Coal-containing wastewater treatment system based on self-filling type water absorption and intelligent control

Abstract

The invention provides a coal-containing wastewater treatment system based on self-filling type water suction and intelligent control, which is characterized in that a vacuum self-filling type water suction module is arranged in the system, and a negative pressure environment is formed in a vacuum container by utilizing a vacuum generating device, so that self-filling type suction of coal-containing wastewater is realized; the multi-source data sensing module is used for acquiring operation parameters of the vacuum self-filling type water absorption module; the data analysis module receives the operation parameters and corrects PID control parameters according to the operation parameters, so that an adjusting instruction can accurately adapt to the real-time working condition of the system, the problem that fixed parameter control is difficult to adapt to flow and pressure fluctuation of coal-containing wastewater is avoided, and the stability and high efficiency of system operation are guaranteed; besides, the remote operation and maintenance module extracts characteristic parameters of the operation parameters, outputs equipment health scores by using a health evaluation model and generates a maintenance report, so that the defects of insufficient control precision and low operation and maintenance efficiency of the coal-containing wastewater treatment system are overcome.

Description

Technical Field

[0001] This invention relates to the field of coal-containing wastewater treatment technology, and in particular to a coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control. Background Technology

[0002] The coal mining, washing, and processing processes generate large amounts of coal-containing wastewater. This wastewater contains pollutants such as coal dust, suspended solids, and heavy metal ions. Direct discharge of this wastewater can cause serious pollution to soil and water bodies, and also waste coal resources. Therefore, efficient treatment of coal-containing wastewater and the recycling of coal slime are crucial steps for the coal industry to achieve green environmental protection and resource recycling.

[0003] Currently, coal-containing wastewater treatment systems typically employ a combination of processes such as sedimentation, filtration, and purification. One of the core components is the use of pumps to transport wastewater from the collection tank to the treatment unit. Traditional wastewater transport often relies on a bottom valve-type suction structure, where a bottom valve is installed at the bottom of the pump inlet pipe to create a sealed space, and water is manually introduced into the pump unit. In terms of control, simple start-stop control or single-parameter PID (Proportional-Integral-Derivative) regulation is often used, relying on manual inspection and monitoring of equipment operation. In terms of operation and maintenance management, the main approach is periodic inspections and post-failure repairs, lacking the ability to conduct real-time assessments and predictions of equipment health status. Summary of the Invention

[0004] This invention provides a coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control, which solves the problems of insufficient control accuracy and low operation and maintenance efficiency in coal-containing wastewater treatment systems.

[0005] On one hand, the present invention provides a coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control, comprising:

[0006] A vacuum self-filling water suction module includes a vacuum container for creating a negative pressure environment, an inlet pipe connected to the vacuum container, a pump unit connected to the vacuum container, and a vacuum generator for providing negative pressure to the vacuum container.

[0007] A multi-source data sensing module connected to the vacuum self-priming water absorption module, the multi-source data sensing module being used to collect the operating parameters of the vacuum self-priming water absorption module;

[0008] The data analysis module is used to receive the operating parameters, correct the control parameters of the PID control according to the operating parameters, and output adjustment commands to the vacuum self-priming water suction module.

[0009] The remote operation and maintenance module is used to extract the feature parameters of the operating parameters, input the feature parameters into the health assessment model, output the equipment health score, and generate a maintenance report based on the equipment health score; the health assessment model is a prediction model trained based on the random forest algorithm.

[0010] Optionally, the multi-source data sensing module includes a liquid level sensor, a flow sensor, and a pressure sensor; the liquid level sensor is disposed inside the vacuum container and is used to collect the liquid level height of the vacuum container; the flow sensor is disposed in the inlet pipe and is used to collect the flow parameters of the medium in the inlet pipe; the pressure sensor includes a first pressure sensor disposed at the pump outlet and a second pressure sensor disposed in the vacuum container, the first pressure sensor collecting the pump outlet pressure; the second pressure sensor collecting the vacuum pressure inside the vacuum container;

[0011] The multi-source data sensing module is used to sort the liquid level height, the flow rate parameter, the outlet pressure and the vacuum pressure in chronological order, and divide the liquid level height, the flow rate parameter, the outlet pressure and the vacuum pressure with the same timestamp into data groups to generate operating parameters.

[0012] Optionally, the liquid inlet pipeline is equipped with a filter device and a vacuum protection device. The filter device is used to intercept particulate impurities in the coal-containing wastewater, and the vacuum protection device is used to reduce the vacuum pressure in the vacuum container when the vacuum pressure in the vacuum container exceeds a preset threshold.

[0013] Optionally, the vacuum generating device includes a Venturi jet injector and a liquid ring vacuum pump;

[0014] The data analysis module is used for:

[0015] If the vacuum pressure is greater than a preset pressure threshold, a control command to start the Venturi jet is generated.

[0016] If the vacuum pressure is less than the preset pressure threshold, a control command is generated to start the liquid ring vacuum pump.

[0017] Optionally, the data analysis module includes a fuzzy rule base;

[0018] The fuzzy rule base is used to receive the operating parameters collected by the multi-source data sensing module, and by analyzing the deviation and rate of change of the operating parameters, adjust the proportional coefficient, integral time and derivative time of the PID controller to generate corrected control parameters, and output adjustment commands to the vacuum self-priming water suction module for the corresponding pump group operating frequency and inlet pipeline valve opening.

[0019] Optionally, constructing a health assessment model includes:

[0020] Collect historical monitoring data, fault records and maintenance logs of the equipment, and extract characteristic parameters including pressure fluctuation amplitude, flow rate decay rate and liquid level adjustment frequency;

[0021] An initial model is constructed using the random forest algorithm. The feature parameters are divided into a training set and a test set. The number of decision trees and the splitting threshold are iteratively optimized using the training set. The accuracy of the initial model is verified and the hyperparameters are adjusted using the test set.

