Mine wastewater treatment system and method

Abstract

The invention belongs to the technical field of mine wastewater treatment, and discloses a mine wastewater treatment system and method, and the method comprises the steps: collecting M groups of historical mine wastewater data, environment temperature data and corresponding actually measured agent addition amount data; preprocessing the historical mine wastewater data to obtain mine wastewater characteristic data; constructing a temperature influence coefficient model for calculating an agent addition amount influence coefficient based on the environment temperature data, and constructing an agent prediction model for predicting the agent addition amount based on the mine wastewater characteristic data and the corresponding actually measured agent addition amount data; real-time mine wastewater data and real-time environment temperature data are collected, the final dose is calculated through a temperature influence coefficient model and a medicament prediction model, and the final dose serves as a parameter of a dose control terminal for dose control; and the labor cost is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of mine wastewater treatment, and more specifically, to a mine wastewater treatment system and method. Background Art

[0002] The patent with application publication number CN117819767A discloses a mine wastewater denitrification and phosphorus removal treatment system and a denitrification and phosphorus removal treatment method. The mine wastewater denitrification and phosphorus removal treatment system includes a grid channel, a water collection tank, a coagulation and sedimentation unit, a magnetic coagulation and sedimentation unit, a magnetic separation recovery unit, a deep treatment unit, a clear water tank and a disinfection tank; the coagulation and sedimentation unit includes a first coagulation tank, a first flocculation tank and a horizontal flow sedimentation tank; the magnetic coagulation and sedimentation unit includes a second coagulation tank, a magnetic seed stirring tank and a second flocculation tank; the magnetic separation recovery unit includes a super magnetic separator and a magnetic recovery machine; the deep treatment unit includes an aerated biological filter and a sulfur autotrophic denitrification filter; the clear water tank is connected to the sulfur autotrophic denitrification filter, the disinfection tank is connected to the clear water tank, and a disinfection device is provided upstream of the disinfection tank; it can realize effective purification treatment of mine wastewater, has a good removal effect on suspended matter, nitrogen, phosphorus and heavy metals in the wastewater, and has high treatment efficiency and good treatment effect on mine wastewater.

[0003] Mine wastewater is industrial wastewater generated during the mining process that contains a large amount of suspended solids, heavy metal ions, organic matter and other harmful substances; its main sources include groundwater in the mining process, wastewater in the ore crushing and washing process, and surface water infiltrated by mine drainage and rainfall; the discharge of mine wastewater poses a serious threat to the environment and human health, so its effective treatment is of great environmental and economic significance.

[0004] However, in the traditional mine wastewater treatment process, it is difficult to effectively remove fine suspended matter and soluble organic matter in the wastewater, and the treatment efficiency is low; the dosage of the reagent is difficult to accurately control, resulting in excessive or insufficient use of the reagent, and the cost of the reagent becomes the main expense in the treatment process; the system operation is complex and requires frequent manual monitoring and adjustment, which greatly increases the management difficulty and labor cost.

[0005] In view of this, the present invention proposes a mine wastewater treatment system and method to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solutions: a mine wastewater treatment system and method; A mine wastewater treatment system according to the present invention comprises: a data acquisition module for collecting M groups of historical mine wastewater data, environmental temperature data and corresponding measured reagent addition amount data; Data processing module: pre-process historical mine wastewater data to obtain mine wastewater characteristic data; Model building module: a temperature influence coefficient model for calculating the influence coefficient of reagent addition amount is constructed based on the ambient temperature data, and a reagent prediction model for predicting the reagent addition amount is constructed based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data; Wastewater treatment module: collects real-time mine wastewater data and real-time ambient temperature data, calculates the final dosage through the temperature influence coefficient model and the dosage prediction model, and uses the final dosage as the parameter of the dosage control terminal for dosage control.

[0007] Furthermore, the mine wastewater data includes: suspended matter concentration, water quality pH value and wastewater flow rate.

