Intelligent carbon source feeding control system based on biological denitrification process

Abstract

The invention discloses an intelligent carbon source feeding control system based on a biological denitrification process. The intelligent carbon source feeding control system comprises an upper computer, a server, a PLC control cabinet, a dosing pump, a dosing pipeline, a process instrument and an algorithm engine. Wherein the algorithm engine is deployed on a server and comprises a front control module and / or a rear control module and a database, and the front control module and / or the rear control module are / is used for predicting the carbon source adding amount according to water inlet data, conducting iterative correction on the predicted carbon source adding amount according to water outlet indexes and process parameters and outputting the corrected carbon source adding amount; the PLC control cabinet is connected with the server, the dosing pump and the process instrument and used for uploading monitoring data of the process instrument to the algorithm engine and issuing the carbon source dosage output by the algorithm engine to the dosing pump. According to the system, carbon source feeding can be more refined and intelligent, and the operation stability of biological denitrification is ensured.

Description

Technical Field

[0001] This invention relates to the field of wastewater treatment technology, specifically to an intelligent control system for carbon source dosing based on biological denitrification technology. Background Technology

[0002] Carbon source cost is a major component of wastewater treatment plant reagent costs. In biological denitrification processes, additional carbon sources are often required to enhance the process and ensure that the total nitrogen in the effluent meets discharge standards. Currently, carbon source dosing control systems on the market are mainly divided into two categories: one is an automatic control method based on a secondary development of the International Water Association model, and the other is a method that calculates the carbon source based on the theoretical denitrification carbon-to-nitrogen ratio.

[0003] Chinese patent application CN2017104160681 discloses a method for optimizing and controlling wastewater treatment by adding a carbon source. The method includes: incorporating an added kinetic expression into simulation software containing an ASM2d model, and constructing an ASM2e model based on the ASM2d model; constructing a wastewater treatment process using process component units from the ASM2e model library; converting the influent water quality concentration into component concentrations in the ASM2e model, and performing a preliminary steady-state simulation of the wastewater treatment process; calibrating the kinetic and stoichiometric parameters of the ASM2e model based on the results of the preliminary steady-state simulation and sensitivity analysis to achieve steady-state simulation; and using the steady-state simulation results as the initial input values ​​for dynamic simulation to perform dynamic simulation analysis of the dynamic influent water quality. This method is applicable to the simulation and optimization of wastewater treatment production lines with biological denitrification.

[0004] This secondary development method of the model has three main problems: 1. The water coordination model in the secondary development still considers the operating conditions when the microorganisms are running normally, which is difficult to deal with abnormal operating conditions; 2. When the total nitrogen in the effluent exceeds or is close to exceeding the standard due to insufficient dosing, the treatment plan is unclear. Because the water coordination model has no early warning response mechanism, in order to prevent the secondary development from conflicting with the original system, in such cases, it is always switched to manual treatment, which poses a significant production risk; 3. Since the base model is a mathematical algorithm, this type of control does not have a learning function and is difficult to meet the intelligent requirements.

[0005] Chinese patent application CN2019114201526 discloses a method for online calculation of carbon source dosage based on the influent carbon-nitrogen ratio, including: calculating the influent carbon-nitrogen ratio based on the influent COD concentration and total nitrogen concentration, then determining whether the carbon source in the influent is sufficient, insufficient or excessive based on the influent carbon-nitrogen ratio, calculating the carbon source dosage correction amount using different formulas for different carbon source levels, and then calculating the required carbon source dosage amount at the current moment based on the carbon source dosage correction amount.

[0006] The carbon source dosage calculated using this method can respond to fluctuations in the influent carbon-to-nitrogen ratio, providing an appropriate dosage and avoiding substandard water quality due to insufficient dosage or wasteful and secondary pollution from excessive dosage. Based on the mass ratio of nitrogen to oxygen atoms required for denitrification, the denitrification BOD / TN = 2.86 / 1 is derived. This ratio is used to infer whether the influent carbon source is sufficient. If the influent carbon source is insufficient, the carbon source deficit is calculated, and the dosage is determined accordingly.

[0007] Although this method is simple in principle and has high data reliability, it has poor adaptability. Different projects require specific settings for the K-value (saturation dynamic correction coefficient), which needs frequent adjustments based on operational data from the bioreactor. Even within the same project, the bioreactor's operating status varies at different times, causing the K-value to change. This necessitates manual input of the updated K-value from the backend. Because of this reliance on manual input, such systems have significant inherent risks. If operators fail to update the K-value in a timely manner, there is a risk that the required dosing concentration may be insufficient, leading to an increase in total nitrogen levels in the effluent. Summary of the Invention

[0008] To make carbon source dosing more precise and intelligent, this solution proposes an intelligent control system for carbon source dosing based on biological denitrification processes. This system can adjust carbon source dosing in a timely manner according to the influent water quality and the state of sludge in the biological tank, ensuring that precise and timely process adjustments are made when water quality changes occur. This system is adaptable to sudden abnormalities in influent water quality or abnormal operation of the biological tank in various biological denitrification processes, thereby improving the operational stability of the biological reactor.

[0009] According to one aspect of the present invention, a smart control system for carbon source dosing based on a biological denitrification process is provided, comprising a host computer, a server, a PLC control cabinet, a dosing pump, dosing pipelines, process instruments, and a wastewater treatment system based on the biological denitrification process. The host computer and server have a built-in algorithm engine, which includes a front control module and / or a back control module and a database. The front control module and / or back control module are used to predict the carbon source dosage based on the influent data, and iteratively correct the predicted carbon source dosage based on the effluent indicators and process parameters, and output the corrected carbon source dosage. The PLC control cabinet is connected to the host computer, server, dosing pump and process instrument, respectively, and is used to upload the monitoring data of the process instrument to the algorithm engine, and send the carbon source dosage output by the algorithm engine to the dosing pump, so that the dosing pump can add carbon source to the sewage treatment system through the dosing pipeline.

[0010] Optionally, in the intelligent carbon source dosing control system based on biological denitrification process provided by the present invention, when the biological denitrification process is AAO process, oxidation ditch process, SBR and derivative process, the algorithm engine only includes a front control module; when the biological denitrification process is Bartonpo process or multi-stage AO process, the algorithm engine includes a front control module and one or more post-control modules, and the front control module and one or more post-control modules share a database; when the biological denitrification process only includes a denitrification filter, the algorithm engine only includes a post-control module; the number of dosing pumps and their dosing locations in the wastewater treatment system, the type of process instruments and their installation locations in the wastewater treatment system are determined according to the control method of the algorithm engine.

[0011] Optionally, in the intelligent control system for carbon source addition based on biological denitrification process provided by the present invention, the pre-control module is used to predict the amount of carbon source added to the primary denitrification system based on the influent data, calculate the carbon source addition time based on the influent flow rate, internal reflux ratio and external reflux ratio, and correct the amount of carbon source added to the primary denitrification system based on the total nitrogen index of the effluent from the pre-anoxic tank and the process instrument data. The post-control module is used to predict the carbon source dosage of the secondary or multiple denitrification system based on the difference between the nitrile nitrogen data and the nitrile nitrogen index of the effluent from the pre-aerobic tank, and to correct the carbon source dosage of the secondary or multiple denitrification system based on the difference between the nitrile nitrogen data and the nitrile nitrogen index of the effluent from the post-aerobic tank and the process parameters.

