A monitoring system and method for wastewater treatment processes

Abstract

This application relates to the field of wastewater treatment technology and discloses a monitoring system and method for wastewater treatment processes. The system includes a multi-parameter real-time monitoring module, a data fusion and analysis module, an automated dosing control module, and a feedback optimization module. The method is applied to this system. The monitoring system and method for wastewater treatment processes of this application, through the synergistic effect of the multi-parameter real-time monitoring module, data fusion and analysis module, automated dosing control module, and closed-loop feedback optimization module, achieves dynamic and precise monitoring of the wastewater treatment process. This solves the problems of single-parameter limitations, control lag, and high energy consumption in existing technologies, thereby improving the effluent compliance rate while reducing reagent costs and equipment maintenance costs.

Description

TECHNICAL FIELD

[0001] The application relates to the sewage treatment technology field, and particularly relates to a monitoring system and method for a sewage treatment process. BACKGROUND

[0002] At present, the monitoring system for the sewage treatment process has the following technical problems:

[0003] Single real-time monitoring parameter: the existing system usually only monitors a few water quality parameters such as COD and pH, and cannot comprehensively reflect the dynamic changes of sewage components (such as heavy metals and suspended solids) and process parameters such as flow and flow rate, resulting in low dosing control precision.

[0004] Dosing control depends on fixed threshold: the traditional system uses a fixed threshold (such as fixed proportion dosing when COD exceeds the threshold), which cannot adapt to water quality mutations or flow fluctuations, resulting in waste of reagents or substandard treatment effect.

[0005] Lack of dynamic optimization and closed-loop control: the existing algorithm (such as PID control) does not dynamically adjust the dosing strategy combined with machine learning, and does not form a closed-loop feedback mechanism, resulting in a lag in the system's response to abnormal working conditions (such as instantaneous increase in COD) and high energy consumption.

[0006] High energy consumption and equipment maintenance cost: the dosing pump has a fixed start-stop frequency, and the start-stop strategy is not optimized according to real-time working conditions, resulting in accelerated equipment wear and tear, high maintenance cost, and no dynamic constraint on reagent dosage, which poses a risk of secondary pollution.

[0007] These problems make it difficult for the existing technology to achieve precise control, energy saving and consumption reduction, and long-term stable operation of the sewage treatment. SUMMARY

[0008] The application aims to provide a monitoring system and method for a sewage treatment process to solve the technical problems in the background.

[0009] To achieve the above-mentioned purpose, the application discloses the following technical solutions:

[0010] In a first aspect, the application discloses a monitoring system for a sewage treatment process, comprising:

[0011] A multi-parameter real-time monitoring module comprising a water quality sensor group and a flow monitoring device, the water quality sensor group being used to collect real-time chemical oxygen demand, biochemical oxygen demand, pH value, heavy metal concentration and suspended solids concentration data of sewage; the flow monitoring device being used to collect real-time flow, flow rate and temperature data of sewage;

[0012] a data fusion and analysis module configured to receive feedback data from the multi-parameter real-time monitoring module and perform dynamic correlation analysis on water quality parameters through a machine learning algorithm to generate a dosing control instruction;

[0013] an automated dosing control module configured to dose one or more of a coagulant, a flocculant, or a disinfectant into the sewage treatment system through a dosing pump and a flow regulating valve according to the dosing control instruction and adjust the dosage in real time;

[0014] a feedback optimization module configured to perform iterative optimization on the dosing control instruction through an adaptive algorithm based on the feedback data from the multi-parameter real-time monitoring module to form a closed-loop control.

[0015] Preferably, the machine learning algorithm comprises a space-time coupling regression model, a dosing amount optimization model, and a fuzzy PID control algorithm.

[0016] The space-time coupling regression model is configured to extract features of the space-time distribution of chemical oxygen demand, biochemical oxygen demand, and pH value in the water quality parameters through a convolutional neural network, and predict the water quality trend in the future time by combining a long short-term memory network, and its expression is as follows:

[0017]

[0018] wherein X t is an input data tensor of time step t, including feature data extracted from the space-time distribution of chemical oxygen demand, biochemical oxygen demand, and pH value in the water quality parameters; f CNN is a convolutional neural network configured to extract local features in the spatial dimension; f LSTM is a long short-term memory network configured to model dynamic changes in the time dimension; is a predicted water quality parameter value at future time t+Δt;

[0019] The dosing amount optimization model is configured to calculate the dosing amount based on the multi-parameter nonlinear relationship, and the calculation formula of the dosing amount is as follows:

[0020]

[0021] wherein Q t is the dosing amount at time t; α i , β i , and γ i are time-varying coefficients trained through historical data, and correspond to the weights of chemical oxygen demand, biochemical oxygen demand, and pH value, respectively; COD i , BOD i , and pH iCorresponding to chemical oxygen demand, biochemical oxygen demand and pH value respectively, λ is the decay coefficient 0.05≤λ≤1, which is used to reflect the attenuation trend of water quality parameters with time; n is the number of water quality parameters participating in the calculation;

[0022] The proportional coefficient K of the fuzzy PID control algorithm p , the integral time T i and the differential time T d According to the real-time flow Q 流量 and the flow rate v, adaptive adjustment is carried out; the specific adjustment formula is:

[0023]

[0024] T d =T d0 ·(1+μ·ΔQ)

[0025] Wherein, K p0 , T i0 and T d0 are the initial proportional coefficient, the initial integral time and the initial differential time respectively, Q 阈值 and v 阈值 are the preset flow threshold and flow rate threshold, μ is the proportional coefficient gain factor, and ΔQ is the dosage change amount.