[0008] Furthermore, the specific method of preprocessing the historical mine wastewater data includes: Determine the optimal grouping value for suspended solids concentration data. The optimal grouping value can be obtained by: Defining grouping values , , represents a grouping set, is a positive integer and satisfies the condition and ,Right now Divide by The remainder is , is the total amount of suspended matter concentration data; Traverse the grouping set in turn and assign values ​​to the grouping values , sort the suspended matter concentration data in ascending order from small to large, and divide the sorted suspended matter concentration data into data sets; Define the grouping coefficient function, the grouping coefficient function formula is: ;in Representative The grouping coefficient of the data, Representative The average of the difference between a data point and other data points in the group. For the Data and distance The average of the differences between the data in the group closest to the data group; Define the average grouping coefficient function, the formula of the average grouping coefficient function is: ;in represents the average grouping coefficient; Records are different The average grouping coefficient under the value, the corresponding value when the average grouping coefficient is the largest The value is the best grouping value , and group the suspended matter concentration data according to the optimal grouping value; The determination Data and distance The specific method for the most recent data group is to calculate the The difference between the data and the mean of each data group, and the data group with the smallest difference is taken as the The most recent data set of data.

[0009] Furthermore, the specific method of grouping the suspended matter concentration data includes: According to the best grouping value , and group the suspended matter concentration data according to the optimal grouping value; The determination Data and distance The specific method for the most recent data group is to calculate the The difference between the data and the mean of each data group, and the data group with the smallest difference is taken as the The most recent data set of data.

[0010] Furthermore, the specific method of grouping the suspended matter concentration data includes: According to the best grouping value , selected from suspended matter concentration data Data are used as the initial cluster center, and the selected data meets the conditions: ;in Indicates the data interval; Each data in the suspended matter concentration data is clustered, and the clustering steps include: Step 1: Calculate the data points in the suspended solids concentration data The distance to each initial cluster center , ;in For data points The corresponding value, For the Initial cluster centers; and each data point in the suspended matter concentration data is assigned to the nearest initial cluster center to form clusters; Step 2: Calculate the mean of the data in each cluster as the value of the new initial cluster center; Step 3: Repeat steps 1 and 2 until the value of the initial cluster center no longer changes; The allocated clusters, and calculate the cluster mean through the initial cluster center and cluster standard deviation ,Will The clusters are removed as invalid clusters to obtain valid data clusters. For the Initial cluster centers; Extract cluster features of suspended solids concentration data from valid data clusters, the cluster features including initial cluster center, cluster standard deviation, maximum initial cluster center difference, minimum initial cluster center difference, maximum initial cluster center and minimum initial cluster center of each cluster; Similarly, the cluster characteristics of water quality pH value and wastewater flow rate are obtained; The mine wastewater data is subjected to feature derivation to obtain extended features. The feature derivation methods include: product interaction of cluster features of suspended solids concentration data and cluster features of wastewater flow, and product interaction of cluster features of water quality pH value and cluster features of wastewater flow; The cluster characteristics of suspended matter concentration data, the cluster characteristics of water quality pH value, the cluster characteristics of wastewater flow and the extended characteristics are connected to obtain the characteristic data of mine wastewater.

[0011] Furthermore, the temperature influence coefficient model is constructed in the following manner: Based on the ambient temperature data, a temperature influence coefficient model for calculating the influence coefficient of the agent addition amount is constructed. The temperature influence coefficient model is: ;in For the actual temperature The influence coefficient of the amount of agent added is is the standard ambient temperature, and is the adjustment factor, It is pi.

[0012] Furthermore, the adjustment coefficient is obtained by: Collect historical influence coefficient data of group Z, including: predicted influence coefficient of reagent addition, measured influence coefficient of reagent addition and historical actual temperature; Presets , and learning rate The initial value of Define the influence coefficient loss function, the function formula is: ;in is the coefficient loss, For the The measured influence coefficient of the amount of reagent added is For the A coefficient for predicting the influence of the amount of added agent, is the total amount of historical impact coefficient data; Calculate the influence coefficient loss function and The gradient of The gradient calculation formula is: ; The gradient calculation formula is: ;in For the Historical actual temperature; According to the learning rate ,right and Perform a gradient descent update, The update method is: Calculate the updated value , the calculation formula is: , will update the value Assign to ; The update method is: Calculate the updated value , the calculation formula is: , will update the value Assign to ; By continuously updating the adjustment coefficient, when the value of the influence coefficient loss function is the smallest, stop updating. and is the optimal adjustment factor.

