A method for precise control of dissolved oxygen in a sewage aerobic treatment system

Abstract

The present application belongs to the technical field of sewage treatment, more particularly, relates to a precise control method for dissolved oxygen in a sewage aerobic treatment system. The present application combines online monitoring of system parameters of a sewage treatment plant with dissolved oxygen control functions, can collect dissolved oxygen concentration, air valve opening, fan air volume, water inflow and water quality data of the sewage treatment system in real time, the server will automatically calculate the ideal value of fan air volume and air valve opening according to the target value of dissolved oxygen concentration, and select the best adjustment mode based on certain conditions, finally send the command to the fan or air valve, realize the precise control of dissolved oxygen in the sewage treatment system. Through machine learning, the present application controls the dissolved oxygen in the sewage treatment system within a reasonable range, reduces the waste of aeration fan power, greatly reduces the labor cost, and guarantees the efficient and low-cost operation of the biological treatment process in the sewage treatment system.

Description

Technical Field

[0001] This invention belongs to the field of wastewater treatment technology, and more specifically, relates to a method for precise control of dissolved oxygen in an aerobic wastewater treatment system. Background Technology

[0002] Wastewater treatment plants play a vital role in modern society. However, traditional wastewater treatment methods often face challenges such as high energy consumption, high costs, and complex operations. To address these challenges, the application of machine learning technology is bringing new solutions to wastewater treatment plants. The use of machine learning models in wastewater treatment plants has become an important means of improving operational efficiency, reducing costs, and optimizing processes.

[0003] In existing technologies, machine learning methods are widely used in wastewater treatment plants for operational monitoring: by collecting large amounts of data from devices such as sensors, flow meters, and water quality testing instruments, machine learning models can monitor and analyze the wastewater treatment process in real time. These models can identify anomalies and provide early warnings, thereby helping operators respond quickly and reduce potential losses and risks.

[0004] Energy consumption in wastewater treatment plants has always been a focus of industry attention. The aeration process in aerobic wastewater treatment consumes a significant proportion of the electricity. Furthermore, traditional aeration systems require constant manual adjustment, which is not only time-consuming and labor-intensive but also carries the risk of operational errors. Introducing machine learning models to replace manual adjustment of the aeration process would bring numerous advantages.

[0005] Furthermore, wastewater treatment plants often operate with multiple blowers running in parallel. The airflow generated by these blowers is collected in a single duct and distributed to each biological treatment tank via valves. When the dissolved oxygen concentration deviates slightly from the set value, the valve openings in each biological treatment tank are adjusted to control the dissolved oxygen concentration. However, since the blower airflow is constant, changes in the opening of a single valve will affect the airflow to other valves, thus affecting the dissolved oxygen content in other biological treatment tanks. Moreover, when the dissolved oxygen concentration deviates significantly from the set value, the blower airflow will be adjusted. Therefore, controlling the blower airflow and valve openings is crucial and complex. Adjustments made by technicians based on experience are prone to inaccuracies and significant errors.

[0006] Chinese patent document CN201620880870.7 discloses a dissolved oxygen monitoring system for wastewater treatment. This system, through the cooperation of a feedback module and a microprocessor, automatically activates a buzzer and warning lights to trigger on-site alarms, facilitating timely oxygenation measures by on-site management personnel. However, this patent does not provide a technical solution to the problem of the need for manual control of dissolved oxygen concentration, nor does it offer any technical inspiration. Summary of the Invention

[0007] 1. The problem to be solved

[0008] To address the problem that the dissolved oxygen concentration in the biological tank of an aerobic wastewater treatment system needs to be manually controlled in existing technologies, this invention provides a method for precise control of dissolved oxygen in an aerobic wastewater treatment system that uses a deep learning model to intelligently control the dissolved oxygen concentration.

[0009] 2. Technical Solution

[0010] This invention provides a method for precise control of dissolved oxygen in an aerobic wastewater treatment system, comprising the following steps:

[0011] S1. Use the data acquisition and control module to collect data on the influent water quality, influent water volume, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and air volume of the blower in the aerobic wastewater treatment system, and transmit them to the server.

[0012] S2. A machine learning matrix is ​​constructed based on the influent water quality, influent water volume, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and air volume of the blower in the aerobic wastewater treatment system.

[0013] S3. Establish a deep learning model between the influent water quality, influent water volume, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and blower air volume of the aerobic wastewater treatment system; and train the deep learning model based on the data matrix constructed in step S2.

[0014] S4. Set the target values ​​for dissolved oxygen concentration in the first and second biochemical pools in the server;

[0015] S5. When the dissolved oxygen concentration of the first biochemical tank and the second biochemical tank deviates from the target value, the recommended adjustment values ​​of the opening degree of the first air valve, the opening degree of the second air valve, and the air volume of the fan are calculated based on the model constructed in step S3, and the dissolved oxygen concentration of the first biochemical tank and the second biochemical tank is maintained at the target value by adjusting the opening degree of the first air valve, the opening degree of the second air valve, or the air volume of the fan.

