Sewage treatment accurate dosing control system and method based on intelligent multi-dimensional prediction model

Abstract

The invention discloses a sewage treatment accurate dosing control system and method based on an intelligent multi-dimensional prediction model, and the system comprises a coarse screen and water inlet pump room, a fine screen and aeration sedimentation tank, an AAO biochemical tank, a secondary sedimentation tank, an efficient sedimentation tank, a cloth filter, a disinfection tank and a water outlet pump room which are sequentially communicated. The water inlet of the AAO biochemical pool, the water outlet of the AAO biochemical pool, the water outlet of the secondary sedimentation pool and the water outlet of the efficient sedimentation pool are respectively provided with a water quality TP total phosphorus concentration detection device, and the water inlet of the AAO biochemical pool, the water inlet of the secondary sedimentation pool, the water inlet of the efficient sedimentation pool and the water outlet of the efficient sedimentation pool are respectively provided with a water flow detection device. The system has the advantages that multi-dimensional influence factors can be integrated for accurate prediction, the dosing strategy is dynamically adjusted through real-time monitoring data, the sewage treatment efficiency is effectively improved, the operation cost is reduced, and the effluent quality is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of automatic control technology for wastewater treatment, and in particular to a precise dosing control system and method for wastewater treatment based on an intelligent multidimensional prediction model. Background Technology

[0002] In the field of wastewater treatment, reducing the phosphorus content in wastewater is a key step in ensuring that effluent quality meets standards. Traditional phosphorus removal methods mainly include biological enhanced phosphorus removal and coagulation sedimentation phosphorus removal processes. Among them, coagulation sedimentation phosphorus removal is widely used in most wastewater treatment plants in China due to its simple operation and significant phosphorus removal effect. The core principle of this process is to add a coagulant (such as iron salts or aluminum salts) to the wastewater. The coagulant reacts chemically with the phosphorus in the water to form insoluble precipitates, which are then separated and removed from the water through a sedimentation process.

[0003] However, traditional methods of chemical dosing in coagulation and sedimentation processes have significant drawbacks. Determining the dosage often relies on the operator's past experience and historical monitoring data, a method lacking precision and highly susceptible to operator subjectivity. This directly leads to two main problems: either excessive dosage results in wasted chemicals and increased treatment costs, or insufficient dosage leads to poor phosphorus removal, failing to guarantee stable phosphorus concentrations in the effluent.

[0004] Furthermore, chemical phosphorus removal is a crucial process in wastewater treatment, and its effectiveness directly impacts the final effluent quality. Traditional dosing control methods, whether relying on manual adjustments based on experience or employing simple PID control strategies, are ill-suited to effectively handle the frequent and drastic fluctuations in influent water quality (such as phosphorus concentration, pH, and organic matter content) and flow rate during wastewater treatment. This inadequacy easily leads to overdosing or underdosing of chemicals. Overdosing not only increases treatment costs but may also introduce new chemical substances that could negatively impact subsequent treatment units or the ecological environment; underdosing, on the other hand, fails to ensure effective phosphorus removal, resulting in the risk of effluent quality exceeding standards.

[0005] Current wastewater treatment chemical dosing control systems generally suffer from the following deficiencies: Reliance on human experience: The dosage of chemicals is set based on the operator's experience, which cannot cope with the dynamic fluctuations in water quality and quantity, and is prone to overdosing (increased costs) or underdosing (excessive discharge standards). Single-parameter control lag: Traditional PID control is based solely on feedback of effluent TP concentration, lacking the ability to predict changes in influent water quality, and the lag in response leads to water quality fluctuations; Resource waste: The lack of a multi-process parameter correlation model resulted in low matching between reagent dosage and actual demand, leading to high operating costs.

[0006] In summary, current technologies for chemical dosing control in coagulation sedimentation phosphorus removal processes generally lack an intelligent control system that can organically combine feedforward prediction (based on predictions of future operating conditions) and real-time feedback (based on current actual treatment results). This technological gap makes it difficult to achieve precise and dynamic dosing of chemicals and to optimize dosing strategies in real time according to complex and ever-changing influent conditions. Summary of the Invention

[0007] The purpose of this invention is to provide a precise dosing control system and method for wastewater treatment based on an intelligent multidimensional prediction model. This system can integrate multiple influencing factors for accurate prediction and dynamically adjust the dosing strategy through real-time monitoring data, thereby effectively improving wastewater treatment efficiency, reducing operating costs, and ensuring effluent quality.

[0008] The above-mentioned technical objective of the present invention is achieved through the following technical solution: A precise dosing control system for wastewater treatment based on an intelligent multidimensional prediction model is characterized by comprising, in sequence, a coarse screen and inlet pump station, a fine screen and aeration sedimentation tank, an AAO biological treatment tank, a secondary sedimentation tank, a high-efficiency sedimentation tank, a filter cloth filter, a disinfection tank, and an effluent pump station. The inlet and outlet of the AAO biological treatment tank, the outlet of the secondary sedimentation tank, and the outlet of the high-efficiency sedimentation tank are all equipped with total phosphorus (TP) concentration detection devices. The inlet of the AAO biological treatment tank, the inlet of the secondary sedimentation tank, the inlet of the high-efficiency sedimentation tank, and the high-efficiency sedimentation tank are all connected in sequence. The outlet of the water tank is equipped with a water flow detection device. Both the water quality TP total phosphorus concentration detection device and the water flow detection device are connected to a core control unit. The inlet of the high-efficiency sedimentation tank is equipped with an automatic dosing device for PAC polyaluminum chloride and an automatic dosing device for PAM polyacrylamide. The inlet of the secondary sedimentation tank is equipped with an automatic dosing device for PAFC polyferric chloride. The automatic dosing devices for PAC polyaluminum chloride, PAM polyacrylamide, and PAFC polyferric chloride are all controlled by the core control unit.

[0009] Preferably, the sludge outlet of the secondary sedimentation tank is also connected in sequence to a sludge pumping station, a gravity thickening tank, a sludge storage tank, and a sludge dewatering machine room. The sludge in the sludge dewatering machine room is transported off-site for treatment, while the supernatant is returned to the coarse screen and the inlet pumping station.

[0010] Preferably, the AAO biological treatment tank is also connected to a blower room, and the return sludge from the sludge pump room is connected to the AAO biological treatment tank.

[0011] Preferably, the chemical sludge outlet of the high-efficiency sedimentation tank is connected to the gravity thickening tank through a sludge lifting well, the sludge outlet of the filter cloth filter is connected to the sludge dewatering room, and the disinfection tank is also connected to a chlorination and dosing room.

