Vibrating membrane bioreactor sewage treatment automatic monitoring system and method

Abstract

The invention relates to the technical field of sewage treatment and automatic control thereof, in particular to an automatic monitoring system and method for sewage treatment of a vibrating membrane bioreactor. The method comprises the following steps: monitoring key indexes in real time, and carrying out standardization processing and interpolation on sensor data to ensure the accuracy of the data; an information code generation and verification mechanism is adopted to ensure data integrity and security; a physical model and a deep learning model are combined to predict sewage treatment efficiency, and space-time relevance is analyzed through a space-time convolutional network; operating parameters of the system are optimized through a deep reinforcement learning algorithm, and the frequency, the amplitude, the aeration power and the medicament dosage of a vibrating membrane are adjusted to achieve an optimal treatment effect; through real-time adjustment of the vibrating membrane and the aeration equipment, the sewage treatment efficiency is improved; the real-time treatment state is displayed, the sewage removal rate is monitored, and an alarm mechanism is triggered under the abnormal condition. According to the method, the data security and the system optimization are considered while the processing effect is ensured.

Description

Technical Field

[0001] The invention relates to the technical field of sewage treatment and automatic control thereof, and in particular to an automatic monitoring system and method for sewage treatment by a vibrating membrane bioreactor. Background Art

[0002] With the continuous acceleration of industrialization and urbanization, the amount of sewage discharged worldwide has increased year by year. Organic pollutants, nitrogen and phosphorus compounds and other substances in sewage pose a serious threat to the aquatic ecosystem. Traditional sewage treatment technology has certain limitations in treatment efficiency, energy consumption and operating costs, and it is difficult to meet increasingly stringent environmental protection requirements. As a new sewage treatment technology, vibrating membrane bioreactor has the advantages of efficient separation, biodegradation and membrane cleaning integration, but its operation is affected by factors such as pollutant concentration fluctuations and membrane pollution, which makes it difficult to maintain a stable treatment effect.

[0003] A Chinese invention patent with announcement number CN118131631B discloses a dynamic optimization scheduling system and method based on sewage treatment, wherein the dynamic optimization scheduling system includes a data acquisition module, a data processing module, a scheduling decision module, an optimization execution module, an optimization monitoring module and a remote terminal; the data acquisition module collects various data in the sewage treatment process in real time, the data processing module processes and analyzes the data, the scheduling decision module determines the best scheduling plan based on the processed data and combined with the dynamic scheduling algorithm, and the optimization execution module controls the operation of the sewage treatment equipment according to the scheduling plan. The scheduling plan in the sewage treatment process can be adjusted and optimized in real time, and the scheduling analysis execution process and sewage treatment performance can be analyzed to ensure the subsequent sewage treatment efficiency and treatment effect while reducing the treatment cost, significantly reducing the management difficulty of management personnel, and having a high degree of intelligence and automation.

[0004] Most of the existing control methods rely on empirical rules and lack intelligent dynamic optimization mechanisms, making it difficult to achieve efficient and low-consumption operation goals. To address the above problems, the present invention combines cutting-edge technologies such as sensor networks, big data analysis, artificial intelligence, and deep reinforcement learning to propose an automatic monitoring method for wastewater treatment using a vibrating membrane bioreactor. By real-time monitoring of key indicator data, intelligent analysis of operating status, and dynamic optimization of treatment parameters, it achieves refined management of the wastewater treatment process and global performance optimization, providing technical support and theoretical basis for the practical application of vibrating membrane bioreactors in complex water treatment scenarios. Summary of the invention

[0005] The purpose of the present invention is to address the problems existing in the background technology and to propose a vibrating membrane bioreactor sewage treatment automatic monitoring system and method.

[0006] The technical solution of the present invention is a method for automatically monitoring wastewater treatment in a vibrating membrane bioreactor, comprising the following specific implementation steps: S1. Perform initialization settings, calibrate sensors, set initial operating parameters, set initial values ​​of vibration membrane amplitude and frequency, and load historical data; S2, sensor data denoising method based on wavelet transform and Kalman filter, first use wavelet decomposition to remove high-frequency noise, then suppress low-frequency drift through Kalman filter to obtain denoised data, then perform Z-score normalization on the denoised data, and use linear interpolation to supplement missing data, and then use random numbers and hash functions to generate information codes; S3. Verify data integrity and rationality by parsing information codes, and conduct intelligent analysis and optimization of sewage treatment efficiency: Combine the hybrid model of physical model and deep learning, establish a physical model based on reaction kinetics, and use the Transformer model to process time series data to generate hybrid prediction values, and optimize the model output through the spatiotemporal convolutional network, and use the deep reinforcement learning algorithm to dynamically adjust the operating parameters, including the vibration membrane frequency, amplitude, aeration power and dosage of the reagent; S4. According to the optimized vibration membrane frequency and the optimized amplitude, the amplitude and frequency are adjusted in real time, and the operating parameters of the aeration equipment and the reagent dosing equipment are adjusted according to the optimized aeration power and the optimized reagent dosage; S5. Displays real-time processing status data, sewage parameters and vibration membrane operation data, and calculates the sewage removal rate. If the sewage removal rate is less than the set threshold, the vibration membrane is considered to be seriously polluted, triggering the alarm mechanism, and providing the cause of the fault and suggested solutions.