[0012] Optionally, in the intelligent carbon source dosing control system based on biological denitrification process provided by the present invention, the host computer and server are reserved with manual input ports for receiving manually input data such as influent BOD5, internal reflux ratio, sludge organic content ratio, and carbon source COD equivalent; the front control module is used to collect influent data, effluent index data, process instrument data, and objective flow data, including influent COD. Cr The influent total nitrogen index, influent pH, and influent BOD5; process instrument data including water temperature, anoxic influent ORP, MLSS in the biological tank, DO in the aerobic tank's final compartment, ammonia nitrogen in the biological tank effluent, and COD in the biological tank effluent. Cr The data includes the organic content ratio of sludge, the COD equivalent of carbon source; the effluent index data includes the total nitrogen index of the effluent from the pre-anoxic tank and the nitrate nitrogen index of the effluent from the pre-aerobic tank; the objective flow rate data includes the influent flow rate, the external recirculation ratio, and the internal recirculation ratio. Predict the carbon source dosage based on the influent data, adjust the required internal reflux ratio based on the total nitrogen index of the influent or prompt for manual adjustment of the internal reflux ratio, and calculate the carbon source dosage time based on the influent volume, external reflux ratio, and internal reflux ratio. After the carbon source addition time is reached, the total nitrogen index of the effluent from the pre-anoxic tank is read, and a correction is generated based on the difference between the total nitrogen index of the effluent from the pre-anoxic tank and the target value. After the first correction, process instrument data is read to assess the sludge condition and operating status of the biological tank, and a second correction is generated. The dosage of chemicals for the primary denitrification system is output based on the predicted carbon source dosage, primary correction, and secondary correction. The total nitrogen data of the pre-anoxic tank effluent updated after dosing according to the dosage will be fed back to the database, forming a closed loop with the previous set of influent data and corrections.

[0013] Optionally, in the intelligent carbon source addition control system based on biological denitrification process provided by the present invention, the pre-control module is used to verify the total nitrogen data of the effluent from the pre-anoxic tank through the nitrate nitrogen data of the effluent from the pre-aerobic tank during the dosing correction stage; generate a suggested internal recirculation ratio value based on the total nitrogen index of the influent and the nitrate nitrogen data of the effluent from the pre-aerobic tank; and compare the suggested internal recirculation ratio value with the actual internal recirculation ratio so that it can be manually adjusted in real time. The front control module is also used to read the latest ammonia nitrogen data of the effluent from the aerobic tank before each dosing. When the current ammonia nitrogen data of the effluent from the aerobic tank is lower than the warning value, the system operates normally. When the current ammonia nitrogen data of the effluent from the aerobic tank is higher than the warning value, the system enters a warning state and issues an alarm message to prompt that it needs to switch to manual control mode.

[0014] Optionally, in the intelligent control system for carbon source addition based on biological denitrification process provided by the present invention, the front control module is used to analyze whether the carbon source in the influent is sufficient based on the carbon-nitrogen ratio of the influent: if the dosage is equal to the minimum flow rate of the dosing pump, the system is fed back to confirm that the carbon source in the influent is sufficient; if the dosage is greater than the minimum flow rate of the dosing pump, the system is fed back to confirm that the carbon source in the influent is insufficient. If the influent carbon source is sufficient, the carbon source dosage predicted in the previous group will be used, and iterative corrections will be made to output the corrected dosage. If the influent carbon source is insufficient, calculate the amount of carbon source to be added, perform iterative correction, and output the corrected dosage. If insufficient carbon source is detected in the influent for the first time, carbon source dosing is initiated, and the system enters normal operation after a preset time. When the dosing pump controlled by the system output remains at the lowest frequency for a preset time, the system asks the operator whether to stop the dosing, and enters the stop state after obtaining consent.

[0015] Optionally, in the intelligent carbon source dosing control system based on biological denitrification technology provided by the present invention, the front control module is also used to monitor abnormal influent water quality, abnormal operation of the biological reactor, abnormal total nitrogen or nitrate nitrogen levels in the effluent, and instrument malfunction status, and to execute corresponding control measures according to the abnormal status: When the difference between the influent index and the previous data exceeds the preset safety threshold, the influent water quality abnormal condition is entered, the iterative correction is reduced, and the total nitrogen data of the effluent from the anoxic tank is always lower than the target value. When an increase in ammonia nitrogen data is detected in the effluent of the aerobic tank, an alarm prompts whether to switch to manual mode. If not switched to manual mode, the alarm prompts to update the total nitrogen target value and improve the iterative correction. When the output carbon source dosage is converted to the influent COD Cr When COD Cr When / TN>20, the denitrification abnormal condition is entered, and the alarm is switched to manual mode. When the slope of the trend of the process instrument data in the biological reactor is greater than the preset value, the biological reactor enters the impact condition, and the alarm switches to manual mode. When the internal recirculation ratio in the biological tank does not meet the system's required internal recirculation ratio, an abnormal internal recirculation condition is triggered. An alarm will be activated, and manual adjustment of the internal recirculation flow rate will be performed in a timely manner. The updated internal recirculation ratio will then be input into the system. When the instrument deviation of total nitrogen in anoxic effluent or nitrate nitrogen in aerobic effluent reaches 10% or more, or when the nitrate nitrogen in aerobic effluent is 5% or more higher than the total nitrogen in anoxic effluent, or when the instrument has not been maintained for an extended period of time or when the instrument standard solution verification fails, the instrument enters an abnormal operating condition and an alarm is triggered to manually switch to maintenance mode. If the data remains abnormal after manual adjustment and maintenance of the instruments, the bioreactor will enter an abnormal operating condition, triggering an alarm for timely manual adjustment of the bioreactor process.

[0016] Optionally, in the intelligent control system for carbon source addition based on biological denitrification process provided by the present invention, when the post-control module is applied alone to the wastewater treatment system, the post-control module is used to predict the amount of carbon source added to the post-anoxic tank based on the difference between the influent nitrile nitrogen data of the post-anoxic tank and the target value of nitrile nitrogen in the effluent of the post-anoxic tank, and generate a correction based on the difference between the effluent nitrile nitrogen data of the post-anoxic tank and the target value of nitrile nitrogen in the effluent of the post-anoxic tank. When a pre-control module exists in the system, the second-round correction follows the second-round correction of the pre-control module. When a pre-control module does not exist in the system, the correction is based on the temperature value and the total effluent COD. Cr Value generation and two-round correction; The dosage of chemicals for secondary or multiple denitrification systems is output based on the predicted carbon source addition amount in the post-anoxic tank, the first round correction, and the second round correction. The updated nitrate and nitrogen data of the post-anoxic tank effluent after dosing according to the output dosage will be fed back to the database, forming a closed loop with the previous set of nitrate and nitrogen data of the post-anoxic tank influent and iterative corrections.

[0017] Optionally, in the intelligent control system for carbon source addition based on biological denitrification process provided by the present invention, the post-control module is also used to start carbon source addition when the nitrile nitrogen data of the effluent from the aerobic tank is detected to be higher than the target value of nitrile nitrogen in the effluent from the aerobic tank for the first time; when the dosing pump controlled by the system output is at the lowest frequency for a preset time, the system asks the technician whether to stop the dosing, and after obtaining consent, the system enters the stop state. When the system detects that the nitrate nitrogen data in the effluent of the aerobic tank exceeds the design limit of the denitrification filter, it enters an overload condition. Based on the actual situation, the weight of the predicted carbon source addition in the anoxic tank is increased, the iterative correction is reduced, and the output dosage is ensured to be sufficient, so that the nitrate nitrogen data in the effluent of the aerobic tank is lower than the target value of nitrate nitrogen in the effluent of the aerobic tank. After adding medication, the COD is calculated comprehensively. Cr If the ratio of (nitrile nitrogen data in the effluent of the aerobic tank - target nitrile nitrogen value in the effluent of the aerobic tank) to ≥20, the water quality or bacteria species are abnormal, and the denitrification filter enters an abnormal operating condition. The alarm switches to manual mode. When the instrument has not been maintained for an extended period of time, or when the instrument fails the standard fluid verification, it enters an abnormal operating condition and triggers an alarm to manually switch to maintenance mode.

[0018] Optionally, in the intelligent carbon source dosing control system based on biological denitrification process provided by the present invention, the pre-control module is used to compare the total nitrogen data of the pre-anoxic effluent with the target value at preset intervals, and record the comparison results into the database for learning. If the total nitrogen data of the pre-anoxic effluent is greater than the target value, the dosage in the previous round is too low, and subsequent iterations will reduce and correct it; if the total nitrogen data of the pre-anoxic effluent is less than the target value, the dosage in the previous round is too high, and subsequent iterations will increase and correct it. The post-control module is used to compare the effluent nitrate nitrogen data of this unit with the target value at preset intervals, and record the comparison results into the database for learning. If the effluent nitrate nitrogen data of this unit is greater than the target value, the dosage in the previous round was too low, and subsequent iterations will reduce it for correction; if the effluent nitrate nitrogen data of this unit is less than the target value, the dosage in the previous round was too high, and subsequent iterations will increase it for correction. The front control module and the back control module are used to find the closest data set from historical data after each database update.