[0013] Furthermore, the drug prediction model is constructed in the following manner: Based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data, the mine wastewater characteristic data is used as the input of the reagent prediction model, the predicted reagent addition amount is used as the output of the reagent prediction model, and the measured reagent addition amount is used as the prediction target of the reagent prediction model to construct the reagent prediction model. The reagent prediction model is a support vector machine model, and the target loss function of the reagent prediction model is: ;in represents the model complexity, Representative The difference between the predicted amount of added agent and the measured amount of added agent, is the balance parameter, Represents the total number of mine wastewater characteristic data, Represents the mine wastewater characteristic data index; The objective loss function constraints are: ; For the The measured amount of reagent added corresponding to the characteristics of mine wastewater, Representative Characteristics of mine wastewater: represents the mine wastewater characteristic bias, represents the characteristic weight coefficient of mine wastewater; The mine wastewater characteristic weight coefficient and mine wastewater characteristic bias are updated through the sequential minimum optimization algorithm to minimize the target loss function value, and the reagent prediction model with the minimum target loss function value is obtained.

[0014] Furthermore, the real-time mine wastewater data includes: real-time suspended matter concentration, real-time water quality pH value and real-time wastewater flow.

[0015] Furthermore, the specific method of calculating the final dosage by the temperature influence coefficient model and the dosage prediction model includes: Based on real-time mine wastewater data, the amount of reagent added is predicted through the reagent prediction model to obtain the amount of reagent added , based on the real-time ambient temperature data, the influence coefficient of the amount of reagent added is calculated through the temperature influence coefficient model to obtain the influence coefficient of the amount of reagent added , according to the amount of drug added And the influence coefficient of the amount of agent added Calculate the final dose , ; Use the final drug dosage as the parameter of the drug dosage control terminal to control the drug dosage.

[0016] A mine wastewater treatment method according to the present invention comprises: collecting M groups of historical mine wastewater data, environmental temperature data and corresponding measured reagent addition amount data; Preprocess the mine wastewater data to obtain mine wastewater characteristic data; A temperature influence coefficient model for calculating the influence coefficient of reagent addition is constructed based on the ambient temperature data, and a reagent prediction model for predicting the reagent addition amount is constructed based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data; Collect real-time mine wastewater data and real-time ambient temperature data, calculate the final dosage through the temperature influence coefficient model and the dosage prediction model, and use the final dosage as the parameter of the dosage control terminal for dosage control; The technical effects and advantages of a mine wastewater treatment system and method of the present invention are as follows: The present invention predicts the amount of reagent addition by combining ambient temperature data and wastewater characteristic data. The temperature influence coefficient model enables the amount of reagent addition to be dynamically adjusted according to changes in ambient temperature, thereby avoiding unstable reagent reaction effects caused by temperature changes, thereby improving the effect of wastewater treatment; by preprocessing mine wastewater data, data noise is effectively reduced, data quality is improved, and thus the accuracy of the prediction model is improved; through an automated control system, the final dosage is calculated based on the predicted amount of reagent addition and the temperature influence coefficient, and the final dosage is transmitted to a dosage control terminal, thereby reducing manual intervention and improving the degree of automation of the processing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of a mine wastewater treatment system of the present invention; Figure 2 It is a schematic diagram of a mine wastewater treatment method of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example

[0019] See also Figure 1 As shown, a mine wastewater treatment system described in this embodiment includes: a data acquisition module: collecting M groups of historical mine wastewater data, environmental temperature data and corresponding measured reagent addition amount data; Data processing module: pre-process the mine wastewater data to obtain mine wastewater characteristic data; Model building module: a temperature influence coefficient model for calculating the influence coefficient of reagent addition amount is constructed based on the ambient temperature data, and a reagent prediction model for predicting the reagent addition amount is constructed based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data; Wastewater treatment module: collects real-time mine wastewater data and real-time ambient temperature data, calculates the final dosage through the temperature influence coefficient model and the dosage prediction model, and uses the final dosage as the parameter of the dosage control terminal for dosage control; Each module is connected by wired and / or wireless means to achieve data transmission between modules; Mine wastewater data include: suspended solids concentration, water pH value and wastewater flow; The suspended matter concentration is used to reflect the content of solid matter in mine wastewater, and the wastewater flow rate is the amount of wastewater that needs to be treated per unit time.

