Monitoring equipment for depositing polymetallic deposit tailing pollutants

Abstract

The invention provides monitoring equipment for sedimentary polymetallic deposit tailing pollutants, and relates to the technical field of metal ore pollutant monitoring, and the monitoring equipment comprises a data collection module for collecting historical data and real-time data, a model training module for constructing a pollutant concentration prediction model, and a pollution correction module. The system comprises a data acquisition module used for acquiring a corrected radioactivity intensity data set of each position, an overall radioactivity intensity correction module used for acquiring an overall radioactivity intensity correction coefficient, and a region early warning module used for carrying out early warning on a region around an ore deposit tail. And then the change condition is further corrected according to the diffusivity, the radioactivity condition in the overall range is judged in combination with high-altitude radioactivity data, the risk is judged according to the diffusion condition and the surrounding local condition, and the prediction is more accurate and accords with the actual condition.

Description

Technical Field

[0001] The present invention relates to the technical field of metal mine pollutant monitoring, and particularly to a monitoring device for tailing pollutants of sedimentary polymetallic deposits. Background Technique

[0002] Black shale-type polymetallic deposits contain various useful metal elements, with large reserves and wide distribution, which are of great significance for supporting the economic construction and resource security of our country. However, it should also be noted that the ore-forming elements of this type of deposit are complex and the development pollution is large. In particular, the content of heavy metal elements and radioactive elements such as U and Th is relatively high, seriously threatening production safety, having a high environmental risk, and affecting the economic potential of the mine. There is an urgent need for a reliable and effective environmental monitoring facility to detect and control early. At present, there are the following problems in the monitoring and pollution prediction means for pollutants in the tailing area: First, it is difficult for the existing technology to collect pollutant data in a timely and comprehensive manner. Especially under complex meteorological conditions, it is difficult to accurately grasp the diffusion law of pollutants. Second, there is a lack of a systematic correction mechanism, and the influence of meteorological conditions (such as temperature, rainfall) on pollutant diffusion is not fully considered, resulting in insufficient accuracy of pollutant concentration prediction results. Third, the existing early warning mechanism cannot effectively combine the regional environmental carrying capacity and population density for zonal assessment, and it is difficult to achieve targeted early warning for high-risk areas. Therefore, there is an urgent need for a technical means that can comprehensively and accurately monitor and predict the diffusion of pollutants in the tailing area of the deposit, and at the same time achieve effective early warning to reduce environmental risks and health hazards.

[0003] In the prior art, the publication number CN202411131848.8 discloses a multi-source mine water quality prediction method, device, equipment, medium and product. The method includes obtaining the water source of the working face affected by mining fissures; collecting samples of the water source of the working face according to the set sample collection time interval; analyzing the collected water source samples by using the pearson algorithm to obtain the dataset of the main characteristic indexes of the mine water quality monitored in each source layer of the water source of the working face; and based on the dataset of the main characteristic indexes of the mine water quality monitored in each source layer, using the trained ant colony optimization algorithm-convolutional neural network-long short-term memory prediction model to predict the water quality of each source layer to obtain the water quality prediction data of each source layer.

[0004] Although the above information disclosed in the background art part realizes the monitoring of substances in the mining area, it only considers the detection within a fixed range and does not consider the influence brought by diffusion, resulting in insufficient practical significance. Summary of the Invention

[0005] The purpose of the present invention is to provide a monitoring device for tailing pollutants of sedimentary polymetallic deposits to solve the problems raised in the above background art.

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

[0007] A monitoring device for depositing tailings pollutants in polymetallic ore deposits, specifically including:

[0008] A data collection module that collects historical and real-time data of pollution data and meteorological data at multiple positions in the ore deposit tailings, and respectively forms a historical data set and a real-time data set. The historical data set and the real-time data set are respectively preprocessed. The high-altitude historical radioactive intensity of the ore deposit tailings area is obtained through a remote sensing satellite and a high-altitude radioactive intensity historical data set is formed. The meteorological prediction data of the ore deposit tailings area is obtained through a weather station, and the population density and the maximum allowable radioactive intensity of the surrounding area of the ore deposit tailings are obtained;

[0009] A model training module that trains a pollutant concentration prediction model by using the historical data set as the training set and the radioactive intensity of the historical data set as the label, and constructs a pollutant concentration prediction model;

[0010] A pollution correction module that inputs the real-time data set into the trained pollutant concentration prediction model, obtains the prediction results of the radioactive intensity at each position and forms a radioactive intensity prediction data set, and corrects the prediction results of the radioactive intensity by constructing a radioactive intensity correction formula based on the meteorological prediction data and forms a corrected radioactive intensity data set;

[0011] An overall radioactive intensity correction module that extracts the radioactive intensity characteristics from all historical data sets, processes them with the high-altitude historical radioactive intensity in the high-altitude radioactive intensity historical data set, and forms an overall radioactive intensity correction coefficient;

[0012] A regional warning module that divides the area around the ore deposit tailings, obtains the radioactive intensity of each partition according to the overall radioactive intensity correction coefficient, the diffusion principle and the corrected radioactive intensity set, then obtains the risk value of each partition according to the maximum radioactive intensity and the population density, and issues a warning for each partition according to the risk value.