[0019] The intelligent carbon source dosing control system based on biological denitrification technology provided in this solution can achieve at least the following technical effects: (1) The carbon source addition of the primary denitrification system and the secondary / multi-stage denitrification system in various biological denitrification processes can be controlled by the independent front and rear control modules respectively. Appropriate modules or module combinations can be selected according to the specific process, which can be adapted to the production of different wastewater treatment plants and improve the universality of the algorithm engine.

[0020] (2) Using the total nitrogen index of the anoxic tank as the internal control value of the front control module, controlling the total nitrogen index of the anoxic effluent can ensure that the effluent meets the standards, and can more quickly understand the denitrification effect of the biological tank and the effluent data, so as to take measures in advance.

[0021] (3) By reserving a data input port, the required data can be manually input when it is difficult to monitor data such as internal recirculation ratio and five-day oxygen demand online. In addition, for sewage treatment plants that cannot control internal recirculation online in the traditional way, the interface for manually inputting internal recirculation flow can be retained, which can reduce the cost of transformation.

[0022] (4) Add ammonia nitrogen and nitrate nitrogen monitoring equipment to the effluent of the aerobic tank. When the ammonia nitrogen level rises, the system will issue an alarm and the technicians will take timely measures to maintain the normal operation of the system and ensure that the ammonia nitrogen level of the biological tank effluent meets the discharge standards.

[0023] (5) By setting start-up and shutdown logic and safety boundary logic according to the common characteristics of different sewage treatment plants, the total nitrogen index of sewage treatment plants can be guaranteed to meet the standards and the ability to respond to abnormal situations in an emergency to the greatest extent.

[0024] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0025] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A process flow diagram of the Patonp process according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of a multi-stage AO process according to an embodiment of the present invention is shown; Figure 3 A schematic diagram of an intelligent control system for carbon source addition based on a biological denitrification process according to an embodiment of the present invention is shown. Figure 4 A schematic diagram of the operation of the front control module according to an embodiment of the present invention is shown; Figure 5 A schematic diagram of the operation of the post-control module according to an embodiment of the present invention is shown; Figure 6 A schematic diagram of system hierarchy division according to an embodiment of the present invention is shown. Detailed Implementation

[0026] In wastewater treatment denitrification processes, the addition of carbon sources can promote denitrification, thereby facilitating the microbial degradation of nitrogen oxides in the wastewater. When adding carbon sources, it is crucial to ensure accurate control of the dosage and timing, as well as the impact on system pH changes and sediment formation. Timely adjustments are necessary to achieve optimal denitrification results.

[0027] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0028] The intelligent control system for carbon source dosing based on biological denitrification process provided by this invention includes: host computer, server, PLC control cabinet, dosing pump, dosing pipeline, process instrumentation and algorithm engine.

[0029] The algorithm engine is deployed on a server and includes a front control module and / or a back control module and a database. The front control module and / or back control module are used to predict the carbon source dosage based on the influent data, and iteratively correct the predicted carbon source dosage based on the effluent indicators and process parameters, and output the corrected carbon source dosage. The PLC control cabinet is connected to the server, the dosing pump and the process instruments, respectively, and is used to upload the monitoring data of the process instruments to the algorithm engine, and send the carbon source dosage output by the algorithm engine to the dosing pump, so that the dosing pump can add carbon source to the sewage treatment system through the dosing pipeline.

[0030] This system is widely applicable to various biological nitrogen removal processes, achieving full-process control of nitrogen removal through two independent control modules: a pre-control module, a post-control module, or a combination of both, depending on the specific process.

[0031] For example, when the biological nitrogen removal process is an AAO process, oxidation ditch process, SBR (Sequencing Batch Reactor), or derivative processes (such as CASS, CAST, etc.), the algorithm engine only includes a pre-control module. When the biological nitrogen removal process is a Bartonp process or a multi-stage AO process, the algorithm engine includes a pre-control module and one or more post-control modules (the number of post-control modules is determined by the number of subsequent denitrification systems), and the pre-control module and one or more post-control modules share a single database. When the biological nitrogen removal process only includes a denitrification filter, the algorithm engine only includes a post-control module.

[0032] Among them, the Battenpo process or multi-stage AO (aerobic-anaerobic) process adds one or more denitrification systems to the back end of the traditional AAO (anaerobic-anoxic-aerobic) process or oxidation ditch (anaerobic-anoxic-aerobic) process to achieve multiple denitrifications, thereby reducing the total nitrogen index of the effluent to a lower level.

[0033] Figure 1 A process flow diagram of the Paton process according to an embodiment of the present invention is shown. Figure 1 As shown, the Barton process includes: an anaerobic tank for nitrogen removal, converting ammonia nitrogen into nitrite; a pre-anoxic tank and a pre-aerobic tank for organic matter removal; and a post-anoxic tank and a post-aerobic tank for oxidizing nitrite to nitrate, while further removing residual organic matter and nitrogen. The outlet of the pre-aerobic tank is connected to the outlet of the anaerobic tank via a return pipe, and the inlet of the anaerobic tank also includes sludge return from the secondary sedimentation tank.

[0034] For the Barton process, the pre-anoxic tank and pre-aerobic tank are controlled by the pre-control module, while the post-anoxic tank and post-aerobic tank are controlled by the post-control module.

[0035] Figure 2 A schematic diagram of a multi-stage AO process according to an embodiment of the present invention is shown. Figure 2 As shown, the multi-stage AO process consists of an anaerobic tank and multiple anoxic and aerobic tanks (primary anoxic tank, primary aerobic tank, secondary anoxic tank, secondary aerobic tank... n-stage anoxic tank, n-stage aerobic tank). The outlet of the primary aerobic tank is connected to the outlet of the anaerobic tank via a return pipe, and the inlet of the anaerobic tank also includes sludge return from the secondary sedimentation tank.

[0036] For multi-stage AO processes, the first-stage anoxic tank and the first-stage aerobic tank are controlled by the pre-control module, while the subsequent anoxic and aerobic stages are controlled by the post-control module.

[0037] like Figure 1 and 2 As shown, in the Barton process and the multi-stage AO process, the denitrification system added after the biological reactor is highly separated from the denitrification system before the biological reactor: when the biological reactor is running normally, the main components of total nitrogen in the primary denitrification influent are ammonia nitrogen and some nitrate nitrogen, and the main component of total nitrogen in the effluent is ammonia nitrogen. The main form of total nitrogen in the influent and effluent of the subsequent denitrification units is nitrate nitrogen, and the correlation between the subsequent denitrification and the primary denitrification is weak.

[0038] To enhance the versatility of the algorithm engine, this system is designed with independent operation of the front and rear modules, employing corresponding control modules for each of the front and rear denitrification units. Thanks to this independent operation development approach, this system can be widely applied to AAO processes, oxidation ditch processes, Patonp processes, multi-stage AO processes, denitrification filter processes, SBR processes, and more.

[0039] Figure 3 A schematic diagram of an intelligent control system for carbon source addition based on a biological denitrification process according to an embodiment of the present invention is shown. Figure 3 As shown, the system includes a host computer, server, PLC control cabinet, dosing pump, dosing pipeline, multiple process instruments, and algorithm engine.

[0040] The algorithm engine is built into the server and includes a front control module and / or a back control module and a database. The front control module and / or back control module are used to predict the carbon source dosage based on the influent data, and iteratively correct the predicted carbon source dosage based on the effluent indicators and process parameters, and output the corrected carbon source dosage.

[0041] The PLC control cabinet connects to the server, dosing pump, and process instruments. It uploads the monitoring data from the process instruments to the algorithm engine and sends the carbon source dosage output by the algorithm engine to the dosing pump, so that the dosing pump can add carbon source to the wastewater treatment system through the dosing pipeline.

[0042] The number of dosing pumps and their dosing locations in the wastewater treatment system, as well as the type of process instruments and their installation locations in the wastewater treatment system, are determined based on the control method of the algorithm engine.