[0012] A method for a precision dosing control system for wastewater treatment based on an intelligent multidimensional prediction model includes the following steps: S1. Collect the total phosphorus (TP) concentration at the inlet of the AAO biological treatment tank and the outlet of the high-efficiency sedimentation tank over several time periods. Let the TP concentration at the inlet be TP. in The total phosphorus concentration (TP) at the outlet is TP out The water flow rates at the inlet of the AAO biological treatment tank and the outlet of the high-efficiency sedimentation tank are given by Q. in The water flow rate at the outlet is Q. out The dosage of PAFC (polyferric chloride), PAC (polyaluminum chloride), and PAM (polyacrylamide) was collected for several time periods. The dosage of PAFC (polyferric chloride) was A1, the dosage of PAC (polyaluminum chloride) was A2, and the dosage of PAM (polyacrylamide) was A3. S2. Train the multidimensional regression prediction model based on automatic machine learning using data collected over several time periods to obtain the trained multidimensional regression prediction model based on automatic machine learning; the input variable in the core parameters of the multidimensional regression prediction model is TP. in TP out Q in and Q out The output variables are A1, A2, and A3; S3. Based on the total phosphorus concentration (TP) and flow rate of the inlet water of the AAO biochemical tank during the current time period, a trained multidimensional regression prediction model based on automatic machine learning is used to predict the dosage of PAFC (polyferric chloride), PAC (polyaluminum chloride), and PAM (polyacrylamide) during the current time period. S4. Calculate the standardized root mean square error (NRMSE) based on the drug dosage prediction results obtained in S3. When the calculated standardized root mean square error (NRMSE) is greater than the set error threshold, input the difference between the calculated standardized root mean square error (NRMSE) and the set error threshold into the trained multidimensional regression prediction model based on automatic machine learning, and return to S3. Continue until the calculated standardized root mean square error (NRMSE) of the prediction results is less than the set error threshold. Finally, adjust the obtained predicted dosage in real time through the core control unit to control the PAC (polyaluminum chloride) automatic dosing device, the PAM (polyacrylamide) automatic dosing device, and the PAFC (polyferric chloride) automatic dosing device. S5 uses a water quality TP total phosphorus concentration detection device to detect the total phosphorus concentration of water at the outlet of the AAO biological treatment tank and the outlet of the secondary sedimentation tank, and a water flow detection device to detect the water flow at the inlet of the secondary sedimentation tank and the inlet of the high-efficiency sedimentation tank. Then, it controls the dosage, rate, start time and frequency of the dosing of the PAC polyaluminum chloride automatic dosing device, PAM polyacrylamide automatic dosing device and PAFC polyferric chloride automatic dosing device in real time.

[0013] Preferably, the functional relationship of the multidimensional regression prediction model can be defined as: A 1预测 =f(TP) in Q in ); A 2预测 =g(TP) in Q in ); A 3预测 =p(TP) in Q in ); Where f, g, and p represent the independent mapping relationships obtained from training the multidimensional regression model, which are based on the input variable TP from data collected over several time periods in S1. in Q in This is derived from the relationship between the output variables A1, A2, and A3.

[0014] Preferably, the root mean square error (NRMSE) is the square root of the ratio of the square of the deviation between the predicted value and the true value to the number of observations n. The formula for calculating the root mean square error (NRMSE) is defined as follows:

[0015]

[0016] in, : The first data point in several time periods collected in S1 The actual values ​​of each drug dosage for each time period, including A1, A2, and A3, are calculated separately for A1, A2, and A3. : These are the predicted values ​​of drug dosage for each time period in S3; : The total number of data samples collected in S1 over several time periods; : The average of the actual dosage of each drug in the data collected over several time periods in S1.

[0017] Preferably, in S5, the core control unit regulates the dosage, rate, start time and frequency of the PAFC polyferric chloride automatic dosing device, compares the total phosphorus concentration TP1 at the outlet of the AAO biochemical tank with the first total phosphorus concentration threshold range, and compares the water flow rate Q1 at the inlet of the secondary sedimentation tank with the set water flow rate threshold range. Specifically, the upper and lower limits of the first total phosphorus concentration threshold are defined as TP. 1,max and TP 1,minThe upper and lower limits of the water flow threshold are Q, respectively. 1,max and Q 1,min k1 represents the "number of stages" for TP1 exceeding the threshold. That is, when the total phosphorus concentration TP1 exceeds the upper limit of the first total phosphorus concentration threshold by one stage, the amount of PAFC (polyferric chloride) used is A1±a1, then k1=1. Similarly, if the amount of PAFC is A1±a1±a1, then k1=2, and so on. k2 represents the "number of stages" for Q1 exceeding the threshold. Similar to k1, if the PAFC drug delivery rate is VB1±V1, then k2=1; if the PAFC drug delivery rate becomes VB1±V1±V1, then k2=2, and so on. A1' is the actual amount of PAFC drug delivered based on the predicted drug delivery rate A1 in S3, adjusted according to the number of stages for TP1 exceeding the threshold. VB1' is the actual delivery rate of PAFC drug delivered based on the baseline rate VB1, adjusted according to the number of stages for Q1 exceeding the threshold. The above adjustment logic satisfies: The dosage adjustment function relationship for PAFC (polyferric chloride) is as follows: When TP1>TP 1,max At that time, A1' = A1 + k1 × a1; When TP1 <TP 1,min At that time, A1' = A1 - k1 × a1; The regulatory function relationship for the PAFC (polyferric chloride) drug delivery rate is as follows: When Q1>Q 1,max At that time, VB1' = VB1 + k2 × V1; When Q1 1,min At that time, VB1' = VB1 - k2 × V1;

[0018] Preferably, in S5, the core control unit regulates the dosage, rate, start time, and frequency of drug dosing by the PAC (polyaluminum chloride) automatic dosing device and the PAM (polyacrylamide) automatic dosing device, compares the total phosphorus concentration (TP2) at the outlet of the secondary sedimentation tank with the second total phosphorus concentration threshold range, and compares the water flow rate (Q2) at the inlet of the high-efficiency sedimentation tank with the set water flow rate threshold range. Specifically, the upper and lower limits of the second total phosphorus concentration threshold are defined as TP. 2,max and TP 2,min The upper and lower limits of the water flow threshold are Q, respectively. 2,max and Q 2,min ​k3 represents the "number of stages" for TP2 exceeding the threshold. Specifically, when the total phosphorus concentration TP2 exceeds the upper limit of the second total phosphorus concentration threshold by two total phosphorus concentration stages, the dosage of PAC (polyaluminum chloride) is A2 ± a2, then k3 = 1. Similarly, if the dosage of PAC is A2 ± a2 ± a2, then k3 = 2, and so on. k4 represents the "number of stages" for Q2 exceeding the threshold. Similar to k2, if the dosage of PAC is VB2 ± V2, then k4 = 1; if the dosage of PAC changes to VB2 ± V2 ± V2, then k4 = 2, and so on. A2' is the actual dosage of PAC after adjusting for the number of stages exceeding the TP2 threshold based on the dosage A2 predicted by S3. VB2' is the dosage of PAC after adjusting for the number of stages exceeding the TP2 threshold based on the baseline rate VB2. Based on the actual PAC (polyaluminum chloride) drug delivery rate adjusted according to the number of stages exceeding the threshold in Q2, and since the dosage of PAM (polyacrylamide) follows the dosage of PAC, and the dosage A2 of the PAC automatic dosing device increases or decreases by a fixed amount a2, the dosage A3 of the PAM automatic dosing device increases or decreases by a fixed amount a3, then k3 and k4 can also be used to adjust PAM. Define A3' as the actual PAM dosage adjusted according to the number of stages exceeding the threshold in TP2 based on the drug dosage A3 predicted in S3, and VB3' as the actual PAM dosage adjusted according to the number of stages exceeding the threshold in Q2 based on the baseline rate VB3. The above adjustment logic satisfies: The dosage adjustment function relationship between PAC (polyaluminum chloride) and PAM (polyacrylamide) is as follows: When TP2>TP 2,max When, A2'=A2+k3×a2, A3'=A3+k3×a3; When TP2 <TP 2,min When, A2'=A2-k3×a2, A3'=A3-k3×a3; The relationship between the drug delivery rate regulation function of polyaluminum chloride (PAC) and polyacrylamide (PAM) is as follows: When Q2>Q 2,max At that time, VB2' = VB2 + k4 × V2, VB3' = VB3 + k4 × V3; When Q2 2,min At that time, VB2' = VB2 - k4 × V2, VB3' = VB3 - k4 × V3.