[0022] If the accuracy of the initial model in predicting the health status of the equipment reaches a preset threshold, a health assessment model is generated.

[0023] Optionally, the feature parameters further include environmental correlation factors, including rainfall, ambient temperature, and coal content in wastewater;

[0024] The health assessment model is used to correct the mapping relationship between the characteristic parameters and the health status of the equipment based on the environmental correlation factors.

[0025] Optionally, the data analysis module further includes a digital twin engine;

[0026] The digital twin engine is used to receive the operating parameters and map the physical state of the vacuum self-filling water absorption module according to the operating parameters.

[0027] Optionally, building the digital twin engine includes:

[0028] Collect the physical parameters of the vacuum self-filling water absorption module and the historical operating data of the multi-source data sensing module to establish a standardized dataset;

[0029] A geometric model of the equipment is constructed using 3D modeling tools, and physical properties are assigned to the geometric model of the equipment through finite element analysis.

[0030] Set up a data interface to establish a dynamic mapping relationship between the vacuum self-priming water absorption module and the operating parameters;

[0031] Set up a reinforcement learning model, with equipment energy consumption and operational stability as constraints, input the adjustment command to the reinforcement learning model, and output a set of control parameters.

[0032] Optionally, the vacuum container is provided with a slag discharge device at the bottom;

[0033] The data analysis module is used to generate control commands to open or close the slag discharge device based on the changing trends of liquid level and vacuum pressure.

[0034] As can be seen from the above technical solutions, this application provides a coal-containing wastewater treatment system based on self-priming water intake and intelligent control. The system includes a vacuum self-priming water intake module, which includes a vacuum container for creating a negative pressure environment, an inlet pipe connected to the vacuum container, a pump group connected to the vacuum container, and a vacuum generator for providing negative pressure to the vacuum container; a multi-source data sensing module connected to the vacuum self-priming water intake module, which is used to collect the operating parameters of the vacuum self-priming water intake module; a data analysis module, which is used to receive the operating parameters, correct the control parameters of PID control according to the operating parameters, and output adjustment commands to the vacuum self-priming water intake module; and a remote operation and maintenance module, which is used to extract the feature parameters of the operating parameters, input the feature parameters into a health assessment model, output a device health score, and generate a maintenance report based on the device health score; the health assessment model is a prediction model trained based on a random forest algorithm. The system employs a vacuum self-priming water intake module, utilizing a vacuum generator to create a negative pressure environment within a vacuum container, enabling the self-priming intake of coal-containing wastewater. A multi-source data sensing module collects the operating parameters of the vacuum self-priming water intake module. A data analysis module receives these operating parameters and adjusts the PID control parameters accordingly, ensuring that the adjustment commands accurately adapt to the system's real-time operating conditions. This avoids the problem of fixed parameter control being unable to adapt to fluctuations in the flow and pressure of coal-containing wastewater, thus guaranteeing the stability and efficiency of the system operation. Furthermore, a remote operation and maintenance module extracts characteristic parameters from the operating parameters, uses a health assessment model to output equipment health scores, and generates maintenance reports, addressing the shortcomings of insufficient control precision and low operation and maintenance efficiency in coal-containing wastewater treatment systems. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0036] Figure 1 This is a schematic diagram of a coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control provided in an embodiment of the present invention;

[0037] Figure 2 This is a schematic diagram of the vacuum self-filling water absorption module provided in an embodiment of the present invention;

[0038] Figure 3 This is a schematic diagram of the multi-source data sensing module provided in an embodiment of the present invention;

[0039] Figure 4This is a schematic diagram of the data analysis module provided in an embodiment of the present invention.

[0040] Figure label:

[0041] 456 Among them, 101-vacuum self-filling water suction module; 102-multi-source data sensing module; 103-data analysis module; 104-remote operation and maintenance module; 1011-vacuum container; 1012-liquid inlet pipeline; 1013-pump set; 1014-vacuum generating device; 1021-liquid level sensor; 1022-flow sensor; 1023-pressure sensor; 1024-first pressure sensor; 1025-second pressure sensor; 1031-fuzzy rule base; 1032-digital twin engine; 10111-slag discharge device; 10121-filtration device; 10122-vacuum protection device; 10141-Venturi jet; 10142-liquid ring vacuum pump. Detailed Implementation

[0042] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0043] Figure 1 This is a schematic diagram of a coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control, provided in an embodiment of the present invention.

[0044] like Figure 1 As shown, the coal-containing wastewater treatment system based on self-irrigation water suction and intelligent control provided in this embodiment of the invention includes:

[0045] The vacuum self-filling water suction module 101 includes a vacuum container 1011 for creating a negative pressure environment, an inlet pipe 1012 connected to the vacuum container 1011, a pump set 1013 connected to the vacuum container 1011, and a vacuum generator 1014 for providing negative pressure to the vacuum container 1011.

[0046] When the vacuum self-priming water suction module 101 is working, the vacuum generator 1014 is activated, creating a negative pressure environment for the vacuum container 1011. This negative pressure environment in the vacuum container 1011 causes wastewater from the inlet pipe 1012 to be drawn into the vacuum container 1011, where it is then pumped by the pump unit 1013. Under the influence of negative pressure, the pump unit 1013 can more efficiently extract coal-containing wastewater, avoiding the problems of clogging or insufficient suction that occur in traditional suction methods. For example, in the process of treating coal-containing wastewater, when the wastewater contains a large amount of coal slag, the vacuum self-priming water suction module 101 can use its powerful negative pressure suction to draw in both the wastewater and the coal slag, reducing the occurrence of clogging.

