Intelligent control system and method for oxidation ditch with adaptive inflow carbon-nitrogen ratio fluctuation

Abstract

The application discloses an oxidation ditch intelligent regulation and control system and method adaptive to fluctuation of carbon-nitrogen ratio of inflow, and belongs to the technical field of sewage treatment. The technical scheme provided by the application realizes the innovation of an integrated installation structure of an aeration-pushing flow device, breaks through the technical bottleneck of separation installation of an existing aerator and a pusher, adopts an integrated installation structure of a bottom full-paving aeration unit and an upper coaxial pushing flow, and realizes more than 30% improvement of aeration uniformity at the bottom of the oxidation ditch. Moreover, the operation parameters of the aeration unit and the pushing flow unit are regulated and controlled through a cooperative control module, the system is switched between an anoxic state and an aerobic state, for example, when the aeration unit is controlled to be closed and the pushing flow unit is controlled to be opened, the system is in the anoxic state; when the aeration unit is controlled to be opened and the pushing flow unit is controlled to be closed, the system is in the aerobic state, so as to adapt to the requirements of different denitrification / nitrification reactions when the carbon-nitrogen ratio of the inflow fluctuates, and the reconstruction does not need to reconstruct the oxidation ditch pool body, and the construction is simple.

Description

Technical Field

[0001] This application relates to the field of wastewater treatment technology, and in particular to an intelligent control system and method for oxidation ditch that adapts to fluctuations in the influent carbon-nitrogen ratio. Background Technology

[0002] Currently, many existing wastewater treatment plants in southern China widely adopt the AAO (Anaerobic-Anoxic-Oxic) micro-aeration oxidation ditch process. This process, with its simple structure, stable operation, and moderate energy consumption, is widely used in municipal wastewater and small-to-medium-scale industrial wastewater treatment. However, with the acceleration of urbanization, industrial restructuring, and the impact of combined sewer overflows, the influent water quality of wastewater treatment plants in southern China exhibits significant and frequent fluctuations in the carbon-to-nitrogen ratio (C / N). For example, the C / N ratio can be as low as 2.0 during the dry season, but can surge to over 8.0 during the rainy season or when industrial wastewater is introduced, becoming a core bottleneck restricting the treatment effectiveness of existing AAO micro-aeration oxidation ditch processes. Summary of the Invention

[0003] This application provides an intelligent control system and method for oxidation ditches that adapts to fluctuations in the influent carbon-nitrogen ratio. It can overcome the shortcomings of existing AAO micro-aeration oxidation ditch processes, such as insufficient flexibility, low accuracy of aeration volume calculation, poor equipment coordination, and high difficulty in modification when treating water with large fluctuations in the influent carbon-nitrogen ratio. Through innovation in equipment installation structure, multimodal machine learning for aeration volume calculation, and equipment collaborative control, it can achieve dynamic control of anoxic / aerobic areas and precise supply of aeration volume, thereby improving the treatment effect under scenarios with large fluctuations in the influent carbon-nitrogen ratio and reducing modification costs.

[0004] According to one aspect of the embodiments of this application, an intelligent control system for oxidation ditch that adapts to fluctuations in the influent carbon-nitrogen ratio is provided. The intelligent control system includes an integrated aeration and propulsion module and a collaborative control module. The integrated aeration and propulsion module is disposed inside the oxidation ditch and includes an aeration unit and a propulsion unit. The aeration unit is laid at the bottom of the oxidation ditch, and the propulsion unit is disposed above the aeration unit. The collaborative control module is connected to the integrated aeration and propulsion module and is used to regulate the operating parameters of the aeration unit and the propulsion unit to control the oxidation ditch to switch between anoxic and aerobic states. In the anoxic state, the aeration unit is turned off and the propulsion unit is turned on; in the aerobic state, the aeration unit is turned on and the propulsion unit is turned off.

[0005] In an exemplary embodiment, the collaborative control module is further configured to regulate the residence time of the oxidation ditch in the anoxic state and the aerobic state.

[0006] In an exemplary embodiment, the aeration unit employs a micro-aerator, which is evenly distributed across the bottom of the oxidation ditch. The micro-aerator includes multiple independent aeration branch pipes, wherein the start / stop state and aeration intensity of each aeration branch pipe are independently controlled. The flow propulsion unit employs a submersible flow propulsion device, which is installed directly above the micro-aerator and coaxially arranged with it. The submersible flow propulsion device is fixed to the micro-aerator via a bracket, and its rotation speed is infinitely adjustable and controlled in conjunction with the aeration branch pipes.

[0007] In an exemplary embodiment, the intelligent control system for oxidation ditch further includes: a multi-source data acquisition module, used to acquire multi-dimensional data of the entire wastewater treatment process, the multi-dimensional data including fused water quality parameter data, equipment operating parameter data, and real-time operating condition image data; wherein, the fused water quality parameter data is used to characterize the change in the carbon-nitrogen ratio of the wastewater entering the oxidation ditch, the equipment operating parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch; the collaborative control module is also used to output control commands to the intelligent control system based on the multi-dimensional data; wherein, the multi-source data acquisition module includes: a water quality parameter acquisition module, including a first sensor disposed at the inlet, outlet, and middle of the oxidation ditch, the first sensor including a chemical oxygen demand sensor and a total nitrogen sensor. The system includes sensors such as an ammonia nitrogen sensor, a nitrate nitrogen sensor, a dissolved oxygen sensor, a pH sensor, and a sludge concentration sensor. The integrated water quality parameter data includes collected data corresponding to the influent carbon-to-nitrogen ratio, carbon-to-nitrogen ratio fluctuation slope, dissolved oxygen concentration, and sludge concentration. The equipment operation parameter acquisition module includes second sensors installed on the aeration branch pipes, the blower, and the submersible jet mixer. These second sensors include a flow sensor, a pressure sensor, a speed sensor, and a current sensor. The equipment operation parameter data includes collected data corresponding to the flow rate of each aeration branch pipe, the air volume and / or pressure of the blower, and the speed of the submersible jet mixer. The operating condition image acquisition module includes an industrial camera installed above the oxidation tank. This industrial camera is used to acquire real-time images of the aeration bubble distribution in the oxidation tank. The real-time operating condition image data includes multiple images of the aeration bubble distribution.

[0008] In an exemplary embodiment, the intelligent control system further includes: a multimodal intelligent computing module, which incorporates a multimodal fusion machine learning model optimized for carbon-nitrogen ratio fluctuations. The multimodal fusion machine learning model is used to process multimodal data in the multidimensional data and output the required aeration rate and the defined virtual anoxic / aerobic zones. The required aeration rate and the virtual anoxic / aerobic zones are adapted to the carbon-nitrogen ratio conditions of the wastewater currently entering the oxidation ditch. The multi-source data acquisition module further includes an environmental parameter acquisition module. The multidimensional data also includes environmental parameter data acquired by the environmental parameter acquisition module. The environmental parameter data is used to correct the required aeration rate.

[0009] In an exemplary embodiment, the collaborative control module is also connected to the multimodal intelligent computing module; the collaborative control module is also used to output control commands corresponding to the current operating conditions of the oxidation ditch intelligent control system based on the fused water quality parameter data and the output results of the multimodal intelligent computing module, so as to control the operating parameters of the aeration unit and the propulsion unit.

[0010] In an exemplary embodiment, the collaborative control module is specifically used to determine the carbon-nitrogen ratio classification conditions corresponding to the integrated water quality parameter data and output aeration-flow staggered operation control commands according to the carbon-nitrogen ratio classification conditions; wherein, the aeration-flow staggered operation control commands include: timing parameters and dynamic correction coefficients corresponding to aeration duration, aeration stop duration, flow propulsion duration, and flow stop duration, the stepless speed regulation value corresponding to the flow propulsion unit, and an aeration / flow propulsion interlock signal; the collaborative control module is also specifically used to output precise aeration volume zoning control according to the required aeration volume and the virtual anoxic / aerobic zone. The control module provides control commands; wherein, the precise control commands for aeration volume zoning include a precise allocation value for total aeration volume, a start / stop status value for a single aeration branch pipe, an aeration intensity adjustment value for a single aeration branch pipe, blower airflow / pressure adjustment parameters, and an isolation signal for a faulty aeration branch pipe, wherein the faulty aeration branch pipe is identified based on the aeration bubble distribution image; the collaborative control module is also specifically used to output emergency working condition backup control commands to provide preset control parameters to the integrated aeration propulsion module in emergency situations such as sudden changes in carbon-nitrogen ratio, sensor failure, or failure of the multimodal intelligent computing module.