[0019] ​In summary, this invention has the following beneficial effects: It predicts the dosage of various drugs through a multi-dimensional dosing prediction model, and adjusts the dosage of various drugs based on the real-time detection of total phosphorus (TP) concentration and water flow rate, thereby making dosing in wastewater treatment more precise. Simultaneously, the core control unit controls the dosing rate of various drugs according to the water flow rate, ensuring that the added drugs react quickly and evenly with phosphorus-containing substances in the wastewater. Through the multi-dimensional dosing prediction model and the core control unit, it achieves precise dosing by combining feedforward and feedback mechanisms, effectively improving wastewater treatment efficiency, reducing operating costs, and ensuring effluent quality. Attached Figure Description

[0020] Figure 1 This is a schematic diagram of the structure of a precision dosing control system for wastewater treatment based on an intelligent multidimensional prediction model, according to the present invention. Figure 2 This is a flowchart of a precise dosing control method for wastewater treatment based on an intelligent multidimensional prediction model, according to the present invention. Detailed Implementation

[0021] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings. These embodiments do not constitute a limitation on the present invention.

[0022] like Figure 1 The wastewater treatment precision dosing control system shown includes a coarse screen and influent pump station, a fine screen and aerated sedimentation tank, an AAO biological treatment tank, a secondary sedimentation tank, a high-efficiency sedimentation tank, a filter cloth filter, a disinfection tank, and an effluent pump station, all connected in sequence. The effluent pump station pumps the treated water into the urban water supply system for use in landscape artificial wetlands, etc. The inlet and outlet of the AAO biological treatment tank, the outlet of the secondary sedimentation tank, and the outlet of the high-efficiency sedimentation tank are all equipped with total phosphorus (TP) concentration detection devices. The inlet of the AAO biological treatment tank and the inlet of the secondary sedimentation tank are also equipped with TP concentration detection devices. Water flow detection devices are installed at the inlet and outlet of the high-efficiency sedimentation tank. Both the water quality TP total phosphorus concentration detection device and the water flow detection device are connected to the core control unit. Automatic dosing devices for PAC polyaluminum chloride and PAM polyacrylamide are installed at the inlet of the high-efficiency sedimentation tank, and automatic dosing devices for PAFC polyferric chloride are installed at the inlet of the secondary sedimentation tank. The automatic dosing devices for PAC polyaluminum chloride, PAM polyacrylamide, and PAFC polyferric chloride are all controlled by the core control unit.

[0023] In the coarse screen and inlet pumping station, the coarse screen isolates large pollutants in the wastewater, removing and transporting the screenings. The inlet pumping station then pumps the wastewater into the fine screen and aerated sedimentation tank. In the fine screen and aerated sedimentation tank, the fine screen isolates solid pollutants, which settle in the aerated grit chamber, removing the screenings and settled grit. The wastewater from the aerated grit chamber is then fed into the AAO biological treatment tank, where the metabolic activity of microorganisms removes organic matter, nitrogen, and phosphorus pollutants. Simultaneously, air is blown into the AAO biological treatment tank through the blower room to provide a suitable environment for the microorganisms. The treated wastewater in the precipitator is fed into the secondary sedimentation tank, where polyferric chloride (PAFC) is added at the inlet. After dosing, the wastewater is allowed to remain for a period of time. The treated wastewater is then fed into the high-efficiency sedimentation tank, where polyaluminum chloride (PAC) and polyacrylamide (PAM) are added at the inlet. After dosing, the wastewater is allowed to remain for a period of time. The treated wastewater is then fed into a filter cloth filter for filtration. The filtered wastewater is then fed into a disinfection tank for disinfection. Chlorine dioxide is added to the disinfection tank through a chlorination and dosing room to disinfect the wastewater. After disinfection, the wastewater is fed into the effluent pumping station and then transported to other water-using areas.

[0024] When water flow is detected at the inlet of the secondary sedimentation tank, the core control unit activates the automatic PAFC (polyferric chloride) dosing device to begin adding the reagent. Similarly, when water flow is detected at the inlet of the high-efficiency sedimentation tank, the core control unit activates the automatic PAC (polyaluminum chloride) and PAM (polyacrylamide) dosing devices to begin adding the reagent. The automatic PAFC dosing device is positioned at the inlet of the secondary sedimentation tank. The PAFC is added at a controlled rate; the water flow at the inlet acts as agitation, allowing the reagent to react more quickly and evenly with pollutants in the wastewater. Similarly, the automatic PAC and PAM dosing devices are positioned at the inlet of the high-efficiency sedimentation tank. Both PAC and PAM are added at controlled rates, and the water flow at the inlet acts as agitation, allowing the reagents to react more quickly and evenly with pollutants in the wastewater.

[0025] The sludge outlet of the secondary sedimentation tank is sequentially connected to a sludge pumping station, a gravity thickener, a sludge storage tank, and a sludge dewatering room. Sludge from the dewatering room is transported off-site for processing, while the supernatant is returned to the coarse screen and the influent pumping station. A portion of the settled sludge from the secondary sedimentation tank is returned to the AAO biological treatment tank via the sludge pumping station, and the remaining sludge is fed into the gravity thickener. The high-efficiency sedimentation tank feeds chemical sludge into the sludge lift well, which in turn feeds sludge into the gravity thickener. The gravity thickener feeds sludge into the sludge storage tank, which then feeds sludge into the dewatering room. The filter cloth filter feeds filtered sludge into the dewatering room. The dewatering room returns the supernatant to the coarse screen and the influent pumping station, and the dewatered sludge is transported off-site for further processing.

[0026] The AAO biological treatment tank is also connected to a blower room, and the return sludge from the sludge pump room is connected to the AAO biological treatment tank.

[0027] The chemical sludge outlet of the high-efficiency sedimentation tank is connected to the gravity thickening tank through a sludge lifting well, the sludge outlet of the filter cloth filter is connected to the sludge dewatering room, and the disinfection tank is also connected to the chlorination and chemical dosing room.

[0028] like Figure 2 The method for a precise dosing control system for wastewater treatment based on an intelligent multidimensional prediction model, as shown, includes the following steps: S1. Collect the total phosphorus (TP) concentration at the inlet of the AAO biological treatment tank and the outlet of the high-efficiency sedimentation tank over several time periods. Let the TP concentration at the inlet be TP. in The total phosphorus concentration (TP) at the outlet is TP out The water flow rates at the inlet of the AAO biological treatment tank and the outlet of the high-efficiency sedimentation tank are given by Q. in The water flow rate at the outlet is Q. out The dosage of PAFC (polyferric chloride), PAC (polyaluminum chloride), and PAM (polyacrylamide) was collected for several time periods. The dosage of PAFC (polyferric chloride) was A1, the dosage of PAC (polyaluminum chloride) was A2, and the dosage of PAM (polyacrylamide) was A3. S2. Based on the collected data from several time periods, train the multidimensional regression prediction model based on automatic machine learning to obtain the trained multidimensional regression prediction model based on automatic machine learning; the input variable in the core parameters of the multidimensional regression prediction model is TP. in TP out Q in and Q out The output variables are A1, A2, and A3; S3. Based on the total phosphorus concentration (TP) and flow rate of the inlet water of the AAO biochemical tank during the current time period, a trained multidimensional regression prediction model based on automatic machine learning is used to predict the dosage of PAFC (polyferric chloride), PAC (polyaluminum chloride), and PAM (polyacrylamide) during the current time period. S4. Calculate the standardized root mean square error (NRMSE) based on the drug dosage prediction results obtained in S3. When the calculated standardized root mean square error (NRMSE) is greater than the set error threshold, input the difference between the calculated standardized root mean square error (NRMSE) and the set error threshold into the trained multidimensional regression prediction model based on automatic machine learning, and return to S3. Continue until the calculated standardized root mean square error (NRMSE) of the prediction results is less than the set error threshold. Finally, adjust the obtained predicted dosage in real time through the core control unit to control the PAC (polyaluminum chloride) automatic dosing device, the PAM (polyacrylamide) automatic dosing device, and the PAFC (polyferric chloride) automatic dosing device. S5 uses a water quality TP total phosphorus concentration detection device to detect the total phosphorus concentration of water at the outlet of the AAO biological treatment tank and the outlet of the secondary sedimentation tank, and a water flow detection device to detect the water flow at the inlet of the secondary sedimentation tank and the inlet of the high-efficiency sedimentation tank. Then, it controls the dosage, rate, start time and frequency of the dosing of the PAC polyaluminum chloride automatic dosing device, PAM polyacrylamide automatic dosing device and PAFC polyferric chloride automatic dosing device in real time.