[0007] Preferably, the implementation process of the sensor data denoising method based on wavelet transform and Kalman filtering is as follows: S21, analyze sensor noise and set sensor measurement value x measured (t)=x true (t)+n H (t)+n L (t); Among them, x measured (t) represents the sensor data collected at time t; x true (t) represents the true value; n H (t) and n L (t) represent high-frequency noise and low-frequency noise respectively; S22, sensor measurement value x measured (t) Perform multi-layer wavelet decomposition: ; Where D j(t) represents the wavelet detail component of the jth layer, corresponding to the high-frequency part of the signal; A J (t) represents the wavelet approximation component of the Jth layer, corresponding to the low-frequency part of the signal; J represents the number of layers of wavelet decomposition; S23, for each layer's detail component D j (t) Threshold processing is performed to suppress high-frequency noise. The soft threshold method is used. The formula is as follows: ; ; In the formula, represents the detail component after threshold processing; λ' represents the threshold, which is usually determined by the noise standard deviation σ n and the number of data points N: sign(D j (t)) is used to preserve the original detail component D j The sign of (t); sign() represents the sign function; S24, based on this, the processed detail components and approximate component A J (t) Reconstruct and obtain the preliminary denoised signal ; S25. Establish state space model: ; Among them, x k represents the state of the real signal at time k; z k represents the observed value, that is ; F represents the state transfer matrix; H represents the observation matrix; w k represents process noise; v k represents the observation noise; S26, Kalman filtering is divided into two steps: prediction and update: predict: ; renew: ; In the formula, Indicates the status of the prediction; Indicates the updated status; P k|k-1 and P k|k Represent the predicted and updated covariance matrices respectively; K k represents the Kalman gain; Q and R represent the covariance matrices of process noise and observation noise respectively; S27, after wavelet transform and Kalman filtering, the denoised signal is obtained .

[0008] Preferably, the information code generation process is as follows: S31, select a random number k∈[1,q-1], and calculate the first-order auxiliary information code parameter fi1 =(Cg×k mod p) modq; Where q is a 160-bit prime number, p is a 1024-bit prime number, and satisfies q|p-1; Cg is the auxiliary information code, Cg=[r(p-1) / q] mod p, and satisfies Cg>1; r is the A number selected from the above; q|p-1 means that q is a factor of p-1, that is, p-1 can be divided by q; S32, calculate the second-order auxiliary information code parameter fi 2 ={(k-1)×[H(BX t )+Pg×fi 1 ]} mod q; Where H() represents a hash function; BX t Represents the processed key indicator data X t The binary string, BX t =COD real ||BOD real ;COD real Indicates key indicator data X t COD data in; BOD real Indicates key indicator data X t BOD data; || indicates cascade operation; Pg indicates randomly selecting information code generation parameters, based on which information code parsing parameters Pa = Cg × Pg mod p are calculated, and {p, q, Cg, Pa} is made public, where Pg is an integer in [1, q-1]; S33, if 2 =0, then fi 2 -1 mod q does not exist, then return to S31; S34, generate information code IC={fi 1 ,fi 2}.

[0009] Preferably, the verification process of verifying data integrity and rationality by parsing the information code is as follows: S41, extracting parameters {prime number p, prime number q, auxiliary information code Cg, information code parsing parameter Pa}; S42, judgement fi 1 、fi 2 Is it an integer between [1,q-1]? If yes, calculate the parsing code Ca=fi 2 -1 modq; if not true, a warning is given to the user; S43, calculating the following auxiliary analysis parameters: Pp 1 =[H(BX t) × Ca] mod q; Pp 2 =(fi 1 ×Ca) mod q; Where H() represents a hash function; BX t Indicates the key indicator data X of the verification t The binary string, BX t =COD real ||BOD real ;COD real Indicates the key indicator data X of the verification t COD data in; BOD real Indicates the key indicator data X of the verification t BOD data; || indicates cascade operation; S44. Calculate the test factor Fa = (Cg × Pp 1 ×Pa×Pp 2 mod p) mod q; S45. If Fa=fi 1 , the verification passes; if not, a warning is given to the user.

[0010] Preferably, the analysis and optimization process of intelligent analysis and optimization of sewage treatment efficiency is as follows: S51. Based on key indicator data X t , build an adaptive hybrid model AHM, integrate physical model and deep learning model, predict key indicator data of sewage treatment, and output time series data y t ; S52, introduce the spatiotemporal correlation analysis model, and convert the time series data y output by the hybrid model into t Combined with the spatial distribution information of the sensor, the spatiotemporal analysis results are output; S53. Based on the results of spatiotemporal analysis, a deep reinforcement learning algorithm is used to optimize the operating parameters of sewage treatment to achieve the optimal global goal.

