A LSP sludge reduction biological treatment system

Abstract

The invention discloses an LSP sludge reduction biological treatment system, and relates to the technical field of sludge treatment. The invention predicts the demand of microbial communities in real time through an MPC algorithm, dynamically regulates oxygen concentration, substrate concentration and pH value and pressure conditions, realizes accurate control of microbial communities in various reaction stages, stabilizes the sludge reduction effect, reduces the risk of metabolic disorders, adopts a combination of an LSTM deep learning model and MPC, predicts and adjusts the operating state of a reactor in real time, quickly responds to environmental changes, improves adaptability to fluctuations in process conditions, identifies sludge generation trends and microbial community changes in advance through intelligent prediction, adjusts control parameters in advance, maintains stable sludge reduction performance, and dynamically regulates the microbial growth environment at the same time, combines real-time monitoring with the predictive regulation of the MPC model, realizes accurate enrichment of simultaneous nitrification and denitrification (SND) and anaerobic ammonia oxidation (Anammox) microbial communities, and can still maintain a stable sludge reduction effect in the process of complex wastewater treatment.

Description

Technical Field

[0001] The invention relates to the technical field of sludge treatment, in particular to an LSP sludge reduction biological treatment system. Background Art

[0002] Domestic sewage and industrial sewage pose a threat to the ecological environment and human health. The activated sludge method is one of the main urban sewage treatment technologies, with the advantages of good treatment effect and low investment. As sewage discharge standards become increasingly stringent, traditional water treatment processes are gradually being replaced by the sludge reduction multi-stage process LSP (Less Sludge Process) due to limited design capacity, increased sludge generation and high treatment costs. LSP uses folded plates to separate aerobic aeration tanks and uses fiber fillers to replace traditional biofilms to improve biofilm efficiency, effectively enrich microorganisms in layers, and optimize pollutant treatment.

[0003] The traditional multi-stage gradient-controlled sequencing batch reactor (SBR) system achieves sludge reduction through staged operations such as filling, reaction, sedimentation, and discharge, but it does not carry out targeted control of different reaction stages, making it difficult to accurately match the dynamic needs of microorganisms, resulting in unstable system efficiency. During the stage switching process, the reaction conditions, dissolved oxygen, pH value, and temperature change dramatically, which can easily cause fluctuations in microbial activity. The alternating operation of aerobic and anaerobic microorganisms cannot be accurately adjusted, affecting the sludge reduction effect. In addition, the current microbial community structure is unstable and mainly relies on natural selection and artificially regulated pressure conditions, such as oxygen concentration and substrate concentration to screen microbial communities. It is easily affected by environmental disturbances and is difficult to achieve precise enrichment and control of specific functional microorganisms, resulting in unsatisfactory sludge reduction effects and unstable effects when treating complex wastewater.

[0004] At present, in order to solve the problem of alternating operation regulation, most of them adopt time control or feedback control based on parameters such as dissolved oxygen and redox potential (ORP) to adjust the reaction stage. However, there is still a response lag problem, and the response speed to environmental changes is slow, which can easily cause metabolic disorders in the sludge reduction process, and it is impossible to finely control the interaction of microbial communities and it is difficult to optimize the operating conditions of each stage. Some selective pressure control technologies such as redox potential and substrate limitation are used for natural enrichment and screening of microorganisms, but they still lack specificity and accuracy. The screening process is slow and easily disturbed by environmental changes. Therefore, there is an urgent need for an LSP sludge reduction biological treatment system to solve such problems. Summary of the invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a LSP sludge reduction biological treatment system to solve the problems of insufficient control accuracy, delayed system response and unstable microbial community structure in the current LSP sludge reduction biological treatment method in SBR system, multi-stage control and microbial selective pressure control.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0008] In one aspect, the present invention provides a LSP sludge reduction biological treatment system, comprising:

[0009] Microbial community screening and enrichment module, screening and enriching the dominant microbial communities with high sludge reduction efficiency, including simultaneous nitrification and denitrification (SND) and anaerobic ammonia oxidation (Anammox) microorganisms; controlling the reaction conditions, optimizing the growth environment of microorganisms, and forming microbial communities with high sludge reduction efficiency;

[0010] The multi-stage SBR reactor optimization module optimizes the operating conditions of the SBR reactor based on the characteristics of the dominant microbial community enriched in the microbial community screening and enrichment module, dynamically adapts to the metabolic needs of microorganisms; dynamically adjusts the reaction conditions to coordinate with the metabolic activities of microorganisms to maximize the sludge reduction efficiency;

[0011] The intelligent control and regulation module uses deep learning technology to predict and adjust the operating status of the reactor in real time, combining the prediction ability of LSTM with the control feedback of MPC to form an adaptive control system to improve the operating efficiency of the reactor and the sludge reduction effect.