[0020] The specific methods for preprocessing mine wastewater data include: Determine the optimal grouping value for suspended solids concentration data. The optimal grouping value can be obtained by: Defining grouping values , , represents a grouping set, is a positive integer and satisfies the condition and ,Right now Divide by The remainder is , is the total amount of suspended matter concentration data; Traverse the grouping set in turn and assign values ​​to the grouping values , sort the suspended matter concentration data in ascending order from small to large, and divide the sorted suspended matter concentration data into data sets; Define the grouping coefficient function, the grouping coefficient function formula is: ;in Representative The grouping coefficient of the data, Representative The average of the difference between a data point and other data points in the group. For the Data and distance The average of the differences between the data in the group closest to the data group; Define the average grouping coefficient function, the formula of the average grouping coefficient function is: ;in represents the average grouping coefficient; Records are different The average grouping coefficient under the value, the corresponding value when the average grouping coefficient is the largest The value is the best grouping value , and group the suspended matter concentration data according to the optimal grouping value; The determination Data and distance The specific method for the most recent data group is to calculate the The difference between the data and the mean of each data group, and the data group with the smallest difference is taken as the The most recent data set of data; Through optimized grouping, large-scale suspended solids concentration data can be processed more quickly, saving computing resources and improving data processing efficiency; According to the best grouping value , selected from suspended matter concentration data Data are used as the initial cluster center, and the selected data meets the conditions: ;in Indicates the data interval; Each data in the suspended matter concentration data is clustered, and the clustering steps include: Step 1: Calculate the data points in the suspended solids concentration data The distance to each initial cluster center , ;in For data points The corresponding value, For the Initial cluster centers; and each data point in the suspended matter concentration data is assigned to the nearest initial cluster center to form clusters; Step 2: Calculate the mean of the data in each cluster as the value of the new initial cluster center; Step 3: Repeat steps 1 and 2 until the value of the initial cluster center no longer changes; The allocated clusters, and calculate the cluster mean through the initial cluster center and cluster standard deviation ,Will The clusters are removed as invalid clusters to obtain valid data clusters. For the Initial cluster centers; Extract cluster features of suspended solids concentration data from valid data clusters, the cluster features including initial cluster center, cluster standard deviation, maximum initial cluster center difference, minimum initial cluster center difference, maximum initial cluster center and minimum initial cluster center of each cluster; Similarly, the cluster characteristics of water quality pH value and wastewater flow rate are obtained (i.e., the same way as the cluster characteristics of suspended solids concentration data are obtained through suspended solids concentration data); The cluster features of suspended solids concentration data, the cluster features of water quality pH value and the cluster features of wastewater flow are connected to obtain mine wastewater characteristic data; The mine wastewater data is subjected to feature derivation to obtain extended features. The feature derivation methods include: product interaction of cluster features of suspended solids concentration data and cluster features of wastewater flow, and product interaction of cluster features of water quality pH value and cluster features of wastewater flow, so as to capture the interaction effect between features. The cluster features of suspended solids concentration data, the cluster features of water quality pH value, the cluster features of wastewater flow and the extended features are connected to obtain the mine wastewater characteristic data; The construction methods of the temperature influence coefficient model include: Based on the ambient temperature data and the corresponding reagent addition amount data, a temperature influence coefficient model for calculating the reagent addition amount influence coefficient is constructed. The temperature influence coefficient model is: ;in For the actual temperature The influence coefficient of the amount of agent added is is the standard ambient temperature, and is the adjustment factor, is pi; By considering the temperature influencing factor, the dosage of the agent can be optimized to avoid overdosage or underdosage and improve cost-effectiveness; The