[0011] Preferably, the time series data y t The output process is as follows: S61. Establish a physical model based on the process mechanism according to the reaction kinetics formula: ; In the formula, y phy,t represents the physical prediction value, that is, the concentration prediction value of the target pollutant at a certain time point t; k represents the reaction rate constant; C reactant,t represents the concentration of reactants; n represents the reaction order; S62. Use the deep learning model Transformer to process time series data: ; In the formula, f NN () represents the neural network prediction function; X t represents the input data vector, i.e., the sensor time series; θ represents the neural network parameters; S63, combining the physical model and the neural network model, and generating a hybrid prediction value y by dynamically adjusting the weights t : ; In the formula, α and β represent weight coefficients, which are dynamically adjusted through Bayesian optimization.

[0012] Preferably, the time series data y output by the hybrid model t The combination process with the spatial distribution information of the sensor is as follows: S71. Define the spatial correlation strength matrix A between sensors based on the actual sewage flow path and sensor layout: ; In the formula, A ij represents the association strength between sensors i and j; d ij represents the physical distance between sensors; σ' represents the adjustment parameter; S72. Combining time series and spatial distribution characteristics, constructing a spatiotemporal convolutional network ST , that is, a hybrid model including temporal attention and spatial convolution: ; In the formula, Indicates the results of spatiotemporal analysis; X t Indicates the processed key indicator data.

[0013] Preferably, the optimization process of optimizing the operating parameters of sewage treatment using a deep reinforcement learning algorithm is as follows: S81. Define state, action and reward functions: Status t :Current sewage treatment status, key indicator data, including: chemical oxygen demand, biological oxygen demand COD and BOD; Action a t : The adjusted operating parameter vector includes: vibration membrane frequency, amplitude, aeration power and dosage of reagent; Reward function R: ; In the formula, Eff(x) represents the processing efficiency; Energy(x) represents the energy consumption; Cost(x) represents the operating cost; S82. Use the deep deterministic policy gradient algorithm to generate an optimization strategy: ; In the formula, π represents the optimization strategy function; Indicates that in state s t Next select action a t The probability of; E[] represents the expected value, which is used to quantify the average value of the reward that a certain strategy may obtain after selecting an action in the current state; Indicates that in the current state s t Next, perform action a t , the expected value of the reward R obtained; θ' represents the policy parameters, which are optimized through gradient updates.

[0014] The technical solution of the present invention is: a vibrating membrane bioreactor wastewater treatment automatic monitoring system, which is used to execute the above-mentioned vibrating membrane bioreactor wastewater treatment automatic monitoring method, comprising: Vibration membrane control module, used to adjust the amplitude, frequency and operation time of the vibration membrane to optimize the sewage filtration efficiency; A multi-parameter sensor network for monitoring parameters in sewage, including chemical oxygen demand (COD) and biological oxygen demand (BOD); Data processing and transmission module, used to collect and process sensor data, and transmit the data to the intelligent analysis and control module using industrial Internet of Things technology; Intelligent analysis and control module, used to dynamically adjust the working parameters of the diaphragm based on artificial intelligence algorithms; The user interface and alarm module is used to provide a remote operation interface to achieve visual monitoring of the processing status and alarm of abnormal status.

[0015] Compared with the prior art, the above technical solution of the present invention has the following beneficial technical effects: The present invention designs a vibrating membrane bioreactor sewage treatment automatic monitoring system and method, which significantly improves the efficiency, accuracy and economy of sewage treatment through the combination of multi-layer data processing, dynamic optimization and intelligent control technology, and has wide application value and significant technical advantages: (1) Realize real-time monitoring and optimization of the entire sewage treatment process: Through a distributed multi-parameter sensor network, key indicator data (such as COD and BOD) in the sewage treatment process are collected in real time, and combined with data processing methods such as data standardization, missing data cleaning and interpolation, the integrity and accuracy of key data are ensured, providing a reliable basis for system optimization; (2) Ensuring data integrity and reliability: The proposed information code generation and verification mechanism is based on multi-layer hash functions and random number generation technology to encrypt and verify the processed data; this mechanism can effectively prevent data from being tampered with during transmission and ensure the integrity and rationality of key data; (3) Improving sewage treatment efficiency and prediction accuracy: The system constructs an adaptive hybrid model (AHM) that combines a physical model and a deep learning model, dynamically adjusts weights to generate prediction results, and further explores the global correlation of sewage treatment through a spatiotemporal analysis model. This method takes into account the accuracy of the process mechanism and the adaptive ability of deep learning, improving the accuracy of prediction and the global optimization effect. (4) Realize intelligent optimization of sewage treatment parameters: Introduce the deep reinforcement learning (DRL) algorithm, define the state, action and reward function, and dynamically optimize the frequency, amplitude, aeration power and dosage of the vibration membrane according to the comprehensive goals of treatment efficiency, energy consumption and cost to ensure the efficiency and economy of the system operation; (5) Provide flexible and reliable closed-loop control: Through the optimized parameters, the vibrating membrane control module, aeration equipment and reagent dosing equipment are adjusted in real time to achieve closed-loop control of the sewage treatment process, which can quickly respond to changes in different sewage concentrations and operating conditions, ensuring the stability and efficiency of the treatment process; (6) Enhanced system visualization and alarm capabilities: Real-time display of treatment status, key indicators and operating parameters, and calculation of sewage removal rate; when the treatment efficiency is lower than the set threshold, an alarm is triggered in time and cause analysis and solution suggestions are provided, thus improving system security and user experience; (7) Energy saving and environmental protection benefits: Through the dynamic adjustment of parameters such as the vibration membrane frequency, amplitude, and aeration power, the energy consumption of sewage treatment is finely controlled, and the operating cost is reduced to the greatest extent while ensuring the treatment effect, which has a significant energy-saving and environmental protection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a system architecture diagram of an automatic monitoring system for wastewater treatment using a vibrating membrane bioreactor proposed by the present invention; Figure 2 This is a method flow chart of an automatic monitoring method for wastewater treatment using a vibrating membrane bioreactor proposed by the present invention. DETAILED DESCRIPTION