[0016] Furthermore, in step S1, the influent water quality data includes influent chemical oxygen demand concentration data and influent ammonia nitrogen concentration data.

[0017] Furthermore, in step S1, the data acquisition time interval for the influent water quality, influent water volume, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and air volume of the blower in the aerobic wastewater treatment system is a fixed value.

[0018] The time interval for collecting the influent water quality and influent water volume data shall not exceed 60 minutes, and the time interval for collecting the dissolved oxygen concentration of the first and second biological treatment tanks, the opening degree of the first and second air valves, and the air volume of the blower shall be 20-60 seconds.

[0019] Furthermore, in step S2, constructing the machine learning matrix specifically involves:

[0020]

[0021] t1~t m For each recording time, OP11~OP1 m To record the opening degree of the first air valve corresponding to each time, OP21~OP2 m To record the opening degree of the corresponding second air valve each time, FL1~FL m To record the corresponding fan air volume each time, COD1~COD m To record the corresponding influent chemical oxygen demand (COD) for each time, N1 to N m To record the corresponding influent ammonia nitrogen concentration each time, intFL1~intFL m To record the corresponding influent volume each time, DO11~DO1 m To record the dissolved oxygen concentration (DO21~DO2) in the first biological treatment tank for each time, m Record the dissolved oxygen concentration in the second biological tank for each time.

[0022] Furthermore, the deep learning model in step S3 is a long short-term memory model built based on a long short-term memory (LSTM) network.

[0023] Furthermore, the long short-term memory model requires training with at least 6 days of data before it can be used.

[0024] Furthermore, the long short-term memory model is retrained every 30 minutes based on updated data during use.

[0025] By adopting the above technical solution, the model can be retrained every 30 minutes based on the updated data, enabling it to continuously adapt to the current aerobic wastewater treatment system and more accurately adjust the dissolved oxygen concentration in the first and second biological treatment tanks.

[0026] Furthermore, in step S4, the target range of dissolved oxygen concentration for the first and second biological treatment tanks is 1-5 mg / L; when the difference between the measured dissolved oxygen concentration and the target value is greater than 0.2 mg / L, it is considered that the dissolved oxygen concentration of the first and second biological treatment tanks deviates from the target value.

[0027] Furthermore, in step S4, the method for setting the target dissolved oxygen concentration values ​​of the first and second biochemical pools in the server is to input the set target values ​​into the server using a user terminal.

[0028] Using the above technical solution, the server provides adjustment suggestions for the blower and valves based on the current influent water quality, influent water volume and dissolved oxygen concentration of the aerobic wastewater treatment system. The optimal adjustment command obtained is transmitted to the blower and / or valve through the data acquisition and control module, thereby enabling the dissolved oxygen concentration in the biological tank to meet the requirements.

[0029] Furthermore, in step S5, the specific method for calculating the adjustment values ​​of the first air valve opening, the second air valve opening, and the fan air volume is as follows:

[0030] When the dissolved oxygen concentration detected by the first or second online dissolved oxygen detector deviates from the target value, the long short-term memory model is input with the following data: influent water quality and flow rate, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and blower airflow data from time 0 to T. The target dissolved oxygen concentration values ​​for the first and second biological treatment tanks at time T+1 are then used for prediction. The required adjustment values ​​for the opening degree of the first and second air valves or the blower airflow at time T+1 are calculated accordingly. A time step separates T from T+1, and the data acquisition time intervals of multiple monitoring devices are typically different. When the time step of the input data matrix is ​​less than the parameter acquisition interval, the value of that parameter within a single acquisition interval is repeatedly applied to multiple time steps; when the time step is greater than the parameter acquisition interval, the average value of that parameter within multiple acquisition intervals is applied to that time step.

[0031] Furthermore, in step S5, when the fan air volume needs to be increased, the opening of the first air valve and the second air valve are adjusted first. When the suggested values ​​for adjusting the opening of the first air valve and the second air valve given by the server exceed the adjustable range, the fan air volume is then adjusted. This solution is more energy-efficient and reduces consumption compared to directly increasing the fan air volume.

[0032] Furthermore, in step S5, when the fan air volume needs to be reduced, the fan air volume is adjusted first. When the fan air volume adjustment suggestion value given by the server exceeds the adjustable range, the opening of the first air valve or the second air valve is adjusted. Compared with prioritizing the adjustment of the first air valve or the second air valve, this scheme can reduce the aeration volume of the sewage treatment plant, thereby achieving the effect of energy saving.