[0043] The front control module controls the carbon source addition process in the pre-anoxic-aerobic phase; the rear control module controls the carbon source addition process in the post-anoxic-aerobic phase. The front and rear control modules share a database, but their operating environments and data acquisition are independent. Their operating conditions, instrument settings, and signal acquisition differ, and the front control module runs first, followed by the rear control module.

[0044] The front control module can predict the carbon source dosage of the primary denitrification system based on the influent data, calculate the carbon source dosage time based on the influent flow rate, internal reflux ratio, and external reflux ratio, and correct the carbon source dosage of the primary denitrification system based on the total nitrogen index of the effluent from the anoxic tank and the process instrument data.

[0045] The post-control module can predict the carbon source dosage of the secondary or multiple denitrification system based on the difference between the nitrile nitrogen data and the nitrile nitrogen index of the effluent from the pre-aerobic tank, and adjust the carbon source dosage of the secondary or multiple denitrification system based on the difference between the nitrile nitrogen data and the nitrile nitrogen index of the effluent from the post-aerobic tank and the process parameters.

[0046] This system allows for the selection and combination of pre- and post-control modules based on the specific needs of the denitrification process in a wastewater treatment plant, meeting the carbon source dosing control requirements of the mainstream wastewater treatment market. For example, when the system is applied to a traditional AAO process or oxidation ditch process, only the pre-control module is required for core control, and only three instruments (total nitrogen, nitrate nitrogen, and ammonia nitrogen) are needed per set of core instruments. When the system is applied to a denitrification filter, only the post-control module is required for core control, and only two nitrate nitrogen instruments are needed per set of core instruments.

[0047] Since BOD5, internal recirculation ratio, sludge organic content ratio, and carbon source COD equivalent are crucial for biological nitrogen removal, but all of them are difficult to monitor online, this system has reserved a manual input port on the host computer and server interface to receive manually input data on influent BOD5, internal recirculation ratio, sludge organic content ratio, and carbon source COD equivalent.

[0048] Specifically, the front control module can collect influent data, effluent indicator data, process instrument data, and objective flow data. Influent data includes influent COD. Cr The data includes influent total nitrogen, influent pH, manually input influent BOD5, and process instrument data including water temperature, anoxic influent ORP, MLSS in the biological tank, DO in the aerobic tank's final compartment, ammonia nitrogen in the biological tank effluent, and COD in the biological tank effluent. Cr The data includes the organic content ratio of sludge, the COD equivalent of carbon source, and the effluent index data, including the total nitrogen index of the effluent from the pre-anoxic tank and the nitrate nitrogen index of the effluent from the pre-aerobic tank. The objective flow data includes the influent flow rate, the external recirculation ratio, and the internal recirculation ratio. Predict the carbon source dosage based on the influent data, adjust the required internal reflux ratio based on the total nitrogen index of the influent or prompt for manual adjustment of the internal reflux ratio, and calculate the carbon source dosage time based on the influent volume, external reflux ratio, and internal reflux ratio. After the carbon source addition time is reached, the total nitrogen index of the effluent from the pre-anoxic tank is read, and a first correction is generated based on the difference between the total nitrogen index of the pre-anoxic tank effluent and the target value. After the first correction, the process instrument data is read to assess the sludge status and operating conditions of the biological tank, and a second correction is generated. Based on the predicted carbon source addition amount and iterative correction, the dosage of the first denitrification system is output. The updated total nitrogen data of the pre-anoxic tank effluent after adding the dosage is fed back to the database, forming a closed loop with the previous set of influent data and correction information for intelligent learning.

[0049] Figure 4 A schematic diagram of the operation of the pre-control module according to an embodiment of the present invention is shown. The pre-control module can collect process instrument data through the PLC cabinet, calculate the carbon source dosage and carbon source addition time of the pre-anoxic tank based on the collected data, and add the carbon source to the pre-anoxic tank through the dosing pump.

[0050] The data collected by the front control module can be divided into four categories: feedforward prediction, feedback target, process reference, and objective data. Among them, feedforward prediction-type influent data includes influent COD. Cr The influent total nitrogen, influent pH, and influent BOD5 are measured. The influent data update frequency complies with the requirements of the "Technical Specification for Operation of Online Monitoring System for Water Pollution Sources" (HJ 355-2019), which is once every 2 hours. Therefore, this system collects influent data once every 2 hours.

[0051] The feedforward target instrument data includes total nitrogen in the anoxic effluent and nitrate nitrogen in the aerobic effluent. When the biological tank is operating normally, the total nitrogen index of the effluent from the anoxic tank is equal to the target value of nitrate nitrogen in the effluent from the aerobic tank.

[0052] The target data from this type of instrument is the core of the front control module, serving two purposes: ① as the core control target, ensuring stable system operation; ② evaluating the accuracy of the previous group's chemical dosing and adjusting iteratively based on the evaluation results. The total nitrogen instrument measures at a maximum frequency of 30 minutes / time, and the nitrate nitrogen instrument measures at a maximum frequency of 15 seconds / time. To standardize the data update frequency and facilitate data grouping and summarization, this system collects total nitrogen target data for anoxic effluent at a frequency of 30 minutes / time, and nitrate nitrogen target data for aerobic effluent at a frequency of 5 minutes / time.

[0053] Process instrumentation data includes water temperature, anoxic influent ORP, MLSS in the biological tank, DO in the aerobic tank's final compartment, ammonia nitrogen in the biological tank effluent, and COD in the biological tank effluent. Cr The instruments measure the organic content of the sludge and the COD equivalent of the carbon source. Their function is to assess the sludge condition in the biological treatment tank and adjust the parameters based on the assessment results through a second round of iterative correction. Because the process instrument parameters in the biological treatment tank are updated frequently and the data is highly consistent, while the ammonia nitrogen instrument measures at least once every 5 minutes, the COD of the effluent from the biological treatment tank... Cr The frequency is 2 hours per instance. To avoid introducing errors and to facilitate data grouping and summarization, the system collects these process instrument data every 5 minutes.

[0054] Objective flow data includes influent flow rate, external recirculation ratio, and internal recirculation ratio. The purpose of this type of instrument data is to calculate the carbon source dosing time; it does not directly affect the denitrification process. Influent flow rate and internal recirculation flow rate directly affect the dosing time; therefore, the system needs to collect objective process data in real time to ensure timely and accurate dosing.

[0055] Specifically, the system reads the total nitrogen index of the anoxic effluent and generates a correction ΔV1. Since the current total nitrogen in the anoxic effluent is the result after the previous group was dosing, this correction comes from the feedback of the previous group's dosing behavior.

[0056] After one correction, the system reads the process data of the biological tank (water temperature, ORP of anoxic influent, MLSS of biological tank, DO of the aerobic tank's final compartment, ammonia nitrogen in biological tank effluent, and COD of biological tank effluent). Cr The analysis of sludge organic content ratio and carbon source COD equivalent (including sludge organic content ratio and carbon source COD equivalent) is used to determine sludge condition and operating conditions, and to generate a secondary correction ΔV2.

[0057] The final output dosage V = V0 - △V1 - △V2. The updated total nitrogen data of the anoxic effluent after dosing is fed back to the database, forming a closed loop with the previous set of feedforward influent parameters and corrections, so as to facilitate intelligent learning by the system.

[0058] According to one embodiment of the present invention, V0 update frequency: based on the sampling frequency of each instrument, the carbon source dosage V0 is updated every 2 hours; △V1 update frequency: the first round correction frequency follows the total nitrogen measurement frequency of the anoxic effluent, which is 30 minutes / time; △V2 update frequency: the second round correction and the first round correction work together at 30 minutes / time, reading 6 sets of process instrument data within 30 minutes, and combining historical trends to analyze the operating status of the biological tank to generate correction values.

[0059] Based on the frequency of data updates, the dosage is adjusted every 2 hours according to significant changes in influent indicators and every half hour according to the total nitrogen index of the pre-anoxic effluent.

[0060] The pre-control module can verify the nitrate nitrogen data of the pre-aerobic tank effluent to the total nitrogen data of the pre-anoxic tank effluent during the dosing correction phase. Based on the total nitrogen influent and nitrate nitrogen influent of the pre-aerobic tank, a suggested internal recirculation ratio is generated, and the suggested internal recirculation ratio is compared with the actual internal recirculation ratio for real-time manual adjustment.