[0012] In another aspect, the present invention provides a method for biological treatment of LSP sludge reduction, comprising:

[0013] Step S1, screening and enrichment of microbial communities based on stress conditions,

[0014] Screen and enrich the dominant microbial communities with high sludge reduction capacity, including simultaneous nitrification and denitrification (SND) and anaerobic ammonium oxidation (Anammox) microorganisms;

[0015] Step S2, multi-stage gradient SBR reactor optimization,

[0016] Based on the dominant microbial community enriched in step S1, the operating conditions of the SBR reactor are optimized to adapt to the dynamic metabolic needs of the microorganisms;

[0017] Step S3, intelligent control and regulation of sludge reduction process,

[0018] Deep learning technology is used to intelligently control the multi-stage SBR reactor to predict and regulate the sludge reduction process.

[0019] Furthermore, in step S1, pressure conditions are regulated to adjust oxygen concentration, substrate concentration and pH pressure conditions to create a growth environment conducive to the target microorganisms, and the concentrations of dissolved oxygen, ammonia nitrogen and nitrate are monitored in real time. The real-time monitoring data is used to adopt the model predictive control MPC (Model Predictive Control) algorithm to predict the dynamic needs of the microbial community, control the pressure conditions, stabilize the enrichment process of the target microorganisms, and form a dominant microbial community with high sludge reduction performance.

[0020] Furthermore, the pressure condition regulation method in step S1 is:

[0021] In step S1, the oxygen concentration, substrate concentration and pH pressure conditions are regulated, key parameters are monitored in real time, and the MPC algorithm is used for prediction and control to form a dominant microbial community, and real-time monitoring and data collection are performed. The real-time monitoring parameters include:

[0022] Dissolved oxygen concentration (DO), ammonia nitrogen concentration Nitrate concentration The MPC model is established and the microbial demand is predicted. The MPC goal is set to minimize the sludge production while maintaining the reaction efficiency of the system. The objective function is: Where J represents the optimization target, i.e., total operating cost, comprehensive sludge reduction and energy consumption, Q sludge (k) represents the sludge production at the kth moment, λ represents the weight coefficient, balancing the sludge production and energy consumption, and E(k) represents the energy consumption, including aeration energy consumption and stirring energy consumption;

[0023] Dynamic control variables include:

[0024] Oxygen concentration control, O set (k) = O meas (k)+K O ·(DO set -DO meas ), where O set (k) represents the oxygen concentration setting value of the current time step, O meas (k) represents the actual value of oxygen concentration in the current time step, K O Indicates the oxygen concentration control gain coefficient, DO set Indicates the target dissolved oxygen concentration, DO meas Indicates the measured dissolved oxygen concentration;

[0025] Substrate concentration control, Among them, S set (k) represents the substrate concentration setting value of the current time step, S meas (k) represents the actual value of the substrate concentration at the current time step, K S represents the substrate concentration control gain coefficient, Indicates the target ammonia nitrogen concentration, Indicates the measured ammonia nitrogen concentration;

[0026] pH control, pH set (k) = pH meas (k)+K pH ·(pH target -pH meas ), where pH set (k) represents the pH setting value of the current time step, pH meas (k) represents the actual pH value of the current time step, K pH Indicates the pH control gain coefficient, pH target Indicates the target pH value.

[0027] Furthermore, the pressure condition control method in step S1 also includes:

[0028] To regulate the pressure conditions, U(k)=[O set (k),S set (k), pH set (k)], where U(k) represents the control input, including the oxygen concentration set value, the substrate concentration set value and the pH set value,

[0029] The enrichment effect feedback is: X(k+1)=A·X(k)+B·U(k), where X(k+1) represents the microbial concentration state at the next moment, A represents the microbial growth coefficient matrix, and B represents the influence matrix of the control input on microbial growth;

[0030] The MPC prediction and control model dynamically adjusts the oxygen concentration, substrate concentration and pH value in the reactor to match the metabolic needs of microorganisms, forming and stably enriching a dominant microbial community with high sludge reduction performance.