[0029] The functional relationship of a multidimensional regression prediction model can be defined as: A 1预测 =f(TP) in Q in ); A 2预测 =g(TP) in Q in ); A 3预测 =p(TP) in Q in ); Where f, g, and p represent the independent mapping relationships obtained from training the multidimensional regression model, which are based on the input variable TP from data collected over several time periods in S1. in Q in This is derived from the relationship between the output variables A1, A2, and A3.

[0030] The root mean square error (NRMSE) is the square root of the ratio of the square of the deviation between the predicted and actual values ​​to the number of observations (n). The formula for calculating NRMSE is defined as follows:

[0031]

[0032] in, : The first data point in several time periods collected in S1 The actual values ​​of each drug dosage for each time period, including A1, A2, and A3, are calculated separately for A1, A2, and A3. : These are the predicted values ​​of drug dosage for each time period in S3; : The total number of data samples collected in S1 over several time periods; : The average of the actual dosage of each drug in the data collected over several time periods in S1.

[0033] The core control unit in S5 regulates the dosage, rate, start time, and frequency of the PAFC polyferric chloride automatic dosing device. It compares the total phosphorus concentration (TP1) at the outlet of the AAO biological tank with the first total phosphorus concentration threshold range, and compares the water flow rate (Q1) at the inlet of the secondary sedimentation tank with the set water flow rate threshold range.

[0034] When the total phosphorus concentration TP1 does not exceed the first total phosphorus concentration threshold range, and the water flow rate Q1 does not exceed the set water flow rate threshold range, the dosage A1 of the PAFC (polyferric chloride) automatic dosing device is the PAFC dosage predicted by a multidimensional regression prediction model based on automatic machine learning. The dosing rate is the baseline rate VB1, and the number of dosings is the standard number of dosings N1. When the total phosphorus concentration TP1 exceeds the upper limit of the first total phosphorus concentration threshold or falls below the lower limit of the first total phosphorus concentration threshold by one total phosphorus concentration stage value, the dosage A1 of the PAFC automatic dosing device increases or decreases by one fixed dosage a1. When the water flow rate Q1 exceeds the upper limit of the set water flow rate threshold or falls below the lower limit of the set water flow rate threshold by one water flow rate stage value, the dosing rate VB1 of the PAFC automatic dosing device increases or decreases by one fixed rate V1. The total phosphorus concentration stage values ​​include no less than five total phosphorus concentration stage values, and the water flow rate stage values ​​include no less than five water flow rate stage values.

[0035] When the total phosphorus concentration TP1 exceeds the upper limit of the first total phosphorus concentration threshold by one stage value (i.e., the difference between the total phosphorus concentration TP1 and the upper limit of the first total phosphorus concentration threshold is greater than the first total phosphorus concentration stage value), the amount of PAFC polyferric chloride A1 increases by a fixed amount a1, and the amount of PAFC polyferric chloride becomes A1+a1; when the difference between the total phosphorus concentration TP1 and the upper limit of the first total phosphorus concentration threshold is greater than the second total phosphorus concentration stage value, the amount of PAFC polyferric chloride A1+a1 increases by a fixed amount a1, and the amount of PAFC polyferric chloride becomes A1+a1+a1; and so on, until the difference between the total phosphorus concentration TP1 and the upper limit of the first total phosphorus concentration threshold is greater than the fifth total phosphorus concentration stage value, then the amount of PAFC polyferric chloride A1+a1+a1+a1+a1 increases by a fixed amount a1, and the amount of PAFC polyferric chloride becomes A1+a1+a1+a1+a1+a1.

[0036] When the total phosphorus concentration TP1 falls below the lower limit of the first total phosphorus concentration threshold by one total phosphorus concentration stage value (i.e., the difference between the lower limit of the first total phosphorus concentration threshold and the total phosphorus concentration TP1 is greater than the first total phosphorus concentration stage value), the dosage A1 of PAFC (polyferric chloride) is reduced by a fixed dosage a1, and the dosage of PAFC becomes A1-a1; when the difference between the lower limit of the first total phosphorus concentration threshold and the total phosphorus concentration TP1 is greater than the second total phosphorus concentration stage value, then PAF... The dosage of polyferric chloride (PFC) is reduced by a fixed amount a1, and the dosage of PFC becomes A1-a1-a1; this process continues until the difference between the lower limit of the first total phosphorus concentration threshold and the total phosphorus concentration TP1 is greater than the fifth total phosphorus concentration stage value. At this point, the dosage of PFC A1-a1-a1-a1-a1 is reduced by a fixed amount a1, and the dosage of PFC becomes A1-a1-a1-a1-a1-a1.

[0037] When the water flow rate Q1 exceeds the upper limit of the set water flow rate threshold by one water flow stage value (i.e., the difference between the water flow rate Q1 and the upper limit of the water flow rate threshold is greater than the first water flow stage value), the dosing rate VB1 of the PAFC automatic dosing device is increased by a fixed rate V1, and the dosing rate becomes VB1+V1. When the difference between the water flow rate Q1 and the upper limit of the water flow rate threshold is greater than the second water flow stage value, the dosing rate VB1+V1 of the PAFC automatic dosing device is increased by a fixed rate V1, and the dosing rate becomes VB1+V1+V1. This continues until the difference between the water flow rate Q1 and the upper limit of the water flow rate threshold is greater than the fifth water flow stage value, at which point the dosing rate VB1+V1+V1+V1+V1 is increased by a fixed rate V1, and the dosing rate becomes VB1+V1+V1+V1+V1+V1.

[0038] When the water flow rate Q1 falls below the lower limit of the set water flow rate threshold by one water flow stage value (i.e., the difference between the lower limit of the set water flow rate threshold and the water flow rate Q1 is greater than the first water flow stage value), the dosing rate VB1 of the PAFC automatic dosing device is reduced by a fixed rate V1, and the dosing rate becomes VB1-V1. If the difference between the lower limit of the set water flow rate threshold and the water flow rate Q1 is greater than the second water flow stage value, the dosing rate VB1-V1 of the PAFC automatic dosing device is reduced by a fixed rate V1, and the dosing rate becomes VB1-V1-V1. This continues until the difference between the lower limit of the set water flow rate threshold and the water flow rate Q1 is greater than the fifth water flow stage value, at which point the dosing rate VB1-V1-V1-V1-V1 is reduced by a fixed rate V1, and the dosing rate becomes VB1-V1-V1-V1-V1-V1. Specifically, the upper and lower limits of the first total phosphorus concentration threshold are defined as TP. 1,max and TP 1,min The upper and lower limits of the water flow threshold are Q, respectively. 1,max and Q 1,mink1 represents the "number of stages" for TP1 exceeding the threshold. That is, when the total phosphorus concentration TP1 exceeds the upper limit of the first total phosphorus concentration threshold by one stage, the amount of PAFC (polyferric chloride) used is A1±a1, then k1=1. Similarly, if the amount of PAFC is A1±a1±a1, then k1=2, and so on. k2 represents the "number of stages" for Q1 exceeding the threshold. Similar to k1, if the PAFC drug delivery rate is VB1±V1, then k2=1; if the PAFC drug delivery rate becomes VB1±V1±V1, then k2=2, and so on. A1' is the actual amount of PAFC drug delivered based on the predicted drug delivery rate A1 in S3, adjusted according to the number of stages for TP1 exceeding the threshold. VB1' is the actual delivery rate of PAFC drug delivered based on the baseline rate VB1, adjusted according to the number of stages for Q1 exceeding the threshold. The above adjustment logic satisfies: The dosage adjustment function relationship for PAFC (polyferric chloride) is as follows: When TP1>TP 1,max At that time, A1' = A1 + k1 × a1; When TP1 <TP 1,min At that time, A1' = A1 - k1 × a1; The regulatory function relationship for the PAFC (polyferric chloride) drug delivery rate is as follows: When Q1>Q 1,max At that time, VB1' = VB1 + k2 × V1; When Q1 1,min At that time, VB1' = VB1 - k2 × V1;

[0039] The core control unit in S5 regulates the dosage, rate, start time, and frequency of the PAC (polyaluminum chloride) and PAM (polyacrylamide) automatic dosing devices. It also compares the total phosphorus concentration (TP2) at the outlet of the secondary sedimentation tank with the second total phosphorus concentration threshold range, and compares the water flow rate (Q2) at the inlet of the high-efficiency sedimentation tank with the set water flow rate threshold range.