[0011] According to one aspect of the embodiments of this application, an intelligent control method for oxidation ditch that adapts to influent carbon-nitrogen ratio fluctuations is provided. The method is applied to the intelligent control system for oxidation ditch described in any of the above claims. The method includes: acquiring integrated water quality parameter data, equipment operating parameter data, and real-time operating condition image data; wherein the integrated water quality parameter data is used to characterize the carbon-nitrogen ratio change of the wastewater entering the oxidation ditch, the equipment operating parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch; inputting the integrated water quality parameter data, the equipment operating parameter data, and the real-time operating condition image data into a multimodal fusion machine learning model for processing, and outputting the required aeration volume and the divided virtual anoxic / aerobic zones, wherein the required aeration volume and the virtual anoxic / aerobic zones are adapted to the current carbon-nitrogen ratio operating condition of the wastewater entering the oxidation ditch; and outputting control commands corresponding to the current operating condition of the intelligent control system for oxidation ditch based on the integrated water quality parameter data and the output results of the multimodal intelligent calculation module, so as to control the operating parameters of the aeration unit and the propulsion unit.

[0012] In an exemplary embodiment, the multimodal fusion machine learning model includes a time-series data processing sub-model and a visual data processing sub-model. The step of inputting the fused water quality parameter data, the equipment operating parameter data, and the real-time operating condition image data into the multimodal fusion machine learning model for processing, and outputting the required aeration volume and the defined virtual anoxic / aerobic zones, includes: preprocessing the fused water quality parameter data and the equipment operating parameter data and performing feature extraction to obtain standardized sequence feature data; preprocessing the real-time operating condition image data and performing feature extraction to obtain visual feature data; and inputting the sequence feature data into the time-series data processing sub-model for processing to obtain the required aeration volume and the aeration volume of each... The system collects target start / stop status data, target aeration intensity, and virtual anoxic / aerobic areas corresponding to the branch pipes. Visual feature data is input into the visual data processing sub-model for detection, yielding aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration area location information. When the aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration area location information indicate that the oxidation ditch intelligent control system is operating normally, the system outputs the required aeration volume, target start / stop status data, target aeration intensity, and virtual anoxic / aerobic areas corresponding to each aeration branch pipe. The target open / closed status of each aeration branch pipe is associated with the virtual anoxic / aerobic area.

[0013] In an exemplary embodiment, the method further includes: when the aeration uniformity level data, the aeration pipe operating condition judgment result, and the abnormal aeration area location information indicate that the operating condition of the oxidation ditch intelligent control system is abnormal, removing the aeration volume calculation weight of the faulty aeration branch indicated by the aeration pipe operating condition judgment result, redistributing a portion of the aeration volume corresponding to the faulty aeration branch, and updating the target start / stop status data, the target aeration intensity, and the virtual anoxic / aerobic area corresponding to each aeration branch.

[0014] According to one aspect of the embodiments of this application, an intelligent control device for oxidation ditch that adapts to influent carbon-nitrogen ratio fluctuations is provided. The device includes: a multi-source data acquisition module, used to acquire fused water quality parameter data, equipment operating parameter data, and real-time operating condition image data; wherein, the fused water quality parameter data is used to characterize the carbon-nitrogen ratio change of the wastewater entering the oxidation ditch, the equipment operating parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch; a data fusion processing module, used to input the fused water quality parameter data, equipment operating parameter data, and real-time operating condition image data into a multimodal fusion machine learning model for processing, and output the required aeration volume and the divided virtual anoxic / aerobic zones, wherein the required aeration volume and the virtual anoxic / aerobic zones are adapted to the current carbon-nitrogen ratio operating condition of the wastewater entering the oxidation ditch; and a control module, which outputs control commands corresponding to the current operating condition of the intelligent control system for oxidation ditch based on the fused water quality parameter data and the output results of the multimodal intelligent calculation module, so as to control the operating parameters of the aeration unit and the propulsion unit.

[0015] According to one aspect of the embodiments of this application, a computer device is provided, the computer device including a processor and a memory, the memory storing at least one instruction, at least one program, code set or instruction set, the at least one instruction, the at least one program, the code set or instruction set being loaded and executed by the processor to realize the above-described intelligent control method for oxidation ditch adaptive influent carbon-nitrogen ratio fluctuation.

[0016] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein at least one instruction, at least one program, code set or instruction set is stored in the storage medium, wherein the at least one instruction, the at least one program, the code set or instruction set is loaded and executed by a processor to realize the above-described intelligent control method for oxidation ditch with adaptive influent carbon-nitrogen ratio fluctuation.

[0017] According to one aspect of the embodiments of this application, a computer program product is provided, the computer program product including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the computer device to perform the above-described intelligent control method for adaptive influent carbon-nitrogen ratio fluctuations in oxidation ditches.

[0018] The technical solution provided in this application provides the following benefits: It achieves an innovative integrated installation structure for aeration and propulsion equipment, overcoming the technical bottleneck of separate installation of aerators and propulsion devices. By adopting an integrated installation structure with a bottom layer of fully-covered aeration units and an upper layer of coaxial propulsion, the uniformity of aeration at the bottom of the oxidation ditch is improved by more than 30%. Furthermore, the operating parameters of the aeration and propulsion units are adjusted through a collaborative control module, enabling the system to switch between anoxic and aerobic states. For example, when the aeration unit is turned off and the propulsion unit is turned on, the system is in anoxic mode; when the aeration unit is turned on and the propulsion unit is turned off, the system is in aerobic mode. This adapts to the different denitrification / nitrification reaction requirements when the influent carbon-nitrogen ratio fluctuates. Moreover, the modification does not require reconstruction of the oxidation ditch, simplifying construction.

[0019] In addition, the technical solution provided in this application embodiment adopts a modular design for the intelligent control system of oxidation ditch. The aeration unit, the flow propulsion unit, and the collaborative control module can all be directly connected to the equipment interface of the existing AAO micro-aeration oxidation ditch, realizing modular transformation innovation adapted to old plant areas. The construction cycle can be shortened to less than 7 days, and the transformation cost can be reduced by more than 40%. This solves the problems of large engineering workload and high cost of the existing transformation scheme, and is suitable for the actual application scenario of sewage treatment plants in the south that have been built to cope with the aggravated fluctuation of influent carbon-nitrogen ratio. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of an intelligent control method for oxidation ditch that adapts to fluctuations in the influent carbon-nitrogen ratio, provided in an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0023] The inventors of this application have discovered, through their research, that the existing AAO micro-aeration oxidation ditch process has the following key defects, making it unable to adapt to the flexible treatment needs of large fluctuations in the carbon-nitrogen ratio of influent: 1. Insufficient flexibility in regional control: The existing process has a fixed division between anoxic and aerobic zones. The anoxic zone is equipped with a flow booster, and the aerobic zone is equipped with a dedicated aerator. The two are installed at a reasonable distance according to the design specifications. This fixed equipment installation layout means that the maximum effective retention time of the anoxic and aerobic zones is fixed. It is impossible to flexibly adjust the retention time and aeration intensity of the two zones according to the real-time fluctuation of the influent carbon-nitrogen ratio. When the carbon-nitrogen ratio fluctuates greatly, it is difficult to balance the utilization efficiency of denitrification carbon source and the sufficiency of nitrification reaction. The stability of total nitrogen (TN) and chemical oxygen demand (COD) in the effluent is poor, and there is a frequent risk of exceeding the standard. 2. Low accuracy and poor adaptability of aeration volume calculation: The mainstream aeration volume calculation relies on traditional mechanism models based on parameters such as influent volume, COD, and ammonia nitrogen, which are difficult to adapt to nonlinear operating conditions with large fluctuations in carbon-nitrogen ratio; moreover, it is heavily dependent on online monitoring data, and scenarios such as sludge suction from the influent pipe and equipment standard solution calibration can easily lead to distortion of monitoring data. The system cannot identify invalid data, which further aggravates the calculation deviation, often exceeding 15%, resulting in an imbalance between aeration supply and demand, energy waste, or substandard treatment effect; 3. Low level of equipment coordination and control: The operation and control of aerators and flow generators are independent of each other and lack a unified coordination and control logic. The operating status can only be adjusted by manual experience or simple timing control. It is impossible to dynamically optimize the operation strategy based on real-time water quality parameters, such as the fluctuation of key water quality parameters such as carbon-nitrogen ratio and equipment operating load. This not only further aggravates the problem of unstable treatment effect caused by water quality fluctuations, but also leads to high equipment energy consumption and shortened service life. 4. High difficulty and cost of renovation: For existing sewage treatment plants, if traditional renovation methods are adopted, such as adding civil engineering, not only will the project be large in scale and have a long construction period, but it will also affect the normal operation of the sewage treatment plant. The renovation cost is high, making it difficult to promote and apply on a large scale, and it cannot fundamentally adapt to the working conditions with large fluctuations in carbon-nitrogen ratio.

[0024] To overcome the shortcomings of existing AAO micro-aeration oxidation ditch processes, such as insufficient flexibility, low accuracy of aeration volume calculation, poor equipment coordination, and high difficulty in modification when treating water with large fluctuations in influent carbon-nitrogen ratio, this application provides an intelligent control system and method for AAO micro-aeration oxidation ditch that adapts to fluctuations in influent carbon-nitrogen ratio. Through innovative equipment installation structure, multimodal machine learning for aeration volume calculation, and equipment collaborative control, it achieves dynamic control of anoxic / aerobic zones and precise aeration volume supply, thereby improving the treatment effect under scenarios with large fluctuations in influent carbon-nitrogen ratio and reducing modification costs.