[0031] Furthermore, in step S2, a multi-stage SBR reactor is used. Based on the characteristics of the dominant microbial community screened in step S1 and the real-time monitoring data, an MPC algorithm is used to predict the operating status of each stage, and the control data in the multi-stage SBR reactor, namely, aeration intensity, reaction time, water inlet volume and sludge discharge cycle, are dynamically adjusted to match the metabolic needs of the microorganisms.

[0032] Further, in step S2, the multi-stage gradient SBR reactor is optimized as follows:

[0033] In step S2, based on the characteristics of the dominant microbial community screened in step S1 and the real-time monitoring data, the MPC algorithm is used to predict the operating status of each stage of the multi-stage SBR (Sequencing Batch Reactor) reactor, and dynamically adjust the control parameters including aeration intensity, reaction time, water inflow and sludge discharge cycle to match the metabolic needs of the microorganisms and maximize the sludge reduction effect;

[0034] To predict the operation status, the objective function is to minimize the system energy consumption and sludge output.

[0035] Where J represents the overall optimization target, including energy consumption, sludge production and reaction efficiency, E(k) represents the total energy consumption at the kth moment, including aeration and mixing operations, and Q sludge (k) represents the sludge production at the kth moment, R eff (k) represents the reaction efficiency at the kth moment, that is, the efficiency of pollutant removal, and α, β, and γ represent weight coefficients;

[0036] Perform dynamic control variable adjustments, including:

[0037] Aeration intensity control, A set (k) = A meas (k)+K A ·(DO target -DO meas (k)), where A set (k) represents the aeration intensity setting value at the kth moment, A meas (k) represents the actual aeration intensity at the kth moment, K A represents the aeration control gain coefficient, DO target Indicates the target dissolved oxygen concentration, DO meas (k) represents the actual dissolved oxygen concentration at the kth moment;

[0038] Reaction time control, T set (k) = T meas (k)+K T ·(R eff (k)-R target ), where T set (k) represents the reaction time setting value at the kth moment, T meas (k) represents the actual reaction time at the kth moment, K T Represents the reaction time control gain coefficient, R eff (k) represents the actual reaction efficiency at the kth moment, R target represents the target reaction efficiency;

[0039] Water inlet control, V set (k) = V meas(k)+K V ·(S target -S meas (k)), where V set (k) represents the water inlet setting value at the kth moment, V meas (k) represents the actual water inflow at the kth moment, K V Indicates the water inlet control gain coefficient, S target represents the target substrate concentration, S meas (k) represents the actual substrate concentration at the kth moment;

[0040] Sludge discharge cycle control, M set (k) = M meas (k)+K M ·(Q sludge (k)-Q target ), where M set (k) represents the mud discharge cycle setting value at the kth moment, M meas (k) represents the actual mud discharge cycle at the kth moment, K M It represents the gain coefficient of mud discharge cycle control, Q sludge (k) represents the sludge production at the kth moment, Q target Indicates the target sludge production.

[0041] Further, in step S2, the multi-stage gradient SBR reactor optimization method also includes:

[0042] MPC optimization and feedback control are performed, and the state feedback control equation is X(k+1)=A·X(k)+B·U(k), where X(k+1) represents the state variables at the k+1th moment, including microbial concentration and pollutant concentration, A represents the system state transfer matrix, and B represents the influence matrix of the control input on the system state;

[0043] U(k)=[A set (k),T set (k),V set (k),M set (k)], represents the control input vector, including the set values ​​of each stage;

[0044] The status of each operation stage is predicted in real time, and the operating conditions of the reactor are dynamically adjusted according to the monitoring data and MPC optimization results, so that the control data of each stage can match the dynamic metabolic needs of the microbial community and optimize the sludge reduction process.

[0045] Furthermore, in step S3, the LSTM deep learning model is used to perform real-time analysis on the operating data of the multi-stage SBR reactor to predict the sludge generation trend and the dynamic changes of the microbial community;

[0046] The reaction conditions are adjusted in combination with the multi-stage control data fed back by the MPC algorithm in step S2.