[0017] Embodiment 1, as Figure 1 As shown, the vibrating membrane bioreactor wastewater treatment automatic monitoring system proposed in the present invention includes: a vibrating membrane control module, a multi-parameter sensor network, a data processing and transmission module, an intelligent analysis and control module, and a user interface and alarm module.

[0018] The vibration membrane control module is used to adjust the amplitude, frequency and operation time of the vibration membrane to optimize the sewage filtration efficiency; A multi-parameter sensor network monitors parameters in sewage, including but not limited to COD (chemical oxygen demand), BOD (biological oxygen demand); The data processing and transmission module collects and processes sensor data and transmits the data to the intelligent analysis and control module using the Industrial Internet of Things (IIoT) technology; The intelligent analysis and control module dynamically adjusts the working parameters of the diaphragm based on artificial intelligence algorithms; The user interface and alarm module provide a remote operation interface to achieve visual monitoring of the processing status and alarm of abnormal status.

[0019] Embodiment 2, as Figure 2 As shown, the vibrating membrane bioreactor wastewater treatment automatic monitoring method proposed in the present invention is applied to the vibrating membrane bioreactor wastewater treatment automatic monitoring system proposed in Example 1, and its specific implementation steps are as follows: S1. System initialization, start the multi-parameter sensor network and vibration membrane control module, calibrate all sensors, including but not limited to COD, BOD, set initial operating parameters, including but not limited to the vibration membrane amplitude A 初始 and frequency f 初始 , and load historical data.