[0025] The intelligent control system for oxidation ditch that adapts to fluctuations in influent carbon-nitrogen ratio provided in this application includes: an integrated aeration and flow propulsion module and a collaborative control module.

[0026] The integrated aeration and propulsion module is installed inside the oxidation ditch, for example, inside an existing AAO micro-aeration oxidation ditch. The integrated aeration and propulsion module includes an aeration unit and a propulsion unit. The aeration unit is laid at the bottom of the oxidation ditch, and the propulsion unit is positioned above the aeration unit.

[0027] In one possible implementation, the aeration unit uses a micro-aerator, which is evenly spread across the bottom of the oxidation ditch. The micro-aerator includes multiple independent aeration branch pipes, each of which corresponds to a control node. The start / stop status and aeration intensity of each aeration branch pipe can be independently controlled.

[0028] In one possible implementation, the propulsion unit employs a submersible propulsion unit, which is installed directly above and coaxially with the micro-aerator. The submersible propulsion unit and the micro-aerator are fixed together by a bracket to prevent water flow interference. The rotational speed of the submersible propulsion unit is infinitely adjustable to drive water flow. Furthermore, the submersible propulsion unit, in conjunction with aeration branch pipe control, can support dynamic adjustment of the wastewater retention time in the area, adapting to fluctuations in the carbon-to-nitrogen ratio.

[0029] The collaborative control module is connected to the integrated aeration and propulsion module. The collaborative control module is used to regulate the operating parameters of the aeration unit and the propulsion unit to control the oxidation ditch to switch between anoxic and aerobic states.

[0030] In the oxidation ditch, the aeration unit is closed and the propulsion unit is open when the oxidation ditch is in anoxic condition; when the oxidation ditch is in aerobic condition, the aeration unit is open and the propulsion unit is closed.

[0031] The technical solution provided in this application embodiment achieves an innovative integrated installation structure for aeration and propulsion equipment, breaking through the technical bottleneck of separate installation of aerators and propulsion devices in existing systems. It adopts an integrated installation structure with a bottom layer of fully-covered aeration units and an upper layer of coaxial propulsion units, achieving an aeration uniformity improvement of over 30% at the bottom of the oxidation ditch. Furthermore, it uses a collaborative control module to regulate the operating parameters of the aeration and propulsion units, enabling the system to switch between anoxic and aerobic states. For example, when the aeration unit is controlled to be off and the propulsion unit is on, the system is in anoxic state; when the aeration unit is controlled to be on and the propulsion unit is controlled to be off, the system is in aerobic state. This adapts to the needs of different denitrification / nitrification reactions when the influent carbon-nitrogen ratio fluctuates, and the modification does not require reconstruction of the oxidation ditch tank, simplifying construction.

[0032] In addition, the technical solution provided in this application embodiment adopts a modular design for the intelligent control system of oxidation ditch. The aeration unit, the flow propulsion unit, and the collaborative control module can all be directly connected to the equipment interface of the existing AAO micro-aeration oxidation ditch, realizing modular transformation innovation adapted to old plant areas. The construction cycle can be shortened to less than 7 days, and the transformation cost can be reduced by more than 40%. This solves the problems of large engineering workload and high cost of the existing transformation scheme, and is suitable for the actual application scenario of sewage treatment plants in the south that have been built to cope with the aggravated fluctuation of influent carbon-nitrogen ratio.

[0033] In an exemplary embodiment, the collaborative control module is also used to regulate the residence time of the oxidation ditch in anoxic and aerobic states, thereby achieving automated and precise control of the residence time of the system in anoxic and aerobic states, and more accurately adapting to the needs of different denitrification / nitrification reactions when the carbon-nitrogen ratio of the influent fluctuates.

[0034] In an exemplary embodiment, the oxidation ditch intelligent control system also includes a multi-source data acquisition module.

[0035] The multi-source data acquisition module is used to collect multi-dimensional data from the entire wastewater treatment process, with a focus on strengthening the acquisition and preprocessing of carbon-nitrogen ratio fluctuation data, providing data support for aeration calculation and collaborative control.

[0036] The aforementioned multi-source data acquisition modules include, but are not limited to: a water quality parameter acquisition module, an equipment operation parameter acquisition module, and an operating condition image acquisition module. Specifically, the water quality parameter acquisition module is used to collect and integrate water quality parameter data, the equipment operation parameter acquisition module is used to collect equipment operation parameter data, and the operating condition image acquisition module is used to collect real-time operating condition image data. Correspondingly, the aforementioned multi-dimensional data includes integrated water quality parameter data, equipment operation parameter data, and real-time operating condition image data. The integrated water quality parameter data is used to characterize the change in the carbon-to-nitrogen ratio of the wastewater entering the oxidation ditch, the equipment operation parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch.

[0037] The water quality parameter acquisition module includes a first sensor installed at the inlet, outlet, and center of the oxidation ditch. This first sensor includes, but is not limited to, a chemical oxygen demand (COD) sensor, a total nitrogen sensor, an ammonia nitrogen sensor, a nitrate nitrogen sensor, a dissolved oxygen (DO) sensor, a pH sensor, and a sludge concentration sensor. These sensors can collect and integrate water quality parameter data in real time, including but not limited to the influent C / N ratio, C / N ratio fluctuation slope, dissolved oxygen concentration, and sludge concentration. The dissolved oxygen concentration includes the dissolved oxygen (DO) concentration in each area of ​​the oxidation ditch. The C / N ratio fluctuation slope can be used to determine the trend of abrupt changes in the C / N ratio. Optionally, the sampling period of the first sensor is 1-5 minutes to ensure rapid capture of abrupt changes in the C / N ratio.

[0038] The equipment operating parameter acquisition module includes second sensors installed on the aeration branch pipes, blowers, and submersible jet mixers. These second sensors include, but are not limited to, flow sensors, pressure sensors, speed sensors, and current sensors. Through these sensors, real-time equipment operating parameter data can be collected, including the flow rate of each aeration branch pipe, the air volume and / or pressure of the blower, and the corresponding data for the speed of the submersible jet mixer.

[0039] The operating condition image acquisition module includes an industrial camera installed above the oxidation tank. This camera is used to acquire real-time images of the aeration bubble distribution within the oxidation tank. The real-time operating condition image data includes multiple aeration bubble distribution images. These images are used for visual feature extraction in a subsequent multimodal fusion machine learning model to determine the operating condition of the aeration unit.

[0040] Correspondingly, the collaborative control module is connected to the multi-source data acquisition module, and the collaborative control module is also used to output control commands to the intelligent control system based on multi-dimensional data.

[0041] In an exemplary embodiment, the oxidation ditch intelligent control system also includes a multimodal intelligent computing module, which serves as the core of the system and incorporates a multimodal fusion machine learning model optimized for carbon-nitrogen ratio fluctuations. The multimodal intelligent computing module receives all data collected by the multi-source data acquisition module to accurately calculate aeration volume, determine the number of aeration branch pipes to be opened, and calculate blower airflow adjustment parameters. The multimodal fusion machine learning model processes multimodal data from multiple dimensions and outputs the required aeration volume and the defined virtual anoxic / aerobic zones. The required aeration volume and virtual anoxic / aerobic zones are adapted to the current carbon-nitrogen ratio of the wastewater entering the oxidation ditch. Furthermore, the multimodal fusion machine learning model also outputs blower airflow adjustment data.

[0042] In an exemplary embodiment, since some environmental parameters of the oxidation pond, such as water temperature, have a significant impact on the nitrification rate, the multi-source data acquisition module also includes an environmental parameter acquisition module for acquiring environmental parameter data. Correspondingly, the multi-dimensional data also includes the environmental parameter data acquired by the environmental parameter acquisition module. The environmental parameter data is used to correct the required aeration rate and eliminate the interference of environmental factors on the carbon-nitrogen ratio adaptation and control.

[0043] The aforementioned collaborative control module is also connected to the multimodal intelligent computing module. This module, based on the integrated water quality parameter data and the output of the multimodal intelligent computing module, outputs control commands corresponding to the current operating conditions of the oxidation ditch intelligent control system to control the operating parameters of the aeration unit and the propulsion unit. The collaborative control module provided in this embodiment is connected to the multi-source data acquisition module, the multimodal intelligent computing module, the integrated aeration and propulsion module, and the blower, thereby achieving three core control functions based on the real-time value and fluctuation trend of the influent carbon-nitrogen ratio: aeration-propulsion staggered operation control, aeration volume zoned precise control, and emergency backup control. Correspondingly, the control commands include aeration-propulsion staggered operation control commands, aeration volume zoned precise control commands, and emergency backup control commands.

[0044] In an exemplary embodiment, the collaborative control module is specifically used to determine the carbon-nitrogen ratio classification conditions corresponding to the integrated water quality parameter data and output aeration-flow staggered operation control commands according to the carbon-nitrogen ratio classification conditions.

[0045] The aeration-flow staggered operation control command includes: the timing parameters and dynamic correction coefficients corresponding to aeration duration, aeration stop duration, flow propulsion duration, and flow stop duration, the stepless speed regulation value corresponding to the flow propulsion unit, and the aeration / flow propulsion interlock signal.