[0040] ​When the total phosphorus concentration TP2 does not exceed the second total phosphorus concentration threshold range, and the water flow rate Q2 does not exceed the set water flow rate threshold range, the dosage A2 of PAC (polyaluminum chloride) and the dosage A3 of PAM (polyacrylamide) automatically added by the automatic dosing device are the PAC and PAM dosages predicted by a multidimensional regression prediction model based on automatic machine learning, respectively. The dosing rate of the PAC automatic dosing device is the baseline rate VB2, and the number of dosings is the standard number of dosings N2. The dosing rate of the PAM automatic dosing device is the baseline rate VB3, and the number of dosings is the standard number of dosings N3. When the total phosphorus concentration TP2 exceeds the second total phosphorus concentration threshold range... When the upper limit threshold of the total phosphorus concentration (TP2) or the total phosphorus concentration (TP2) falls below the lower limit threshold of the second total phosphorus concentration threshold by one total phosphorus concentration stage value, the dosage A2 of the PAC (polyaluminum chloride) automatic dosing device increases or decreases by one fixed dosage a2, and the dosage A3 of the PAM (polyacrylamide) automatic dosing device increases or decreases by one fixed dosage a3. When the water flow rate (Q2) exceeds the upper limit threshold of the set water flow rate threshold or falls below the lower limit threshold threshold of the set water flow rate threshold by one water flow rate stage value, the dosing rate VB2 of the PAC (polyaluminum chloride) automatic dosing device increases or decreases by one fixed rate V2, and the dosing rate VB3 of the PAM (polyacrylamide) automatic dosing device increases or decreases by one fixed rate V3.

[0041] The dosage of polyaluminum chloride (PAC) will change accordingly, as will the dosage of polyacrylamide (PAM). Although PAM cannot remove phosphates or phosphorus-containing substances from wastewater, the amount of PAC reacting with phosphorus-containing substances in the wastewater will also increase, thus requiring an increase in PAM to convert all the increased reactants into sludge. If the amount of PAM is not increased, some of the reactants from the PAC will flow into the filter tank, failing to achieve the purpose of wastewater treatment. Therefore, the dosage of PAM follows the dosage of PAC.

[0042] When the total phosphorus concentration TP2 exceeds the upper limit of the second total phosphorus concentration threshold by one total phosphorus concentration stage value (i.e., the difference between the total phosphorus concentration TP2 and the upper limit of the second total phosphorus concentration threshold is greater than the first total phosphorus concentration stage value), the dosage of PAC polyaluminum chloride (A2) increases by a fixed amount a2, becoming A2+a2, and the dosage of PAM polyacrylamide (A3) increases by a fixed amount a3, becoming A3+a3. Similarly, when the difference between the total phosphorus concentration TP2 and the upper limit of the second total phosphorus concentration threshold is greater than the second total phosphorus concentration stage value, the dosage of PAC polyaluminum chloride (A2+a2) increases by a fixed amount a2, becoming A3+a3. The dosage of PAM (polyacrylamide) increases by a fixed amount a3 from A2+a2+a2, and the dosage of PAC (polyaluminum chloride) increases by A3+a3. This process continues until the difference between the total phosphorus concentration TP2 and the upper limit of the second total phosphorus concentration threshold is greater than the fifth total phosphorus concentration stage value. At this point, the dosage of PAC (polyaluminum chloride) increases by a fixed amount a2 from A2+a2+a2+a2+a2, and the dosage of PAM (polyacrylamide) increases by a fixed amount a3 from A3+a3+a3+a3+a3, and the dosage of PAC (polyaluminum chloride) increases by A3+a3+a3+a3+a3+a3.

[0043] When the total phosphorus concentration TP2 falls below the lower limit of the second total phosphorus concentration threshold by one total phosphorus concentration stage value (i.e., the difference between the lower limit of the second total phosphorus concentration threshold and the total phosphorus concentration TP2 is greater than the first total phosphorus concentration stage value), the dosage of PAC polyaluminum chloride (A2) decreases by a fixed amount a2, becoming A2-a2, and the dosage of PAM polyacrylamide (A3) decreases by a fixed amount a3, becoming A3-a3. Similarly, when the difference between the lower limit of the second total phosphorus concentration threshold and the total phosphorus concentration TP2 is greater than the second total phosphorus concentration stage value, the dosage of PAC polyaluminum chloride (A2-a2) decreases by a fixed amount a2, becoming A3-a3. The dosage of PAM (polyacrylamide) is reduced by a fixed amount a3 from A2-a2-a2, and the dosage of PAC (polyaluminum chloride) is reduced by a fixed amount a3. This process continues until the difference between the lower limit of the second total phosphorus concentration threshold and the total phosphorus concentration TP2 is greater than the fifth total phosphorus concentration stage value. At this point, the dosage of PAC (polyaluminum chloride) is reduced by a fixed amount a2 from A2-a2-a2-a2-a2, and the dosage of PAM (polyacrylamide) is reduced by a fixed amount a3 from A3-a3-a3-a3-a3, and the dosage of PAC (polyaluminum chloride) is reduced by a fixed amount a3.

[0044] When the water flow rate Q2 exceeds the upper limit of the set water flow rate threshold by one water flow stage value (i.e., the difference between the water flow rate Q2 and the upper limit of the water flow rate threshold is greater than the first water flow stage value), the dosing rate VB2 of the PAC (polyaluminum chloride) automatic dosing device is increased by a fixed rate V2, and the PAC dosing rate becomes VB2 + V2. Similarly, the dosing rate VB3 of the PAM (polyacrylamide) automatic dosing device is increased by a fixed rate V3, and the PAM dosing rate becomes VB3 + V3. When the difference between the water flow rate Q2 and the upper limit of the water flow rate threshold is greater than the second water flow stage value, the dosing rate VB2 + V2 of the PAC (polyaluminum chloride) automatic dosing device is increased by a fixed rate V2, and the PAC dosing rate becomes VB2 + V2. +V2+V2, the dosing rate of the PAM automatic dosing device VB3+V3 is increased by a fixed rate V3, and the PAM dosing rate becomes VB3+V3+V3; similarly, if the difference between the water flow rate Q2 and the upper limit of the water flow rate threshold is greater than the fifth water flow rate stage value, then the dosing rate of the PAC automatic dosing device VB2+V2+V2+V2+V2 is increased by a fixed rate V2, and the PAC dosing rate becomes VB2+V2+V2+V2+V2+V2. Similarly, the dosing rate of the PAM automatic dosing device VB3+V3+V3+V3+V3+V3 is increased by a fixed rate V3, and the PAM dosing rate becomes VB3+V3+V3+V3+V3+V3+V3.