[0013] Further, historical data of pollution data are collected by evenly arranging pollution detectors in the area where the tail of the polymetallic deposit is located. The coverage range of the pollution detectors is larger than the area occupied by the deposit tail. The pollution data include the number of pollutant types, the radioactive intensity, and the concentration of heavy metal pollutants. Each pollution detector is numbered in matrix form. Taking the pollution detector at the corner point as the origin, the positive directions of the row and the column are set respectively, and the position of each pollution detector is represented by a two-dimensional vector composed of the row and the column. Each detector collects pollution data at equal time intervals and adds a timestamp. Meteorological monitors are arranged around each pollution detector to collect historical data of meteorological data. The meteorological monitors use the same numbering as the pollution detectors. The meteorological data include wind speed, wind direction, temperature, and rainfall. Each meteorological monitor collects meteorological data at equal time intervals and adds a timestamp. The historical data set of meteorological data is aligned with the historical data set of pollution data in time;

[0014] The historical data sets of pollution data and meteorological data are aggregated into a historical data set, and preprocessing and normalization are performed. The radioactive intensity detected by each pollutant is extracted as a feature label, and the historical data set is input into a long short-term memory network for training. The trained model is calibrated as a pollutant concentration prediction model.

[0015] Further, the pollution detectors and meteorological monitors continuously collect real-time data of pollution data and real-time data of meteorological data to form a real-time data set and input it into the pollutant concentration prediction model. The radioactive intensity at the location of each pollution detector is obtained and a radioactive intensity prediction data set is formed. The formula is as follows:

[0016] FSYQD i，j ={YFd 1 ,YFd 2 ,YFd 3 ,…,YFd t ,…,YFd n}

[0017] Wherein, FSYQD i，j is the radioactive intensity prediction data set of the pollution detector numbered i, j. i is the row number of the pollution detector, j is the column number of the pollution detector, i ∈ N, 1 ≤ i ≤ n, j ∈ N, 1 ≤ j ≤ m, and YFd t is the predicted value of the radioactive intensity at the t-th moment. t is the moment retrieval variable, t ∈ N, 1 ≤ t ≤ K.

[0018] Further, meteorological prediction data of the deposit tail area are obtained through a weather station. The meteorological prediction data include wind speed, wind direction, temperature, and rainfall. The predicted values of the radioactive intensity of each pollution detector are corrected according to the meteorological data. The formula for radioactive intensity correction is as follows:

[0019]

[0020] Among them, YFd t ′ is the corrected radioactive intensity at time t, and YFd t is the predicted value of the radioactive intensity at time t, v wind-t is the wind speed at time t, θ wind-t is the wind direction at time t, T t is the temperature at time t, T 0 is the temperature at the current time, P t is the rainfall at time t, β T is the temperature correction coefficient, β P is the rainfall correction coefficient, P is the rainfall, f(T t , P t ) is the correction function considering the effects of temperature and rainfall on pollutant diffusion. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K, Q t is the predicted value of the radioactive intensity of the pollution source, and the pollution source is the previous pollutant detector in the wind direction, σ y is the diffusion coefficient of the wind direction relative to the row direction of the pollutant detector, σ x is the diffusion coefficient of the wind direction relative to the column direction of the pollutant detector, σ y 2 + σ x 2 = 1;

[0021] The formula on which the correction function considering the effects of temperature and rainfall on pollutant diffusion is based is as follows:

[0022] f(T, P) = α1 T · T + α2 P · P + β1 T · T 2 + β2 P · P 2

[0023] Among them, T is the temperature, P is the rainfall, α1 T , α2 P and β1 T , β2 P are the linear and quadratic correction coefficients of temperature and rainfall;

[0024] Obtain the temperature characteristics and rainfall characteristics from the meteorological dataset, substitute the historical data of the temperature characteristics and rainfall characteristics into the formula on which the correction function considering the effects of temperature and rainfall on pollutant diffusion is based, and obtain α1 T , α2 P and β1 T , β2 P, so as to obtain the correction function of temperature and rainfall on pollutant diffusion;

[0025] The temperature correction coefficient β T and the rainfall correction coefficient β P The acquisition logic is as follows: obtain the radioactive intensity data characteristics in the pollution dataset, obtain the temperature characteristics and humidity characteristics in the meteorological dataset, converge the temperature characteristics and humidity characteristics into a coefficient dataset, add the radioactive intensity characteristics as labels into the coefficient dataset, and input the coefficient dataset into a regression analysis model for analysis. The formula is as follows:

[0026] C t = β 0 + β T ·T t + β P ·P t + ∈

[0027] where C t is the radioactive intensity at time t, T t is the temperature at time t, P t is the rainfall at time t, t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K, β 0 is the initial coefficient, β T is the temperature correction coefficient, β P is the humidity correction coefficient, ∈ is the error term;

[0028] Obtain the temperature correction coefficient β T and the humidity correction coefficient β P through regression analysis, and substitute them into the formula for radioactive intensity correction;

[0029] Sort out the corrected radioactive intensities of each pollutant detector to form a corrected radioactive intensity dataset. The formula is as follows:

[0030]

[0031] where FD i，j is the corrected radioactive intensity dataset of the pollutant detector numbered i, j. i is the row number of the pollutant detector, j is the column number of the pollutant detector, i ∈ N, 1 ≤ i ≤ n, j ∈ N, 1 ≤ j ≤ m, Fd t is the corrected radioactive intensity value at the t-th moment. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K;

[0032] Summarize the corrected radioactive intensity datasets of all pollutant detectors to form the total radioactive intensity prediction dataset FQ in the tail area of the ore deposit. The formula is as follows:

[0033] FQ = {Fq 1 , Fq2 , Fq 3 , …, Fq t , …, Fq n}

[0034] Among them, Fq t is the overall radiation intensity in the area at time t, where t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ n.

[0035] Furthermore, radiation intensity features are extracted from the historical data of the pollutant datasets of all pollutant detectors, and the basis formula is as follows:

[0036]

[0037] Among them, Lfq t is the overall historical radiation intensity in the area at time t, and lfq i，j，t is the historical radiation intensity data recorded by the pollutant monitors numbered i and j in the area at time t. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K, i is the row number of the pollutant detector, j is the column number of the pollutant detector, i ∈ N, 1 ≤ i ≤ I, j ∈ N, 1 ≤ j ≤ J;

[0038] The overall historical radiation intensity data in the tail area of the ore deposit is aggregated to form an overall historical radiation intensity dataset;

[0039] The location of the tail area of the ore deposit is obtained through the geographic information system, and the high-altitude historical radiation intensity data of the tail area of the ore deposit is obtained through a remote sensing satellite to form a high-altitude historical radiation intensity dataset. The overall radiation intensity correction coefficient is obtained from the high-altitude historical radiation intensity dataset obtained by the remote sensing satellite, and the basis formula is as follows:

[0040]

[0041] Among them, τ is the overall radiation intensity correction coefficient, and Wfq t is the high-altitude historical radiation intensity obtained by the remote sensing satellite at time t, and Lfq t is the overall historical radiation intensity in the area at time t. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ n, and n is the number of samples.

[0042] Furthermore, the surrounding location division logic is as follows: A circle is drawn with the location of the tail area of the ore deposit as the center, and the ring excluding the tail area of the ore deposit is equally divided into eight parts to form 8 partitions. The radiation intensity data in the eight partitions are calculated respectively according to the diffusion principle, and the basis formula is as follows:

[0043]

[0044] Among them: BQD i，tis the radioactive intensity of the partition numbered i at time t, Fq t is the overall radioactive intensity in the area at time t, σ y is the diffusion coefficient of the wind direction relative to the row direction of the pollutant detector, σ x is the diffusion coefficient of the wind direction relative to the column direction of the pollutant detector, σ y 2 +σ x 2 = 1, where R is the distance from the center of the partition to the center of the tail area of the ore deposit;

[0045] Obtain the maximum allowable radioactive intensity data and population density data stipulated by local regulations through the network, and obtain the environmental carrying capacity of the tail area of the ore deposit. The formula is as follows:

[0046]

[0047] where C ca-i，t is the carrying capacity of the partition numbered i at time t, C max is the maximum allowable radioactive intensity stipulated by local environmental regulations, BQD i，t is the radioactive intensity of the partition numbered i at time t;

[0048] Calculate the risk value for each partition respectively. The formula is as follows:

[0049] R i，t = w 1 ·C ca-i，t -1 + w 2 ·C cap-i

[0050] where: R i，t is the risk value of the partition numbered i at time t, C cap-i is the population density of the partition numbered i, w 1 , w 2 are weights respectively, w 1 2 + w 2 2 = 1;

[0051] Set the risk value threshold. When the risk value in the ith partition at time t exceeds the threshold, a risk warning is issued.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] The present invention first trains a pollution prediction model using historical data, obtains the change status of radioactive pollution within the tail area of the ore deposit through the model, then further corrects the change status according to diffusivity, and combines the high-altitude radioactive data to judge the radioactive status within the overall range. Again, it judges the risk according to the diffusion situation and the actual local conditions, achieving more accurate prediction and conforming to the actual situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a schematic diagram of the overall structure of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with specific embodiments.