[0046] In one possible implementation, the output of the aeration-flow staggered operation control command includes the following: timing parameters for aeration duration, aeration stop duration, flow propulsion duration, and flow stop duration, based on the carbon-to-nitrogen ratio (C / N ratio) tiered operating conditions, as well as the stepless speed regulation value of the flow propeller; simultaneously, an aeration / flow propulsion interlock signal is output to prevent the aeration unit and the flow propulsion unit from operating simultaneously; and when the C / N ratio changes abruptly, a timing dynamic correction coefficient is output to adjust the duration of each stage in real time. The abrupt change in C / N ratio is determined by a C / N ratio abrupt change judgment threshold, for example, a C / N ratio fluctuation slope ≥ ±0.5 / hour. There is a preset correspondence between the C / N ratio and the timing dynamic correction coefficient, which can be determined by those skilled in the art based on actual conditions or experiments.

[0047] The criteria for determining a sudden change in the carbon-nitrogen ratio are: the fluctuation of the current influent C / N ratio relative to the average influent C / N ratio over the past 2 hours reaches or exceeds ±40%, that is, it meets any of the following conditions: current C / N ≤ average C / N over the past 2 hours × 60% (sudden decrease); current C / N ≥ average C / N over the past 2 hours × 140% (sudden increase).

[0048] The time-series dynamic correction coefficient is determined according to the following rules: when the fluctuation range is between ±40% and ±60%, the correction coefficient is 0.5 (sudden decrease) or 1.5 (sudden increase); when the fluctuation range is greater than ±60%, the system alarms and manual intervention is required; this correction coefficient is used to multiply the aeration duration and push flow duration in the original time-series parameters, thereby achieving adaptive and rapid adjustment of the off-peak operation duration.

[0049] The corresponding control actions triggered by each carbon-nitrogen ratio classification condition include, but are not limited to: if the carbon-nitrogen ratio classification condition determined based on integrated water quality parameter data is a low carbon-nitrogen ratio condition, for example, C / N < 4.5, then a long plug flow and short aeration sequence is triggered to prolong the wastewater residence time in the anoxic zone, improve the carbon source utilization efficiency for denitrification, and aeration is only used to maintain the basic nitrification reaction requirements. Here, long plug flow and short aeration refers to a sequence where the plug flow duration is longer than the aeration duration.

[0050] If the carbon-to-nitrogen ratio (C / N) classification condition determined based on integrated water quality parameter data is a medium C / N ratio condition, such as 4.5 ≤ C / N ≤ 6, then the aeration / flow equalization sequence is triggered to balance the efficiency of denitrification and nitrification reactions, maintaining a dynamic balance between anoxic and aerobic states. The aeration / flow equalization sequence can be either equal in duration to the flow duration or maintained within a threshold range.

[0051] If the carbon-to-nitrogen ratio (C / N) classification condition determined based on integrated water quality parameter data is a high C / N ratio condition, such as C / N > 6, then a short plug flow and long aeration sequence is triggered to prolong the wastewater's residence time in the aerobic zone and enhance COD degradation and nitrification. Here, short plug flow and long aeration refer to a sequence where the plug flow duration is shorter than the aeration duration.

[0052] If the carbon-nitrogen ratio (C / N) grading condition determined based on integrated water quality parameter data is a sudden change in C / N ratio, for example, a C / N ratio fluctuation ≥ 40% of the average C / N ratio over the past 2 hours, then an emergency timing correction action is triggered. This action rapidly adjusts the aeration / flow propulsion duration ratio according to the sudden increase / decrease trend. This emergency timing correction action can be accomplished by sending a dynamic timing correction coefficient to modify the timing parameters of aeration duration, aeration stop duration, flow propulsion duration, and flow stop duration. The aeration unit and the flow propulsion unit can then operate according to the timing parameters corrected using the dynamic timing correction coefficient, thereby rapidly adjusting the aeration / flow propulsion duration ratio.

[0053] In an exemplary embodiment, the collaborative control module is further configured to output precise control commands for aeration volume zoning based on the required aeration volume and the virtual hypoxic / aerobic zones.

[0054] Among them, the precise control command for aeration volume zoning includes the precise allocation value of total aeration volume, the start / stop status value of a single aeration branch pipe, the aeration intensity adjustment value of a single aeration branch pipe, the blower air volume / pressure adjustment parameters, and the isolation signal of a faulty aeration branch pipe. The faulty aeration branch pipe is identified based on the aeration bubble distribution image.

[0055] In one possible implementation, the output of the aeration volume zone precise control command includes the following: It outputs the precise allocation value of total aeration volume, the start / stop status of individual aeration branch pipes, the aeration intensity adjustment value of individual aeration branch pipes, and the blower airflow / pressure adjustment parameters. Simultaneously, it outputs a faulty branch pipe isolation signal, such as an isolation signal for damaged / blocked aeration branch pipes identified by the visual data processing sub-model. The aeration intensity adjustment value for individual aeration branch pipes can be 0-100%.

[0056] The aeration volume zoned precise control commands include zoned aeration regulation commands, blower linkage regulation commands, faulty branch pipe emergency control commands, and carbon-nitrogen ratio gradient regulation commands. Each command triggers a different control action.

[0057] The control actions corresponding to the zoned aeration control command include: triggering the start / stop and intensity adjustment of the corresponding aeration branch pipes based on the virtual anoxic / aerobic zones divided by the multimodal intelligent calculation module; closing all branch pipes in the anoxic zone; and adjusting the aeration intensity of the branch pipes in the aerobic zone according to the calculated value, thereby achieving spatially differentiated control of aeration intensity within the oxidation ditch. The control actions corresponding to the blower linkage control command include: triggering the blower air volume based on the total aeration demand, matching the total air intake demand of the aeration branch pipes, avoiding aeration supply and demand imbalance, and reducing blower energy consumption.

[0058] The control actions corresponding to the emergency control command for a faulty branch pipe include: when a signal of damage to an aeration branch pipe (i.e., an aerator) is received, if the damage signal is less than the limit, the faulty branch pipe is immediately shut down, and the aeration demand of the branch pipe is redistributed to other normal branch pipes in the same area to ensure that the total aeration volume in the area meets the calculated requirements and avoids a decrease in treatment effect due to local aeration abnormalities; if the damage signal is greater than the limit, an alarm is triggered to indicate that the aeration branch pipe or aerator needs to be replaced. The control actions corresponding to the carbon-nitrogen ratio gradient control command include: for the gradient change of the influent carbon-nitrogen ratio along the oxidation ditch inlet to outlet, the aeration intensity is adjusted along the gradient, with the aeration branch pipes in the inlet area operating at high intensity and the outlet area operating at low intensity to achieve precise matching of the nitrification reaction.

[0059] In an exemplary embodiment, the collaborative control module is further configured to output emergency operating condition backup control commands to provide preset control parameters to the integrated aeration propulsion module in emergency situations such as sudden changes in carbon-nitrogen ratio, sensor failure, or multimodal intelligent computing module failure.

[0060] In one possible implementation, the output of the emergency standby control command includes the following: outputting the emergency aeration volume baseline value, emergency aeration / flow sequence, and uniform aeration command for the entire pool, providing preset control parameters for emergency scenarios such as extreme changes in carbon-nitrogen ratio, sensor failure, and multimodal intelligent computing module failure.

[0061] Different emergency situations trigger different control actions. Specifically, in the event of extreme changes in the carbon-to-nitrogen ratio, such as C / N < 4.5 or C / N > 8.0, a preset emergency sequence is triggered. Specifically, when C / N < 4.5, the flow propulsion time is maximized and the aeration intensity is reduced; if the plant is equipped with a carbon source dosing system, the carbon source dosing linkage is also activated. When C / N > 8.0, the aeration time is maximized and aerobic treatment is enhanced. In the event of sensor or computing module failure, basic aeration of the entire tank and uniform flow propulsion are triggered, operating according to the plant's historical best stable operating parameters to avoid process paralysis due to data interruption. Simultaneously, an emergency alarm is sent to the remote monitoring module to remind manual intervention.

[0062] In the event of an emergency where the effluent water quality exceeds the standard, such as when total nitrogen / COD / ammonia nitrogen is close to the discharge standard, the aeration / flow timing enhancement and correction action is triggered. If the excess item is total nitrogen, the flow time is extended; if the excess item is COD / ammonia nitrogen, the aeration time is extended, so as to quickly bring the effluent water quality back to the standard range.

[0063] In an exemplary embodiment, the oxidation ditch intelligent control system also includes a remote monitoring module, which uses Internet of Things (IoT) technology to upload all collected data, such as the highlighted carbon-nitrogen ratio fluctuation curve, equipment operating status, relevant calculation results of required aeration volume, and treatment effect data, to a cloud platform. It supports remote monitoring, parameter setting, and fault alarms on computers and mobile phones, and can also store historical carbon-nitrogen ratio fluctuation data to provide data support for model self-learning and process optimization, adapting to the unmanned operation requirements of wastewater treatment plants.