[0020] S2. Real-time monitoring of key indicator data in the sewage treatment process through a distributed multi-parameter sensor network, including but not limited to: Chemical oxygen demand (COD) real , Biological oxygen demand BOD real , and perform the following processing through the data processing and transmission module: S21, based on the joint denoising method of wavelet transform and Kalman filter, high-frequency noise is removed by wavelet transform and low-frequency drift is suppressed by Kalman filter, which realizes comprehensive noise filtering, ensures the accuracy and reliability of sensor data, and provides a solid data foundation for the efficient operation of the system. The specific implementation process is as follows: S2101, analyze sensor noise and set sensor measurement value x measured (t)=x true (t)+n H (t)+n L (t); Among them, x measured (t) represents the sensor data collected at time t; x true (t) represents the true value; n H (t) and n L (t) respectively represent high-frequency noise (introduced by, but not limited to, electrical interference and environmental vibration, with a higher frequency) and low-frequency noise (caused by sensor aging or environmental changes, with a lower frequency); S2102, sensor measurement value x measured (t) Perform multi-layer wavelet decomposition: ; Where Dj (t) represents the wavelet detail component of the jth layer, corresponding to the high-frequency part of the signal; A J (t) represents the wavelet approximation component of the Jth layer, corresponding to the low-frequency part of the signal; J represents the number of layers of wavelet decomposition, and the appropriate number of layers is selected according to the frequency characteristics of the signal; For each layer’s detail component D j (t) Threshold processing is performed to suppress high-frequency noise. The soft threshold method is used. The formula is as follows: ; ; In the formula, represents the detail component after threshold processing; λ' represents the threshold, which is usually determined by the noise standard deviation σ n and the number of data points N: sign(D j (t)) is used to preserve the original detail component D j The sign (positive or negative) of (t) ensures the directionality of the denoised data (i.e., positive numbers remain positive and negative numbers remain negative); sign() represents the sign function; According to this, the processed detail components and approximate component A J (t) Reconstruct and obtain the preliminary denoised signal ; S2103. Establish state space model: ; Among them, x k represents the state of the real signal at time k; z k represents the observed value, that is ; F represents the state transfer matrix; H represents the observation matrix; w k represents process noise; v k represents the observation noise; Kalman filtering is divided into two steps: prediction and update: predict: ; renew: ; In the formula, Indicates the status of the prediction; Indicates the updated status; P k|k-1 and P k|k Represent the predicted and updated covariance matrices respectively; K k represents the Kalman gain, which is used to balance the weight of prediction and observation; Q and R represent the covariance matrix of process noise and observation noise respectively; Based on this: After wavelet transform and Kalman filtering, the denoised signal is obtained ; S22. Since the data units collected by different sensors are different and the dimensional differences may affect the subsequent analysis results, data standardization is required. The standardized data have the same scale so that different data can be compared and integrated. The standardization formula uses the Z-score standardization method, that is: ; In the formula, x represents the denoised signal ; μ represents the historical mean value of the real-time data x collected by the sensor, that is, the average value of the real-time data x collected by the sensor within a certain time range; σ represents the historical standard deviation of the real-time data x collected by the sensor; x 标准化 Represents the normalized sensor data; S23. To ensure the integrity of the data, the missing data were cleaned and interpolated, and the linear interpolation method was used to supplement them; Assuming that the data changes are linear, interpolation is performed through adjacent valid data points. The linear interpolation formula is: ; In the formula, x t represents the supplementary interpolation at time t; x t-1 and x t+1 Represent the standardized data at time t-1 and t+1 respectively; S24. To ensure that the intelligent analysis and control module can timely discover the missing or unreasonable key indicator data, an information code IC is generated for the processed key indicator data. The generation process is as follows: S2401, select a random number k∈[1,q-1], calculate the first-order auxiliary information code parameter fi 1 =(Cg×k mod p)mod q; Where q is a 160-bit prime number, p is a 1024-bit prime number, and satisfies q|p-1; Cg is the auxiliary information code, Cg=[r(p-1) / q] mod p; r is the A number selected from the above, and satisfying Cg>1; based on this, the information code generation parameter Pg is randomly selected, the information code parsing parameter Pa=Cg×Pg mod p is calculated, and {p,q,Cg,Pa} is made public to the intelligent analysis and control module, where Pg is an integer in [1,q-1]; q|p-1 means that q is a factor of p-1, that is, p-1 is divisible by q; S2402, calculate the second-order auxiliary information code parameter fi 2 ={(k-1)×[H(BX t )+Pg×fi 1 ]} mod q; Where H() represents a hash function; BXt Represents the processed key indicator data X t The binary string, BX t =COD real ||BOD real ;COD real Indicates key indicator data X t COD data in; BOD real Indicates key indicator data X t BOD data; || indicates cascade operation; S2403, if 2 =0, then fi 2 -1 mod q does not exist, then return S2401; S2404, generate information code IC={fi 1 ,fi 2}; S25, {IC, X t}Transmitted to the intelligent analysis and control module.