[0056] It should be noted that unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meaning understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second", and similar terms used in the present invention do not indicate any order, quantity, or importance, but are only used to distinguish different components. The terms such as "comprising" or "including" mean that the elements or objects appearing before this term cover the elements or objects listed after this term and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0057] Embodiment:

[0058] Please refer to Figure 1 , the present invention provides a technical solution:

[0059] A monitoring device for depositing tailings pollutants of a polymetallic ore deposit, specifically including:

[0060] A data collection module that collects historical data and real-time data of pollution data and meteorological data at multiple locations in the tail of the ore deposit, and respectively forms a historical data set and a real-time data set. The historical data set and the real-time data set are respectively preprocessed, and the high-altitude historical radioactive intensity of the tail area of the ore deposit is obtained through a remote sensing satellite and a high-altitude radioactive intensity historical data set is formed. The meteorological prediction data of the tail area of the ore deposit is obtained through a weather station, and the population density and the maximum allowable radioactive intensity of the surrounding area of the tail of the ore deposit are obtained;

[0061] A model training module trains a pollutant concentration prediction model by using a historical data set as a training set and the radioactive intensity of the historical data set as a label, and constructs a pollutant concentration prediction model;

[0062] A pollution correction module is used to input a real-time data set into the trained pollutant concentration prediction model, obtain the prediction results of the radioactive intensity at each location and form a radioactive intensity prediction data set, and correct the prediction results of the radioactive intensity by constructing a radioactive intensity correction formula based on meteorological prediction data to form a corrected radioactive intensity data set;

[0063] An overall radioactive intensity correction module is used to extract the radioactive intensity characteristics from all historical data sets, and process them with the high-altitude historical radioactive intensity in the high-altitude radioactive intensity historical data set to form an overall radioactive intensity correction coefficient;

[0064] A regional warning module is used to divide the area around the tail of the ore deposit into regions, obtain the radioactive intensity of each region according to the overall radioactive intensity correction coefficient, the diffusion principle and the corrected radioactive intensity set, then obtain the risk value of each region according to the maximum radioactive intensity and the population density, and issue a warning to each region according to the risk value.

[0065] As a preferred embodiment, pollutant detectors are evenly arranged in the area where the tail of the polymetallic ore deposit is located to collect historical data of pollution data. The coverage range of the pollutant detectors is larger than the area occupied by the tail of the ore deposit. The coverage range of the pollutant detectors encompasses all areas of the tail of the ore deposit and the boundary of the coverage range of the pollutant detectors is equidistant from the boundary of the area occupied by the tail of the ore deposit. The pollution data includes the number of pollutant types, radioactive intensity and heavy metal pollutant concentration. Each pollutant detector is numbered in matrix form. Taking the pollutant detector at the corner point as the origin, the positive directions of the row and the column are set respectively, and the position of each pollutant detector is represented by a two-dimensional vector composed of the row and the column. Each detector collects pollutant data at equal time intervals and adds a time stamp. Meteorological monitors are arranged around each pollutant detector to collect historical data of meteorological data. The meteorological monitors use the same numbering as the pollutant detectors. The meteorological data includes wind speed, wind direction, temperature and rainfall. Each meteorological monitor collects meteorological data at equal time intervals and adds a time stamp, and aligns the historical data set of meteorological data with the historical data set of pollutant data in terms of time;

[0066] The historical data sets of pollution data and meteorological data are summarized into a historical data set, preprocessed and normalized. The radioactive intensity detected by each pollutant is extracted as a feature label, and the historical data set is input into a long short-term memory network for training. The trained model is calibrated as a pollutant concentration prediction model.

[0067] Pollutant data and meteorological data are both time - series data. They are continuous and dependent over time. The short - term memory network can effectively handle the long - distance dependence problem in time series. A well - trained long short - term memory network can generalize well to unseen data and predict the trend of future pollutant concentrations.

[0068] As a preferred embodiment, pollutant detectors and meteorological monitors continuously collect real - time data of pollution data and real - time data of meteorological data to form a real - time data set and input it into the pollutant concentration prediction model, obtaining the radioactive intensity at the location of each pollutant detector and forming a radioactive intensity prediction data set. The formula is as follows:

[0069] FSYQD i，j ={YFd 1 ,YFd 2 ,YFd 3 ,…,YFd t ,…,YFd n}

[0070] Among them, FSYQD i，j is the radioactive intensity prediction data set of pollutant detectors numbered i and j. i is the row number of the pollutant detector, j is the column number of the pollutant detector, i ∈ N, 1 ≤ i ≤ n, j ∈ N, 1 ≤ j ≤ m, and YFd t is the predicted radioactive intensity value at the t - th moment. t is the moment retrieval variable, t ∈ N, 1 ≤ t ≤ K.