[0045] When the water flow rate Q2 falls below the lower limit of the set water flow rate threshold by one water flow stage value (i.e., the difference between the lower limit of the water flow rate threshold and the water flow rate Q2 is greater than the first water flow stage value), the dosing rate VB2 of the PAC (polyaluminum chloride) automatic dosing device is reduced by one fixed rate V2, and the PAC dosing rate becomes VB2-V2. Similarly, the dosing rate VB3 of the PAM (polyacrylamide) automatic dosing device is reduced by one fixed rate V3, and the PAM dosing rate becomes VB3-V3. When the difference between the lower limit of the water flow rate threshold and the water flow rate Q2 is greater than the second water flow stage value, the dosing rate VB2-V2 of the PAC (polyaluminum chloride) automatic dosing device is reduced by one fixed rate V2, and the PAC dosing rate becomes VB2-V2. V2-V2, the dosing rate of the PAM automatic dosing device is reduced by a fixed rate V3 from VB3-V3, and the PAM dosing rate becomes VB3-V3-V3; and so on, until the difference between the lower limit of the water flow threshold and the water flow rate Q2 is greater than the fifth water flow stage value, then the dosing rate of the PAC automatic dosing device is reduced by a fixed rate V2 from VB2-V2-V2-V2-V2, and the dosing rate of the PAM automatic dosing device is reduced by a fixed rate V3 from VB3-V3-V3-V3-V3, and the PAM dosing rate becomes VB3-V3-V3-V3-V3-V3.

[0046] Specifically, the upper and lower limits of the second total phosphorus concentration threshold are defined as TP. 2,max and TP 2,min The upper and lower limits of the water flow threshold are Q, respectively. 2,max and Q 2,mink3 represents the "number of stages" for TP2 exceeding the threshold. Specifically, when the total phosphorus concentration TP2 exceeds the upper limit of the second total phosphorus concentration threshold by two total phosphorus concentration stages, the dosage of PAC (polyaluminum chloride) is A2 ± a2, then k3 = 1. Similarly, if the dosage of PAC is A2 ± a2 ± a2, then k3 = 2, and so on. k4 represents the "number of stages" for Q2 exceeding the threshold. Similar to k2, if the dosage of PAC is VB2 ± V2, then k4 = 1; if the dosage of PAC changes to VB2 ± V2 ± V2, then k4 = 2, and so on. A2' is the actual dosage of PAC after adjusting for the number of stages exceeding the TP2 threshold based on the dosage A2 predicted by S3. VB2' is the dosage of PAC after adjusting for the number of stages exceeding the TP2 threshold based on the baseline rate VB2. Based on the actual PAC (polyaluminum chloride) drug delivery rate adjusted according to the number of stages exceeding the threshold in Q2, and since the dosage of PAM (polyacrylamide) follows the dosage of PAC, and the dosage A2 of the PAC automatic dosing device increases or decreases by a fixed amount a2, the dosage A3 of the PAM automatic dosing device increases or decreases by a fixed amount a3, then k3 and k4 can also be used to adjust PAM. Define A3' as the actual PAM dosage adjusted according to the number of stages exceeding the threshold in TP2 based on the drug dosage A3 predicted in S3, and VB3' as the actual PAM dosage adjusted according to the number of stages exceeding the threshold in Q2 based on the baseline rate VB3. The above adjustment logic satisfies: The dosage adjustment function relationship between PAC (polyaluminum chloride) and PAM (polyacrylamide) is as follows: When TP2>TP 2,max When, A2'=A2+k3×a2, A3'=A3+k3×a3; When TP2 <TP 2,min When, A2'=A2-k3×a2, A3'=A3-k3×a3; The relationship between the drug delivery rate regulation function of polyaluminum chloride (PAC) and polyacrylamide (PAM) is as follows: When Q2>Q 2,max At that time, VB2' = VB2 + k4 × V2, VB3' = VB3 + k4 × V3; When Q2 2,min At that time, VB2' = VB2 - k4 × V2, VB3' = VB3 - k4 × V3.

[0047] ​Both the secondary sedimentation tank and the high-efficiency sedimentation tank are equipped with water quality TP (total phosphorus) concentration monitoring devices. After the automatic PAFC (polyferric chloride) dosing device at the inlet of the secondary sedimentation tank is put into operation, the wastewater needs to remain in the secondary sedimentation tank for a period of time. Therefore, after a period of time T1, the water quality TP concentration monitoring device inside the secondary sedimentation tank begins to detect the water quality TP concentration. When the water quality TP concentration exceeds the total phosphorus concentration threshold of the secondary sedimentation tank, the core control unit controls the automatic PAFC dosing device to add one unit of PAFC. After a period of time... After time T2, the total phosphorus (TP) concentration monitoring device inside the secondary sedimentation tank re-detects the TP concentration. When the TP concentration exceeds the threshold for total phosphorus concentration in the secondary sedimentation tank, the core control unit again controls the automatic PAFC (polyferric chloride) dosing device to administer one unit of PAFC. This process is repeated every T2, cyclically detecting the TP concentration and comparing it to the threshold. The core control unit then controls the automatic PAFC dosing device to administer one unit of PAFC until the TP concentration in the secondary sedimentation tank no longer exceeds the threshold. Time T1 is relatively long; although adding chemicals at the inlet creates a stirring effect and accelerates the reaction, it still requires time for the chemicals and wastewater to react. Time T2 is very short because the dosage is small, resulting in a rapid reaction and eliminating the need for long intervals to detect the TP concentration.

[0048] After the automatic dosing devices for PAC (polyaluminum chloride) and PAM (polyacrylamide) at the inlet of the high-efficiency sedimentation tank are completed, the wastewater needs to remain in the sedimentation tank for a period of time. Therefore, after a period of time T3, the total phosphorus (TP) concentration detection device inside the high-efficiency sedimentation tank begins to monitor the TP concentration. When the TP concentration exceeds the threshold for total phosphorus concentration in the high-efficiency sedimentation tank, the core control unit again controls the automatic dosing device for PAC to add one unit of PAFC (polyferric chloride) and the automatic dosing device for PAM to add one unit of PAM. After a period of time T4, the TP concentration inside the high-efficiency sedimentation tank... The detection device re-detects the total phosphorus (TP) concentration in the water inside the high-efficiency sedimentation tank. When the TP concentration exceeds the threshold for total phosphorus concentration in the high-efficiency sedimentation tank, the core control unit again controls the automatic dosing device for polyaluminum chloride (PAC) to add one unit of polyferric chloride (PAFC), and controls the automatic dosing device for polyacrylamide (PAM) to add one unit of polyacrylamide. Then, every time interval T4, the TP concentration in the water inside the high-efficiency sedimentation tank is cyclically detected and compared with the threshold. The core control unit controls the automatic dosing device for PAC to add one unit of polyaluminum chloride (PAC) and the automatic dosing device for PAM to add one unit of polyacrylamide (PAM) until the TP concentration in the water inside the high-efficiency sedimentation tank is no longer above the threshold. Time T3 is relatively long. Although adding chemicals at the inlet can create a stirring effect and accelerate the reaction, it still requires some time for the chemicals and wastewater to react. Time T4 is very short because the dosage of chemicals added is very small, so the reaction is rapid, and there is no need to detect the TP concentration again after a long interval.

[0049] The automatic dosing device for PAFC (polyferric chloride) uses a base rate VB1, which is set based on the dosage of PAFC and the unit water flow rate. Similarly, the automatic dosing device for PAC (polyaluminum chloride) uses a base rate VB2, which is set based on the dosage of PAC and the unit water flow rate. The automatic dosing device for PAM (polyacrylamide) uses a base rate VB3, which is set based on the dosage of PAM and the unit water flow rate.