[0021] S3, the intelligent analysis and control module uses data-driven dynamic analysis and optimization strategies, combined with the actual operation data of the vibrating membrane bioreactor, to perform intelligent analysis and optimization of sewage treatment efficiency. The specific implementation process is as follows: S31, parsing information code IC={fi 1 ,fi 2}, check the processed data X t The integrity and rationality of the verification process are as follows: S3101, extracting parameters {prime number p, prime number q, auxiliary information code Cg, information code parsing parameter Pa}; S3102, judge fi 1 、fi 2 Is it an integer between [1,q-1]? If yes, calculate the parsing code Ca=fi 2 -1mod q; if not established, an alarm is given to the user interface and alarm module; S3103, calculate the following auxiliary analysis parameters: Pp 1 =[H(BX t ) × Ca] mod q; Pp 2 =(fi 1 ×Ca) mod q; Where H() represents a hash function; BX t Indicates the received key indicator data X t The binary string, BX t =CODreal ||BOD real ;COD real Indicates the received key indicator data X t COD data in; BOD real Indicates the received key indicator data X t BOD data; || indicates cascade operation; S3104, calculate the test factor Fa = (Cg × Pp 1 ×Pa×Pp 2 mod p) mod q; S3105, if Fa=fi 1 , the verification is passed, and step S32 is executed; if not, an alarm is given to the user interface and the alarm module; S32, based on the received key indicator data X t , construct an adaptive hybrid model AHM, comprehensively utilize the advantages of physical models and deep learning models, and predict key indicator data of sewage treatment (including but not limited to COD and BOD). The specific implementation process is as follows: S3201. Establish a physical model based on the process mechanism according to the reaction kinetics formula: ; In the formula, y phy,t represents the physical prediction value, that is, the concentration prediction value of the target pollutant (including but not limited to COD and BOD) at a certain time point t; k represents the reaction rate constant; C reactant,t represents the concentration of reactants; n represents the reaction order; S3202, using deep learning model Transformer to process time series data: ; In the formula, f NN () represents the neural network prediction function; X t represents the input data vector, i.e., the sensor time series; θ represents the neural network parameters; S3203, combining the physical model and the neural network model, and generating a hybrid prediction value y by dynamically adjusting the weights t : ; In the formula, α and β represent weight coefficients, which are dynamically adjusted through Bayesian optimization; S33, in order to improve the global accuracy of the prediction, the spatiotemporal correlation analysis model is introduced to convert the time series data y output by the hybrid model into t Combined with the spatial distribution information of the sensor, the global correlation in the sewage treatment process is further explored. The specific implementation process is as follows: S3301. Define the spatial correlation strength matrix A between sensors based on the actual sewage flow path and sensor layout: ; In the formula, A ij represents the association strength between sensors i and j; d ij represents the physical distance between sensors; σ' represents the adjustment parameter; S3302, combining time series and spatial distribution characteristics to build a spatiotemporal convolutional network ST , that is, a hybrid model including temporal attention and spatial convolution: ; In the formula, Indicates the results of spatiotemporal analysis; X t Represents the processed data; S34. Based on the results of spatiotemporal analysis, the deep reinforcement learning (DRL) algorithm is used to optimize the operating parameters of sewage treatment to achieve the optimal global goal. The specific implementation process is as follows: S3401. Define state, action and reward functions: Status t : Current sewage treatment status, key indicator data (including but not limited to COD, BOD); Action a t : vector of adjusted operating parameters (including but not limited to diaphragm frequency or amplitude); Reward function R: ; In the formula, Eff(x) represents the processing efficiency; Energy(x) represents the energy consumption; Cost(x) represents the operating cost; S3402. Use the Deep Deterministic Policy Gradient (DDPG) algorithm to generate an optimization strategy: ; In the formula, π represents the optimization strategy function; Indicates that in state s t Next select action a t The probability of; E[] represents the expected value, which is used to quantify the average value of the reward that a certain strategy may obtain after selecting an action in the current state; Indicates that in the current state s t Next, perform action a t When , the expected value of the reward R is obtained; θ' represents the policy parameter, which is optimized by gradient update; S35. According to the optimization strategy output by reinforcement learning, dynamically adjust the system operating parameters and continuously optimize through closed-loop control. The specific implementation process is as follows: S3501. Dynamically set the frequency f of the diaphragm according to the optimized output vib and amplitude A vib , improve processing efficiency: ; ; In the formula, f vib,opt represents the optimized vibration membrane frequency; f base represents the default vibration membrane frequency; Δf represents the frequency adjustment value, which is generated by the optimization action output by reinforcement learning; A vib,opt Represents the optimized amplitude; A base represents the default vibration membrane amplitude; ΔA represents the amplitude adjustment value, which is generated by the optimization action output by reinforcement learning; S3502, real-time adjustment of aeration power P air To balance energy consumption and dissolved oxygen concentration: ; Where P vib,opt Represents the optimized aeration power; P base Indicates the default aeration power; ΔP indicates the aeration power adjustment value; S3503, dynamically adjust the optimized M according to the sewage concentration chem : ; Where M chem,opt represents the optimized dosage of the reagent, that is, the dosage of the reagent that needs to be added at the current sewage concentration to ensure that the chemical reaction of sewage treatment achieves the optimal effect; k chem Represents the chemical reaction coefficient, that is, the amount of drug required per unit concentration of reactant; C reactant Indicates the concentration of reactants (including but not limited to COD and BOD) in sewage; S36, the optimized vibration membrane frequency f vib,opt , optimized amplitude A vib,opt , optimized aeration power P vib,opt And the optimized dosage of the agent M chem Transmitted to the diaphragm control module and the user interface and alarm module.

[0022] S4, the vibration membrane control module is based on the optimized vibration membrane frequency f vib,opt And the optimized amplitude A vib,opt , adjust the amplitude and frequency in real time, and according to the optimized aeration power P vib,opt And the optimized dosage of the agent M chem Adjust the operating parameters of aeration equipment and chemical dosing equipment to ensure the coordination and efficiency of the sewage treatment process.

[0023] S5, the user interface and alarm module displays real-time processing status data, sewage parameters and vibration membrane operation data, and calculates the sewage removal rate Rr: ; In the formula, COD real Indicates COD data in real-time processing status data; BOD real Indicates BOD data in real-time processing status data; COD initial Indicates the initial COD value, that is, the COD concentration when the sewage enters the reactor; BOD initial It indicates the initial BOD value, that is, the BOD concentration when the sewage enters the reactor; If |Rr| is less than the set threshold, the vibration membrane is considered to be seriously contaminated and the alarm mechanism is triggered. The alarm is notified through multiple channels, including but not limited to SMS, APP push and sound and light signals, and the cause of the fault and suggested solutions are provided.

[0024] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited thereto, and various changes can be made within the knowledge scope of technicians in the relevant technical field without departing from the purpose of the present invention.