[0064] To precisely control the aforementioned intelligent control system for oxidation ditches, this application also provides an adaptive intelligent control method for oxidation ditches based on influent carbon-nitrogen ratio fluctuations. This method is applied to the intelligent control system for oxidation ditches as described in the above embodiments. Please refer to... Figure 1 , Figure 1 This is a flowchart of an intelligent control method for oxidation ditch that adapts to the fluctuation of influent carbon-nitrogen ratio provided in this application embodiment. The method can be applied to computer equipment, which refers to electronic equipment with data computing and processing capabilities. For example, the executing entity of each step can be an electronic equipment with a collaborative control module, an electronic equipment with the above-mentioned multimodal intelligent computing module, or an electronic equipment equipped with both a collaborative control module and a multimodal intelligent computing module. The method may include the following steps (110~130).

[0065] Step 110: Obtain and integrate water quality parameter data, equipment operating parameter data, and real-time operating condition image data.

[0066] Among them, the integrated water quality parameter data is used to characterize the change in the carbon-nitrogen ratio of the wastewater entering the oxidation ditch, the equipment operation parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch.

[0067] The above data can be read or received from the multi-source data acquisition module.

[0068] Step 120: Input the fused water quality parameter data, equipment operation parameter data, and real-time operating condition image data into the multimodal fusion machine learning model for processing, and output the required aeration volume and the divided virtual anoxic / aerobic zones.

[0069] Among them, the required aeration volume and the virtual anoxic / aerobic zone are matched with the carbon-nitrogen ratio of the wastewater currently entering the oxidation ditch.

[0070] The aforementioned multimodal fusion machine learning model can be set within the multimodal intelligent computing module. In the absence of a separate multimodal intelligent computing module, it can also be set within the collaborative control module.

[0071] In an exemplary embodiment, the multimodal fusion machine learning model includes a time-series data processing sub-model and a visual data processing sub-model. When step 120 is executed, the following steps are specifically performed: preprocessing and extracting features from the fused water quality parameter data and equipment operating parameter data to obtain standardized sequence feature data; preprocessing and extracting features from the real-time operating condition image data to obtain visual feature data; and inputting the sequence feature data into the time-series data processing sub-model for processing to obtain the required aeration volume, the target start / stop status data corresponding to each aeration branch, the target aeration intensity, and the virtual aeration deficit. Oxygen / aerobic zones; visual feature data is input into the visual data processing sub-model for detection, obtaining aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration zone location information; under the condition that the aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration zone location information indicate that the oxidation ditch intelligent control system is operating normally, the system outputs the required aeration volume, target start / stop status data corresponding to each aeration branch pipe, target aeration intensity, and virtual anoxic / aerobic zones; among which, the target open / closed status corresponding to each aeration branch pipe is associated with the virtual anoxic / aerobic zone.

[0072] The above process will be explained below with reference to a specific model structure. In an exemplary embodiment, the multimodal fusion machine learning model may include a data preprocessing module, a multimodal fusion model, and a model output module.

[0073] The data preprocessing module is used to preprocess the collected multi-source data, namely, to preprocess the above-mentioned integrated water quality parameter data, equipment operation parameter data, and environmental parameter data (if any). The preprocessing methods include noise reduction and normalization of the above data, and after removing abnormal data, the integrated water quality parameter data, equipment operation parameter data, and environmental parameter data are transformed into standardized sequence feature data, thereby completing the preprocessing and feature extraction of text data.

[0074] The data preprocessing module, on the other hand, is used to preprocess and extract features from real-time operating condition image data to obtain visual feature data. This module can convert the aeration bubble distribution image, which characterizes the system's operating conditions, to grayscale and extract features such as bubble density and foam color, thus obtaining visual feature data.

[0075] The multimodal fusion model includes an LSTM sub-model and a visual sub-model. The LSTM sub-model, serving as the time-series data processing sub-model, is used to process time-series data, such as the dynamic changes in COD, TN, DO, and the carbon-nitrogen ratio, capturing the temporal correlation and lag effects of carbon-nitrogen ratio fluctuations. Therefore, sequential feature data can be input into the LSTM sub-model for processing to obtain the required aeration volume, the target start / stop status data for each aeration branch, the target aeration intensity, and the virtual anoxic / aerobic zones.

[0076] The visual sub-model, serving as a visual data processing sub-model, is used to process image data of aeration bubble distribution under operating conditions and determine aeration uniformity. Therefore, visual feature data can be input into the visual data processing sub-model for detection to obtain aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration area location information.

[0077] The model output module outputs the results of the two sub-models using a dynamic weighted fusion algorithm. The weights of the two sub-models are adjusted based on the fluctuation range of the carbon-to-nitrogen ratio (C / N ratio). The greater the C / N ratio fluctuation, the higher the weight of the LSTM sub-model. This outputs the final total aeration volume requirement, the start / stop status of each aeration branch pipe, and the blower airflow adjustment value. The calculation cycle is synchronized with the data acquisition cycle and has a self-learning function. Every 7 days, it optimizes the model parameters using recent operating data under different C / N ratio fluctuation conditions, improving calculation accuracy and adaptability. The method of dividing the virtual anoxic / aerobic zone based on the C / N ratio can be achieved by controlling the start / stop status of each aeration branch pipe.

[0078] The model output module receives the output results of the LSTM sub-model and the visual sub-model respectively. Based on the dual-model functional positioning, it realizes independent output and linkage feedback of the results. The calculation cycle and data acquisition cycle are synchronized, for example, both are 1-5 minutes. The module has built-in self-learning optimization logic, which optimizes the model parameters every 7 days through the operating data under different carbon-nitrogen ratio fluctuation conditions, continuously improving the calculation accuracy and equipment condition judgment ability.

[0079] The model output module includes an LSTM sub-model result output sub-module, which receives the calculation results of the LSTM sub-model and directly outputs the total required aeration volume, the start / stop status and aeration intensity of each aeration branch pipe, and the blower air volume adjustment value adapted to the current influent carbon-nitrogen ratio. At the same time, based on the real-time value and fluctuation trend of the carbon-nitrogen ratio, it divides the oxidation ditch into virtual anoxic / aerobic areas, providing the core parameters for the collaborative control module to accurately regulate the aeration volume.

[0080] The model output module also includes a visual sub-model result output sub-module, which receives the analysis results of the visual sub-model and outputs the aeration uniformity level, the aeration pipe operating condition judgment results, and the location information of abnormal aeration areas, providing direct basis for equipment operation and maintenance. Simultaneously, it feeds back abnormal signals from the aeration equipment to the LSTM sub-model's computation link in real time. The aeration pipe operating condition judgment results include normal, damaged, or blocked.

[0081] Therefore, under the condition that the aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration area location information indicate that the oxidation ditch intelligent control system is operating normally, it can output the required aeration volume, the target start / stop status data corresponding to each aeration branch pipe, the target aeration intensity, and the virtual anoxic / aerobic area according to normal rules. Among them, the target open / closed status corresponding to each aeration branch pipe is associated with the virtual anoxic / aerobic area.

[0082] However, when the aeration uniformity level data, the aeration pipe operating condition judgment results, and the abnormal aeration area location information indicate that the intelligent control system for oxidation ditch is operating abnormally, abnormal linkage correction is required.

[0083] The abnormal linkage correction process includes: when the visual sub-model detects equipment abnormalities such as uneven aeration or damage / blockage of aeration branch pipes, it triggers the aeration volume allocation correction logic of the LSTM sub-model. The aeration volume calculation weight of the faulty aeration branch pipe indicated by the aeration pipe operating condition judgment result is removed, and a portion of the aeration volume corresponding to the faulty aeration branch pipe is redistributed. This ensures that the aeration volume demand is reasonably allocated to the normally operating aeration branch pipes, matching the actual aeration volume supply with the calculated demand. This updates the target start / stop status data, target aeration intensity, and virtual anoxic / aerobic zones for each aeration branch pipe.

[0084] The aforementioned multimodal fusion machine learning model can also perform self-learning optimization based on data acquired during operation. It can dynamically optimize the aeration calculation parameters of the LSTM sub-model and the operating condition judgment threshold of the visual sub-model based on the recent operational data of the entire wastewater treatment process, including carbon-nitrogen ratio fluctuation data, aeration volume calculation and actual supply data, water quality treatment effect data, and aeration equipment operating condition detection data. At the same time, it can adjust the triggering rules for abnormal linkage correction, thereby improving the model's adaptability to carbon-nitrogen ratio fluctuations and the accuracy of equipment operating condition judgment.

[0085] Step 130: Based on the integrated water quality parameter data and the output results of the multimodal intelligent calculation module, output control commands corresponding to the current operating conditions of the oxidation ditch intelligent control system to control the operating parameters of the aeration unit and the propulsion unit.

[0086] The aforementioned control commands may include the aeration-flow staggered operation control commands, aeration volume zoned precise control commands, and emergency standby control commands mentioned earlier, which will not be elaborated here.