[0047] Furthermore, in step S3, the intelligent control method for adjusting the sludge reduction process is:

[0048] The real-time operation data in the reactor, including dissolved oxygen concentration, ammonia nitrogen concentration, nitrate concentration, aeration intensity, reaction time, water inflow and sludge discharge cycle, are used as input features for model training and prediction;

[0049] LSTM model analysis and prediction are performed, and the memory unit captures the time dependency between various parameters during the operation of the reactor, predicting the sludge generation trend and dynamic changes of the microbial community at future time points;

[0050] According to the prediction results of LSTM and the feedback data of the MPC algorithm in step S2, the control variables of aeration intensity, reaction time, water inflow and sludge discharge cycle are adjusted in real time.

[0051] The beneficial effects of the present invention are:

[0052] The present invention predicts the needs of microbial communities in real time through the MPC algorithm, dynamically regulates oxygen concentration, substrate concentration and pH value and pressure conditions, realizes precise control of microbial communities at each reaction stage, stabilizes the sludge reduction effect, and reduces the risk of metabolic disorders.

[0053] The present invention adopts a combination of LSTM deep learning model and MPC to predict and adjust the operating status of the reactor in real time, quickly respond to environmental changes, significantly improve the adaptability to fluctuations in process conditions, identify sludge generation trends and microbial community changes in advance through intelligent prediction, adjust control parameters in advance, and maintain stable sludge reduction performance.

[0054] The present invention dynamically regulates the microbial growth environment, combines real-time monitoring with the predictive regulation of the MPC model, and achieves precise enrichment of simultaneous nitrification and denitrification (SND) and anaerobic ammonia oxidation (Anammox) microbial communities. It can also maintain a stable sludge reduction effect in complex wastewater treatment processes.

[0055] The present invention optimizes the multi-stage gradient SBR reactor and combines the predictive adjustment of MPC to dynamically adjust the aeration intensity, reaction time, water inlet and sludge discharge cycle parameters to match the metabolic needs of microorganisms, thereby effectively improving the efficiency and stability of sludge reduction.

[0056] The present invention enhances the anti-disturbance capability of the system through intelligent control and adaptive regulation, and can quickly adjust the reaction conditions to maintain efficient operation of the system even in the case of load fluctuations and environmental changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0058] Figure 1 It is a schematic diagram of the structure of the LSP sludge reduction biological treatment system of the present invention;

[0059] Figure 2 It is a schematic diagram of the process of the LSP sludge reduction biological treatment method of the present invention. DETAILED DESCRIPTION

[0060] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0061] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0062] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0063] Example 1, reference Figure 1 This embodiment provides a LSP sludge reduction biological treatment system, comprising:

[0064] Microbial community screening and enrichment module, screening and enriching the dominant microbial communities with high sludge reduction efficiency, including simultaneous nitrification and denitrification (SND) and anaerobic ammonia oxidation (Anammox) microorganisms; controlling the reaction conditions, optimizing the growth environment of microorganisms, and forming microbial communities with high sludge reduction efficiency;

[0065] The multi-stage SBR reactor optimization module optimizes the operating conditions of the SBR reactor based on the characteristics of the dominant microbial community enriched in the microbial community screening and enrichment module, dynamically adapts to the metabolic needs of microorganisms; dynamically adjusts the reaction conditions to coordinate with the metabolic activities of microorganisms to maximize the sludge reduction efficiency;

[0066] The intelligent control and regulation module uses deep learning technology to predict and adjust the operating status of the reactor in real time, combining the prediction ability of LSTM with the control feedback of MPC to form an adaptive control system to improve the operating efficiency of the reactor and the sludge reduction effect.