[0033] Furthermore, in step S5, when the server's suggested adjustments to the fan airflow and the openings of the first and second air valves are both increased or decreased, a multi-parameter adjustment strategy is implemented. All three parameters are simultaneously assessed to determine if they exceed their adjustable range; those within the range are adjusted simultaneously. When the server's suggested adjustments to the fan airflow and the openings of the first and second air valves are both increased or decreased, the dissolved oxygen in both the first and second biological treatment tanks typically reaches its limit simultaneously. Adjusting multiple parameters simultaneously provides a faster response compared to adjusting only a single parameter, preventing emergencies.

[0034] Furthermore, in step S5, when the server suggests increasing the fan airflow, it is impossible for both the first and second valve openings to be decreased; when the server suggests decreasing the fan airflow, it is impossible for both the first and second valve openings to be increased.

[0035] Furthermore, in step S5, when the fan air volume, the opening degree of the first air valve, and the opening degree of the second air valve given by the server all exceed the adjustable range, a manual inspection and confirmation is performed.

[0036] It should be noted that increasing the blower airflow will increase the aeration rate in both the first and second biological treatment tanks, leading to an increase in dissolved oxygen concentration. The same applies when the blower airflow is decreased. Increasing the opening of the first air valve while keeping the opening of the second air valve constant will increase the aeration rate in the first biological treatment tank, resulting in an increase in dissolved oxygen concentration. However, because the blower airflow remains constant, the aeration rate in the second biological treatment tank will decrease, leading to a decrease in dissolved oxygen concentration. The same applies when increasing the opening of the second air valve while keeping the opening of the first air valve constant. In other words, when the blower airflow is constant, the total airflow through the first and second air valves is constant, and therefore the total aeration rate in the first and second biological treatment tanks is constant.

[0037] Furthermore, the adjustable range of the opening degree of the first air valve and the second air valve is 30-100%.

[0038] Furthermore, the adjustable air volume of the fan is 20-90% of the maximum air volume.

[0039] Furthermore, the dissolved oxygen precise control method for the aerobic wastewater treatment system is based on an aerobic wastewater treatment system comprising:

[0040] The first and second biological treatment tanks are used for aerobic wastewater treatment.

[0041] A blower is used to control the total aeration volume of the first and second biological treatment tanks.

[0042] A first air valve is located between the blower and the first biochemical tank and is connected to the blower and the first biochemical tank;

[0043] The second air valve is located between the blower and the second biochemical tank and is connected to the blower and the second biochemical tank;

[0044] The first online dissolved oxygen detector is used to detect the dissolved oxygen concentration in the first biological treatment tank.

[0045] The second online dissolved oxygen detector is used to detect the dissolved oxygen concentration in the second biological treatment tank.

[0046] Online influent water quality and quantity monitoring instruments, including online COD analyzers, online ammonia nitrogen analyzers, and flow meters, are used to monitor the influent water quality and quantity of wastewater treatment systems.

[0047] Furthermore, the wastewater aerobic treatment system also includes a data acquisition and control module connected to the blower, the first air valve, the second air valve, the first dissolved oxygen online detector, the second dissolved oxygen online detector, and the influent water quality and quantity data online detector.

[0048] The data acquisition and control module is used to collect data on the fan air volume, the opening degree of the first air valve, the opening degree of the second air valve, the dissolved oxygen concentration of the first biological treatment tank, the dissolved oxygen concentration of the second biological treatment tank, the influent water quality, and the influent water volume, and to control the fan air volume, the opening degree of the first air valve, and the opening degree of the second air valve.

[0049] Furthermore, the wastewater aerobic treatment system also includes a server connected to the data acquisition and control module.

[0050] After the server obtains and stores the data on fan air volume, first air valve opening, second air valve opening, dissolved oxygen concentration in the first biological treatment tank, dissolved oxygen concentration in the second biological treatment tank, influent water quality, and influent water volume, it calculates the recommended adjustment values ​​for the fan air volume, first air valve opening, and second air valve opening, selects the best adjustment method, and then transmits the adjustment command to the data acquisition and control module.

[0051] Furthermore, the aerobic wastewater treatment system also includes a user terminal connected to the server.

[0052] The user inputs target values ​​for the dissolved oxygen concentrations of the first and second biochemical pools into the user terminal; the user terminal then transmits these target values ​​to the server.

[0053] Furthermore, the connection between the data acquisition and control module and the fan, the first air valve, the second air valve, the first dissolved oxygen online detector, the second dissolved oxygen online detector, and the influent water quality and quantity data online detector, the connection between the server and the data acquisition and control module, and the connection between the user terminal and the server can be wired or wireless.