[0047] A multi-source data sensing module 102 is connected to the vacuum self-priming water absorption module 101. The multi-source data sensing module 102 is used to collect the operating parameters of the vacuum self-priming water absorption module 101.

[0048] The multi-source data sensing module 102 is responsible for collecting the operating parameters of the vacuum self-priming water suction module 101. These operating parameters include, but are not limited to, the rotational speed of the pump set 1013, the negative pressure value of the vacuum container 1011, and the flow rate of the liquid inlet pipe 1012. The operating parameters are acquired in real time through sensors and transmitted to the data analysis module 103.

[0049] The data analysis module 103 is used to receive operating parameters, correct the control parameters of PID control according to the operating parameters, and output adjustment commands to the vacuum self-priming water suction module 101.

[0050] Specifically, after receiving the operating parameters from the multi-source data sensing module 102, the data analysis module 103 immediately processes and analyzes these parameters. The data analysis module 103 can dynamically correct the control parameters of the PID control based on the current operating parameters to ensure the stable operation of the coal-containing wastewater treatment system. PID control is a control algorithm that adjusts the control quantity through proportional, integral, and derivative components based on the deviation between the current state and the target state of the coal-containing wastewater treatment system, gradually bringing the system closer to the target state. In this embodiment of the invention, the data analysis module 103 utilizes the PID control algorithm to dynamically adjust the operating state of the pump unit 1013 based on real-time operating parameters, thereby achieving precise control of the coal-containing wastewater treatment process.

[0051] The remote operation and maintenance module 104 is used to extract the feature parameters of the operating parameters, input the feature parameters into the health assessment model, output the equipment health score, and generate a maintenance report based on the equipment health score; the health assessment model is a prediction model trained based on the random forest algorithm.

[0052] Specifically, the remote maintenance module 104 is primarily responsible for remote monitoring and maintenance of the vacuum self-priming water absorption module 101. The remote maintenance module 104 first extracts the operating parameters collected by the multi-source data sensing module 102 and processes these parameters to obtain characteristic parameters. These characteristic parameters reflect the operating status and health level of the self-priming water absorption module. Then, the remote maintenance module 104 inputs these characteristic parameters into a health assessment model trained using a random forest algorithm. The health assessment model analyzes and predicts the characteristic parameters, outputting a health score for the equipment. Based on the equipment health score, the remote maintenance module 104 generates a corresponding maintenance report, reminding maintenance personnel to perform timely maintenance and upkeep. For example, when the equipment health score falls below a preset threshold, the remote maintenance module 104 automatically triggers an alarm mechanism, notifying maintenance personnel to go to the site to inspect and handle potential faults.

[0053] In some embodiments, such as Figure 3 As shown, the multi-source data sensing module 102 includes a liquid level sensor 1021, a flow sensor 1022, and a pressure sensor 1023. The liquid level sensor 1021 is disposed inside the vacuum container 1011 and is used to collect the liquid level height of the vacuum container 1011. The flow sensor 1022 is disposed in the liquid inlet pipe 1012 and is used to collect the flow parameters of the medium in the liquid inlet pipe 1012. The pressure sensor 1023 includes a first pressure sensor 1024 disposed at the outlet of the pump group 1013 and a second pressure sensor 1025 disposed in the vacuum container 1011. The first pressure sensor 1024 collects the outlet pressure of the pump group 1013. The second pressure sensor 1025 collects the vacuum pressure inside the vacuum container 1011.

[0054] Specifically, a liquid level sensor 1021 is installed inside the vacuum container 1011. The liquid level sensor 1021 is used to collect the liquid level height of the vacuum container 1011 in real time. The liquid level height reflects the storage status of the medium inside the vacuum container 1011. For example, when the liquid level height approaches the upper or lower limit of the vacuum container 1011, the liquid level sensor 1021 can promptly provide feedback, reminding operators to take timely measures to prevent medium overflow or evacuation, thereby ensuring the stable operation of the coal-containing wastewater treatment system.

[0055] The flow sensor 1022 is installed in the inlet pipe 1012 to collect the flow parameters of the medium within the inlet pipe 1012. The flow parameters reflect the treatment capacity of the coal-containing wastewater treatment system and the flow state of the medium. By monitoring the flow data, operators can promptly detect flow anomalies, such as excessive or insufficient flow, and thus determine whether there are problems such as pipe blockage or pump unit 1013 malfunction.

[0056] The pressure sensor 1023 includes a first pressure sensor 1024 disposed at the outlet of the pump unit 1013 and a second pressure sensor 1025 disposed in the vacuum container 1011. The first pressure sensor 1024 collects the outlet pressure of the pump unit 1013, which helps to understand the operating status of the pump unit 1013 and the delivery pressure of the medium. The second pressure sensor 1025 collects the vacuum pressure inside the vacuum container 1011, which can assess the sealing performance of the vacuum container 1011 and the vacuum treatment effect of the medium. For example, when the vacuum pressure rises abnormally, it indicates that the sealing performance of the vacuum container 1011 has decreased or there is a problem such as medium leakage.

[0057] The multi-source data sensing module 102 is used to sort the liquid level height, flow rate parameter, outlet pressure and vacuum pressure in chronological order, and divide the liquid level height, flow rate parameter, outlet pressure and vacuum pressure with the same timestamp into data groups to generate operating parameters.

[0058] The multi-source data sensing module 102 first receives real-time data from the level sensor 1021, flow sensor 1022, and pressure sensor 1023. The real-time data includes liquid level height, flow parameters, pump unit 1013 outlet pressure, and vacuum pressure inside the vacuum container 1011. To ensure the accuracy and reliability of the real-time data, the multi-source data sensing module 102 preprocesses the received data, including data cleaning, noise reduction, and calibration.