[0071] As a preferred embodiment, the meteorological prediction data of the tail area of the ore deposit is obtained through a meteorological station. The meteorological prediction data includes wind speed, wind direction, temperature, and rainfall. The predicted radioactive intensity values of each pollutant detector are corrected according to the meteorological data. The formula for radioactive intensity correction is as follows:

[0072]

[0073] Among them, YFd t ′ is the corrected radioactive intensity at the t - th moment, YFd t is the predicted radioactive intensity value at the t - th moment, v wind-t is the wind speed at the t - th moment, θ wind-t is the wind direction at the t - th moment, T t is the temperature at the t - th moment, T 0 is the temperature at the current moment, P t is the rainfall at the t - th moment, β T is the temperature correction coefficient, β P is the rainfall correction coefficient, P is the rainfall, f(T t ,P t) is a correction function considering the influence of temperature and rainfall on pollutant diffusion. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K, Q t is the predicted emission intensity of the pollution source, and the pollution source is the previous pollutant detector in the wind direction, σ y is the diffusion coefficient of the wind direction relative to the row direction of the pollutant detector, σ x is the diffusion coefficient of the wind direction relative to the column direction of the pollutant detector, σ y 2 +σ x 2 = 1;

[0074] Temperature and humidity affect the diffusion of pollutants, while wind direction, rainfall, and diffusion phenomenon are important factors affecting the radioactive intensity in the tail area of the ore deposit.

[0075] The formula on which the correction function considering the influence of temperature and rainfall on pollutant diffusion is based is as follows:

[0076] f(T, P) = α1 T ·T + α2 P ·P + β1 T ·T 2 + β2 P ·P 2

[0077] where T is the temperature, P is the rainfall, α1 T , α2 P and β1 T , β2 P are the linear and quadratic correction coefficients of temperature and rainfall;

[0078] Both temperature and rainfall affect the diffusion of pollutants. As the temperature rises, the change rate of radioactive substances in pollutants increases, leading to an increase in radioactive intensity. Excessive rainfall causes the effects of radioactivity to re-converge in the environment with rainwater.

[0079] Obtain the temperature characteristics and rainfall characteristics from the meteorological dataset, substitute the historical data of the temperature characteristics and rainfall characteristics into the formula on which the correction function considering the influence of temperature and rainfall on pollutant diffusion is based, and obtain α1 T , α2 P and β1 T , β2 P by fitting according to the least squares method, so as to obtain the correction function of temperature and rainfall on pollutant diffusion;

[0080] The temperature correction coefficient β T and the rainfall correction coefficient β PThe acquisition logic is as follows: Obtain the radioactive intensity data characteristics in the pollution dataset, obtain the temperature characteristics and humidity characteristics in the meteorological dataset, converge the temperature characteristics and humidity characteristics into a coefficient dataset, add the radioactive intensity characteristics as labels to the coefficient dataset, and input the coefficient dataset into a regression analysis model for analysis. The formula is as follows:

[0081] C t =β 0 +β T ·T t +β P ·P t +∈

[0082] Where C t is the radioactive intensity at time t, T t is the temperature at time t, P t is the rainfall at time t, t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K, β 0 is the initial coefficient, β T is the temperature correction coefficient, β P is the humidity correction coefficient, ε is the error term;

[0083] Obtain the temperature correction coefficient β T and the humidity correction coefficient β P through regression analysis, and substitute them into the formula for radioactive intensity correction;

[0084] The linear regression model is relatively simple, easy to understand and implement. It establishes a linear relationship between temperature, humidity and radioactive intensity, making the analysis results easy to interpret, and the calculation is relatively efficient. Especially when the dataset is large, it can quickly obtain results.

[0085] Organize the corrected radioactive intensities of each pollutant detector to form a corrected radioactive intensity dataset. The formula is as follows:

[0086] FD i，j ={Fd 1 , Fd 2 , Fd 3 , …, Fd t , …, YFd n}

[0087] Where FD i，j is the corrected radioactive intensity dataset of the pollutant detector numbered i, j. i is the row number of the pollutant detector, j is the column number of the pollutant detector, i ∈ N, 1 ≤ i ≤ n, j ∈ N, 1 ≤ j ≤ m, Fd t is the corrected radioactive intensity value at the t-th moment. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K;

[0088] Summarize the corrected radioactive intensity data sets of all pollutant detectors to form the total radioactive intensity prediction data set FQ in the tail area of the ore deposit. The basis formula is as follows:

[0089] FQ = {Fq 1 , Fq 2 , Fq 3 , …, Fq t , …, Fq n}

[0090] Among them, Fq t is the overall radioactive intensity in the area at time t. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ n.