[0054] Using the above technical solution, the data acquisition and control module collects real-time data on influent water quality, influent water volume, opening degree of the first and second air valves, and blower air volume; then transmits the data to the server in real time for storage and analysis; based on big data analysis and machine learning methods, a dynamic simulation model of the aerobic wastewater treatment system is constructed, and the server provides suggested adjustment values ​​for blower air volume, first air valve opening degree, and second air valve opening degree; the server selects the optimal adjustment method according to the priority adjustment level and sends the optimal adjustment command to the data acquisition and control module to control the blower air volume, first air valve opening degree, and second air valve opening degree, thereby achieving precise control of dissolved oxygen in the biological treatment tank.

[0055] In existing technologies, dissolved oxygen concentration control in wastewater treatment plants is typically based on manual adjustment, which requires highly experienced workers, resulting in high labor costs and energy consumption. This invention uses a long short-term memory model to calculate and select appropriate valve opening and blower airflow adjustment methods based on a pre-set dissolved oxygen concentration target value, and automatically sends commands to the blower and valves, achieving automated and precise control of dissolved oxygen. This not only saves labor costs in wastewater treatment plants but also maintains the dissolved oxygen concentration in the biological treatment tank within a stable range, improving process stability and reducing accidents such as effluent deterioration caused by dissolved oxygen runaway. Furthermore, it reduces the total aeration volume and saves energy. This invention, using a long short-term memory model, enables intelligent control and improves upon the aforementioned problems.

[0056] 3. Beneficial effects

[0057] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0058] This invention uses machine learning to control dissolved oxygen in wastewater treatment systems within a reasonable range, which reduces power waste from aeration blowers, significantly lowers labor costs, and ensures efficient and low-consumption operation of the biological treatment process in wastewater treatment systems.

[0059] More specifically, it has the following beneficial effects:

[0060] (1) The machine learning model used in this invention can accurately monitor water quality and aeration requirements, and make rapid and accurate adjustments based on real-time data, thereby improving treatment efficiency and reducing energy waste.

[0061] The machine can work continuously and stably, unaffected by factors such as human fatigue or mood, ensuring the stability and reliability of the aeration system. The fully automated control method eliminates the need for human labor costs and reduces the impact of human error on the treatment effect, further improving the overall quality and efficiency of wastewater treatment.

[0062] (2) This invention uses a Long Short-Term Memory (LSTM) model to intelligently regulate dissolved oxygen, enabling precise, adaptive, and intelligent control, bringing new possibilities to the operation and management of wastewater treatment plants. Compared with traditional methods, the LSTM model can achieve precise control, adapt to different operating conditions, generalize predictions for unknown conditions, and realize automated control decisions.

[0063] (3) In this invention, the LSTM model and optimization algorithm are used to realize the optimal control strategy, which further improves the processing efficiency and energy saving. When it is necessary to increase the fan air volume, the opening of the first air valve and the second air valve are adjusted first; when it is necessary to decrease the fan air volume, the fan air volume is adjusted first, which can save the energy consumption of the fan. When the fan air volume, the opening of the first air valve and the opening of the second air valve need to be increased or decreased, multiple parameters are adjusted at the same time, the response is faster, and the occurrence of emergency situations is prevented.

[0064] (4) This invention does not require large-scale structural modifications to the sewage treatment plant, resulting in low construction costs. Attached Figure Description

[0065] Figure 1 This is a schematic diagram of the control flow of the dissolved oxygen precision control method in the aerobic wastewater treatment system of Example 1;

[0066] Figure 2 This is a schematic diagram of the adjustment scheme and priority adjustment stage for the first air valve, the second air valve, and the fan in Example 1;

[0067] Figure 3 This is a schematic diagram illustrating the specific priority adjustment methods for the first air valve, the second air valve, and the blower in Example 1;

[0068] Figure 4 This is a diagram illustrating the implementation effect of the precise dissolved oxygen control method in the aerobic wastewater treatment system of Example 1.

[0069] Figure 5 The diagram shows the effect of manually controlling dissolved oxygen in the aerobic wastewater treatment system in Comparative Example 1.

[0070] In the picture,

[0071] 1. First biological treatment tank; 2. Second biological treatment tank; 3. Blower; 4. First air valve; 5. Second air valve; 6. First online dissolved oxygen detector; 7. Second online dissolved oxygen detector; 8. Online influent water quality and quantity data detector; 9. Data acquisition and control module; 10. Server; 11. User terminal;

[0072] "FL" indicates the fan air volume; "OP1" indicates the opening degree of the first air valve; "OP2" indicates the opening degree of the second air valve; "DO1" indicates the dissolved oxygen concentration of the first biological treatment tank; "DO2" indicates the dissolved oxygen concentration of the second biological treatment tank; "↑" indicates that the server suggests increasing the concentration; "↓" indicates that the server suggests decreasing the concentration. Detailed Implementation

[0073] It should be noted that when a component is referred to as being "mounted" on another component, it can be directly on the other component or the two components can be integrated as one unit; when a component is referred to as being "connected" to another component, it can be directly connected to the other component or the two components can be integrated as one unit. Furthermore, terms such as "upper," "lower," "left," "right," and "middle" used in this specification are merely for clarity of description and are not intended to limit the scope of implementation. Changes or adjustments to their relative relationships, without substantially altering the technical content, should also be considered within the scope of the invention.