[0087] In summary, the technical solution provided in this application achieves an innovative approach to calculating the target aeration rate by processing multimodal oxidation ditch operating data through a multimodal fusion machine learning model. It abandons the existing traditional mechanism model or single LSTM model aeration rate calculation method and proposes a multimodal fusion machine learning calculation method based on text time series and visual models. It integrates water quality parameters, such as the real-time value and fluctuation trend of carbon-nitrogen ratio, equipment operating parameters, and real-time operating condition images, etc., to solve the problem of low calculation accuracy caused by nonlinear fluctuations in water quality when the influent carbon-nitrogen ratio fluctuates greatly. The calculation error is reduced to less than 5%. At the same time, it has self-learning and adaptive capabilities and can dynamically optimize the model according to the carbon-nitrogen ratio fluctuation law of different sewage treatment plants, which is different from the existing single parameter or single algorithm calculation scheme.

[0088] Furthermore, to facilitate understanding by those skilled in the art, the above-mentioned intelligent control system and method are summarized here. When implementing the above-mentioned control method based on the intelligent control system, the specific process includes the following steps.

[0089] Step 1: System initialization and device docking and debugging Complete the initial parameters of the model, such as weights and thresholds; based on the wastewater treatment plant's process design and historical operating data, preset the basic timing parameters for aeration-plug flow staggered operation, such as aeration, aeration stop, plug flow, and plug stop durations; set three-level control thresholds for the carbon-nitrogen ratio, such as <4.5, 4.5-6, and >6, as well as a threshold for judging sudden changes in the carbon-nitrogen ratio, such as fluctuations ≥ 40% of the average C / N ratio over the past 2 hours; complete the hardware interface and communication protocol integration between the integrated aeration-plug flow module and the collaborative control module of this system and the existing equipment in the wastewater treatment plant, such as blowers, oxidation ditch tanks, and aeration branch pipes; conduct full system joint debugging and testing to ensure smooth data transmission between modules and precise equipment execution; and complete modular installation and commissioning.

[0090] A multimodal fusion machine learning model optimized for carbon-nitrogen ratio fluctuations is loaded into an embedded chip. Initial parameters of the model are set, such as weights and thresholds. Parameters for staggered operation are set, such as aeration, aeration stop, flow push, and stop-push time. CNOOC ratio graded control thresholds are set, such as <4.5, 4.5-6, and >6. The model is then connected to existing equipment in wastewater treatment plants, such as blowers and oxidation ditch tanks, and modular installation and commissioning are completed. Step 2: Real-time acquisition and feature extraction of multi-source data The multi-source data acquisition module continuously collects multi-dimensional data from the entire wastewater treatment process with a sampling cycle of 1-5 minutes: water quality parameters such as COD, total nitrogen, ammonia nitrogen, nitrate nitrogen, dissolved oxygen, pH, and sludge concentration are collected at the inlet, middle of the tank, and outlet of the oxidation ditch. The influent carbon-nitrogen ratio (C / N) and the slope of C / N ratio fluctuation are calculated in real time, where C is the COD concentration and N is the total nitrogen concentration, and the sudden increase and decrease trends of C / N ratio are quickly captured. Equipment operating parameters such as flow rate, pressure, speed, and current are collected from the aeration branch pipes, blowers, and flow promoters. Images of aeration bubble distribution are collected by an industrial camera above the tank, and water temperature environmental parameters are collected by a water temperature sensor. The collected raw data are initially screened to remove invalid data from scenarios such as sensor failure and equipment maintenance, and key features such as C / N ratio fluctuation trends and bubble distribution characteristics are initially extracted.

[0091] The system continuously collects water quality parameters, environmental parameters, equipment operating parameters, and operational image data from the inlet, the middle of the tank, and the outlet, with a sampling cycle of 1-5 minutes. It also calculates the influent C / N ratio and fluctuation slope to quickly detect sudden increases and decreases in the C / N ratio. Water quality parameters at the outlet include, but are not limited to, COD, TN, NH3-N, HNO3-, DO, pH, and MLSS. Environmental parameters include water temperature. Equipment operating parameters include aeration branch flow rate, blower air volume, and propeller speed.

[0092] Step 3: Calculation of multimodal fusion aeration volume and output of aeration parameters The multimodal intelligent computing module receives the multi-source data collected in step 2 and performs fine preprocessing: it performs noise reduction and normalization on water quality, equipment, and environmental parameters to convert them into standardized input data; it performs grayscale conversion and feature extraction on the working condition images to extract visual features such as bubble distribution density, bubble size uniformity, and blank areas of aeration on the pool surface.

[0093] Standardized water quality time-series parameters, equipment operating parameters, and environmental parameters are input into the LSTM sub-model. The calculation results are corrected by combining water temperature parameters. The model outputs the total aeration volume requirement adapted to the current carbon-nitrogen ratio conditions, the open / closed status and aeration intensity of each aeration branch pipe, and the blower airflow / pressure adjustment values. At the same time, based on the real-time carbon-nitrogen ratio value and fluctuation trend, virtual anoxic / aerobic areas are divided in the oxidation ditch. The visual features of the operating condition image are input into the visual sub-model to complete the determination of aeration uniformity level, identification of aeration pipe operating conditions, and location of abnormal aeration areas, and output abnormal signals of aeration equipment.

[0094] If the visual sub-model detects an abnormality in the aeration equipment, it immediately triggers the aeration volume allocation correction logic of the LSTM sub-model, removes the aeration volume calculation weight of the faulty branch pipe, and redistributes the aeration volume demand to the normal aeration branch pipe in the same area to ensure that the aeration volume calculation result matches the actual operating status of the equipment.

[0095] The multimodal intelligent computing module preprocesses the collected multi-source data, focusing on extracting the carbon-nitrogen ratio fluctuation trend characteristics; it inputs the standardized water quality parameters and equipment parameters into the LSTM sub-model, outputs the preliminary aeration volume calculation results, the opening status of each aeration branch pipe, and the blower air volume adjustment value, and at the same time combines the water temperature parameters to correct the calculation results to ensure accuracy; it inputs the operating condition image features into the visual sub-model to judge the aeration uniformity. The water temperature parameter in the environmental parameter data is not directly input into the LSTM sub-model. Instead, it is post-processed and corrected by an independent temperature correction function to obtain the final required aeration rate.

[0096] The temperature correction function is as follows: ;in: The final required aeration rate is the revised value (unit: m³ / h). The basic required aeration rate output by the LSTM sub-model; This is a temperature correction factor, and its specific value is determined based on empirical values ​​of the temperature sensitivity of the activated sludge microbial nitrification reaction.

[0097] Step 4: Coordinated Control Execution The collaborative control module receives aeration volume control parameters and aeration equipment operating status signals output by the multimodal intelligent computing module. It then combines these with the real-time value of the influent carbon-nitrogen ratio and the fluctuation type, such as steady state or sudden change, to execute multi-dimensional collaborative control actions.

[0098] Aeration-flow staggered operation control: The aerator and flow generator are controlled according to the "aeration-stop aeration-flow-stop flow" cyclic staggered operation mode, and an aeration / flow interlock signal is output to prevent the two from operating simultaneously; under low C / N conditions, such as C / N < 4.5, a long flow and short aeration sequence is triggered; under medium C / N conditions, such as 4.5 ≤ C / N ≤ 6, a balanced sequence of aeration and flow duration is triggered; under high C / N conditions, such as C / N > 6, a short flow and long aeration sequence is triggered; when the carbon-nitrogen ratio changes suddenly, a dynamic correction coefficient for the sequence is output, such as 0.5-1.5 times, to quickly adjust the duration ratio of each stage and adapt to the needs of sudden changes in water quality.

[0099] The criteria for determining a sudden change in the carbon-nitrogen ratio are: the fluctuation of the current influent C / N ratio relative to the average influent C / N ratio over the past 2 hours reaches or exceeds ±40%, that is, it meets any of the following conditions: current C / N ≤ average C / N over the past 2 hours × 60% (sudden decrease); current C / N ≥ average C / N over the past 2 hours × 140% (sudden increase).

[0100] The time-series dynamic correction coefficient is determined according to the following rules: when the fluctuation range is between ±40% and ±60%, the correction coefficient is 0.5 (sudden decrease) or 1.5 (sudden increase); when the fluctuation range is greater than ±60%, the system alarms and manual intervention is required; this correction coefficient is used to multiply the aeration duration and push flow duration in the original time-series parameters, thereby achieving adaptive and rapid adjustment of the off-peak operation duration.

[0101] Precise aeration volume control by zone: Based on the virtual anoxic / aerobic zone division, the start / stop and aeration intensity of each aeration branch are independently controlled. All branches in the anoxic zone are closed, and the aeration intensity of the branches in the aerobic zone is adjusted according to the calculated value. The blower air volume is adjusted in conjunction with the overall aeration volume demand. When a faulty branch isolation signal is received, the damaged / blocked aeration branch is immediately closed, and its aeration volume demand is redistributed to the normal branches in the same area. If the damage signal exceeds the threshold, an alarm is triggered.