[0091] As a preferred embodiment, extract the radioactive intensity characteristics from the historical data of the pollutant data sets of all pollutant detectors. The basis formula is as follows:

[0092]

[0093] Among them, Lfq t is the overall historical radioactive intensity in the area at time t, lfq i，j，t is the historical radioactive intensity data recorded by the pollutant monitors numbered i and j in the area at time t. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ K, i is the row number of the pollutant detector, j is the column number of the pollutant detector, i ∈ N, 1 ≤ i ≤ I, j ∈ N, 1 ≤ j ≤ J;

[0094] Converge the overall historical radioactive intensity data in the tail area of the ore deposit to form the overall historical radioactive intensity data set;

[0095] Obtain the location of the tail area of the ore deposit through the geographic information system, and obtain the high-altitude historical radioactive intensity data of the tail area of the ore deposit through the remote sensing satellite to form the high-altitude radioactive intensity historical data set. Obtain the overall radioactive intensity correction coefficient through the high-altitude historical radioactive intensity data set obtained by the remote sensing satellite. The basis formula is as follows:

[0096]

[0097] Among them, τ is the overall radioactive intensity correction coefficient, Wfq t is the high-altitude historical radioactive intensity obtained by the remote sensing satellite at time t, Lfq t is the overall historical radioactive intensity in the area at time t. t is the time retrieval variable, t ∈ N, 1 ≤ t ≤ n, and n is the number of samples.

[0098] By comparing the radioactive intensity data obtained from the surface dispersion with the high-altitude radioactive intensity data obtained by remote sensing satellites, the representation range of the overall radioactive intensity is extended from a two-dimensional plane to a three-dimensional state, enhancing the accuracy of subsequent data prediction.

[0099] As a preferred embodiment, the peripheral position division logic is as follows: draw a circle with the position of the tail area of the ore deposit as the center, divide the ring excluding the tail area of the ore deposit into eight equal parts to form 8 partitions, and calculate the radioactive intensity data in the eight partitions respectively according to the diffusion principle. The formula is as follows:

[0100]

[0101] Where: BQD i，t is the radioactive intensity of the partition numbered i at time t, Fq t is the overall radioactive intensity in the area at time t, σ y is the diffusion coefficient of the wind direction relative to the row direction of the pollutant detector, σ x is the diffusion coefficient of the wind direction relative to the column direction of the pollutant detector, σ y 2 +σ x 2 = 1, R is the distance from the center of the partition to the center of the tail area of the ore deposit;

[0102] Judge the radioactive intensity of the surrounding partitions through the diffusion phenomenon. The farther the distance, the smaller the influence of diffusion, and the closer the distance, the greater the influence of diffusion. The greater the radioactive intensity of the tail area of the ore deposit, the greater the influence on the surrounding area.

[0103] Obtain the maximum allowable radioactive intensity data and population density data stipulated by local regulations through the network, and obtain the environmental carrying capacity of the tail area of the ore deposit. The formula is as follows:

[0104]

[0105] Where, C ca-i，t is the carrying capacity of the partition numbered i at time t, C max is the maximum allowable radioactive intensity stipulated by local environmental regulations, BQD i，t is the radioactive intensity of the partition numbered i at time t;

[0106] The maximum allowable radioactive intensity stipulated by local regulations represents the local carrying capacity. The greater the maximum allowable radioactive intensity, the stronger the local carrying capacity.

[0107] Calculate the risk value for each partition respectively. The formula is as follows:

[0108] R i，t = w1 ·°C ca-i，t -1 + w 2 ·°C cap-i

[0109] Where: R i，t is the risk value of the partition numbered i at time t, C cap-i is the population density of the partition numbered i, w 1 , w 2 are weights respectively, w 1 2 + w 2 2 = 1;

[0110] The impact of radioactive pollutants exists in both the environment and the humanities. If either aspect exceeds the specified value, it will have an adverse impact. The more people there are, the weaker the carrying capacity of the area, and the greater the risk.

[0111] Set the risk value threshold. When the risk value in the i-th partition at time t exceeds the threshold, a risk warning is issued.

[0112] The above formulas are all dimensionless and take their numerical values for calculation. The formula is obtained by collecting a large amount of data and performing software simulation to get a formula that is closest to the actual situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0113] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0114] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0115] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all should be covered by the protection scope of the present application.

Claims

1. A monitoring device for tailings pollutants in deposited polymetallic ore deposits, characterized in that: Specifically include: The data collection module collects historical and real-time data of pollution data and meteorological data at multiple locations of the tail of the ore deposit, and forms a historical data set and a real-time data set respectively, pre-processes the historical data set and the real-time data set respectively, obtains the high-altitude historical radioactive intensity of the tail of the ore deposit through remote sensing satellites and forms a high-altitude radioactive intensity historical data set, obtains meteorological forecast data of the tail of the ore deposit through meteorological stations, and obtains the population density and maximum allowable radioactive intensity of the area surrounding the tail of the ore deposit; The model training module uses the historical data set as the training set and the radioactivity intensity of the historical data set as the label to train the pollutant concentration prediction model, thereby constructing a pollutant concentration prediction model; The pollution correction module is used to input the real-time data set into the trained pollutant concentration prediction model, obtain the prediction results of the radioactivity intensity at each location and form a radioactivity intensity prediction data set, and construct a radioactivity intensity correction formula through meteorological prediction data to correct the prediction results of the radioactivity intensity and form a corrected radioactivity intensity data set; The overall radioactivity intensity correction module is used to extract the radioactivity intensity features in all historical data sets and process them with the high-altitude historical radioactivity intensity in the high-altitude radioactivity intensity historical data set to form an overall radioactivity intensity correction coefficient; The regional early warning module is used to divide the area around the tail of the ore deposit into regions, obtain the radioactivity intensity of each partition according to the overall radioactivity intensity correction coefficient, diffusion principle and corrected radioactivity intensity set, and then obtain the risk value of each partition according to the maximum radioactivity intensity and population density, and issue early warnings for the partitions based on the risk values.