[0059] After the multi-source data sensing module 102 preprocesses the received data, it sorts the real-time data according to time sequence. This sorting can be based on the timestamps of the collected data, which, as a crucial data attribute, ensures accuracy and consistency during the sorting process. The sorted data will then be arranged in chronological order, forming a continuous data stream.

[0060] Based on the sorting, the multi-source data sensing module 102 groups liquid level height, flow rate parameters, outlet pressure, and vacuum pressure with the same timestamp into a data group. Each data group represents the operating status of the coal-containing wastewater treatment system at a specific point in time. Each data group contains all the key operating parameters for that point in time, providing comprehensive information support for subsequent data analysis and processing.

[0061] For example, at a certain time point T, the level sensor 1021 measures the liquid level height as H1, the flow sensor 1022 measures the flow rate parameter as F1, the pressure sensor 1023 measures the outlet pressure of the pump unit 1013 as P1, and the vacuum pressure inside the vacuum container 1011 is V1. The data at time point T will then be divided into a data set, represented as {T, H1, F1, P1, V1}, indicating the operating status of the coal-containing wastewater treatment system at time point T.

[0062] Through the above steps, a series of data sets arranged in chronological order can be obtained, comprehensively reflecting the operating status of the coal-containing wastewater treatment system. After arranging and integrating the data sets in chronological order, operating parameters are obtained. These operating parameters are then uploaded to the data analysis module 103 for further analysis and processing to support the intelligent control and optimized operation of the coal-containing wastewater treatment system.

[0063] In some embodiments, such as Figure 2 As shown, the liquid inlet pipeline 1012 is equipped with a filter device 10121 and a vacuum protection device 10122. The filter device 10121 is used to intercept particulate impurities in the coal-containing wastewater; the vacuum protection device 10122 is used to reduce the vacuum pressure in the vacuum container 1011 when the vacuum pressure in the vacuum container 1011 exceeds a preset threshold.

[0064] The filter device 10121 can effectively prevent large particles of impurities from entering the vacuum container 1011, avoid damage to the vacuum pump, and ensure the stable operation of the coal-containing wastewater treatment system.

[0065] In addition, the vacuum protection device 10122 can also protect the vacuum container 1011. When the vacuum pressure inside the vacuum container 1011 rises abnormally, the vacuum protection device 10122 can respond quickly and effectively reduce the pressure inside the vacuum container 1011 by introducing external gas, thereby protecting the vacuum pump from damage.

[0066] In some embodiments, the filter device 10121 is detachably connected to the vacuum container 1011, which facilitates regular cleaning and replacement of the filter device 10121 to ensure that the filter device 10121 continuously and effectively intercepts particulate impurities.

[0067] Meanwhile, the vacuum protection device 10122 employs fast-response valve technology. When the vacuum pressure inside the vacuum container 1011 reaches or exceeds a preset safety threshold, the pressure sensor 1023 immediately triggers a signal, and the fast-response valve opens rapidly, allowing external gas to safely enter the vacuum container 1011, thereby quickly balancing the internal pressure and effectively preventing the vacuum pump from being damaged due to overpressure.

[0068] In some embodiments, such as Figure 2As shown, the vacuum generating device 1014 includes a Venturi jet injector 10141 and a liquid ring vacuum pump 10142.

[0069] Data analysis module 103 is used for:

[0070] If the vacuum pressure is greater than the preset pressure threshold, a control command to start the Venturi jet 10141 is generated.

[0071] If the vacuum pressure is less than the preset pressure threshold, a control command is generated to start the liquid ring vacuum pump 10142.

[0072] The Venturi jet 10141 operates based on Bernoulli's principle in fluid mechanics. When high-speed fluid passes through the constricted section of the Venturi tube, the flow velocity increases, the static pressure decreases, and a negative pressure region is formed. This negative pressure region can extract external gas or liquid, thus achieving a vacuum effect. In the coal-containing wastewater treatment system, when the vacuum pressure exceeds a preset pressure threshold, the data analysis module 103 determines that a larger pumping capacity is needed and generates a control command to activate the Venturi jet 10141. The Venturi jet 10141 responds quickly, utilizing its efficient pumping capacity to effectively reduce the pressure inside the vacuum container 1011.

[0073] The liquid ring vacuum pump 10142 achieves its pumping function by changing the space between the rotating liquid ring and the pump casing. As the liquid ring rotates, the space between the liquid ring and the pump casing first expands and then contracts, creating a negative pressure to extract external gas. When the vacuum pressure is lower than a preset pressure threshold, the data analysis module 103 determines that a stable vacuum environment is needed and therefore generates a control command to start the liquid ring vacuum pump 10142. The liquid ring vacuum pump 10142 provides stable pumping performance and a low noise level, providing a continuous vacuum environment for coal-containing wastewater treatment systems.

[0074] Through intelligent control of the vacuum generator 1014, the Venturi ejector 10141 or the liquid ring vacuum pump 10142 can be flexibly selected according to actual needs. For example, when treating a large amount of coal-containing wastewater, the pressure inside the vacuum container 1011 may rise rapidly. At this time, the data analysis module 103 will quickly activate the Venturi ejector 10141 to rapidly reduce the pressure. In daily operation and maintenance, when the pressure inside the vacuum container 1011 is kept at a low level, the liquid ring vacuum pump 10142 provides a continuous vacuum environment for the system with its stable performance.

[0075] Through intelligent control, not only is the treatment efficiency improved, but energy consumption and noise levels are also reduced, achieving more environmentally friendly and efficient treatment of coal-containing wastewater.