[0074] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0075] Unless otherwise specified in the examples, the procedures should be performed under standard conditions or conditions recommended by the manufacturer. Reagents or instruments whose manufacturers are not specified are all commercially available products.

[0076] As used herein, the term “about” is used to provide for the flexibility and imprecision associated with a given term, measure, or value. Those skilled in the art can readily determine the degree of flexibility for a particular variable.

[0077] As used herein, “adjacent” means that two structures or elements are close to each other. Specifically, elements identified as “adjacent” may be adjacent or connected. Such elements may also be close to or near each other without necessarily touching. In some cases, the precision of proximity may depend on the specific context.

[0078] As used herein, the term “at least one of…” is intended to be synonymous with “one or more of…”. For example, “at least one of A, B, and C” explicitly includes only A, only B, only C, and combinations thereof.

[0079] Concentration, amount, and other numerical data may be presented in range format herein. It should be understood that such range format is used solely for convenience and brevity and should be flexibly interpreted to include not only the values ​​explicitly stated as the limits of the range, but also all individual values ​​or subranges encompassed within the range, as if each value and subrange were explicitly stated. For example, a range of values ​​from about 1 to about 4.5 should be interpreted to include not only the explicitly stated limits of 1 to 4.5, but also individual numbers (such as 2, 3, 4) and subranges (such as 1 to 3, 2 to 4, etc.). The same principle applies to ranges that describe only a single value, such as “less than about 4.5,” which should be interpreted to include all the aforementioned values ​​and ranges. Furthermore, this interpretation should apply regardless of the breadth of the range or characteristic described.

[0080] Any step described in any method or process claim (e.g., steps S1, S2, S3... or steps (1), (2), (3)... or steps 1), 2), 3)...) may be performed in any order, and is not limited to the order set forth in the claims.

[0081] The limitation of method + function or step + function is used only if all of the following conditions are met in a particular claim: a) it expressly states "a method for..." or "a step for..."; b) it expressly states the corresponding function. The structures, materials, or actions supporting the method + function are expressly described in the description herein. Therefore, the scope of the invention should be determined solely by the appended claims and their legal equivalents, and not by the description and examples given herein. The invention is further described below with reference to specific embodiments.

[0082] Example 1

[0083] This embodiment provides a method for precise control of dissolved oxygen in an aerobic wastewater treatment system, as detailed below:

[0084] In this embodiment, the aerobic wastewater treatment system uses the activated sludge process to treat industrial park wastewater. The influent chemical oxygen demand (COD) ranges from 100-500 mg / L. The aerobic wastewater treatment system includes a first biological treatment tank 1 and a second biological treatment tank 2. A set of variable frequency blowers 3 is used for daily oxygen supply to the first and second biological treatment tanks. A first air valve 4 and a second air valve 5 are installed before the gas flows into each of the two biological treatment tanks. An online influent water quality and quantity data monitor 8 is used to monitor the influent flow rate, COD, and ammonia nitrogen concentration of the aerobic wastewater treatment system in real time. The influent flow rate data acquisition interval is 30 minutes, and the COD and ammonia nitrogen concentration data acquisition interval is 1 hour. The first and second biological treatment tanks 1 and 2 are equipped with a first dissolved oxygen online monitor 6 and a second dissolved oxygen online monitor 7, with a real-time data acquisition interval of 1 minute. The real-time data acquisition interval for the blower 3 airflow and the opening degree of the two air valves is also 1 minute. All monitored data are transmitted in real time to the central control system (user terminal 11) for worker observation and operation.

[0085] In this embodiment, the online influent water quality and quantity data monitoring instrument 8 includes an online COD analyzer, an online ammonia nitrogen analyzer, and a flow meter.

[0086] In this embodiment, the first online dissolved oxygen detector 6, the second online dissolved oxygen detector 7, the online COD analyzer, the online ammonia nitrogen water quality analyzer, and the flow meter are all conventional products that can be purchased commercially, and there are no restrictions on the products and models.

[0087] The method for precise control of dissolved oxygen in the aerobic wastewater treatment system of the present invention is applied to the aerobic wastewater treatment system. The system monitors the influent water quality and quantity data, dissolved oxygen data of the first biological treatment tank 1 and the second biological treatment tank 2, the air volume of the blower 3, and the opening data of the two gas valves and transmits them to the server in real time. Based on the current status of each parameter and the target set value of dissolved oxygen concentration, the server will provide suggested values ​​for the opening of the gas valves of the first biological treatment tank 1 and the second biological treatment tank 2, and the air volume of the blower 3.