2. The monitoring device for tailings pollutants in polymetallic deposits according to claim 1, characterized in that: Evenly arrange pollutant detectors in the area where the tail of the polymetallic ore deposit is located to collect historical data of pollution data. The coverage of the pollutant detectors is larger than the area occupied by the tail of the ore deposit. The pollution data includes the number of pollutant types, radioactivity intensity and heavy metal pollutant concentration. Each pollutant detector is numbered in the form of a matrix. The pollutant detector located at the corner point is used as the origin. The positive direction of the row and the positive direction of the column are set respectively. The position of each pollutant detector is represented by a two-dimensional vector composed of rows and columns. Each detector collects pollutant data at equal time intervals and adds a timestamp. Meteorological monitors are arranged around each pollutant detector to collect historical data of meteorological data. The meteorological monitors use the same number as the pollutant detectors. The meteorological data includes wind speed, wind direction, temperature and rainfall. Each meteorological monitor collects meteorological data at equal time intervals and adds a timestamp. The historical data of meteorological data is aligned with the historical data of pollutant data in time. The historical data sets of pollution data and meteorological data are aggregated into historical data sets, and preprocessed and normalized. The radioactivity intensity of each pollutant detection is extracted as a feature label. The historical data sets are input into the long short-term memory network for training, and the trained model is calibrated as a pollutant concentration prediction model.

3. A monitoring device for tailings pollutants in deposited polymetallic ore deposits according to claim 2, characterized in that: Pollutant detectors and meteorological monitors continuously collect real-time data of pollution data and real-time data of meteorological data to form a real-time data set and input it into the pollutant concentration prediction model to obtain the radioactivity intensity at the location of each pollutant detector and form a radioactivity intensity prediction data set. The formula is as follows: FSYQD i，j ={YFd1,YFd2,YFd3,…,YFd t ,…,YFd n } Among them, FSYQD i，j is the radioactivity intensity prediction data set of the pollutant detector numbered i, j, i is the row number of the pollutant detector, j is the column number of the pollutant detector, i∈N, 1≤i≤n, j∈N, 1≤j≤m, YFd t is the predicted value of radioactivity intensity at the tth moment, t is the moment retrieval variable, t∈N, 1≤t≤K.

4. The monitoring device for tailings pollutants in polymetallic deposits according to claim 3 is characterized in that: The meteorological forecast data of the tail area of ​​the ore deposit is obtained through the meteorological station. The meteorological forecast data includes wind speed, wind direction, temperature and rainfall. The radioactivity intensity forecast value of each pollutant detector is corrected according to the meteorological data. The radioactivity intensity correction is based on the following formula: Among them, YFD t ′ is the corrected radioactivity intensity at time t, YFd t is the predicted value of radioactivity intensity at time t, v wind-t is the wind speed at time t, θ wind-t is the wind direction at time t, T t is the temperature at time t, T0 is the temperature at that time, P t is the rainfall at time t, β T is the temperature correction coefficient, β P is the rainfall correction coefficient, P is the rainfall, f(T t ,P t ) is a correction function that considers the effect of temperature and rainfall on pollutant diffusion, t is the time retrieval variable, t∈N, 1≤t≤K, Q t is the predicted value of the radiation intensity of the pollution source, the pollution source is the previous pollution detector in the wind direction, σ y is the diffusion coefficient of wind direction relative to the direction of pollutant detector, σ x is the diffusion coefficient of wind direction relative to the direction of pollutant detector array, σ y 2 +σ x 2 =1; The formula for the correction function considering the effect of temperature and rainfall on the diffusion of pollutants is as follows: f(T,P)=α1 T ·T+α2 P ·P+β1 T ·T 2 +β2 P ·P 2 Where T is temperature, P is rainfall, α1 T ,α2 p and β1 T ,β2 p are the linear and quadratic correction coefficients for temperature and rainfall; The temperature characteristics and rainfall characteristics are obtained from the meteorological data set, and the historical data of temperature characteristics and rainfall characteristics are substituted into the formula based on the correction function considering the effects of temperature and rainfall on pollutant diffusion. α1 is obtained by fitting according to the least squares method. T , α2 p and β1 T , β2 P , thus obtaining the correction function of temperature and rainfall on pollutant diffusion; The temperature correction factor β T and rainfall correction factor β P The acquisition logic is: obtain the radioactivity intensity data features in the pollution data set, obtain the temperature features and humidity features in the meteorological data set, aggregate the temperature features and humidity features into a coefficient data set, add the radioactivity intensity features as labels to the coefficient data set, and input the coefficient data set into the regression analysis model for analysis. The formula is as follows: C t =β0+β T ·T t +b P ·P t +e Among them, C t is the radioactivity intensity at time t, T t is the temperature at time t, P t is the rainfall at time t, t is the time retrieval variable, t∈N, 1≤t≤K, β0 is the initial coefficient, β T is the temperature correction coefficient, β P is the humidity correction coefficient, ∈ is the error term; The temperature correction coefficient β is obtained by regression analysis T and humidity correction factor β P , and substitute it into the formula based on which the radioactivity intensity correction is based; The corrected radioactivity intensity of each pollutant detector is sorted to form a corrected radioactivity intensity data set based on the following formula: FD i，j ={Fd1,Fd2,Fd3,…,Fd t ,…,YFd n } Among them, FD i，j is the corrected radioactivity intensity data set of the pollutant detector numbered i, j, i is the row number of the pollutant detector, j is the column number of the pollutant detector, i∈N, 1≤i≤b, j∈N, 1≤j≤m, Fd t is the value of the radioactivity intensity after correction at the tth moment, t is the moment retrieval variable, t∈N, 1≤t≤K; The corrected radioactivity intensity data sets of all pollutant detectors are aggregated to form the total radioactivity intensity prediction data set FQ in the tail area of ​​the deposit, based on the following formula: FQ={Fq1,Fq2,Fq3,…,Fq t ,…,Fq n } Among them, Fq t is the overall radiation intensity in the region at time t, t is the time retrieval variable, t∈N, 1≤t≤n.