[0076] In some embodiments, such as Figure 4As shown, the data analysis module 103 includes a fuzzy rule base 1031. The fuzzy rule base 1031 is used to receive the operating parameters collected by the multi-source data sensing module 102, and by analyzing the deviation and rate of change of the operating parameters, adjust the proportional coefficient, integral time and derivative time of the PID controller to generate corrected control parameters, and output adjustment commands to the vacuum self-priming water suction module 101 for the corresponding pump group operating frequency and inlet pipeline valve opening.

[0077] Specifically, the fuzzy rule base 1031 is used to receive the operating parameters collected by the multi-source data sensing module 102 and analyze the deviation and rate of change of the operating parameters, such as the difference between the actual value and the set value, and the rate of change of the deviation over time. By analyzing the deviation and the rate of change of the deviation, the fuzzy rule base 1031 can determine the state of the coal-containing wastewater treatment system and adjust the proportional coefficient, integral time, and derivative time of the PID controller accordingly.

[0078] In the process of adjusting the PID controller, the proportional gain determines the response speed of the coal-containing wastewater treatment system to deviations; the integral time is used to eliminate static errors, ensuring that the coal-containing wastewater treatment system can eventually reach the set value; the derivative time predicts the future trend of the deviation, thereby making adjustments in advance to avoid overshoot or oscillation. The fuzzy rule base 1031 dynamically adjusts these parameters according to the magnitude and direction of the deviation and the rate of change of the deviation to generate corrected control parameters.

[0079] The corrected control parameters are then output to the vacuum self-priming water suction module 101 to guide the adjustment of the corresponding pump group operating frequency and the opening of the inlet pipeline valve, thereby improving the response speed and accuracy of the coal-containing wastewater treatment system, as well as enhancing the adaptability and robustness of the coal-containing wastewater treatment system, enabling the coal-containing wastewater treatment system to maintain stable performance under various operating conditions.

[0080] The fuzzy rule base 1031 first receives operating parameters collected by the multi-source data sensing module 102, compares the operating parameters with preset standard values, and calculates the deviation value and deviation change rate of each parameter. Next, based on preset fuzzy classification standards, the deviation value and deviation change rate are mapped to fuzzy subsets such as positive large, positive medium, zero, negative medium, and negative large, respectively, realizing the conversion of precise numerical values ​​into fuzzy linguistic variables. Subsequently, it calls the logical rules stored in the fuzzy rule base 1031; for example, if the liquid level deviation is positive large and the deviation change rate is positive medium, then the proportional coefficient correction is positive large. Matching and reasoning are performed on the input fuzzy variables to obtain the fuzzy correction amounts of the proportional coefficient, integral time, and derivative time of the PID controller. Then, through a defuzzification algorithm, such as using the centroid method, the fuzzy correction amounts are converted into specific precise values, generating corrected PID control parameters. Finally, based on the corrected control parameters, the adjustment value of the pump group operating frequency and the adjustment amount of the inlet pipeline valve opening are calculated, forming specific adjustment commands and sending them to the vacuum self-priming water suction module 101 to complete the real-time adjustment of the system's operating status.

[0081] For example, the standard liquid level of the vacuum container 1011 in the coal-containing wastewater treatment system is preset to 50cm. When the multi-source data sensing module 102 collects the current liquid level as 70cm, it calculates the deviation and the rate of change of deviation. The liquid level deviation is 70cm-50cm=20cm. If the deviation increases from 15cm to 20cm within 10 seconds, the rate of change of deviation is (20cm-15cm) / 10s=0.5cm / s, indicating an increasing trend in deviation.

[0082] According to preset standards, a 20cm deviation corresponds to the large fuzzy subset, and a 0.5cm / s change rate corresponds to the medium fuzzy subset. The fuzzy rule in the fuzzy rule library 1031 is called, which states that if the liquid level deviation is large and the deviation change rate is medium, then the proportional coefficient Kp correction is large and the integral time Ti correction is medium, thus obtaining the fuzzy correction amount. The large and medium fuzzy correction amounts are converted into precise values, for example, Kp increases by 30% and Ti decreases by 20%. Based on the corrected PID parameters, specific instructions are calculated to increase the pump unit operating frequency by 15Hz and decrease the inlet pipe valve opening by 20%, and sent to the vacuum self-priming water suction module 101 to accelerate the drainage rate and lower the liquid level.

[0083] In some embodiments, constructing a health assessment model includes:

[0084] Collect historical monitoring data, fault records, and maintenance logs from the equipment, and extract characteristic parameters including pressure fluctuation amplitude, flow rate decay rate, and liquid level adjustment frequency.

[0085] An initial model was constructed using the random forest algorithm. The feature parameters were divided into training and test sets. The number of decision trees and the split threshold were iteratively optimized using the training set. The accuracy of the initial model was verified and the hyperparameters were adjusted using the test set.

[0086] If the initial model's accuracy in predicting the health status of the equipment reaches a preset threshold, a health assessment model is generated.

[0087] Specifically, historical monitoring data, fault records, and maintenance logs are derived from the equipment's daily operation records and form the basis for building a health assessment model. To ensure the comprehensiveness and accuracy of the data, information needs to be collected from multiple dimensions, including but not limited to the equipment's pressure fluctuation amplitude, flow rate decay rate, and liquid level regulation frequency characteristic parameters. These characteristic parameters can intuitively reflect the equipment's operating status and health level.

[0088] Then, an initial model is constructed using the random forest algorithm. Random forest is an ensemble learning method that improves the stability and accuracy of the model by constructing multiple decision trees and combining their prediction results. In this embodiment of the invention, the collected feature parameters are divided into a training set and a test set. The training set is used to iteratively optimize the number of decision trees and the split threshold to construct an initial model whose performance meets the requirements. The test set is used to verify the accuracy of the initial model, and hyperparameters, such as the maximum depth of the decision trees and the minimum number of samples, are adjusted based on the verification results to further improve the model's performance.