[0102] Gradient aeration control: In response to the gradient change in the carbon-nitrogen ratio of the influent along the oxidation ditch from the inlet to the outlet, gradient aeration control is implemented along the aeration branch pipes. The branch pipes in the inlet area operate at high intensity, while the branch pipes in the outlet area operate at low intensity, so as to achieve precise matching between nitrification reaction and water quality gradient.

[0103] Based on the calculation results of step 3, combined with the real-time C / N ratio and fluctuation trend, the collaborative control module performs three major control actions: controlling the aerators and propellers to operate in a staggered manner according to a preset timing sequence, dynamically adjusting the staggered timing sequence when the C / N ratio fluctuates greatly to avoid mutual interference; and controlling the start and stop of each aeration branch pipe to achieve zoned aeration regulation.

[0104] Step 5: Model self-learning optimization Every 7 days is a self-learning cycle. The system automatically extracts recent historical operating data, such as multi-source data, aeration calculation results, and treatment effects, and optimizes the weights and thresholds of the multimodal fusion model to improve the accuracy of aeration calculation and the ability to adapt to fluctuations in carbon-nitrogen ratio, thus adapting to the long-term fluctuation patterns of influent water quality.

[0105] Step 6: Remote Monitoring and Emergency Handling of Faults The remote monitoring module uses IoT technology to upload multi-source data, carbon-nitrogen ratio fluctuation curves, real-time equipment operating status, aeration calculation results, and wastewater treatment effluent indicators to the cloud monitoring platform in real time. It supports remote viewing, remote parameter setting, and historical data tracing on computers and mobile phones. The system monitors the equipment operating status and water quality changes in real time. If equipment abnormalities such as aerator blockage / damage, sensor malfunction, or blower failure occur, or if there are extreme changes in the influent carbon-nitrogen ratio, such as C / N < 2.0 or > 8.0, or if the effluent TN / COD / ammonia nitrogen levels are close to the discharge standards, the cloud platform will immediately issue a fault or early warning alarm signal to remind staff to handle the situation on-site in a timely manner. At the same time, the system automatically activates the emergency backup control strategy, outputs the emergency aeration benchmark value, emergency aeration, and push flow sequence to maintain the continuous and stable operation of the wastewater treatment process and prevent effluent from exceeding standards.

[0106] The remote monitoring module uploads all data to the cloud platform, displaying the equipment's operating status, treatment effect, and carbon-nitrogen ratio fluctuation curve in real time. If equipment failure occurs, such as aerator blockage, sensor malfunction, or abnormal water quality, such as extreme changes in C / N ratio, an alarm signal will be issued immediately to remind staff to handle the situation promptly. At the same time, the system will automatically activate the backup control strategy to ensure continuous operation of the wastewater treatment system.

[0107] Therefore, the technical solution provided in this application has the following significant advantages compared to the prior art: 1. Addresses the pain point of large fluctuations in influent carbon-nitrogen ratio, significantly improving treatment efficiency: Through regional dynamic control linked to carbon-nitrogen ratio fluctuations and precise calculation of aeration volume, it can quickly adapt to sudden increases, decreases, and full-range fluctuations in the influent carbon-nitrogen ratio, such as fluctuations within the range of C / N=2.0-8.0, reducing the amount of carbon source added and increasing the total nitrogen removal rate by more than 25%, completely solving the problems of insufficient flexibility, unstable effluent, and large amount of carbon source added caused by carbon-nitrogen ratio fluctuations in existing processes; 2. Avoid sludge deposition and improve the stability of tank operation: Micro aerators are laid at the bottom of the oxidation ditch. With the water propulsion of the flow pusher, the bottom of the tank is aerated without dead corners, which avoids the problem of sludge deposition from the source and ensures the continuous and stable operation of the sewage treatment process. 3. High accuracy in aeration volume calculation and strong resistance to fluctuations: The multimodal fusion machine learning method integrates multi-source data and enhances the capture of carbon-nitrogen ratio fluctuation trends, reducing the calculation error to below 5%. Compared with existing traditional methods, the accuracy is greatly improved, and it has self-learning ability, which can adapt to the carbon-nitrogen ratio fluctuation patterns of different wastewater treatment plants without the need for frequent manual parameter adjustments. 4. Integrating visual recognition technology to achieve precise aeration control: The addition of a visual recognition module for real-time monitoring of aeration conditions incorporates abnormal equipment operation such as damaged aeration branch pipes and uneven aeration into the control considerations, and promptly triggers aeration volume distribution correction and emergency handling of faulty branch pipes, so that the aeration volume supply matches the actual operating status of the aeration equipment, further improving the accuracy and effectiveness of aeration control. 5. Low modification difficulty and strong adaptability: The modular design can be directly connected to the existing AAO micro-aeration oxidation ditch equipment without the need to reconstruct the tank body. The construction period is short and the modification cost is reduced by more than 40%. It is particularly suitable for existing sewage treatment plants in the south to cope with the current situation of increased fluctuations in the influent carbon-nitrogen ratio. It has broad prospects for promotion and application.

[0108] The following are embodiments of the apparatus of this application, which can be used to execute embodiments of the method of this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method of this application.

[0109] One embodiment of this application provides an intelligent control device for oxidation ditch that adapts to fluctuations in the influent carbon-nitrogen ratio. This device has the function of implementing the aforementioned intelligent control method for oxidation ditch that adapts to fluctuations in the influent carbon-nitrogen ratio. This function can be implemented in hardware or by hardware executing corresponding software. The device can be a computer device or can be installed within a computer device. The device may include: a multi-source data acquisition module for acquiring integrated water quality parameter data, equipment operating parameter data, and real-time operating condition image data; wherein, the integrated water quality parameter data is used to characterize the change in the carbon-nitrogen ratio of the wastewater entering the oxidation ditch, the equipment operating parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch; a data fusion processing module for inputting the integrated water quality parameter data, equipment operating parameter data, and real-time operating condition image data into a multimodal fusion machine learning model for processing, and outputting the required aeration volume and the divided virtual anoxic / aerobic zones, which are adapted to the current carbon-nitrogen ratio of the wastewater entering the oxidation ditch; and a control module for outputting control commands corresponding to the current operating condition of the oxidation ditch intelligent control system based on the integrated water quality parameter data and the output results of the multimodal intelligent calculation module, so as to control the operating parameters of the aeration unit and the propulsion unit.

[0110] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules when implementing its functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0111] One embodiment of this application provides a computer device. This computer device can be a computer device housing a collaborative control module, a computer device housing a multimodal intelligent computing module, or a computer device simultaneously equipped with a collaborative control module and a multimodal intelligent computing module. This computer device is used to implement the intelligent control method for adaptive influent carbon-nitrogen ratio fluctuations in oxidation ditches provided in the above embodiments. Specifically, the computer device typically includes a processor and a memory.

[0112] The processor may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor may be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field Programmable Gate Array), or PLA (Programmable Logic Array). The processor may also include a main processor and coprocessors. The main processor, also known as the CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content required to be displayed on the screen. In some embodiments, the processor may also include an AI (Artificial Intelligence) processor, which is used to handle computational operations related to machine learning.

[0113] The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory is used to store at least one instruction, at least one program, code set, or instruction set, configured to be executed by one or more processors to implement the above-described intelligent control method for oxidation ditch adaptive influent carbon-nitrogen ratio fluctuations.

[0114] In some embodiments, the computer device may also optionally include: a peripheral device interface and at least one peripheral device. The processor, memory, and peripheral device interface can be connected via a bus or signal lines. Each peripheral device can be connected to the peripheral device interface via a bus, signal lines, or a circuit board.

[0115] Computer equipment can receive user input to execute the steps in the above method or the operations within those steps.

[0116] Those skilled in the art will understand that the above structure does not constitute a limitation on the computer device, and may include more or fewer components than illustrated, or combine certain components, or employ different component arrangements.

[0117] In an exemplary embodiment, a computer-readable storage medium is also provided, wherein at least one instruction, at least one program, code set, or instruction set is stored therein, wherein the at least one instruction, the at least one program, the code set, or the instruction set, when executed by a processor, implements the above-described intelligent control method for oxidation ditch that adapts to the fluctuation of influent carbon-nitrogen ratio.

[0118] Optionally, the computer-readable storage medium may include: ROM (Read-Only Memory), RAM (Random Access Memory), SSD (Solid State Drives), or optical disc, etc. The random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0119] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned intelligent control method for adaptive influent carbon-nitrogen ratio fluctuations in oxidation ditches.

[0120] It should be understood that "multiple" as used herein refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. Furthermore, the step numbers described herein are merely illustrative of one possible execution order. In some other embodiments, the steps may not be executed in numerical order, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the illustration. This application does not limit this.

[0121] In the description of this application, it should be noted that, in the embodiments of this application, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature.

[0122] In the description of the embodiments of this application, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, "linking" can be a detachable connection or a non-detachable connection; it can be a direct connection or an indirect connection through an intermediate medium. "Fixed connection" refers to a connection where the relative positional relationship remains unchanged after the connection.