[0088] In this embodiment, the user uses user terminal 11 to input the set dissolved oxygen target value into server 10.

[0089] In this embodiment, the Keras framework is used to build a long short-term memory model.

[0090] Specifically, the Long Short-Term Memory (LSTM) model is input with the following data: influent water quality data, influent water volume data, dissolved oxygen concentrations DO1 and DO2 of the first and second biological treatment tanks 1 and 2, opening data OP1 and OP2 of the first and second air valves 4 and 5, air volume data FL of the blower 3, and the target values ​​of dissolved oxygen concentrations DO1(T+1) and DO2(T+1) of the first and second biological treatment tanks 1 and 2 at time T+1. The required adjustment values ​​for the opening of the first air valve 4 OP1(T+1), the opening of the second air valve 5 OP2(T+1), or the air volume FL(T+1) of the blower 3 at time T+1 are calculated respectively.

[0091] Based on the principle of energy conservation, when the airflow of fan 3 needs to be increased, the air valve opening is adjusted first. Only when the suggested air valve openings all exceed the adjustable range is the airflow of fan 3 adjusted. Conversely, when the airflow of fan 3 needs to be decreased, the airflow of fan 3 is adjusted first. Only when the suggested airflow of fan 3 exceeds the adjustable range is the air valve opening adjusted. When the server's suggested adjustments for fan 3 airflow, the opening of the first air valve 4, and the opening of the second air valve 5 are all either increased or decreased, a multi-parameter adjustment strategy is implemented. All three parameters are simultaneously assessed to determine if they exceed the adjustable range. For parameters not exceeding the range... The parameters are adjusted simultaneously. When the server suggests increasing the airflow of fan 3, it's impossible for the opening of the first air valve 4 and the second air valve 5 to both be decreased. Similarly, when the server suggests decreasing the airflow of fan 3, it's impossible for the opening of the first air valve 4 and the second air valve 5 to both be increased. If the server-suggested airflow of fan 3, the opening of the first air valve 4, and the opening of the second air valve 5 all exceed the adjustable range, an alarm will be sent to the central control system for manual inspection and confirmation. The specific adjustment methods and priorities for fan 3 and the air valves are as follows: Figure 2 and Figure 3 As shown.

[0092] Based on the above adjustment priorities, the optimal adjustment method is finally selected, and a command is sent to the air pump electronic valve or fan 3 to achieve precise control of dissolved oxygen concentration.

[0093] Before implementing this method for automated control, a Long Short-Term Memory (LSTM) model was trained using 6 days of manually calibrated data. The machine learning matrix time step was 1 minute, the training set to test set ratio was 9:1, the model training timesteps were 10, the number of training epochs was 50, and the batch size was 100. The target dissolved oxygen concentrations (DO1 and DO2) in the first and second biological treatment tanks were set at 2.5 mg / L. The model was used for automated control, and the changes in dissolved oxygen concentration were tested over 5 hours. The implementation effect of this method is as follows: Figure 4 .

[0094] In this embodiment, the following specific adjustment process is provided as a concrete application of the method for precise control of dissolved oxygen in the aerobic wastewater treatment system:

[0095] Adjustment Process 1: The target dissolved oxygen concentration in the first biological tank 1 and the second biological tank 2 is 2.5 mg / L. DO1 shows an upward trend (2.6 mg / L), while DO2 shows a downward trend (2.3 mg / L). DO2 reaches the target value range threshold of 2.5-0.2 mg / L. The model calculation yielded three adjustment methods for the parameters: (1) increasing the air volume FL of blower 3 from 53% to 65%; (2) decreasing the opening of the first air valve 4 OP1 from 51% to 45%; (3) increasing the opening of the second air valve 5 OP2 from 59% to 72%. The optimal adjustment method is to adjust the opening of the second air valve 5 OP2 from 59% to 72%, which increases the dissolved oxygen concentration DO2 in the second biological tank 2 and decreases the dissolved oxygen concentration DO1 in the first biological tank 1. After adjustment, DO1 and DO2 are maintained within the range of 2.3-2.7 mg / L.

[0096] Adjustment Process 2: The target dissolved oxygen concentrations in the first biological tank 1 and the second biological tank 2 are 2.5 mg / L. DO1 shows an upward trend (2.7 mg / L), and DO2 shows an upward trend (2.4 mg / L). DO1 reaches the target value range threshold of 2.5 + 0.2 mg / L. The model calculation yields three adjustment methods for the parameters: (1) reduce the air volume FL of blower 3 from 34% to 18%; (2) reduce the opening of the first air valve 4 OP1 from 52% to 37%; (3) increase the opening of the second air valve 5 OP2 from 69% to 84%. The air volume of blower 3 is adjusted first, but since the air volume adjustment method FL < 20%, it exceeds the set normal adjustment range. Therefore, the optimal adjustment method is to adjust the opening of the first air valve 4 OP1 from 52% to 37%, so that the dissolved oxygen concentration DO1 in the first biological tank 1 decreases and the dissolved oxygen concentration DO2 in the second biological tank 2 increases. After adjustment, DO1 and DO2 were maintained within the range of 2.3-2.7 mg / L.