[0089] During model training, the predictive ability of the model can be gradually optimized by continuously adjusting the number of decision trees and the split threshold. Simultaneously, methods such as cross-validation are used to fully validate the model, ensuring its stability and generalization ability across different datasets. When the initial model's prediction accuracy for device health status reaches a preset threshold, the model can be considered to have sufficient accuracy and reliability, and can be used to generate a health assessment model.

[0090] Understandably, data preprocessing and feature selection are crucial steps in building a health assessment model. Data preprocessing includes data cleaning, missing value imputation, and outlier detection, aiming to improve data quality and the accuracy of feature parameters. Feature selection, on the other hand, involves filtering from numerous features to identify the key features that contribute most to the model's predictive ability, thereby simplifying the model structure and improving computational efficiency.

[0091] In some embodiments, the characteristic parameters also include environmental factors, such as rainfall, ambient temperature, and coal content in wastewater.

[0092] Health assessment models are used to adjust the mapping relationship between characteristic parameters and equipment health status based on environmental factors.

[0093] Among these, health assessment models that do not consider environmental factors are constructed solely based on traditional characteristic parameters, neglecting environmental factors such as rainfall, ambient temperature, and coal content in wastewater. In practical applications, these models may lead to biased health assessment results due to insufficient consideration of changes in external environmental factors. For example, a sudden increase in rainfall or a drastic change in ambient temperature may affect the operational status of a coal-containing wastewater treatment system, but the health assessment model may fail to accurately reflect such changes.

[0094] In this embodiment of the invention, the health assessment model is not only constructed based on traditional characteristic parameters, but also fully considers environmental correlation factors, including rainfall, ambient temperature, and coal content in wastewater. By introducing environmental correlation factors, the health assessment model corrects the mapping relationship between characteristic parameters and equipment health status. In practical applications, the health assessment model can more accurately reflect the operating status of equipment under different environmental conditions, thereby improving the accuracy and practicality of the health assessment model. For example, with increased rainfall, the health assessment model can predict the increase in equipment load and the decrease in processing efficiency, thus promptly reminding maintenance personnel to make corresponding adjustments and maintenance.

[0095] Comparative analysis reveals that health assessment models incorporating environmental factors exhibit significant advantages in accuracy and practicality. Health assessment models that do not consider environmental factors fail to adequately reflect changes in external environmental factors, leading to biased assessment results. In contrast, models incorporating environmental factors more accurately reflect the operational status of equipment under different environmental conditions, providing a more scientific basis for the maintenance and management of coal-containing wastewater treatment systems.

[0096] In some embodiments, such as Figure 4 As shown, the data analysis module 103 also includes a digital twin engine 1032.

[0097] The digital twin engine 1032 is used to receive operating parameters and map the physical state of the vacuum self-filling water absorption module 101 according to the operating parameters.

[0098] The digital twin engine 1032 simulates the operation of the vacuum self-priming water absorption module 101 to achieve real-time monitoring and prediction of the module. By combining with a health assessment model, the digital twin engine 1032 can more accurately predict the future operating status of the vacuum self-priming water absorption module 101, thereby taking corresponding maintenance measures in advance to avoid equipment failure.

[0099] Specifically, building the digital twin engine 1032 includes:

[0100] Collect the physical parameters of the vacuum self-filling water absorption module 101 and the historical operating data of the multi-source data sensing module 102 to establish a standardized dataset.

[0101] The equipment's geometric model is constructed using 3D modeling tools, and its physical properties are assigned through finite element analysis.

[0102] Set up a data interface to establish a dynamic mapping relationship between the vacuum self-priming water absorption module 101 and the operating parameters.

[0103] Set up a reinforcement learning model, with equipment energy consumption and operational stability as constraints, input adjustment commands to the reinforcement learning model, and output a set of control parameters.

[0104] This involves collecting physical parameters from the vacuum self-priming water intake module 101, such as temperature, pressure, and flow rate, as well as historical operating data from the multi-source data sensing module 102, such as fault records and maintenance logs. The physical parameters and historical operating data are integrated and preprocessed to establish a standardized dataset. This standardized dataset ensures the accuracy and consistency of the data.

[0105] Next, a geometric model of the equipment is constructed using 3D modeling tools. Specifically, a precise model is created based on the actual dimensions and structure of the vacuum self-priming water absorption module 101, reflecting its true physical form. Subsequently, finite element analysis software is used to assign physical properties to the geometric model, including simulating the elastic modulus, density, thermal conductivity, and other physical characteristics of the materials, as well as considering the mechanical and thermal loads that the equipment may experience during operation, thereby more accurately reflecting the actual operating state of the equipment.

[0106] Next, efficient data transmission protocols and interface standards need to be designed to ensure the real-time and accurate acquisition and transmission of equipment operating parameters. Simultaneously, a database needs to be established to store and manage these operating parameters for subsequent analysis and simulation.

[0107] Finally, a reinforcement learning model is set up, with equipment energy consumption and operational stability as constraints. Adjustment commands are input to the reinforcement learning model, and the model outputs a set of control parameters. Through continuous trial and error and optimization, the reinforcement learning model gradually adjusts the control parameters to achieve efficient equipment operation and minimize energy consumption. It is important to note that a suitable reinforcement learning model needs to be selected, and a reasonable reward function and state space need to be designed to ensure that the reinforcement learning model can accurately reflect the actual operating state of the equipment and provide an effective control strategy. At the same time, the reinforcement learning model needs to be thoroughly trained and validated to ensure its stability and reliability in practical applications.