Claims

1. A vibrating membrane bioreactor wastewater treatment automatic monitoring method, characterized in that: The specific implementation steps include the following: S1. Perform initialization settings, calibrate sensors, set initial operating parameters, set initial values ​​of vibration membrane amplitude and frequency, and load historical data; S2, sensor data denoising method based on wavelet transform and Kalman filter, first use wavelet decomposition to remove high-frequency noise, then suppress low-frequency drift through Kalman filter to obtain denoised data, then perform Z-score normalization on the denoised data, and use linear interpolation to supplement missing data, and then use random numbers and hash functions to generate information codes; S3. Verify data integrity and rationality by parsing information codes, and conduct intelligent analysis and optimization of sewage treatment efficiency: Combine the hybrid model of physical model and deep learning, establish a physical model based on reaction kinetics, and use the Transformer model to process time series data to generate hybrid prediction values, and optimize the model output through the spatiotemporal convolutional network, and use the deep reinforcement learning algorithm to dynamically adjust the operating parameters, including the vibration membrane frequency, amplitude, aeration power and dosage of the reagent; S4. According to the optimized vibration membrane frequency and the optimized amplitude, the amplitude and frequency are adjusted in real time, and the operating parameters of the aeration equipment and the reagent dosing equipment are adjusted according to the optimized aeration power and the optimized reagent dosage; S5. Displays real-time processing status data, sewage parameters and vibration membrane operation data, and calculates the sewage removal rate. If the sewage removal rate is less than the set threshold, the vibration membrane is considered to be seriously polluted, triggering the alarm mechanism, and providing the cause of the fault and suggested solutions.

2. The method for automatically monitoring wastewater treatment in a vibrating membrane bioreactor according to claim 1, characterized in that: The implementation process of the sensor data denoising method based on wavelet transform and Kalman filtering is as follows: S21, analyze sensor noise and set sensor measurement value x measured (t)=x true (t)+n H (t)+n L (t); Among them, x measured (t) represents the sensor data collected at time t; x true (t) represents the true value; n H (t) and n L (t) represent high-frequency noise and low-frequency noise respectively; S22, sensor measurement value x measured (t) Perform multi-layer wavelet decomposition: ； Where D j (t) represents the wavelet detail component of the jth layer, corresponding to the high-frequency part of the signal; A J (t) represents the wavelet approximation component of the Jth layer, corresponding to the low-frequency part of the signal; J represents the number of layers of wavelet decomposition; S23, for each layer's detail component D j (t) Threshold processing is performed to suppress high-frequency noise. The soft threshold method is used. The formula is as follows: ； ； In the formula, represents the detail component after threshold processing; λ' represents the threshold, which is usually composed of the noise standard deviation σ n and the number of data points N: sign(D j (t)) is used to preserve the original detail component D j The sign of (t); sign() represents the sign function; S24, based on this, the processed detail components and approximate component A J (t) Reconstruct and obtain the preliminary denoised signal ; S25. Establish state space model: ; Among them, x k represents the state of the real signal at time k; z k represents the observed value, that is ; F represents the state transfer matrix; H represents the observation matrix; w k represents process noise; v k represents the observation noise; S26, Kalman filtering is divided into two steps: prediction and update: predict: ; renew: ; In the formula, Indicates the status of the prediction; Indicates the updated status; P k|k-1 and P k|k Represent the predicted and updated covariance matrices respectively; K k represents the Kalman gain; Q and R represent the covariance matrices of process noise and observation noise respectively; S27, after wavelet transform and Kalman filtering, the denoised signal is obtained .

3. The method for automatic monitoring of wastewater treatment by a vibrating membrane bioreactor according to claim 1, characterized in that: The process of generating the information code is as follows: S31, select a random number k∈[1,q-1], and calculate the first-order auxiliary information code parameter fi1=(Cg×k mod p) mod q; Where q is a 160-bit prime number, p is a 1024-bit prime number, and satisfies q|p-1; Cg is the auxiliary information code, Cg=[r(p-1) / q] mod p, and satisfies Cg>1; r is the A number selected from the above; q|p-1 means that q is a factor of p-1, that is, p-1 can be divided by q; S32, calculate the second-order auxiliary information code parameter fi2={(k-1)×[H(BX t )+Pg×fi1]} mod q; Where H() represents a hash function; BX t Represents the processed key indicator data X t The binary string, BX t =COD real ||BOD real ;COD real Indicates key indicator data X t COD data in; BOD real Indicates key indicator data X t BOD data; || indicates cascade operation; Pg indicates randomly selecting information code generation parameters, based on which information code parsing parameters Pa = Cg × Pg mod p are calculated, and {p, q, Cg, Pa} is made public, where Pg is an integer in [1, q-1]; S33. If fi2=0, fi2-1 mod q does not exist, and the process returns to S31; S34. Generate information code IC={fi1, fi2}.