adjustment coefficient can be obtained by: Collect historical influence coefficient data of group Z, including: predicted influence coefficient of reagent addition, measured influence coefficient of reagent addition and historical actual temperature; Presets , and learning rate The initial value of (i.e. the step size of each learning); Define the influence coefficient loss function, the function formula is: ;in is the coefficient loss, For the The measured influence coefficient of the amount of reagent added is For the A coefficient for predicting the influence of the amount of added agent, is the total amount of historical impact coefficient data; Calculate the influence coefficient loss function and The gradient of The gradient calculation formula is: ; The gradient calculation formula is: ;in For the Historical actual temperature; According to the learning rate ,right and Perform a gradient descent update, The update method is: Calculate the updated value , the calculation formula is: , will update the value Assign to ; The update method is: Calculate the updated value , the calculation formula is: , will update the value Assign to ; By continuously updating the adjustment coefficient, when the value of the influence coefficient loss function is the smallest, stop updating. and To optimize the adjustment coefficient, the temperature influence coefficient model can more accurately reflect the effect of temperature on the amount of reagent added, thereby improving the prediction accuracy; The methods for constructing drug prediction models include: Based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data, the mine wastewater characteristic data is used as the input of the reagent prediction model, the predicted reagent addition amount is used as the output of the reagent prediction model, and the measured reagent addition amount is used as the prediction target of the reagent prediction model to construct the reagent prediction model. The reagent prediction model is a support vector machine model, and the target loss function of the reagent prediction model is: ;in represents the model complexity, Representative The difference between the predicted amount of added agent and the measured amount of added agent, is the balance parameter, Represents the total number of mine wastewater characteristic data, Represents the mine wastewater characteristic data index; The objective loss function constraints are: ; For the The measured amount of reagent added corresponding to the characteristics of mine wastewater, Representative Characteristics of mine wastewater: represents the mine wastewater characteristic bias, represents the characteristic weight coefficient of mine wastewater; The mine wastewater characteristic weight coefficient and the mine wastewater characteristic bias are updated through the sequence minimum optimization algorithm to minimize the target loss function value, and the reagent prediction model with the minimum target loss function value is obtained, so that the regression function of the reagent prediction model is smoother, which helps to improve the prediction accuracy and prevent overfitting; Collect real-time mine wastewater data and real-time ambient temperature data. Real-time mine wastewater data includes suspended solids concentration, water quality pH value and wastewater flow rate. Based on real-time mine wastewater data, the agent addition amount is predicted through the agent prediction model to obtain the agent addition amount. , based on the real-time ambient temperature data, the influence coefficient of the amount of reagent added is calculated through the temperature influence coefficient model to obtain the influence coefficient of the amount of reagent added , according to the amount of drug added And the influence coefficient of the amount of agent added Calculate the final dose , ; The final drug dosage is used as the parameter of the drug dosage control terminal to control the drug dosage; Adjust the dosage of reagents according to the real-time temperature to ensure the best wastewater treatment effect at different temperatures, thereby improving the overall treatment efficiency; This embodiment predicts the amount of reagent addition by combining ambient temperature data and wastewater characteristic data. The temperature influence coefficient model enables the amount of reagent addition to be dynamically adjusted according to changes in ambient temperature, avoiding unstable reagent reaction effects caused by temperature changes, thereby improving the effect of wastewater treatment; by preprocessing mine wastewater data, data noise is effectively reduced, and data quality is improved, thereby improving the accuracy of the prediction model; through the automated control system, the final dosage is calculated based on the predicted amount of reagent addition and the temperature influence coefficient, and transmitted to the dosage control terminal, reducing manual intervention and improving the degree of automation of the processing process. Example