[0067] Example 2, reference Figure 2 This embodiment provides a method for biological treatment of LSP sludge reduction, comprising the following steps:

[0068] Step S1, screening and enrichment of microbial communities based on stress conditions,

[0069] Screen and enrich the dominant microbial communities with high sludge reduction capacity, including simultaneous nitrification and denitrification (SND) and anaerobic ammonium oxidation (Anammox) microorganisms;

[0070] In step S1, pressure conditions are regulated to adjust oxygen concentration, substrate concentration and pH pressure conditions, create a growth environment conducive to the target microorganisms, monitor dissolved oxygen, ammonia nitrogen and nitrate concentrations in real time, and use the real-time monitoring data to adopt the model predictive control MPC (Model Predictive Control) algorithm to predict the dynamic needs of the microbial community, control the pressure conditions, stabilize the enrichment process of the target microorganisms, and form a dominant microbial community with high sludge reduction performance;

[0071] The pressure condition control method in step S1 is:

[0072] In step S1, the oxygen concentration, substrate concentration and pH pressure conditions are regulated, key parameters are monitored in real time, and the MPC algorithm is used for prediction and control to form a dominant microbial community, and real-time monitoring and data collection are performed. The real-time monitoring parameters include:

[0073] Dissolved oxygen concentration (DO), ammonia nitrogen concentration Nitrate concentration The MPC model is established and the microbial demand is predicted. The MPC goal is set to minimize the sludge production while maintaining the reaction efficiency of the system. The objective function is: Where J represents the optimization target, i.e., total operating cost, comprehensive sludge reduction and energy consumption, Q sludge (k) represents the sludge production at the kth moment, λ represents the weight coefficient, balancing the sludge production and energy consumption, and E(k) represents the energy consumption, including aeration energy consumption and stirring energy consumption;

[0074] Dynamic control variables include:

[0075] Oxygen concentration control, O set (k) = O meas (k)+K O ·(DO set -DOmeas ), where O set (k) represents the oxygen concentration setting value of the current time step, O meas (k) represents the actual value of oxygen concentration in the current time step, K O Indicates the oxygen concentration control gain coefficient, DO set Indicates the target dissolved oxygen concentration, DO meas Indicates the measured dissolved oxygen concentration;

[0076] Substrate concentration control, Among them, S set (k) represents the substrate concentration setting value of the current time step, S meas (k) represents the actual value of the substrate concentration at the current time step, K S represents the substrate concentration control gain coefficient, Indicates the target ammonia nitrogen concentration, Indicates the measured ammonia nitrogen concentration;

[0077] pH control, pH set (k) = pH meas (k)+K pH ·(pH target -pH meas ), where pH set (k) represents the pH setting value of the current time step, pH meas (k) represents the actual pH value of the current time step, K pH Indicates the pH control gain coefficient, pH target Indicates the target pH value;

[0078] The pressure condition control method in step S1 also includes:

[0079] To regulate the pressure conditions, U(k)=[O set (k),S set (k), pH set (k)], where U(k) represents the control input, including the oxygen concentration set value, the substrate concentration set value and the pH set value,

[0080] The enrichment effect feedback is: X(k+1)=A·X(k)+B·U(k), where X(k+1) represents the microbial concentration state at the next moment, A represents the microbial growth coefficient matrix, and B represents the influence matrix of the control input on microbial growth;

[0081] The MPC prediction and control model dynamically adjusts the oxygen concentration, substrate concentration and pH value in the reactor to match the metabolic needs of microorganisms, forming and stably enriching a dominant microbial community with high sludge reduction performance;

[0082] Specifically, by regulating key pressure conditions such as oxygen concentration, substrate concentration and pH value, combining real-time monitoring data and MPC algorithm, we screen and enrich simultaneous nitrification and denitrification (SND) and anaerobic ammonia oxidation (Anammox) microbial communities with high sludge reduction capacity, which have efficient conversion capabilities for nitrogen and carbon pollutants, reduce the excess sludge generated in traditional activated sludge systems, and significantly improve the sludge reduction effect.

[0083] Step S2, multi-stage gradient SBR reactor optimization,

[0084] Based on the dominant microbial community enriched in step S1, the operating conditions of the SBR reactor are optimized to adapt to the dynamic metabolic needs of the microorganisms;

[0085] In step S2, a multi-stage SBR reactor is used. Based on the characteristics of the dominant microbial community screened in step S1 and the real-time monitoring data, an MPC algorithm is used to predict the operating status of each stage, and the control data in the multi-stage SBR reactor, i.e., aeration intensity, reaction time, water inflow, and sludge discharge cycle, are dynamically adjusted to match the metabolic needs of the microorganisms.