[0123] The directional terms used in the embodiments of this application, such as "inner" and "outer," are merely for reference to the directions in the accompanying drawings. Therefore, the directional terms used are for better and clearer explanation and understanding of the embodiments of this application, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application. Furthermore, unless otherwise stated in this application, "multiple" as used in this application refers to two or more.

[0124] In the description of embodiments of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0125] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. An intelligent control system for oxidation ditch that adapts to fluctuations in influent carbon-nitrogen ratio, characterized in that, The intelligent control system for oxidation ditch includes an integrated aeration and propulsion module and a collaborative control module. The integrated aeration and propulsion module is installed inside the oxidation ditch and includes an aeration unit and a propulsion unit. The aeration unit is laid at the bottom of the oxidation ditch and the propulsion unit is installed on top of the aeration unit. The collaborative control module is connected to the integrated aeration and propulsion module. The collaborative control module is used to regulate the operating parameters of the aeration unit and the propulsion unit to control the oxidation ditch to switch between anoxic and aerobic states. In the anoxic state, the aeration unit is turned off and the propulsion unit is turned on. In the aerobic state, the aeration unit is turned on and the propulsion unit is turned off.

2. The intelligent control system for oxidation ditches according to claim 1, characterized in that, The collaborative control module is also used to regulate the residence time of the oxidation ditch in the anoxic state and the aerobic state.

3. The intelligent control system for oxidation ditches according to claim 1, characterized in that, The aeration unit employs a micro-aerator, which is evenly distributed across the bottom of the oxidation ditch. Each micro-aerator includes multiple independent aeration branch pipes, with the start / stop status and aeration intensity of each branch pipe independently controlled. The flow propulsion unit employs a submersible flow propulsion device, which is installed directly above and coaxially with the micro-aerator. The submersible flow propulsion device is fixed to the micro-aerator via a bracket, and its rotation speed is infinitely adjustable and controlled in conjunction with the aeration branch pipes.

4. The intelligent control system for oxidation ditches according to claim 3, characterized in that, The intelligent control system for the oxidation ditch further includes: a multi-source data acquisition module for collecting multi-dimensional data of the entire wastewater treatment process, including integrated water quality parameter data, equipment operating parameter data, and real-time operating condition image data; wherein, the integrated water quality parameter data is used to characterize the change in the carbon-nitrogen ratio of the wastewater entering the oxidation ditch, the equipment operating parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch; the collaborative control module is also used to output control commands to the intelligent control system based on the multi-dimensional data; wherein, the multi-source data acquisition module includes: a water quality parameter acquisition module, including a first sensor installed at the inlet, outlet, and middle of the oxidation ditch, the first sensor including a chemical oxygen demand sensor, a total nitrogen sensor, and... The system includes an ammonia nitrogen sensor, a nitrate nitrogen sensor, a dissolved oxygen sensor, a pH sensor, and a sludge concentration sensor. The integrated water quality parameter data includes collected data corresponding to the influent carbon-to-nitrogen ratio, carbon-to-nitrogen ratio fluctuation slope, dissolved oxygen concentration, and sludge concentration. The equipment operation parameter acquisition module includes second sensors installed on the aeration branch pipes, the blower, and the submersible jet mixer. These second sensors include a flow sensor, a pressure sensor, a speed sensor, and a current sensor. The equipment operation parameter data includes collected data corresponding to the flow rate of each aeration branch pipe, the air volume and / or pressure of the blower, and the speed of the submersible jet mixer. The operating condition image acquisition module includes an industrial camera installed above the oxidation tank. This industrial camera is used to acquire real-time images of the aeration bubble distribution in the oxidation tank. The real-time operating condition image data includes multiple images of the aeration bubble distribution.

5. The intelligent control system for oxidation ditches according to claim 4, characterized in that, The intelligent control system further includes: a multimodal intelligent computing module, which has a built-in multimodal fusion machine learning model optimized for carbon-nitrogen ratio fluctuations. The multimodal fusion machine learning model is used to process multimodal data in the multidimensional data and output the required aeration rate and the divided virtual anoxic / aerobic zones. The required aeration rate and the virtual anoxic / aerobic zones are adapted to the carbon-nitrogen ratio conditions of the wastewater currently entering the oxidation ditch. The multi-source data acquisition module also includes an environmental parameter acquisition module. The multidimensional data also includes environmental parameter data acquired by the environmental parameter acquisition module. The environmental parameter data is used to correct the required aeration rate.

6. The intelligent control system for oxidation ditches according to claim 5, characterized in that, The collaborative control module is also connected to the multimodal intelligent computing module; the collaborative control module is also used to output control commands corresponding to the current operating conditions of the oxidation ditch intelligent control system based on the fused water quality parameter data and the output results of the multimodal intelligent computing module, so as to control the operating parameters of the aeration unit and the propulsion unit.

7. The intelligent control system for oxidation ditches according to claim 6, characterized in that, The collaborative control module is specifically used to determine the carbon-nitrogen ratio classification conditions corresponding to the integrated water quality parameter data and output aeration-flow staggered operation control commands according to the carbon-nitrogen ratio classification conditions; wherein, the aeration-flow staggered operation control commands include: the timing parameters and dynamic correction coefficients corresponding to aeration duration, aeration stop duration, flow propulsion duration, and flow stop duration, the stepless speed regulation value corresponding to the flow propulsion unit, and the aeration / flow propulsion interlock signal; the collaborative control module is also specifically used to output aeration volume zoning precise control commands according to the required aeration volume and the virtual anoxic / aerobic zone; In this system, the precise control command for aeration volume zoning includes a precise allocation value for total aeration volume, a start / stop status value for a single aeration branch pipe, an aeration intensity adjustment value for a single aeration branch pipe, blower airflow / pressure adjustment parameters, and an isolation signal for a faulty aeration branch pipe, wherein the faulty aeration branch pipe is identified based on the aeration bubble distribution image; the collaborative control module is also specifically used to output emergency working condition backup control commands to provide preset control parameters to the integrated aeration propulsion module in emergency situations such as sudden changes in carbon-nitrogen ratio, sensor failure, or failure of the multimodal intelligent computing module.

8. A method for intelligent control of oxidation ditch based on adaptive influent carbon-nitrogen ratio fluctuations, characterized in that, The method is applied to the intelligent control system for oxidation ditch as described in any one of claims 1 to 7. The method includes: acquiring integrated water quality parameter data, equipment operating parameter data, and real-time operating condition image data; wherein the integrated water quality parameter data is used to characterize the change in the carbon-nitrogen ratio of the wastewater entering the oxidation ditch, the equipment operating parameter data is used to characterize the operation of the integrated aeration and propulsion module, and the real-time operating condition image data is used to characterize the distribution of aeration bubbles in the oxidation ditch; inputting the integrated water quality parameter data, the equipment operating parameter data, and the real-time operating condition image data into a multimodal fusion machine learning model for processing, outputting the required aeration volume and the divided virtual anoxic / aerobic zones, wherein the required aeration volume and the virtual anoxic / aerobic zones are adapted to the current carbon-nitrogen ratio of the wastewater entering the oxidation ditch; and outputting control commands corresponding to the current operating conditions of the intelligent control system for oxidation ditch based on the integrated water quality parameter data and the output results of the multimodal intelligent calculation module, so as to control the operating parameters of the aeration unit and the propulsion unit.

9. The intelligent control method for oxidation ditch according to claim 8, characterized in that, The multimodal fusion machine learning model includes a time-series data processing sub-model and a visual data processing sub-model. The process of inputting the fused water quality parameter data, the equipment operating parameter data, and the real-time operating condition image data into the multimodal fusion machine learning model for processing, and outputting the required aeration volume and the defined virtual anoxic / aerobic zones, includes: preprocessing and extracting features from the fused water quality parameter data and the equipment operating parameter data to obtain standardized sequence feature data; preprocessing and extracting features from the real-time operating condition image data to obtain visual feature data; and inputting the sequence feature data into the time-series data processing sub-model for processing to obtain the required aeration volume and the corresponding values ​​for each aeration branch pipe. The system obtains target start / stop status data, target aeration intensity, and the virtual anoxic / aerobic region; inputs the visual feature data into the visual data processing sub-model for detection to obtain aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration region location information; when the aeration uniformity level data, aeration pipe operating condition judgment results, and abnormal aeration region location information indicate that the oxidation ditch intelligent control system is operating normally, it outputs the required aeration volume, target start / stop status data corresponding to each aeration branch pipe, target aeration intensity, and the virtual anoxic / aerobic region; wherein, the target open / closed status corresponding to each aeration branch pipe is associated with the virtual anoxic / aerobic region.

10. The intelligent control method for oxidation ditch according to claim 9, characterized in that, The method further includes: when the aeration uniformity level data, the aeration pipe operating condition judgment result, and the abnormal aeration area location information indicate that the operating condition of the oxidation ditch intelligent control system is abnormal, removing the aeration volume calculation weight of the faulty aeration branch indicated by the aeration pipe operating condition judgment result, redistributing part of the aeration volume corresponding to the faulty aeration branch, and updating the target start / stop status data, the target aeration intensity, and the virtual anoxic / aerobic area corresponding to each aeration branch.

Images