[0021] See also Figure 2 As shown, the part not described in detail in this embodiment is described in Example 1, and a method for treating mine wastewater is provided, comprising: Collect M groups of historical mine wastewater data, ambient temperature data and corresponding reagent addition amount data; Preprocess the mine wastewater data to obtain mine wastewater characteristic data; A temperature influence coefficient model for calculating the influence coefficient of reagent addition is constructed based on the ambient temperature data, and a reagent prediction model for predicting the reagent addition amount is constructed based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data; Collect real-time mine wastewater data and real-time ambient temperature data, calculate the final dosage through the temperature influence coefficient model and the dosage prediction model, and use the final dosage as the parameter of the dosage control terminal for dosage control. Example

[0022] This embodiment discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the operation mode of the mine wastewater treatment method provided above is implemented.

[0023] Since the electronic device introduced in this embodiment is an electronic device used to implement a mine wastewater treatment method in the embodiment of this application, based on the mine wastewater treatment method introduced in the embodiment of this application, a person skilled in the art can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application is not described in detail here. As long as a person skilled in the art implements the electronic device used in a mine wastewater treatment method in the embodiment of this application, it belongs to the scope of protection of this application.

[0024] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0025] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technical users in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A mine wastewater treatment system, characterized in that: include: Data collection module: collects M groups of historical mine wastewater data, ambient temperature data and corresponding measured reagent addition amount data; Data processing module: pre-process historical mine wastewater data to obtain mine wastewater characteristic data; Model building module: a temperature influence coefficient model for calculating the influence coefficient of reagent addition amount is constructed based on the ambient temperature data, and a reagent prediction model for predicting the reagent addition amount is constructed based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data; Wastewater treatment module: collects real-time mine wastewater data and real-time ambient temperature data, calculates the final dosage through the temperature influence coefficient model and the dosage prediction model, and uses the final dosage as the parameter of the dosage control terminal for dosage control; The specific method of calculating the final dosage by using the temperature influence coefficient model and the dosage prediction model includes: predicting the dosage of the dosage by using the dosage prediction model based on the real-time mine wastewater data to obtain the dosage of the dosage Based on the real-time ambient temperature data, the influence coefficient of the amount of reagent added is calculated through the temperature influence coefficient model to obtain the influence coefficient of the amount of reagent added , according to the amount of drug added And the influence coefficient of the amount of agent added Calculate the final dose , .

2. A mine wastewater treatment system according to claim 1, characterized in that: The mine wastewater data include: suspended matter concentration, water quality pH value and wastewater flow rate.

3. A mine wastewater treatment system according to claim 2, characterized in that: The specific method of preprocessing the historical mine wastewater data includes: Determine the optimal grouping value for suspended solids concentration data. The optimal grouping value can be obtained by: Defining grouping values , , represents a grouping set, is a positive integer and satisfies the condition and ,Right now Divide by The remainder is , is the total amount of suspended matter concentration data; Traverse the grouping set in turn and assign values ​​to the grouping values , sort the suspended matter concentration data in ascending order from small to large, and divide the sorted suspended matter concentration data into data sets; Define the grouping coefficient function, the grouping coefficient function formula is: ;in Representative The grouping coefficient of the data, Representative The average of the difference between a data point and other data points in the group. For the The average of the differences between a data point and each data point in the group of data points closest to the data point; Define the average grouping coefficient function, the formula of the average grouping coefficient function is: ;in represents the average grouping coefficient; Records are different The average grouping coefficient under the value, the corresponding value when the average grouping coefficient is the largest The best grouping value , and group the suspended matter concentration data according to the optimal grouping value; The said Data and distance The method for determining the most recent data group is to calculate the The difference between the data and the mean of each data group, and the data group with the smallest difference is taken as the first The most recent data set of data.

4. A mine wastewater treatment system according to claim 3, characterized in that: The specific method of grouping the suspended solids concentration data includes: According to the best grouping value , selected from suspended matter concentration data data as the initial cluster center, and the selected data meets the conditions: ;in Indicates the data interval; Each data in the suspended matter concentration data is clustered, and the clustering steps include: Step 1: Calculate the data points in the suspended solids concentration data The distance to each initial cluster center , ;in For data points The corresponding value, For the Initial cluster centers; and each data point in the suspended matter concentration data is assigned to the nearest initial cluster center to form clusters; Step 2: Calculate the mean of the data in each cluster as the value of the new initial cluster center; Step 3: Repeat steps 1 and 2 until the value of the initial cluster center no longer changes; The allocated clusters, and calculate the cluster mean through the initial cluster center and cluster standard deviation ,Will The clusters are removed as invalid clusters to obtain valid data clusters. For the Initial cluster centers; Extract cluster features of suspended solids concentration data from valid data clusters, the cluster features including initial cluster center, cluster standard deviation, maximum initial cluster center difference, minimum initial cluster center difference, maximum initial cluster center and minimum initial cluster center of each cluster; Similarly, the cluster characteristics of water quality pH value and wastewater flow rate are obtained; The mine wastewater data is subjected to feature derivation to obtain extended features. The feature derivation methods include: product interaction of cluster features of suspended solids concentration data and cluster features of wastewater flow, and product interaction of cluster features of water quality pH value and cluster features of wastewater flow; The cluster characteristics of suspended matter concentration data, the cluster characteristics of water quality pH value, the cluster characteristics of wastewater flow and the extended characteristics are connected to obtain the characteristic data of mine wastewater.