[0050] Because a residence time needs to be reserved for the drug in the secondary sedimentation tank, when the dosage of PAFC (polyferric chloride) is large, the baseline drug dosing rate VB1 needs to be accelerated. Simultaneously, the water flow rate also affects the PAFC dosing rate; a larger water flow rate requires a correspondingly higher dosing rate to accelerate the reaction, while a smaller water flow rate requires a correspondingly lower dosing rate to achieve a uniform dosing reaction. Therefore, a drug dosing rate strategy is established. First, the median value between the maximum and minimum dosing rates of the automatic PAFC dosing device is taken as the intermediate rate. The portion exceeding the intermediate rate is divided into several rate intervals, and the portion below the intermediate rate is also divided into several rate intervals. Then, the dosage of PAFC is divided into several dosage intervals, and the water flow rate is divided into several flow rate intervals. A baseline drug dosage interval and a baseline water flow rate interval are selected from each of these dosage and flow rate intervals, with the baseline water flow rate interval being a set water flow rate threshold range.

[0051] When the dosage of PAFC (polyferric chloride) exceeds the baseline dosage range by m intervals, and the water flow rate exceeds the baseline water flow rate range by w intervals, the automatic PAFC dosing device increases the dosing rate by m+w intervals from the intermediate rate. This means moving the intermediate rate upwards by m+w intervals to obtain the baseline rate VB1. Conversely, when the dosage of PAFC exceeds the baseline dosage range by m intervals, and the water flow rate is lower than the baseline water flow rate range by w intervals, the automatic PAFC dosing device increases the dosing rate by m intervals and decreases it by w intervals from the intermediate rate. This means moving the intermediate rate upwards by m intervals and downwards by w intervals to obtain the baseline rate VB1. 1. When the dosage of PAFC (polyferric chloride) is lower than the baseline dosage range by m intervals, and the water flow exceeds the baseline water flow range by w intervals, the dosing rate of the automatic PAFC dosing device decreases by m intervals and increases by w intervals from the intermediate rate. That is, the intermediate rate is moved down by m intervals and up by w intervals to obtain the baseline rate VB1. When the dosage of PAFC is lower than the baseline dosage range by m intervals, and the water flow is lower than the baseline water flow range by w intervals, the dosing rate of the automatic PAFC dosing device decreases by m+w intervals from the intermediate rate. That is, the intermediate rate is moved down by m+w intervals to obtain the baseline rate VB1.

[0052] Simultaneously, when adjusting the dosing rate of the PAFC (polyferric chloride) automatic dosing device through the core control unit, it can also be adjusted according to the set baseline dosing rate VB1 using the aforementioned dosing rate strategy. For example, when the baseline rate VB1 is determined, in subsequent actual wastewater treatment, if the PAFC dosage A1 increases by m baseline dosage intervals, and the water flow exceeds the baseline water flow interval by w intervals, the baseline rate VB1 will increase by m+w rate intervals. When the water flow rate is lower than the baseline water flow rate range by w intervals, the baseline rate VB1 increases by m rate intervals and decreases by w rate intervals. When the dosage of PAFC (polyferric chloride) A1 decreases by m baseline dosage intervals, and the water flow rate exceeds the baseline water flow rate range by w intervals, the baseline rate VB1 decreases by m rate intervals and increases by w rate intervals. When the dosage of PAFC (polyferric chloride) A1 decreases by m baseline dosage intervals, and the water flow rate is lower than the baseline water flow rate range by w intervals, the baseline rate VB1 decreases by m+w rate intervals.

[0053] Because a residence time needs to be reserved for the drugs in the high-efficiency sedimentation tank, when the dosage of PAC (polyaluminum chloride) and PAM (polyacrylamide) is large, the dosage reference rates VB2 and VB3 need to be accelerated. Simultaneously, the water flow rate also affects the dosage rates of PAC and PAM. With a large water flow rate, the dosage rates of PAC and PAM need to be increased accordingly to accelerate the reaction; with a small water flow rate, the dosage rates of PAC and PAM need to be decreased accordingly to achieve a uniform dosing reaction. The setting of the dosage rate VB2 for the automatic PAC dosing device and the dosage rate VB3 for the automatic PAM dosing device can also be obtained based on the above-mentioned drug dosage rate strategy. Similarly, when adjusting the dosage rates of the automatic PAC and PAM dosing devices in the core control unit, real-time adjustments can also be made using the drug dosage rate strategy.

[0054] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art can make various modifications or equivalent substitutions to the present invention within the scope of its essence and protection. Such modifications or equivalent substitutions should also be considered to fall within the protection scope of the present invention.

Claims

1. A precise dosing control system for wastewater treatment based on an intelligent multidimensional prediction model, characterized in that, The system comprises, in sequence, a coarse screen and inlet pump station, a fine screen and aeration sedimentation tank, an AAO biological treatment tank, a secondary sedimentation tank, a high-efficiency sedimentation tank, a filter cloth filter, a disinfection tank, and an effluent pump station. The inlet and outlet of the AAO biological treatment tank, the outlet of the secondary sedimentation tank, and the outlet of the high-efficiency sedimentation tank are all equipped with total phosphorus (TP) concentration detection devices. The inlet and outlet of the AAO biological treatment tank, the secondary sedimentation tank, the high-efficiency sedimentation tank, and the high-efficiency sedimentation tank are all equipped with flow rate detection devices. Both the TP concentration detection devices and the flow rate detection devices are connected to a core control unit. The inlet of the high-efficiency sedimentation tank is equipped with an automatic PAC (polyaluminum chloride) dosing device and an automatic PAM (polyacrylamide) dosing device. The inlet of the secondary sedimentation tank is equipped with an automatic PAFC (polyferric chloride) dosing device. All the automatic PAC, PAM, and PAFC dosing devices are controlled by the core control unit.

2. The wastewater treatment precision dosing control system based on an intelligent multidimensional prediction model according to claim 1, characterized in that: The sludge outlet of the secondary sedimentation tank is also connected in sequence to a sludge pumping station, a gravity thickening tank, a sludge storage tank, and a sludge dewatering machine room. The sludge from the sludge dewatering machine room is transported off-site for treatment, while the supernatant is returned to the coarse screen and the inlet pumping station.

3. The wastewater treatment precision dosing control system based on an intelligent multidimensional prediction model according to claim 2, characterized in that: The AAO biological treatment tank is also connected to a blower room, and the return sludge from the sludge pump room is connected to the AAO biological treatment tank.

4. The wastewater treatment precision dosing control system and method based on an intelligent multidimensional prediction model according to claim 2, characterized in that: The chemical sludge outlet of the high-efficiency sedimentation tank is connected to the gravity thickening tank through a sludge lifting well, the sludge outlet of the filter cloth filter is connected to the sludge dewatering room, and the disinfection tank is also connected to a chlorination and dosing room.