[0108] Understandably, when constructing a reinforcement learning model, the objectives and constraints of the reinforcement learning are first clearly defined. In this embodiment of the invention, the objective is to achieve efficient operation and minimize energy consumption of the vacuum self-filling water intake module 101, and the constraints include equipment energy consumption and operational stability. The objectives and constraints will guide the design and training of the reinforcement learning model.

[0109] Next, the state space, action space, and reward function of the reinforcement learning are designed. The state space describes the possible states of the vacuum self-priming water intake module 101, the action space defines the control actions that can be taken, and the reward function is used to evaluate the effect of taking a certain action. In this invention, the state space may include physical parameters such as temperature, pressure, and flow rate of the equipment, as well as historical operating data, while the action space may include control parameters such as pump operating frequency and inlet pipeline valve opening. The reward function can be set to give a positive reward when the equipment energy consumption decreases and a negative reward when the equipment operation is unstable.

[0110] Next, a suitable reinforcement learning model is selected. In this embodiment of the invention, the DQN (Deep Q-Network) algorithm is chosen as the basis for the reinforcement learning model. The DQN algorithm combines the advantages of deep learning and reinforcement learning, and can handle high-dimensional state spaces and complex action decision problems.

[0111] Next, the reinforcement learning model is trained and validated. During training, a large number of state-action pairs are generated using simulated environments or historical data, and the value of each state-action pair is calculated based on the reward function. Then, a deep learning network is used to fit the state-action pairs to obtain the value estimate of each action in each state. Through continuous iterative training, the network gradually converges to obtain the optimal policy. The validation process involves testing the trained model in a real-world environment to ensure its stability and reliability.

[0112] For example, the state space of the vacuum self-priming water intake module 101 includes three parameters: temperature T, pressure P, and flow rate Q. The action space includes two control parameters: the operating frequency F of the pump unit 1013 and the valve opening V of the inlet pipeline 1012. The goal is to find a strategy π that selects the optimal action under a given state to achieve efficient operation of the vacuum self-priming water intake module 101 and minimize energy consumption.

[0113] First, a reward function R was designed, which calculates the reward value based on the device's energy consumption and operational stability. For example, a positive reward is given when energy consumption decreases, and a negative reward is given when the device operates unstablely, such as when temperature fluctuations are too large.

[0114] Then, the DQN algorithm was chosen as the basis for the reinforcement learning model, and a deep learning network was constructed to fit the value function of the state-action pair. The network input is the state vector [T,P,Q], and the output is the value estimate corresponding to the action vector [F,V].

[0115] During training, a large number of state-action pairs were generated using a simulated environment, and the value of each action was calculated based on the reward function. These state-action pairs were then fed into a deep learning network for training. Through iterative training, the network gradually converged, yielding the optimal policy π.

[0116] Finally, the trained reinforcement learning model is integrated into the digital twin engine 1032. During actual operation, the digital twin engine 1032 selects the optimal control parameters based on the real-time status of the vacuum self-priming water absorption module 101 and dynamically adjusts them through the reinforcement learning model to achieve efficient operation and minimize energy consumption. Simultaneously, the digital twin engine 1032 also provides remote monitoring and fault diagnosis functions, further improving the reliability and maintenance efficiency of the vacuum self-priming water absorption module 101.

[0117] In some embodiments, such as Figure 2 As shown, the vacuum container 1011 is equipped with a slag discharge device 10111 at the bottom.

[0118] The data analysis module 103 is used to generate control commands to open or close the slag discharge device 10111 based on the changing trends of liquid level and vacuum pressure.

[0119] The inclusion of a slag discharge device 10111 helps maintain a stable working environment within the vacuum container 1011, preventing excessive solid particle accumulation from affecting the operational efficiency of the coal-containing wastewater treatment system. When the liquid level reaches a preset threshold or the vacuum pressure shows an upward trend, the data analysis module 103 intelligently determines whether the slag discharge device 10111 needs to be activated to remove excess solid waste. Conversely, when the liquid level decreases and the vacuum pressure stabilizes, the control command will shut down the slag discharge device 10111, reducing energy consumption. This intelligent control method not only improves the efficiency of wastewater treatment but also ensures the long-term stable operation of the coal-containing wastewater treatment system.

[0120] As can be seen from the above technical solutions, this application provides a coal-containing wastewater treatment system based on self-priming water intake and intelligent control. The coal-containing wastewater treatment system includes a vacuum self-priming water intake module 101, which includes a vacuum container 1011 for creating a negative pressure environment, an inlet pipe 1012 connected to the vacuum container 1011, a pump group 1013 connected to the vacuum container 1011, and a vacuum generator 1014 for providing negative pressure to the vacuum container 1011; a multi-source data sensing module connected to the vacuum self-priming water intake module 101. Block 102, the multi-source data sensing module 102 is used to collect the operating parameters of the vacuum self-priming water suction module 101; the data analysis module 103 is used to receive the operating parameters, correct the control parameters of PID control according to the operating parameters, and output adjustment commands to the vacuum self-priming water suction module 101; the remote operation and maintenance module 104 is used to extract the feature parameters of the operating parameters, input the feature parameters into the health assessment model, output the equipment health score, and generate a maintenance report based on the equipment health score; the health assessment model is a prediction model trained based on the random forest algorithm. The coal-containing wastewater treatment system incorporates a vacuum self-priming suction module 101, which uses a vacuum generator 1014 to create a negative pressure environment within a vacuum container 1011, enabling the self-priming suction of coal-containing wastewater. A multi-source data sensing module 102 collects the operating parameters of the vacuum self-priming suction module 101. A data analysis module 103 receives the operating parameters and corrects the PID control parameters accordingly, ensuring that the adjustment commands accurately adapt to the real-time operating conditions of the system. This avoids the problem of fixed parameter control being unable to adapt to fluctuations in the flow and pressure of coal-containing wastewater, thus guaranteeing the stability and efficiency of the system operation. Furthermore, a remote maintenance module 104 extracts characteristic parameters from the operating parameters, uses a health assessment model to output a health score for the equipment, and generates a maintenance report, addressing the shortcomings of insufficient control precision and low maintenance efficiency in the coal-containing wastewater treatment system.