[0061] According to the principle of biological denitrification, when the biological tank is operating stably, the nitrate nitrogen in the aerobic effluent is approximately equal to the total nitrogen in the anoxic effluent. Since the nitrate nitrogen in the aerobic effluent is updated frequently, the accuracy of the correction can be improved by checking the nitrate nitrogen in the aerobic effluent against the total nitrogen data in the anoxic effluent.

[0062] Before each dosing, the current ammonia nitrogen index in the aerobic effluent is compared with the warning value. When the ammonia nitrogen index in the aerobic effluent is lower than the warning value, the system operates normally. When the ammonia nitrogen index in the aerobic effluent is higher than the warning value, the nitrification reaction is restricted, and an alarm message is issued to prompt that the system needs to switch to manual control mode.

[0063] For example, after adding carbon source according to the output carbon source dosage, the COD of the carbon source will be... Cr Equivalent to influent volume, if COD Cr If the ratio of / TN is greater than the preset value, an alarm message will be issued.

[0064] By introducing the activated sludge process to reduce the nitrogen and phosphorus removal ratio (COD) Cr:TN:TP=100:5:1: After adding the carbon source, the COD of the carbon source will be... Cr If the equivalent COD is converted to influent, then... Cr If / TN > 20:1, an alarm will be triggered immediately.

[0065] The carbon-nitrogen ratio of the influent is analyzed to determine whether the carbon source in the influent is sufficient: if the dosage is equal to the minimum flow rate of the dosing pump, feedback is sent to the system to confirm that the carbon source in the influent is sufficient; if the dosage is greater than the minimum flow rate of the dosing pump, feedback is sent to the system to confirm that the carbon source in the influent is insufficient.

[0066] If it is determined that the carbon source in the influent is sufficient, the carbon source prediction dosage V0 corresponding to the previous set of influent data is used, the corrections for ΔV1 and ΔV2 are adjusted normally, and the corrected dosage V is output.

[0067] If the influent carbon source is determined to be insufficient, the carbon source dosage V0 is calculated based on the influent indicators. Corrections to ΔV1 and ΔV2 are adjusted as needed, and the corrected dosage V is output.

[0068] If insufficient carbon source is detected in the influent for the first time, carbon source dosing is initiated, and the system enters normal operation after a preset time. When the dosing pump controlled by the system output remains at the lowest frequency for a preset time, the system asks the operator whether to stop the dosing, and enters the stop state after obtaining consent.

[0069] During the dosage decision correction phase, the front control module can iteratively correct the dosage every 30 minutes based on the updated effluent target data and process instrument data, and output the corrected carbon source dosage.

[0070] In addition to the normal main operating logic of the system, an anomaly detection logic has been added to the system: monitor for abnormal influent water quality, abnormal operation of the biological reactor, abnormal total nitrogen or nitrate nitrogen index of the effluent, and instrument failure status, and execute corresponding control measures according to the abnormal status.

[0071] Specifically, to address situations such as abnormal influent water quality, abnormal operation of the biological reactor, total nitrogen or nitrate nitrogen in the effluent exceeding target values, and instrument malfunctions, the following safety logic is set according to different operating conditions: 1. Abnormal Influent Water Quality Condition: When the system detects that the influent indicators differ from the previous set of data by 20%, or exceed the design standards, it enters the abnormal influent water quality condition. Under this condition, the values ​​of ΔV1 and ΔV2 will be significantly reduced, and if necessary, reduced to negative values, to maintain the total nitrogen in the anoxic effluent below the target value. 2. Abnormal Operation of the Bioreactor: The bioreactor is considered to be in an abnormal operating condition when any of the following situations occur: ① Increased ammonia nitrogen in aerobic effluent: Take the following measures: (1) Issue an alarm and prompt whether to switch to manual mode. If not switched to manual mode, prompt to update the total nitrogen target value. The new total nitrogen target value is TN. 新 =TN 旧 -Ammonia nitrogen in aerobic effluent; (2) Increase ΔV1 and ΔV2, reduce the amount of carbon source added to prevent excessive carbon source from affecting the recovery of nitrifying bacteria; ② Abnormal denitrification in the anoxic tank: When the amount of carbon source added is converted to the influent COD Cr COD appeared Cr When / TN > 20, the denitrification is in an abnormal state. The alarm will sound and you must switch to manual mode to adjust the process and promote the recovery of denitrifying bacteria. ③ Impact on the bioreactor: When the slope of the trend of the process instrument in the bioreactor is greater than the preset value, such as 0.5, the bioreactor enters the impact condition and the alarm will be triggered. It is imperative to switch to manual mode to promote the bioreactor to return to normal as soon as possible. ④ Insufficient internal recirculation ratio: When the internal recirculation ratio of the biological tank does not meet the system's required internal recirculation ratio, an abnormal internal recirculation condition is entered. An alarm is triggered to urge technicians to adjust the internal recirculation flow rate in a timely manner and input the updated internal recirculation ratio into the system to ensure that the total nitrogen in the effluent meets the standard.

[0072] 3. Instrument malfunction: ① When the instrument deviation of total nitrogen in anoxic effluent, nitrate nitrogen in aerobic effluent, and total nitrogen in secondary sedimentation effluent (if any) reaches 10% or more, or when the nitrate nitrogen in aerobic effluent is 5% or more higher than the total nitrogen in anoxic effluent; ② When the instrument has not been maintained for an extended period of time, or when the instrument's standard solution verification fails; In such cases, the system enters an abnormal instrument operation state, triggering an alarm to manually switch to maintenance mode to ensure accurate instrument data and promote stable system operation. If the data remains abnormal after manual adjustment and maintenance, the system enters an abnormal bioreactor operation state, triggering an alarm to promptly adjust the bioreactor process.

[0073] We can assign weights to the three types of data: influent indicator data, effluent target data, and biological tank process data. The overall importance is in the following order: influent indicator data > effluent target data > biological tank process data. The main reason is: ① The influent index data reflects the influent water quality, and the carbon source dosage generated based on this data enables the system to have good predictive capabilities; ② The effluent target data is the control target of this system. Among them, the total nitrogen data of the anoxic effluent directly reflects whether the system output dosage is accurate. ③ The process data of the biological pool is mainly used to reflect the operating status of the biological pool. It cannot be significantly changed by humans in a short period of time, so there is no need to set too high a weight.

[0074] This system will summarize and learn from each medication administration behavior, and group the data into the database according to the time sequence.

[0075] Specifically, the front control module can compare the total nitrogen data of the anoxic effluent with the target data every 30 minutes and enter the comparison results into the database for data learning. If the actual total nitrogen in the anoxic effluent is greater than the target value, the dosage in the previous round was too low, and subsequent iterations need to be reduced to correct it; if the actual total nitrogen in the anoxic effluent is less than the target value, the dosage in the previous round was too high, and subsequent iterations need to correct it.

[0076] Since the main form of total nitrogen in the influent and effluent of secondary or multiple denitrification units is nitrate nitrogen, other indicators will not affect secondary or multiple denitrification. Therefore, the core monitoring indicator of the post-control module is effluent nitrate nitrogen.

[0077] The post-control module is used to predict the carbon source dosage of the secondary or multiple denitrification system based on the difference between the nitrile nitrogen data and the nitrile nitrogen index of the effluent from the pre-aerobic tank, and to correct the carbon source dosage of the secondary or multiple denitrification system based on the difference between the nitrile nitrogen data and the nitrile nitrogen index of the effluent from the post-aerobic tank and the process parameters.

[0078] Figure 5 A schematic diagram of the operation of the rear control module according to an embodiment of the present invention is shown. Figure 5 As shown, the downstream control module controls the carbon source addition process of the secondary (or multiple) denitrification unit. It predicts the carbon source addition amount based on the influent nitrification nitrogen data of this unit, corrects the carbon source addition amount based on the effluent nitrification nitrogen index of this unit and the sludge status assessment results of the upstream control module, outputs the corrected carbon source addition amount, and adds the corrected carbon source addition amount to the downstream anoxic tank through the dosing pump.