5. A method for a precise dosing control system for wastewater treatment based on an intelligent multidimensional prediction model, as described in any one of claims 1 to 4, characterized in that: Includes the following steps: S1. Collect the total phosphorus (TP) concentration at the inlet of the AAO biological treatment tank and the outlet of the high-efficiency sedimentation tank over several time periods. Let the TP concentration at the inlet be TP. in The total phosphorus concentration (TP) at the outlet is TP out The water flow rates at the inlet of the AAO biological treatment tank and the outlet of the high-efficiency sedimentation tank are given by Q. in The water flow rate at the outlet is Q. out The dosage of PAFC (polyferric chloride), PAC (polyaluminum chloride), and PAM (polyacrylamide) was collected for several time periods. The dosage of PAFC (polyferric chloride) was A1, the dosage of PAC (polyaluminum chloride) was A2, and the dosage of PAM (polyacrylamide) was A3. S2. Train the multidimensional regression prediction model based on automatic machine learning using data collected over several time periods to obtain the trained multidimensional regression prediction model based on automatic machine learning; the input variable in the core parameters of the multidimensional regression prediction model is TP. in TP out Q in and Q out The output variables are A1, A2, and A3; S3. Based on the total phosphorus concentration (TP) and flow rate of the inlet water of the AAO biochemical tank during the current time period, a trained multidimensional regression prediction model based on automatic machine learning is used to predict the dosage of PAFC (polyferric chloride), PAC (polyaluminum chloride), and PAM (polyacrylamide) during the current time period. S4. Calculate the standardized root mean square error (NRMSE) based on the drug dosage prediction results obtained in S3. When the calculated standardized root mean square error (NRMSE) is greater than the set error threshold, input the difference between the calculated standardized root mean square error (NRMSE) and the set error threshold into the trained multidimensional regression prediction model based on automatic machine learning, and return to S3. Continue until the calculated standardized root mean square error (NRMSE) of the prediction results is less than the set error threshold. Finally, adjust the obtained predicted dosage in real time through the core control unit to control the PAC (polyaluminum chloride) automatic dosing device, the PAM (polyacrylamide) automatic dosing device, and the PAFC (polyferric chloride) automatic dosing device. S5 uses a water quality TP total phosphorus concentration detection device to detect the total phosphorus concentration of water at the outlet of the AAO biological treatment tank and the outlet of the secondary sedimentation tank, and a water flow detection device to detect the water flow at the inlet of the secondary sedimentation tank and the inlet of the high-efficiency sedimentation tank. Then, it controls the dosage, rate, start time and frequency of the dosing of the PAC polyaluminum chloride automatic dosing device, PAM polyacrylamide automatic dosing device and PAFC polyferric chloride automatic dosing device in real time.

6. The wastewater treatment precision dosing control system and method based on an intelligent multidimensional prediction model according to claim 5, characterized in that: The functional relationship of the multidimensional regression prediction model can be defined as follows: A 1预测 =f(TP in ,Q in ); TO 2预测 =g(TP in ,Q in ); TO 3预测 =p(TP in ,Q in ); Where f, g, and p represent the independent mapping relationships obtained from training the multidimensional regression model, which are based on the input variable TP from data collected over several time periods in S1. in Q in This is derived from the relationship between the output variables A1, A2, and A3.

7. The method for a precise dosing control system for wastewater treatment based on an intelligent multidimensional prediction model according to claim 5, characterized in that: The root mean square error (NRMSE) is the square root of the ratio of the square of the deviation between the predicted and the true values ​​to the number of observations (n). The formula for calculating the root mean square error (NRMSE) is defined as follows: in, : The first data point in several time periods collected in S1 The actual values ​​of each drug dosage for each time period, including A1, A2, and A3, are calculated separately for A1, A2, and A3. : These are the predicted drug dosages for each time period in S3; : The total number of data samples collected in S1 over several time periods; : The average of the actual dosage of each drug in the data collected over several time periods in S1.

8. The method for a precise dosing control system for wastewater treatment based on an intelligent multidimensional prediction model according to claim 5, characterized in that: The core control unit in S5 regulates the dosage, rate, start time and frequency of the PAFC polyferric chloride automatic dosing device, compares the total phosphorus concentration TP1 at the outlet of the AAO biochemical tank with the first total phosphorus concentration threshold range, and compares the water flow rate Q1 at the inlet of the secondary sedimentation tank with the set water flow rate threshold range. Specifically, the upper and lower limits of the first total phosphorus concentration threshold are defined as TP. 1,max and TP 1,min The upper and lower limits of the water flow threshold are Q, respectively. 1,max and Q 1,min k1 represents the "number of stages" for TP1 exceeding the threshold. That is, when the total phosphorus concentration TP1 exceeds the upper limit of the first total phosphorus concentration threshold by one stage, the amount of PAFC (polyferric chloride) used is A1±a1, then k1=1. Similarly, if the amount of PAFC is A1±a1±a1, then k1=2, and so on. k2 represents the "number of stages" for Q1 exceeding the threshold. Similar to k1, if the PAFC drug delivery rate is VB1±V1, then k2=1; if the PAFC drug delivery rate becomes VB1±V1±V1, then k2=2, and so on. A1' is the actual amount of PAFC drug delivered based on the drug delivery rate A1 predicted in S3, adjusted according to the number of stages for TP1 exceeding the threshold. VB1' is the actual delivery rate of PAFC drug delivered based on the baseline rate VB1, adjusted according to the number of stages for Q1 exceeding the threshold. The above adjustment logic satisfies: The dosage adjustment function relationship for PAFC (polyferric chloride) is as follows: When TP1>TP 1,max At that time, A1' = A1 + k1 × a1; When TP1 <TP 1,min At that time, A1' = A1 - k1 × a1; The regulatory function relationship for the PAFC (polyferric chloride) drug delivery rate is as follows: When Q1>Q 1,max At that time, VB1' = VB1 + k2 × V1; When Q1 1,min At that time, VB1' = VB1 - k2 × V1;​ 9. The method for a precise dosing control system for wastewater treatment based on an intelligent multidimensional prediction model according to claim 5, characterized in that: The core control unit in S5 regulates the dosage, rate, start time and frequency of the PAC (polyaluminum chloride) automatic dosing device and the PAM (polyacrylamide) automatic dosing device, compares the total phosphorus concentration (TP2) at the outlet of the secondary sedimentation tank with the second total phosphorus concentration threshold range, and compares the water flow rate (Q2) at the inlet of the high-efficiency sedimentation tank with the set water flow rate threshold range. Specifically, the upper and lower limits of the second total phosphorus concentration threshold are defined as TP. 2,max and TP 2,min The upper and lower limits of the water flow threshold are Q, respectively. 2,max and Q 2,min k3 represents the "number of stages" for TP2 exceeding the threshold. Specifically, when the total phosphorus concentration TP2 exceeds the upper limit of the second total phosphorus concentration threshold by two total phosphorus concentration stages, the dosage of PAC (polyaluminum chloride) is A2±a2, then k3=1. Similarly, if the dosage of PAC is A2±a2±a2, then k3=2, and so on. k4 represents the "number of stages" for Q2 exceeding the threshold. Similar to k2, if the dosage of PAC is VB2±V2, then k4=1; if the dosage of PAC changes to VB2±V2±V2, then k4=2, and so on. A2' is the actual dosage of PAC after adjusting for the number of stages exceeding the TP2 threshold based on the dosage A2 predicted by S3. VB2' is the dosage of PAC after adjusting for the number of stages exceeding the TP2 threshold based on the baseline rate VB2. Based on the actual PAC (polyaluminum chloride) drug delivery rate adjusted according to the number of stages exceeding the threshold in Q2, and since the dosage of PAM (polyacrylamide) follows the dosage of PAC, and the dosage A2 of the PAC automatic dosing device increases or decreases by a fixed amount a2, the dosage A3 of the PAM automatic dosing device increases or decreases by a fixed amount a3, then k3 and k4 can also be used to adjust PAM. Define A3' as the actual PAM dosage adjusted according to the number of stages exceeding the threshold in TP2 based on the drug dosage A3 predicted in S3, and VB3' as the actual PAM dosage adjusted according to the number of stages exceeding the threshold in Q2 based on the baseline rate VB3. The above adjustment logic satisfies: The dosage adjustment function relationship between PAC (polyaluminum chloride) and PAM (polyacrylamide) is as follows: When TP2>TP 2,max When, A2'=A2+k3×a2, A3'=A3+k3×a3; When TP2 <TP 2,min When, A2'=A2-k3×a2, A3'=A3-k3×a3; The relationship between the drug delivery rate regulation function of polyaluminum chloride (PAC) and polyacrylamide (PAM) is as follows: When Q2>Q 2,max At that time, VB2' = VB2 + k4 × V2, VB3' = VB3 + k4 × V3; When Q2 2,min At that time, VB2' = VB2 - k4 × V2, VB3' = VB3 - k4 × V3.​

Images