[0097] Adjustment process three: The target dissolved oxygen concentration in the first biological tank 1 and the second biological tank 2 is 2.5 mg / L. DO1 shows a decreasing trend (2.3 mg / L), and DO2 shows a decreasing trend (2.3 mg / L). Both DO1 and DO2 reach the target value range threshold of 2.5-0.2 mg / L. The model calculation yields the adjustment methods for three parameters: (1) Increase the air volume FL of blower 3 from 43% to 58%; (2) Increase the opening degree OP1 of the first air valve 4 from 87% to 107%; (3) Increase the opening degree OP2 of the second air valve 5 from 64% to 82%. Multiple parameters were adjusted simultaneously. However, since the opening adjustment mode of the first air valve 4, OP1, exceeded 100%, which was outside the normal adjustment range, the optimal adjustment method was to simultaneously adjust the air volume of the blower 3, FL, from 43% to 58% and the opening of the second air valve 5, OP2, from 64% to 82%, so that the dissolved oxygen concentration (DO1) in the first biological treatment tank 1 and the dissolved oxygen concentration (DO2) in the second biological treatment tank 2 would increase simultaneously. After adjustment, DO1 and DO2 were maintained within the range of 2.3-2.7 mg / L.

[0098] Comparative Example 1

[0099] This comparative example provides a method for artificially controlling dissolved oxygen in an aerobic wastewater treatment system, as detailed below:

[0100] In this comparative example, the aerobic wastewater treatment system uses the activated sludge process to treat industrial park wastewater. The influent chemical oxygen demand (COD) ranges from 100-500 mg / L. The aerobic wastewater treatment system includes a first biological treatment tank and a second biological treatment tank. A set of variable frequency blowers is used for daily oxygen supply to the first and second biological treatment tanks. Each biological reactor has a first and a second air valve before the gas flows into it. An online influent water quality and quantity monitoring instrument is used to monitor the influent flow rate, COD, and ammonia nitrogen concentration in real time. The influent flow rate data is collected every 30 minutes, and the COD and ammonia nitrogen concentration data are collected every hour. The first and second biological treatment tanks are equipped with first and second online dissolved oxygen (DO) monitoring instruments, respectively, with data collected every minute. The blower airflow and valve opening data are also collected every minute. All monitored data are transmitted to the central control system in real time for worker observation and operation.

[0101] In this comparative example, the dissolved oxygen in the aerobic wastewater treatment system is adjusted manually in the first and second biological treatment tanks. Workers observe the real-time dynamic changes in dissolved oxygen and adjust the air valves and blower airflow based on experience. Increasing the blower airflow increases the aeration rate in both the first and second biological treatment tanks, leading to an increase in dissolved oxygen concentration; decreasing the blower airflow has the same effect. Increasing the opening of the air valve controlling the first biological treatment tank increases the aeration rate and dissolved oxygen concentration, but because the blower airflow remains unchanged, the aeration rate in the second biological treatment tank decreases, resulting in a decrease in dissolved oxygen concentration. The changes in dissolved oxygen were tested over 5 hours of manual adjustment.

[0102] The results showed that this method could maintain the dissolved oxygen concentrations (DO1 and DO2) in the first and second biological treatment tanks at 1.5-3.0 mg / L. Figure 5 Although manual adjustment is more energy-efficient than no adjustment, overall, the dissolved oxygen fluctuation range in the first and second biological treatment tanks is still relatively large, resulting in high operating costs.

[0103] In Example 1, after connecting to the dissolved oxygen precision control system of the present invention, the dissolved oxygen was controlled between 2.3-2.7 mg / L. Figure 4 This achieved the goal of maintaining dissolved oxygen in both the first and second biological treatment tanks within ±0.2 mg / L of the target set value. Furthermore, the dissolved oxygen control range was significantly narrowed compared to manual adjustment, achieving greater precision and energy savings.

[0104] The above description provides an illustrative overview of the present invention and its embodiments. This description is not restrictive, and the embodiments shown are merely one example of the invention's implementation. Actual implementations are not limited to these examples. Therefore, if those skilled in the art are inspired by this description and design similar implementations and examples without departing from the spirit of the invention, such designs should fall within the scope of protection of the present invention.