[0079] The post-control module primarily collects data including influent nitrate nitrogen, effluent nitrate nitrogen, process instrument data, and influent flow rate for this unit (secondary denitrification unit). The data collected by the post-control module can be categorized into feedforward prediction, feedforward target, and process reference data.

[0080] Among them, the feedforward prediction type is: influent nitrate nitrogen of this unit (when a front control module exists, the influent nitrate nitrogen of this unit is the same as the effluent nitrate nitrogen of the aerobic water). Its update frequency is the same as the target data frequency of aerobic effluent nitrate nitrogen, which is 5 minutes / time; Feedback target type: effluent nitrate nitrogen in this unit can serve as the core control target to ensure stable system operation; it can also assess the accuracy of the feedback dosage to the previous group and adjust iteratively based on the assessment results. Its update frequency is the same as that of influent nitrate nitrogen in this unit, which is once every 5 minutes.

[0081] Process Reference Class: Total Effluent COD CrThe data includes water temperature, ammonia nitrogen, and process parameters collected by the upstream control module (if applicable). This data is used to assess the sludge condition in the biological tank and is used for iterative adjustments based on the assessment results. Data collection is conducted every 5 minutes. Since the residence time in the post-anoxic tank is short, typically 1-2 hours, there is no need to calculate the carbon source addition time.

[0082] The post-control module is used to predict the carbon source addition amount in the post-anoxic tank based on the difference between the nitrate nitrogen data of the influent and the target value of the nitrate nitrogen in the effluent of the post-anoxic tank, and to generate a correction based on the difference between the nitrate nitrogen data of the effluent and the target value of the nitrate nitrogen in the effluent of the post-anoxic tank. When a pre-control module exists in the system, the second-round correction follows the second-round correction of the pre-control module. When a pre-control module does not exist in the system, the correction is based on the temperature value and the total effluent COD. Cr Value generation and two-round correction; The dosage of chemicals for secondary or multiple denitrification systems is output based on the predicted carbon source addition amount in the post-anoxic tank, the first round correction, and the second round correction. The updated nitrate and nitrogen data of the post-anoxic tank effluent after dosing according to the output dosage will be fed back to the database, forming a closed loop with the previous set of nitrate and nitrogen data of the post-anoxic tank influent and iterative corrections, so as to facilitate intelligent learning by the system.

[0083] Because the secondary / multiple denitrification unit has a simple structure, its operating logic is also simplified, and the following control logic exists: 1. Target value setting: Manually set the target values ​​for effluent nitrate and nitrogen; 2. Feedforward prediction: The system calculates the carbon source dosage V1 based on the difference between the influent nitrate nitrogen and the effluent nitrate nitrogen target values ​​of this unit; 3. Feedback Correction: Based on the difference between the effluent nitrile nitrogen data and the target nitrile nitrogen value of this unit, a correction ΔV3 is generated. Since the current effluent nitrile nitrogen data of this unit is the result after the previous group's dosing, this correction comes from the feedback result of the previous group's dosing behavior; 4. Operating condition correction: There are two correction conditions: ① When a pre-control module exists, the correction result △V2 of the pre-control module is used; ② Without a pre-control module, the correction result △V4 is generated as follows: In northern regions where large temperature differences between summer and winter, and extreme low temperatures in winter, affect denitrification, a correction for temperature on nitrogen removal efficiency is added; simultaneously, excessive carbon source addition will lead to increased total effluent COD. Cr The temperature rises. Therefore, based on the temperature value and the total effluent COD... Cr The value generates a second round of correction △V4; 5. Output dosing: Output dosing amount V ， =V1 - △V3 - △V4. The output dosage V. ， Set the upper limit value V max=20 * (Nitrogen in the influent of this unit - Target value of nitrogen in nitrate) * Water volume; 6. Closed-loop learning: The updated effluent nitrate and nitrogen data of this unit after dosing is fed back to the database, forming a closed loop with the feedforward set of influent nitrate and nitrogen data of this unit and the correction value, so as to facilitate intelligent learning of the system. 7. Update frequency: Based on the detection frequency of the nitrate nitrogen meter, the dosage update frequency is synchronized at 5 minutes / time; 8. Auxiliary calibration: Introduce the COD ratio for nitrogen and phosphorus removal using the activated sludge process. Cr Use :TN:TP=100:5:1 as the base calibration and set the upper limit of the dosing flow rate.

[0084] For example, the post-control module generates a carbon source dosage V1 based on the difference between the influent nitrate nitrogen and the target value, and then generates a correction ΔV3 based on the difference between the effluent nitrate nitrogen and the target value, according to the operating conditions (water temperature, effluent COD). Cr The second round of correction ΔV4 is generated, and the carbon source addition V after iterative correction is obtained. ， It can be directly output and added.

[0085] The post-control unit is generally located at the end of the entire wastewater treatment system. Although there may be high levels of influent ammonia nitrogen and total phosphorus, this unit does not have the function of reducing these levels, and these levels do not interfere with nitrate removal. Therefore, the impact of these levels on the system is not considered.

[0086] Because the post-control unit has a short residence time and a high nitrate nitrogen update frequency, the influent nitrate nitrogen, effluent nitrate nitrogen, and dosage V of this unit are more likely to form a data closed loop, thus the system has good stability.

[0087] In addition to the main operating logic, auxiliary operation start / stop logic and safety boundary logic can be added to clarify the conditions for system start and stop, improve the system's operating condition assessment and classification and various alarm functions, and switch to the corresponding emergency mode according to different emergencies to improve the stability of operation.

[0088] If the nitrate nitrogen level in the effluent from the aerobic tank is detected to be higher than the target value for nitrate nitrogen in the aerobic tank for the first time, carbon source dosing will be initiated. When the dosing pump controlled by the system output remains at the lowest frequency for a preset time, the system will ask the technicians whether to stop the dosing and, after obtaining their consent, the system will enter a stop state.

[0089] For example, 1. Start-up logic: When the system is in a state without chemical dosing, if the influent nitrate nitrogen of this unit is detected to be higher than the target value of effluent nitrate nitrogen for the first time, carbon source dosing will be started immediately. At this time, the output dosing amount V = V0 - ΔV1 - ΔV2, and in the next 30 minutes, the 6 sets of ΔV1 and ΔV2 will only take constants less than or equal to 0. After 30 minutes, it will enter the normal operation state. 2. Stop logic: When the dosing pump controlled by the system output is at its lowest frequency for 30 consecutive minutes, a pop-up window will ask the technician whether to stop the dosing. After obtaining consent, the system will enter the stop state.

[0090] Because the retention time in denitrification filters is relatively short, generally 1-2 hours, the removal of nitrate nitrogen is limited. To prevent excessively high influent nitrate nitrogen levels from causing elevated effluent nitrate nitrogen, or water quality abnormalities or instrument malfunctions, the following safety logic is implemented: 3. Safety Logic: Monitor influent nitrate and nitrogen concentration, abnormal denitrification operation status, and abnormal instrument operation status, and issue alarm information for abnormal status so that manual adjustment can be made.

[0091] For example, (1) Influent nitrate concentration overload: When the system detects that the influent nitrate concentration of this unit exceeds the design limit of the denitrification filter, it enters an overload condition. At this time, the system increases the weight of the feedforward prediction and decreases the weight of the two rounds of correction according to the actual situation, ensuring that the output carbon source dosage is sufficient and that the effluent nitrate concentration of this unit is lower than the target value. In addition, if the water flow is adjustable, the system will prompt the technicians to reduce the influent flow of this unit.

[0092] (2) Abnormal operation of denitrification: If the nitrification nitrogen meter readings of the influent of this unit are normal, but the CODCr / (nitrification nitrogen of this unit - target value) after comprehensive calculation of the amount of carbon source added is ≥20, it is considered that the water quality or bacteria are abnormal and the denitrification filter is in an abnormal operation condition. An alarm needs to be set up to switch to manual mode in time.