[0086] In step S2, the multi-stage gradient SBR reactor optimization method is:

[0087] In step S2, based on the characteristics of the dominant microbial community screened in step S1 and the real-time monitoring data, the MPC algorithm is used to predict the operating status of each stage of the multi-stage SBR (Sequencing Batch Reactor) reactor, and dynamically adjust the control parameters including aeration intensity, reaction time, water inflow and sludge discharge cycle to match the metabolic needs of the microorganisms and maximize the sludge reduction effect;

[0088] To predict the operation status, the objective function is to minimize the system energy consumption and sludge output.

[0089] Where J represents the overall optimization target, including energy consumption, sludge production and reaction efficiency, E(k) represents the total energy consumption at the kth moment, including aeration and mixing operations, and Q sludge (k) represents the sludge production at the kth moment, R eff (k) represents the reaction efficiency at the kth moment, that is, the efficiency of pollutant removal, and α, β, and γ represent weight coefficients;

[0090] Perform dynamic control variable adjustments, including:

[0091] Aeration intensity control, A set (k) = A meas (k)+K A ·(DO target -DO meas (k)), where Aset (k) represents the aeration intensity setting value at the kth moment, A meas (k) represents the actual aeration intensity at the kth moment, K A represents the aeration control gain coefficient, DO target Indicates the target dissolved oxygen concentration, DO meas (k) represents the actual dissolved oxygen concentration at the kth moment;

[0092] Reaction time control, T set (k) = T meas (k)+K T ·(R eff (k)-P target ), where T set (k) represents the reaction time setting value at the kth moment, T meas (k) represents the actual reaction time at the kth moment, K T Represents the reaction time control gain coefficient, R eff (k) represents the actual reaction efficiency at the kth moment, R target represents the target reaction efficiency;

[0093] Water inlet control, V set (k) = V meas (k)+K V ·(S target -S meas (k)), where V set (k) represents the water inlet setting value at the kth moment, V meas (k) represents the actual water inflow at the kth moment, K V Indicates the water inlet control gain coefficient, S target represents the target substrate concentration, S meas (j) represents the actual substrate concentration at the kth moment;

[0094] Sludge discharge cycle control, M set (k) = M meas (k)+K M ·(Q sludge (k)-Q target ), where M set (k) represents the mud discharge cycle setting value at the kth moment, M meas (k) represents the actual mud discharge cycle at the kth moment, K M It represents the gain coefficient of mud discharge cycle control, Q sludge (k) represents the sludge production at the kth moment, Q target represents the target sludge production;

[0095] In step S2, the multi-stage gradient SBR reactor optimization method also includes:

[0096] MPC optimization and feedback control are performed, and the state feedback control equation is X(k+1)=A·X(k)+B·U(k), where X(k+1) represents the state variables at the k+1th moment, including microbial concentration and pollutant concentration, A represents the system state transfer matrix, and B represents the influence matrix of the control input on the system state;

[0097] U(k)=[A set (k),T set (k),V set (k),M set (k)], represents the control input vector, including the set values ​​of each stage;

[0098] Real-time prediction of the status of each operating stage, dynamic adjustment of the reactor operating conditions based on monitoring data and MPC optimization results, so that the control data of each stage can match the dynamic metabolic needs of the microbial community and optimize the sludge reduction process;

[0099] Specifically, based on the characteristics of the microorganisms screened in step S1, the operating conditions of the multi-stage SBR reactor are optimized, the operating status of the reactor is predicted by the MPC algorithm, and the aeration intensity, reaction time, water inlet and sludge discharge cycle control parameters are dynamically adjusted to accurately match the metabolic needs of the microorganisms, effectively avoiding the reaction efficiency fluctuations caused by timing control, reducing system energy consumption, and enhancing the sludge reduction effect.