Claims

1. A method for precise control of dissolved oxygen in an aerobic wastewater treatment system, characterized in that, Includes the following steps: S1. The data acquisition and control module is used to collect data on the influent water quality, influent water volume, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and blower air volume of the aerobic wastewater treatment system, and transmits them to the server; the influent water quality data includes influent chemical oxygen demand concentration data and influent ammonia nitrogen concentration data. S2. A machine learning matrix is ​​constructed based on the influent water quality, influent water volume, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and air volume of the blower in the aerobic wastewater treatment system. In step S2, constructing the machine learning matrix specifically involves: t1~t m For each recording time, OP11~OP1 m To record the opening degree of the first air valve corresponding to each time, OP21~OP2 m To record the opening degree of the corresponding second air valve each time, FL1~FL m To record the corresponding fan air volume each time, COD1~COD m To record the corresponding influent chemical oxygen demand (COD) for each time, N1~N m To record the corresponding influent ammonia nitrogen concentration each time, intFL1~intFL m To record the corresponding influent volume each time, DO11~DO1 m To record the dissolved oxygen concentration (DO21~DO2) in the first biological treatment tank for each time, m Record the dissolved oxygen concentration in the second biological treatment tank for each instance; S3. Establish a deep learning model for the relationship between influent water quality, influent water volume, dissolved oxygen concentration in the first and second biological treatment tanks, opening degree of the first and second air valves, and blower air volume in the aerobic wastewater treatment system. The deep learning model is trained based on the data matrix constructed in step S2; the deep learning model in step S3 is a long short-term memory model constructed based on a long short-term memory (LSTM) network. S4. Set target values ​​for dissolved oxygen concentration in the first and second biochemical pools in the server; in step S4, the target value range for dissolved oxygen concentration in the first and second biochemical pools is 1-5 mg / L; when the difference between the measured dissolved oxygen concentration value and the target value is greater than 0.2 mg / L, it is considered that the dissolved oxygen concentration in the first and second biochemical pools deviates from the target value. S5. When the dissolved oxygen concentration of the first biochemical tank and the second biochemical tank deviates from the target value, the adjustment suggestions for the opening degree of the first air valve, the opening degree of the second air valve, and the air volume of the fan are calculated based on the model constructed in step S3, and the dissolved oxygen concentration of the first biochemical tank and the second biochemical tank is maintained at the target value by adjusting the opening degree of the first air valve, the opening degree of the second air valve, or the air volume of the fan. In step S5 When the recommended adjustments for the fan air volume, the opening of the first air valve, and the opening of the second air valve are both to be increased or decreased, simultaneously determine whether the three parameters exceed the adjustable range, and adjust the parameters that do not exceed the range simultaneously. When the fan air volume needs to be increased and the opening of the first air valve and the opening of the second air valve are different when increased or decreased, the opening of the first air valve and the opening of the second air valve should be adjusted first, and the one that needs to be increased should be adjusted first. When the suggested adjustment values ​​of the first air valve opening and the second air valve opening given by the server exceed the adjustable range, the fan air volume should be adjusted then. When the fan air volume needs to be reduced and the opening of the first air valve and the opening of the second air valve are different when they are increased or decreased, the fan air volume should be adjusted first. When the fan air volume adjustment suggestion value given by the server exceeds the adjustable range, the opening of the first air valve and the opening of the second air valve should be adjusted, and the one that needs to be reduced should be adjusted first. When the fan air volume, first air valve opening, and second air valve opening given by the server all exceed the adjustable range, manual inspection and confirmation are required. The adjustable range of the opening degree of the first air valve and the second air valve is 30-100%; The adjustable air volume range of the fan is 20-90% of the maximum air volume; The wastewater aerobic treatment system includes a first biological treatment tank and a second biological treatment tank; a set of variable frequency blowers for daily oxygen supply to the first and second biological treatment tanks; and a first gas valve and a second gas valve are provided before the gas flows to the two biological treatment tanks.

2. The method for precise control of dissolved oxygen in an aerobic wastewater treatment system according to claim 1, characterized in that, In step S1, the data acquisition time interval for the influent water quality, influent water volume, dissolved oxygen concentration of the first and second biological treatment tanks, opening degree of the first and second air valves, and air volume of the blower in the aerobic wastewater treatment system is a fixed value. The time interval for collecting the influent water quality and influent water volume data shall not exceed 60 minutes, and the time interval for collecting the dissolved oxygen concentration of the first and second biological treatment tanks, the opening degree of the first and second air valves, and the air volume of the blower shall be 20-60 seconds.

3. The method for precise control of dissolved oxygen in an aerobic wastewater treatment system according to claim 2, characterized in that, The Long Short-Term Memory (LSTM) model requires training with at least 6 days of data before it can be used.

4. The method for precise control of dissolved oxygen in an aerobic wastewater treatment system according to claim 2, characterized in that, The Long Short-Term Memory (LSTM) model is retrained every 30 minutes based on updated data during use.

Images