[0093] (3) Abnormal instrument operation: When the instrument has not been maintained for an extended period of time or the instrument fails the standard solution verification, it enters an abnormal instrument operation condition, triggering an alarm to manually switch to maintenance mode to ensure accurate instrument data and facilitate stable system operation. If the data is still abnormal after the standard solution verification is passed, it enters an abnormal operation condition, triggering an alarm to manually adjust the denitrification filter process in a timely manner.

[0094] Since denitrification filters are affected by fewer factors, the indicator parameters at both the influent and effluent are nitrate nitrogen, while water temperature generally varies significantly with the seasons and does not fluctuate much in a short period of time. Therefore, during normal operation, the weighting is: influent nitrate nitrogen ≥ effluent nitrate nitrogen ≥ water temperature and effluent COD. Cr When the influent nitrate nitrogen in this unit is overloaded, the weighting is: influent nitrate nitrogen ≤ effluent nitrate nitrogen ≥ water temperature and effluent COD. Cr .

[0095] Based on the principles of system control, the overall control data acquisition and output can be divided into four levels. Figure 6 A schematic diagram of system hierarchy division according to an embodiment of the present invention is shown. Figure 6 As shown, the system consists of an input layer, a hidden layer, an output layer, and a feedback layer from bottom to top: (1) Input layer: Collect all the data required by the system, establish a database, and classify them by group.

[0096] (2) Hidden layer: The model reads historical records based on different categories of data and initiates carbon source addition control and influent prediction. At the same time, this layer also has the function of storing learning data for subsequent reading; and the functions of model testing and model updating to ensure model effectiveness.

[0097] (3) Output layer: Generate correction values ​​based on the historical records read from the upper layer, and finally output the appropriate amount of carbon source to be added.

[0098] (4) Feedback layer: Based on the results generated by the output layer, the difference is evaluated and the results are fed back to the hidden layer for learning.

[0099] The pre-control module is used to compare the total nitrogen data of the pre-anoxic effluent with the target value at preset intervals, and record the comparison results into the database for learning. If the total nitrogen data of the pre-anoxic effluent is greater than the target value, the dosage in the previous round is too low, and subsequent iterations will reduce and correct it. If the total nitrogen data of the pre-anoxic effluent is less than the target value, the dosage in the previous round is too high, and subsequent iterations will increase and correct it. The post-control module is used to compare the effluent nitrate nitrogen data of this unit with the target value at preset intervals, and record the comparison results into the database for learning. If the effluent nitrate nitrogen data of this unit is greater than the target value, the dosage in the previous round was too low, and subsequent iterations will reduce it for correction; if the effluent nitrate nitrogen data of this unit is less than the target value, the dosage in the previous round was too high, and subsequent iterations will increase it for correction. The front control module and the back control module can find the closest data set from historical data after each database update.

[0100] It should be noted that this system can perform offline specialized training based on the characteristic parameters of different wastewater treatment plants, making the model more closely reflect the actual operation of the wastewater treatment plant. After completing the offline specialized training, the system enters the second stage of online training, which begins by connecting on-site input / output signals and setting characteristic parameters.

[0101] As the second phase of training matures, the system can smoothly replace manual carbon source addition without affecting effluent parameters, entering the third phase of stable and intelligent operation, and conducting regular online data training based on backend settings. Furthermore, by adjusting the instrument locations, the system can be adapted to mainstream AAO, oxidation ditch, denitrification filter, and multi-stage AO processes.

[0102] The intelligent carbon source dosing control system based on biological denitrification technology provided in this solution can achieve the following technical effects: (1) The carbon source addition of the primary denitrification system and the secondary / multi-stage denitrification system in various biological denitrification processes can be controlled by the independent front and rear control modules respectively. Appropriate modules or module combinations can be selected according to the specific process, which can be adapted to the production of different wastewater treatment plants and improve the universality of the algorithm engine.

[0103] (2) Using the total nitrogen index of the anoxic tank as the internal control value of the front control module, controlling the total nitrogen index of the anoxic effluent can ensure that the effluent meets the standards, and can more quickly understand the denitrification effect of the biological tank and the effluent data, so as to take measures in advance.

[0104] (3) By reserving a data input port, the required data can be manually input when it is difficult to monitor data such as internal recirculation ratio and five-day oxygen demand online. In addition, for sewage treatment plants that cannot control internal recirculation online in the traditional way, the interface for manually inputting internal recirculation flow can be retained, which can reduce the cost of transformation.

[0105] (4) Add ammonia nitrogen and nitrate nitrogen monitoring equipment to the effluent of the aerobic tank. When the ammonia nitrogen level rises, the system will issue an alarm and the technicians will take timely measures to maintain the normal operation of the system and ensure that the ammonia nitrogen level of the biological tank effluent meets the discharge standards.

[0106] (5) By setting start-up and shutdown logic and safety boundary logic according to the common characteristics of different sewage treatment plants, the total nitrogen index of sewage treatment plants can be guaranteed to meet the standards and the ability to respond to abnormal situations in an emergency to the greatest extent.

[0107] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0108] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments.

[0109] As used herein, unless otherwise specified, the use of ordinal numbers such as “first,” “second,” “third,” etc., to describe ordinary objects merely indicates different instances of similar objects and is not intended to imply that the objects being described must have a given order in time, space, ordering, or any other manner.

[0110] Although the invention has been described with respect to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and edibility purposes, and not for the purpose of explaining or limiting the subject matter of the invention.

[0111] Therefore, many modifications and alterations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. The disclosure of this invention is illustrative rather than restrictive, and the scope of the invention is defined by the appended claims.

Claims

1. A carbon source dosing intelligent control system based on a biological denitrification process, comprising: The host computer, server, PLC control cabinet, dosing pump, dosing pipeline, process instrument and algorithm engine are characterized in that the algorithm engine is deployed on the server and comprises a front control module and / or a rear control module and a database, the front control module and / or the rear control module are used for predicting the carbon source dosage according to the influent data, and iteratively correcting the predicted carbon source dosage according to the effluent indicators and process parameters, and outputting the corrected carbon source dosage; the PLC control cabinet is connected with the server, dosing pump and process instrument respectively, and is used for uploading the monitoring data of the process instrument to the algorithm engine, and delivering the carbon source dosage output by the algorithm engine to the dosing pump, so that the dosing pump adds the carbon source to the sewage treatment system through the dosing pipeline.

2. The carbon source dosing intelligent control system based on a biological denitrification process according to claim 1, characterized in that, When the biological denitrification process is AAO process, oxidation ditch process, SBR and derived process, the algorithm engine only comprises a front control module; when the biological denitrification process is Bardenpho process or multi-stage AO process, the algorithm engine comprises a front control module and one or more rear control modules, and the front control module and the one or more rear control modules share a database; when the biological denitrification process is denitrification filter, the algorithm engine only comprises a rear control module; the number of dosing pumps and the dosing positions in the sewage treatment system, the types of process instruments and the installation positions in the sewage treatment system are determined according to the control mode of the algorithm engine. 3.The carbon source dosing intelligent control system based on a biological denitrification process of claim 1, wherein, The front control module is used for predicting the carbon source dosage of the primary denitrification system according to the influent data, calculating the carbon source dosage time according to the influent water quantity, internal reflux ratio and external reflux ratio, and correcting the carbon source dosage of the primary denitrification system according to the total nitrogen indicator of the front anoxic tank effluent and process instrument data; The rear control module is used for predicting the carbon source dosage of the secondary or multiple denitrification system according to the difference between the front aerobic tank effluent nitrate nitrogen data and the front aerobic tank effluent nitrate nitrogen indicator, and correcting the carbon source dosage of the secondary or multiple denitrification system according to the difference between the rear aerobic tank effluent nitrate nitrogen and the rear aerobic tank effluent nitrate nitrogen indicator and process parameters.