[0100] Step S3, intelligent control and regulation of sludge reduction process,

[0101] Use deep learning technology to intelligently control multi-stage SBR reactors to predict and adjust the sludge reduction process;

[0102] In step S3, the LSTM deep learning model is used to analyze the operating data of the multi-stage SBR reactor in real time to predict the sludge generation trend and the dynamic changes of the microbial community;

[0103] Adjust the reaction conditions in combination with the multi-stage control data fed back by the MPC algorithm in step S2;

[0104] In step S3, the intelligent control method for adjusting the sludge reduction process is:

[0105] The real-time operation data in the reactor, including dissolved oxygen concentration, ammonia nitrogen concentration, nitrate concentration, aeration intensity, reaction time, water inflow and sludge discharge cycle, are used as input features for model training and prediction;

[0106] LSTM model analysis and prediction are performed, and the memory unit captures the time dependency between various parameters during the operation of the reactor, predicting the sludge generation trend and dynamic changes of the microbial community at future time points;

[0107] According to the prediction results of LSTM and the feedback data of the MPC algorithm in step S2, the control variables of aeration intensity, reaction time, water inflow and sludge discharge cycle are adjusted in real time;

[0108] Specifically, the LSTM deep learning model is introduced to analyze and predict the real-time operation data of the reactor, identify the trend of sludge generation and the dynamic changes of the microbial community in advance, and combine with the control data fed back by MPC to adjust the control variables of the reactor in real time, further optimize the reaction conditions, and avoid system performance degradation caused by external disturbances or load changes.