5. The monitoring device for tailings pollutants in deposited polymetallic ore deposits according to claim 4, characterized in that: The radioactivity intensity characteristics are extracted from the historical data of the contamination data set of all contamination detectors according to the following formula: Among them, Lfq t is the overall historical intensity of radioactivity in the region at time t, lfq i，j，t is the historical data of radioactivity intensity recorded by the pollutant monitor numbered i, j in the area at time t, t is the time retrieval variable, t∈N, 1≤t≤K, i is the row number of the pollutant detector, j is the column number of the pollutant detector, i∈N, 1≤i≤I, j∈N, 1≤j≤J; Aggregate the historical overall radioactivity intensity data in the tail area of ​​the ore deposit to form a historical data set of overall radioactivity intensity; The location of the tail area of ​​the ore deposit is obtained through the geographic information system, and the high-altitude historical radioactivity intensity data of the tail area of ​​the ore deposit is obtained through the remote sensing satellite to form a high-altitude radioactivity intensity historical data set. The overall radioactivity intensity correction coefficient is obtained through the high-altitude historical radioactivity intensity data set obtained by the remote sensing satellite. The formula is as follows: Where τ is the overall radioactivity intensity correction factor, Wfq t is the historical high-altitude radioactivity intensity obtained by remote sensing satellite at time t, Lfq t is the overall historical intensity of radioactivity in the region at time t, t is the time retrieval variable, t∈N, 1≤t≤n, and n is the number of samples.

6. The monitoring device for tailings pollutants in polymetallic deposits according to claim 5, characterized in that: The logic of the peripheral location division is: draw a circle with the tail area of ​​the ore deposit as the center, divide the circle excluding the tail area of ​​the ore deposit into eight equal parts, forming 8 partitions, and calculate the radioactivity intensity data in the eight partitions according to the diffusion principle. The formula is as follows: Among them: BQD i，t is the radioactivity intensity of the partition numbered i at time t, Fq t is the overall radiation intensity in the region at time t, σ y is the diffusion coefficient of wind direction relative to the direction of pollutant detector, σ x is the diffusion coefficient of wind direction relative to the direction of pollutant detector array, σ y 2 +σ x 2 =1, R is the distance from the center of the partition to the center of the tail area of ​​the deposit; The maximum allowable radioactivity intensity data and population density data stipulated by local regulations are obtained through the Internet to obtain the environmental carrying capacity of the tail area of ​​the deposit. The formula is as follows: Among them, C ca-i，t is the carrying capacity of the partition numbered i at time t, C max The maximum permissible radioactivity intensity specified by local environmental regulations, BQD i，t is the radioactivity intensity of the partition numbered i at time t; The risk value is calculated for each partition separately according to the following formula: R i，t =w1·C ca-i，t -1 +w2·C cap-i Where: R i，t is the risk value of the partition numbered i at time t, C cap-i is the population density of the partition numbered i, w1 and w2 are weights, w1 2 +w2 2 =1; Set a risk value threshold. When the risk value in the i-th partition exceeds the threshold at time t, a risk warning is issued.

Images