4. The carbon source dosing intelligent control system based on a biological denitrification process according to claim 3, characterized in that, The host computer and the server reserve a manual input port for receiving manual input of water inflow BOD5, internal reflux ratio, sludge organic fraction ratio and carbon source COD equivalent; the front control module is used for collecting water inflow data, water outflow index data, process instrument data and objective flow data, the water inflow data including water inflow COD Cr , water inflow total nitrogen index, water inflow PH and water inflow BOD5, the process instrument data including water temperature, anoxic water inflow ORP, biological pool MLSS, aerobic pool end bin DO, biological pool water outflow ammonia nitrogen, biological pool water outflow COD Cr , sludge organic fraction ratio and carbon source COD equivalent, the water outflow index data including front anoxic pool water outflow total nitrogen index and front aerobic pool water outflow nitrate nitrogen index, and the objective flow data including water inflow volume, external reflux ratio and internal reflux ratio; The carbon source dosage is predicted according to the influent data, the required internal reflux ratio is adjusted according to the influent total nitrogen indicator, or manual adjustment of the internal reflux ratio is prompted, and the carbon source dosage time is calculated according to the influent water quantity, external reflux ratio and internal reflux ratio; After the carbon source dosage time is reached, the front anoxic tank effluent total nitrogen indicator is read, and a primary correction is generated according to the difference between the front anoxic tank effluent total nitrogen indicator and the target value; After the primary correction, the process instrument data is read, the sludge state and operating conditions of the biological tank are evaluated, and a secondary correction is generated; The dosing amount of the primary denitrification system is output based on the predicted carbon source dosage, the primary correction and the secondary correction; The updated front anoxic tank effluent total nitrogen data after dosing according to the dosing amount is fed back to the database, and a closed loop is formed with the last group of influent data and the correction condition.

5. The carbon source dosing intelligent control system based on a biological denitrification process according to claim 4, characterized in that, The front control module is configured to check the total nitrogen data of the front anoxic tank effluent by the nitrate nitrogen data of the front aerobic tank effluent in the dosing amount correction stage, generate a recommended value of the internal reflux ratio according to the total nitrogen index of the influent and the nitrate nitrogen data of the front aerobic tank effluent, and compare the recommended value of the internal reflux ratio with the actual internal reflux ratio, so as to manually adjust in real time. The front control module is also configured to read the latest ammonia nitrogen data of the front aerobic tank effluent before each dosing, and if the ammonia nitrogen data of the front aerobic tank effluent is lower than a warning value, the system is in normal operation, and if the ammonia nitrogen data of the front aerobic tank effluent is higher than the warning value, a warning state is in a pre-alarm state, and an alarm information is sent to prompt switching to a manual control mode. 6.The carbon source dosing intelligent control system based on a biological denitrification process of claim 1, wherein, The front control module is configured to analyze whether the influent carbon source is sufficient according to the influent carbon-nitrogen ratio: if the dosing amount is equal to the minimum flow of the dosing pump, the system is fed back to confirm that the influent carbon source is sufficient; if the dosing amount is greater than the minimum flow of the dosing pump, the system is fed back to confirm that the influent carbon source is insufficient; If the influent carbon source is sufficient, the predicted carbon source dosing amount of the previous group is used, and iterative correction is performed, and the corrected dosing amount is output; If the influent carbon source is insufficient, the carbon source dosing amount is calculated and generated, and iterative correction is performed, and the corrected dosing amount is output; In the case of first detecting that the influent carbon source is insufficient, carbon source dosing is started, and after a preset time, a normal operation state is entered; When the dosing pump controlled by the system is at the minimum frequency for a continuous preset time, it is inquired whether the manual dosing is stopped, and after obtaining the consent, a stop state is entered.

7. The carbon source dosing intelligent control system based on a biological denitrification process according to claim 6, characterized in that, The front control module is also configured to monitor the influent water quality anomaly, biological reaction tank operation anomaly, effluent total nitrogen or nitrate nitrogen index anomaly, instrument fault state, and execute corresponding control measures according to the abnormal state: When it is detected that the difference between the influent index and the data of the previous group exceeds a preset safety threshold, the influent water quality anomaly working condition is entered, the iterative correction is reduced, and the total nitrogen data of the front anoxic tank effluent is maintained to be always lower than the target value; When it is detected that the ammonia nitrogen data of the front aerobic tank effluent is increased, it is prompted whether to switch to the manual mode, and if it is not switched to the manual mode, it is prompted to update the total nitrogen target value and improve the iterative correction; When the output carbon source dosage is converted to the influent COD Cr When COD Cr When TN > 20, the denitrification enters an abnormal state, and an alarm is given to switch to manual mode. When the trend slope of the biological tank process instrument data changes is greater than a preset value, the biological reaction tank is in an impact working condition, and an alarm is given to the manual to switch to the manual mode; When the internal reflux ratio of the biological tank does not meet the required internal reflux ratio of the system, an internal reflux anomaly working condition is entered, and an alarm is given to the manual to timely adjust the internal reflux flow, and the updated internal reflux ratio is input to the system; When the total nitrogen of the anoxic effluent, the nitrate nitrogen of the aerobic effluent, or the instrument is out of order for a long time without maintenance, or the instrument standard liquid verification is not passed, an instrument operation anomaly working condition is entered, and an alarm is given to the manual to switch to the maintenance mode; If the data of the manually adjusted and maintained instrument is still in an abnormal state, a biological reaction tank operation anomaly working condition is entered, and an alarm is given to the manual to timely adjust the biological tank process. 8.The intelligent control system for carbon source dosing based on a biological denitrification process according to claim 1, characterized in that, When the rear control module is applied to a sewage treatment system alone, the rear control module is configured to predict the carbon source dosing amount of the rear anoxic tank according to the difference between the nitrate nitrogen data of the rear anoxic tank influent and the nitrate nitrogen target value of the rear anoxic tank effluent, and generate a round of correction according to the difference between the nitrate nitrogen data of the rear anoxic tank effluent and the nitrate nitrogen target value of the rear anoxic tank effluent. When the system has a front control module, the second round correction follows the secondary correction of the front control module, and when the system does not have a front control module, the second round correction is generated according to the temperature value and the total effluent COD Cr value. According to the predicted carbon source dosage of the post-anoxic tank, one round of correction and two rounds of correction, the dosing amount of the secondary or multiple denitrification systems is outputted; The updated post-anoxic tank effluent nitrate data after dosing according to the outputted dosing amount is fed back to the database, and a closed loop is formed with the previous post-anoxic tank influent nitrate data and iterative correction. 9.The carbon source dosing intelligent control system based on a biological denitrification process of claim 8, wherein, The post-control module is also used to start carbon source addition when the pre-aerobic tank effluent nitrate data is first detected to be higher than the pre-aerobic tank effluent nitrate target value; when the dosing pump controlled by the system is at the lowest frequency for a preset time, the system inquires whether the technician stops dosing, and after obtaining the consent, the system enters the stop state; When the system detects that the pre-aerobic tank effluent nitrate data exceeds the design upper limit of the denitrification filter, enters the overload condition, according to the real situation, the weight of the predicted carbon source dosage of the post-anoxic tank is increased, the iterative correction is reduced, the outputted dosing amount is ensured to be sufficient, and the post-aerobic tank effluent nitrate data is ensured to be lower than the post-aerobic tank effluent nitrate target value; COD after adding chemicals and comprehensive calculation Cr If (nitrate data of water out of the former aerobic tank - target value of nitrate data of water out of the former aerobic tank) ≥ 20, the water quality or bacteria species is abnormal, the denitrification filter enters the abnormal working condition, and the alarm is switched to manual mode. When the instrument is not maintained for a long time, or the instrument standard liquid verification fails, the instrument enters an abnormal operating condition, and an alarm is sent to the manual switch to the maintenance mode. 10.The carbon source dosing intelligent control system based on a biological denitrification process of claim 1, wherein, The pre-control module is used to compare the pre-anoxic effluent total nitrogen data with the target value every preset time, and record the comparison result into the database for learning; if the pre-anoxic effluent total nitrogen data is greater than the target value, the previous dosing amount is low, and the subsequent iterative correction is reduced; if the pre-anoxic effluent total nitrogen data is less than the target value, the previous dosing amount is high, and the subsequent iterative correction is increased; The post-control module is used to compare the unit effluent nitrate data with the target value every preset time, and record the comparison result into the database for learning; if the unit effluent nitrate data is greater than the target value, the previous dosing amount is low, and the subsequent iterative correction is reduced; if the unit effluent nitrate data is less than the target value, the previous dosing amount is high, and the subsequent iterative correction is increased; The pre-control module and the post-control module are used to find the closest data group from the historical data after each database update.

Images