicates that the mine water sample to be identified is in the [number]th [year]. The effective principal component is relative to the first The degree of standardized deviation of the discrete range of the principal components of the candidate water source itself; Indicates the first The eigenvalues ​​corresponding to each effective principal component. Degree of deviation from water source. The smaller the value, the closer the mine water sample to be identified is to the first... Candidate water sources.

[0056] The water source consistency is calculated based on the degree of deviation of each candidate water source, and the water source identification result is output by combining the maximum allowable water source deviation threshold and the identification margin threshold. The specific calculation is as follows:

[0057]

[0058] in, For the mine water sample to be identified relative to the first The water source consistency of the candidate water sources is within the range of [value range missing]. ; denominator Add 1 to make When the value is zero, the water source consistency takes a finite maximum value of 1. The smaller the deviation of the water source, the greater the water source consistency.

[0059] According to the consistency of water source The water sources are sorted from largest to smallest, with the largest candidate source designated as the first candidate source, and the second largest as the second candidate source. This is based on the consistency of the water sources. Depending on the degree of deviation from the water source Monotonically decreasing, the first candidate water source also corresponds to the minimum water source deviation. .

[0060] when When, output "Unknown water source status"; when At that time, the difference in water source consistency between the first candidate water source and the second candidate water source is calculated as the identification margin. .like The first candidate water source will be used as the primary identification source; if The system outputs the first and second candidate water sources and marks them as "composite water source pending judgment". The "composite water source pending judgment" status only indicates that the current water source identification indicators are insufficient to reliably distinguish the two candidate water sources, and does not indicate that the two types of water sources are necessarily actually mixed.

[0061] The final output includes the sample number of the mine water inflow to be identified, the sampling location, the sampling time, the hydrological period, the degree of deviation and consistency of each candidate water source, the first candidate water source, the second candidate water source, the minimum degree of deviation of the water source, the identification margin, and the final water source identification status.

[0062] In summary, a method for identifying water sources in high-chlorine gold mines based on principal component analysis has been developed.

[0063] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for identifying water sources in high-chloride gold mines based on principal component analysis, characterized in that, Specifically, the following steps are included: S1. Identify candidate water sources, collect regional background reference samples and known water source samples, obtain water chemical detection data and stable isotope detection data, and construct a water source identification index sequence; based on the regional background reference samples, process the known water source samples, calculate the characteristic response value of high chlorine water sources, and construct a high chlorine water source characteristic matrix. S2. Perform principal component analysis on the feature matrix of high-chlorine water sources, extract effective principal components, construct the principal component feature centers and principal component discrete ranges for each candidate water source, and generate the identification margin threshold and the maximum allowable water source deviation threshold. S3. Obtain the water source identification index of the mine water sample to be identified, calculate the characteristic response value of high chlorine water source and project it to the principal component space, calculate the water source deviation degree and water source consistency degree between it and each candidate water source; combine the maximum allowable water source deviation threshold and the identification margin threshold to identify the water source.

2. The method for identifying water sources in high-chloride gold mines based on principal component analysis according to claim 1, characterized in that, S1 specifically includes: The water chemistry detection data includes K + Na + Ca 2+ Mg 2+ Cl - SO4 2- HCO3 - CO3 2- and the mass concentration of total dissolved solids; the stable isotope detection data includes the parts per thousand deviation values ​​of stable isotopes of hydrogen, oxygen, and chlorine in water; based on water chemistry detection data, stable isotope detection data, and Na... + Cl - Mg 2+ and Ca 2+ Na mass concentration calculated + With Cl - mg equivalent concentration ratio, Mg 2+ With Ca 2+ The milligram equivalent concentration ratio and Ca 2+ with Na + The milligram equivalent concentration ratio was used to construct a sequence of water source identification indicators.

3. The method for identifying water sources in high-chloride gold mines based on principal component analysis according to claim 2, characterized in that, S1 specifically includes: Based on the regional background reference sample, calculate the regional background reference parameters; for all samples except Cl... - And Na + With Cl - In addition to the milligram equivalent concentration ratio, water source identification indicators are used. Combined with regional background reference parameters, the standardized shift of the water source identification indicators is calculated. Then, combined with the high-chlorine co-enhancing factor, the characteristic response values ​​of high-chlorine water sources for known water samples are calculated. For Cl... - And Na + With Cl - The milligram equivalent concentration ratio is used as the standardized offset of the water source identification index as the characteristic response value of high chlorine water source for known water source samples.

4. The method for identifying water sources in high-chloride gold mines based on principal component analysis according to claim 3, characterized in that, S1 specifically includes: The high-chlorine co-enhancing factor is constructed by calculating the positive high-chlorine anomaly degree of known water source samples and the anomaly degree of the water source identification index itself.

5. The method for identifying water sources in high-chloride gold mines based on principal component analysis according to claim 1, characterized in that, S2 specifically includes: Principal component analysis was performed on the characteristic matrix of high-chlorine water sources after centering, and the cumulative contribution rate was calculated. Based on the cumulative contribution rate, the effective principal components were determined, and the scores of the effective principal components were calculated.

6. The method for identifying water sources in high-chloride gold mines based on principal component analysis according to claim 5, characterized in that, S2 specifically includes: Known water source samples are grouped according to water source category. The mean and standard deviation of the effective principal component scores of each group of known water source samples are calculated and used as the principal component feature center and principal component dispersion range of the candidate water source, respectively. The identification margin threshold and the maximum allowable water source deviation threshold are determined by leave-one-out verification.

7. The method for identifying water sources in high-chloride gold mines based on principal component analysis according to claim 5, characterized in that, S3 specifically includes: Based on the various water source identification indicators of the mine water inflow sample to be identified, the characteristic response value and effective principal component score of the high chlorine water source are calculated according to the methods described in steps S1 and S2. Based on the difference between the effective principal component score and the principal component feature center of the candidate water source and the principal component dispersion range of the candidate water source, the degree of water source deviation of the mine water inflow sample to be identified relative to the candidate water source is calculated.

8. The method for identifying water sources in high-chloride gold mines based on principal component analysis according to claim 7, characterized in that, S3 specifically includes: Based on the degree of deviation of the mine water sample to be identified from the candidate water sources, the water source consistency is calculated, and the first and second candidate water sources are determined. Combining the maximum allowable water source deviation threshold and the identification margin threshold, the water source is identified, and the status of the main identified water source, the composite water source to be judged, or the unknown water source status is output.

Images