[0109] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for biological treatment of LSP sludge reduction, characterized in that: The following steps are included: Step S1, screening and enrichment of microbial communities based on stress conditions, screening and enriching dominant microbial communities with high sludge reduction capabilities, including simultaneous nitrification and denitrification (SND) and anaerobic ammonium oxidation (Anammox) microorganisms; Step S2, multi-stage gradient SBR reactor optimization, based on the dominant microbial community enriched in step S1, optimizing the operating conditions of the SBR reactor; Step S3, intelligently controlling and adjusting the sludge reduction process, using deep learning technology to intelligently control the multi-stage SBR reactor, predicting and adjusting the sludge reduction process; In step S1, pressure conditions are regulated to regulate oxygen concentration, substrate concentration and pH pressure conditions, and the concentrations of dissolved oxygen, ammonia nitrogen and nitrate are monitored in real time. The dynamic needs of the microbial community are predicted by the model predictive control (MPC) algorithm using the real-time monitoring data, and the pressure conditions are controlled to form a dominant microbial community. The pressure condition control method in step S1 is: Carry out real-time monitoring and data collection. The real-time monitoring parameters include: Dissolved oxygen concentration (DO), ammonia nitrogen concentration Nitrate concentration The MPC model is established and the microbial demand is predicted. The MPC goal is set to minimize the sludge production while maintaining the reaction efficiency of the system. The objective function is: Where J represents the optimization target, i.e., total operating cost, comprehensive sludge reduction and energy consumption, Q sludge (k) represents the sludge production at the kth moment, λ represents the weight coefficient, balancing the sludge production and energy consumption, and E(k) represents the energy consumption, including aeration energy consumption and stirring energy consumption; Dynamic control variables include: Oxygen concentration control, O set (k) = O meas (k)+K O ·(DO set -DO meas ), where O set (k) represents the oxygen concentration setting value of the current time step, O meas (k) represents the actual value of oxygen concentration in the current time step, K O Indicates the oxygen concentration control gain coefficient, DO set Indicates the target dissolved oxygen concentration, DO meas Indicates the measured dissolved oxygen concentration; Substrate concentration control, Among them, S set (k) represents the substrate concentration setting value of the current time step, S meas (k) represents the actual value of the substrate concentration at the current time step, K S represents the substrate concentration control gain coefficient, Indicates the target ammonia nitrogen concentration, Indicates the measured ammonia nitrogen concentration; pH control, pH set (k) = pH meas (k)+K pH ·(pH target -pH meas ), where pH set (k) represents the pH setting value of the current time step, pH meas (k) represents the actual pH value of the current time step, K pH Indicates the pH control gain coefficient, pH target Indicates the target pH value, pH meas Indicates the measured pH value; The pressure condition control method in step S1 also includes: To regulate the pressure conditions, U(k)=[O set (k),S set (k), pH set (k)], where U(k) represents the control input, including the oxygen concentration set value, the substrate concentration set value and the pH set value, The enrichment effect feedback is: X(k+1)=A·X(k)+B·U(k), where X(k+1) represents the microbial concentration state at the next moment, A represents the microbial growth coefficient matrix, and B represents the influence matrix of the control input on microbial growth; In step S2, a multi-stage SBR reactor is used. Based on the characteristics of the dominant microbial community screened in step S1 and the real-time monitoring data, an MPC algorithm is used to predict the operating status of each stage, and the control data in the multi-stage SBR reactor, i.e., aeration intensity, reaction time, water inflow, and sludge discharge cycle, are dynamically adjusted; In step S2, the multi-stage gradient SBR reactor optimization method is: To predict the operation status, the objective function is to minimize the system energy consumption and sludge output. Where J represents the overall optimization target, including energy consumption, sludge production and reaction efficiency, E(k) represents the total energy consumption at the kth moment, including aeration and mixing operations, and Q sludge (k) represents the sludge production at the kth moment, R eff (k) represents the reaction efficiency at the kth moment, that is, the efficiency of pollutant removal, and α, β, and γ represent weight coefficients; Perform dynamic control variable adjustments, including: Aeration intensity control, A set (k) = A meas (k)+K A ·(DO target -DO meas (k)), where A set (k) represents the aeration intensity setting value at the kth moment, A meas (k) represents the actual aeration intensity at the kth moment, K A represents the aeration control gain coefficient, DO target Indicates the target dissolved oxygen concentration, DO meas (k) represents the actual dissolved oxygen concentration at the kth moment; Reaction time control, T set (k) = T meas (k)+K T ·(R eff (k)-R target ), where T set (k) represents the reaction time setting value at the kth moment, T meas (k) represents the actual reaction time at the kth moment, K T Represents the reaction time control gain coefficient, R eff (k) represents the actual reaction efficiency at the kth moment, R target represents the target reaction efficiency; Water inlet control, V set (k) = V meas (k)+K V ·(S target -S meas (k)), where V set (k) represents the water inlet setting value at the kth moment, V meas (k) represents the actual water inflow at the kth moment, K V Indicates the water inlet control gain coefficient, S target represents the target substrate concentration, S meas (k) represents the actual substrate concentration at the kth moment; Sludge discharge cycle control, M set (k) = M meas (k)+K M ·(Q sludge (k)-Q target ), where M set (k) represents the mud discharge cycle setting value at the kth moment, M meas (k) represents the actual mud discharge cycle at the kth moment, K M It represents the gain coefficient of mud discharge cycle control, Q sludge (k) represents the sludge production at the kth moment, Q target represents the target sludge production; In step S2, the multi-stage gradient SBR reactor optimization method also includes: MPC optimization and feedback control are performed, and the state feedback control equation is X(k+1)=A·X(k)+B·U(k), where X(k+1) represents the state variables at the k+1th moment, including microbial concentration and pollutant concentration, A represents the system state transfer matrix, and B represents the influence matrix of the control input on the system state; U(k)=[A set (k),T set (k),V set (k),M set (k)], represents the control input vector, including the set values ​​of each stage; In step S3, the LSTM deep learning model is used to analyze the operating data of the multi-stage SBR reactor in real time to predict the sludge generation trend and the dynamic changes of the microbial community; Adjust the reaction conditions in combination with the multi-stage control data fed back by the MPC algorithm in step S2; In step S3, the intelligent control method for adjusting the sludge reduction process is: The real-time operation data in the reactor, including dissolved oxygen concentration, ammonia nitrogen concentration, nitrate concentration, aeration intensity, reaction time, water inflow and sludge discharge cycle, are used as input features for model training and prediction; LSTM model analysis and prediction are performed, and the memory unit captures the time dependency between various parameters during the operation of the reactor, predicting the sludge generation trend and dynamic changes of the microbial community at future time points; According to the prediction results of LSTM and the feedback data of the MPC algorithm in step S2, the control variables of aeration intensity, reaction time, water inflow and sludge discharge cycle are adjusted in real time.

2. A LSP sludge reduction biological treatment system, characterized in that: A LSP sludge reduction biological treatment method according to claim 1, comprising: Microbial community screening and enrichment module, which screens and enriches the dominant microbial communities for efficient sludge reduction, including simultaneous nitrification and denitrification (SND) and anaerobic ammonium oxidation (Anammox) microorganisms; The multi-stage SBR reactor optimization module optimizes the operating conditions of the SBR reactor based on the characteristics of the dominant microbial community enriched in the microbial community screening and enrichment module, and dynamically adapts to the metabolic needs of microorganisms; The intelligent control and regulation module uses deep learning technology to predict and adjust the operating status of the reactor in real time.

Images