4. The method for automatically monitoring wastewater treatment in a vibrating membrane bioreactor according to claim 1, characterized in that: The verification process of verifying data integrity and rationality by parsing the information code is as follows: S41, extracting parameters {prime number p, prime number q, auxiliary information code Cg, information code parsing parameter Pa}; S42, determine whether fi1 and fi2 are integers between [1, q-1]. If yes, calculate the parsing code Ca=fi2-1 mod q; if no, warn the user; S43, calculating the following auxiliary analysis parameters: Pp1=[H(BX t )×Ca] mod q; Pp2 = (fi1 × Ca) mod q; Where H() represents a hash function; BX t Indicates the key indicator data X of the verification t The binary string, BX t =COD real ||BOD real ;COD real Indicates the key indicator data X of the verification t COD data in; BOD real Indicates the key indicator data X of the verification t BOD data; || indicates cascade operation; S44. Calculate the test factor Fa = (Cg × Pp1 × Pa × Pp2 mod p) mod q; S45. If Fa=fi1, the verification is passed; if not, a warning is given to the user.

5. The method for automatically monitoring wastewater treatment in a vibrating membrane bioreactor according to claim 1, characterized in that: The analysis and optimization process of intelligent analysis and optimization of sewage treatment efficiency is as follows: S51. Based on key indicator data X t , build an adaptive hybrid model AHM, integrate physical model and deep learning model, predict key indicator data of sewage treatment, and output time series data y t ; S52, introduce the spatiotemporal correlation analysis model, and convert the time series data y output by the hybrid model into t Combined with the spatial distribution information of the sensor, the spatiotemporal analysis results are output; S53. Based on the results of spatiotemporal analysis, a deep reinforcement learning algorithm is used to optimize the operating parameters of sewage treatment to achieve the optimal global goal.

6. The method for automatically monitoring wastewater treatment in a vibrating membrane bioreactor according to claim 5, characterized in that: Time series data y t The output process is as follows: S61. Establish a physical model based on the process mechanism according to the reaction kinetics formula: ； In the formula, y phy,t represents the physical prediction value, that is, the concentration prediction value of the target pollutant at a certain time point t; k represents the reaction rate constant; C reactant,t represents the concentration of reactants; n represents the reaction order; S62. Use the deep learning model Transformer to process time series data: ； In the formula, f NN () represents the neural network prediction function; X t represents the input data vector, i.e., the sensor time series; θ represents the neural network parameters; S63, combining the physical model and the neural network model, and generating a hybrid prediction value y by dynamically adjusting the weights t : ； In the formula, α and β represent weight coefficients, which are dynamically adjusted through Bayesian optimization.

7. The method for automatically monitoring wastewater treatment in a vibrating membrane bioreactor according to claim 5, characterized in that: The time series data y output by the mixed model t The combination process with the spatial distribution information of the sensor is as follows: S71. Define the spatial correlation strength matrix A between sensors based on the actual sewage flow path and sensor layout: ； In the formula, A ij represents the association strength between sensors i and j; d ij represents the physical distance between sensors; σ' represents the adjustment parameter; S72. Combining time series and spatial distribution characteristics, constructing a spatiotemporal convolutional network ST , that is, a hybrid model including temporal attention and spatial convolution: ； In the formula, Indicates the results of spatiotemporal analysis; X t Indicates the processed key indicator data.

8. The method for automatically monitoring wastewater treatment in a vibrating membrane bioreactor according to claim 5, characterized in that: The optimization process of optimizing the operating parameters of sewage treatment using deep reinforcement learning algorithm is as follows: S81. Define state, action and reward functions: Status t :Current sewage treatment status, key indicator data, including: chemical oxygen demand, biological oxygen demand COD and BOD; Action a t : The adjusted operating parameter vector includes: vibration membrane frequency, amplitude, aeration power and dosage of reagent; Reward function R: ； In the formula, Eff(x) represents the processing efficiency; Energy(x) represents the energy consumption; Cost(x) represents the operating cost; S82. Use the deep deterministic policy gradient algorithm to generate an optimization strategy: ； In the formula, π represents the optimization strategy function; Indicates that in state s t Next select action a t The probability of ; E[ ] represents the expected value, which is used to quantify the average value of the reward that a strategy may obtain after selecting an action in the current state; Indicates that in the current state s t Next, perform action a t , the expected value of the reward R obtained; θ' represents the policy parameters, which are optimized through gradient updates.

9. An automatic monitoring system for wastewater treatment in a vibrating membrane bioreactor, which is used to execute an automatic monitoring method for wastewater treatment in a vibrating membrane bioreactor according to any one of claims 1 to 8, characterized in that: include: Vibration membrane control module, used to adjust the amplitude, frequency and operation time of the vibration membrane to optimize the sewage filtration efficiency; A multi-parameter sensor network for monitoring parameters in sewage, including chemical oxygen demand (COD) and biological oxygen demand (BOD); Data processing and transmission module, used to collect and process sensor data, and transmit the data to the intelligent analysis and control module using industrial Internet of Things technology; Intelligent analysis and control module, used to dynamically adjust the working parameters of the diaphragm based on artificial intelligence algorithms; The user interface and alarm module is used to provide a remote operation interface to achieve visual monitoring of the processing status and abnormal status alarm.

Images