[0121] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control, characterized in that, include: A vacuum self-filling water suction module includes a vacuum container for creating a negative pressure environment, an inlet pipe connected to the vacuum container, a pump unit connected to the vacuum container, and a vacuum generator for providing negative pressure to the vacuum container. A multi-source data sensing module connected to the vacuum self-priming water absorption module, the multi-source data sensing module being used to collect the operating parameters of the vacuum self-priming water absorption module; The data analysis module is used to receive the operating parameters, correct the control parameters of the PID control according to the operating parameters, and output adjustment commands to the vacuum self-priming water suction module. The remote operation and maintenance module is used to extract the feature parameters of the operating parameters, input the feature parameters into the health assessment model, output the equipment health score, and generate a maintenance report based on the equipment health score; the health assessment model is a prediction model trained based on the random forest algorithm.

2. The coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control according to claim 1, characterized in that, The multi-source data sensing module includes a liquid level sensor, a flow sensor, and a pressure sensor. The liquid level sensor is installed inside the vacuum container to collect the liquid level height of the vacuum container. The flow sensor is installed in the inlet pipe to collect the flow rate parameters of the medium in the inlet pipe. The pressure sensor includes a first pressure sensor installed at the pump outlet and a second pressure sensor installed in the vacuum container. The first pressure sensor collects the outlet pressure of the pump, and the second pressure sensor collects the vacuum pressure inside the vacuum container. The multi-source data sensing module is used to sort the liquid level height, the flow rate parameter, the outlet pressure and the vacuum pressure in chronological order, and divide the liquid level height, the flow rate parameter, the outlet pressure and the vacuum pressure with the same timestamp into data groups to generate operating parameters.

3. The coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control according to claim 1, characterized in that, The liquid inlet pipeline is equipped with a filter device and a vacuum protection device. The filter device is used to intercept particulate impurities in coal-containing wastewater, and the vacuum protection device is used to reduce the vacuum pressure in the vacuum container when the vacuum pressure in the vacuum container exceeds a preset threshold.

4. The coal-containing wastewater treatment system based on self-irrigation water suction and intelligent control according to claim 2, characterized in that, The vacuum generating device includes a Venturi jet injector and a liquid ring vacuum pump; The data analysis module is used for: If the vacuum pressure is greater than a preset pressure threshold, a control command to start the Venturi jet is generated. If the vacuum pressure is less than the preset pressure threshold, a control command is generated to start the liquid ring vacuum pump.

5. The coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control according to claim 1, characterized in that, The data analysis module includes a fuzzy rule base; The fuzzy rule base is used to receive the operating parameters collected by the multi-source data sensing module, and by analyzing the deviation and rate of change of the operating parameters, adjust the proportional coefficient, integral time and derivative time of the PID controller to generate corrected control parameters, and output adjustment commands to the vacuum self-priming water suction module for the corresponding pump group operating frequency and inlet pipeline valve opening.

6. The coal-containing wastewater treatment system based on self-irrigation water suction and intelligent control according to claim 1, characterized in that, The construction of a health assessment model includes: Collect historical monitoring data, fault records and maintenance logs of the equipment, and extract characteristic parameters including pressure fluctuation amplitude, flow rate decay rate and liquid level adjustment frequency; An initial model is constructed using the random forest algorithm. The feature parameters are divided into a training set and a test set. The number of decision trees and the splitting threshold are iteratively optimized using the training set. The accuracy of the initial model is verified and the hyperparameters are adjusted using the test set. If the accuracy of the initial model in predicting the health status of the equipment reaches a preset threshold, a health assessment model is generated.

7. The coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control according to claim 6, characterized in that, The characteristic parameters also include environmental factors, which include rainfall, ambient temperature, and coal content in wastewater. The health assessment model is used to correct the mapping relationship between the characteristic parameters and the health status of the equipment based on the environmental correlation factors.

8. The coal-containing wastewater treatment system based on self-irrigation water suction and intelligent control according to claim 1, characterized in that, The data analysis module also includes a digital twin engine; The digital twin engine is used to receive the operating parameters and map the physical state of the vacuum self-filling water absorption module according to the operating parameters.

9. The coal-containing wastewater treatment system based on self-irrigation water suction and intelligent control according to claim 8, characterized in that, Building the digital twin engine includes: Collect the physical parameters of the vacuum self-filling water absorption module and the historical operating data of the multi-source data sensing module to establish a standardized dataset; A geometric model of the equipment is constructed using 3D modeling tools, and physical properties are assigned to the geometric model of the equipment through finite element analysis. Set up a data interface to establish a dynamic mapping relationship between the vacuum self-priming water absorption module and the operating parameters; Set up a reinforcement learning model, with equipment energy consumption and operational stability as constraints, input the adjustment command to the reinforcement learning model, and output a set of control parameters.

10. The coal-containing wastewater treatment system based on self-irrigation water intake and intelligent control according to claim 2, characterized in that, The vacuum container is equipped with a slag discharge device at the bottom. The data analysis module is used to generate control commands to control the opening or closing of the slag discharge device based on the changing trends of the liquid level and the vacuum pressure.

Images