5. A mine wastewater treatment system according to claim 4, characterized in that: The construction method of the temperature influence coefficient model includes: Based on the ambient temperature data, a temperature influence coefficient model for calculating the influence coefficient of the agent addition amount is constructed. The temperature influence coefficient model is: ;in For the actual temperature The influence coefficient of the amount of agent added is is the standard ambient temperature, and is the adjustment factor, It is pi.

6. A mine wastewater treatment system according to claim 5, characterized in that: The method for obtaining the adjustment coefficient includes: Collect historical influence coefficient data of group Z, including: predicted influence coefficient of reagent addition, measured influence coefficient of reagent addition and historical actual temperature; Presets , and learning rate The initial value of Define the influence coefficient loss function, the function formula is: ;in is the coefficient loss, For the The measured influence coefficient of the amount of added agent is For the A coefficient for predicting the influence of the amount of added agent, is the total amount of historical impact coefficient data; Calculate the influence coefficient loss function and The gradient of The gradient calculation formula is: ； The gradient calculation formula is: ;in For the Historical actual temperature; According to the learning rate ,right and Perform a gradient descent update, The update method is: Calculate the updated value , the calculation formula is: , will update the value Assign to ; The update method is: Calculate the updated value , the calculation formula is: , will update the value Assign to ; By continuously updating the adjustment coefficient, when the value of the influence coefficient loss function is the smallest, stop updating. and is the optimal adjustment factor.

7. A mine wastewater treatment system according to claim 6, characterized in that: The specific method of constructing the drug prediction model includes: Based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data, the mine wastewater characteristic data is used as the input of the reagent prediction model, the predicted reagent addition amount is used as the output of the reagent prediction model, and the measured reagent addition amount is used as the prediction target of the reagent prediction model to construct the reagent prediction model. The reagent prediction model is a support vector machine model, and the target loss function of the reagent prediction model is: ;in represents the model complexity, Representative The difference between the predicted amount of added agent and the measured amount of added agent, is the balance parameter, Represents the total number of mine wastewater characteristic data, Represents the mine wastewater characteristic data index; The objective loss function constraints are: ; For the The measured amount of reagent added corresponding to the characteristics of mine wastewater, Representative Characteristics of mine wastewater: represents the mine wastewater characteristic bias, represents the characteristic weight coefficient of mine wastewater; The mine wastewater characteristic weight coefficient and mine wastewater characteristic bias are updated through the sequential minimum optimization algorithm to minimize the target loss function value, and the reagent prediction model with the minimum target loss function value is obtained.

8. A mine wastewater treatment system according to claim 7, characterized in that: The real-time mine wastewater data includes: real-time suspended solids concentration, real-time water quality pH value and real-time wastewater flow.

9. A mine wastewater treatment method, which is implemented based on a mine wastewater treatment system according to any one of claims 1 to 8, characterized in that: include: Collect M groups of historical mine wastewater data, ambient temperature data and corresponding measured reagent addition amount data; Preprocess the mine wastewater data to obtain mine wastewater characteristic data; A temperature influence coefficient model for calculating the influence coefficient of reagent addition is constructed based on the ambient temperature data, and a reagent prediction model for predicting the reagent addition amount is constructed based on the mine wastewater characteristic data and the corresponding measured reagent addition amount data; Collect real-time mine wastewater data and real-time ambient temperature data, calculate the final dosage through the temperature influence coefficient model and the dosage prediction model, and use the final dosage as the parameter of the dosage control terminal for dosage control.

Images