[0026] As preferred, the space-time coupling regression model is also used to compensate the time delay and noise of the sensor data by Kalman filtering algorithm.

[0027] As preferred, the automatic dosing control module further comprises a multi-stage dosing distributor and a medicament concentration correction unit.

[0028] The multi-stage dosing distributor is used to realize segmented dosing of medicament through pressure sensor and electromagnetic valve, adjust the dosing position of medicament according to the depth of sewage, and optimize the flocculation effect through dynamic mixing ratio adjustment. The dynamic mixing ratio formula is: P r =1:(1+k·Conc 悬浮物 ·sin(ω·t)), wherein P r is the calculated mixing ratio, k is the empirical coefficient, Conc 悬浮物 is the suspended solids concentration, ω is the adjustment frequency, and ω=2π·0.5Hz.

[0029] The medicament concentration correction unit is used to correct the dosing ratio of medicament in real time through ion selective electrode according to the heavy metal concentration data, and the correction formula is: Wherein, C 校正 is the corrected medicament concentration, C 初始 is the initial medicament concentration, Conc 重金属 is the heavy metal concentration data, and Conc重金属阈值 is a preset heavy metal concentration threshold value.

[0030] As preferred, the feedback optimization module comprises an abnormal working condition identification unit and an energy consumption optimization submodule, the abnormal working condition identification unit detects mutation or abnormal fluctuation of water quality parameters through cluster analysis, and triggers an emergency dosing mode when detecting that COD instantaneously increases by more than 20%, and the energy consumption optimization submodule is used to optimize the start-stop frequency of the dosing pump.

[0031] As preferred, in the emergency dosing mode, the dosing amount is adjusted through a dosing amount adjustment formula, and the dosing amount adjustment formula is specifically:

[0032]

[0033] wherein, Q 当前 is the dosing amount before triggering the emergency dosing mode, ΔParm is the change amount of any one of chemical oxygen demand, biochemical oxygen demand and pH value in the water quality parameters, Parm 阈值 is a preset trigger threshold value of any one of chemical oxygen demand, biochemical oxygen demand and pH value in the water quality parameters, and σ is a nonlinear gain coefficient.

[0034] As preferred, the objective function of the energy consumption optimization submodule is: wherein, E(t) is the energy consumption at time t, τ is a penalty coefficient for balancing energy consumption and dosing accuracy, W(t) is the actual dosing amount at time t, W 目标 is a preset dosing amount.

[0035] As preferred, the energy consumption optimization submodule is configured with a dynamic constraint algorithm, and the constraint conditions include dynamic limitation of dosing pump start-stop frequency and dynamic constraint of reagent dosing amount.

[0036] The dynamic limitation of dosing pump start-stop frequency is specifically: wherein, ΔS t is the start-stop frequency of the dosing pump at time t, N max (t) is a dynamic start-stop frequency threshold value, Q 流量 (t) and v(t) are real-time flow and flow rate, Q 阈值 and v 阈值 are preset flow threshold value and flow rate threshold value, N 基准 is a preset reference start-stop frequency;

[0037] The dynamic constraint of reagent dosing amount is specifically: W min (t)≤W(t)≤W max (t), W min (t) is the minimum dosing amount, W max(t) is the maximum dosage amount.

[0038] As preferred, the water quality sensor group comprises an electrochemical sensor array for real-time monitoring of pH, dissolved oxygen and oxidation-reduction potential, a spectral analysis unit for online qualitative and quantitative analysis of organic pollutants by Raman spectroscopy or fluorescence spectroscopy technology, and a multi-sensor data fusion unit for fusion of sensor data by Bayesian filtering algorithm to eliminate noise interference.

[0039] In a second aspect, the application discloses a monitoring method for a sewage treatment process, applied to the monitoring system for the sewage treatment process as described above, and the method comprises the following steps:

[0040] Through the multi-parameter sensor group and the flow monitoring device, the chemical oxygen demand, the biochemical oxygen demand, the pH value, the heavy metal concentration, the suspended substance concentration, the flow, the flow rate and the temperature data of the sewage are collected;

[0041] According to the collected data, the water quality parameters are dynamically correlated and analyzed by a machine learning algorithm to generate a dosing control instruction;

[0042] According to the dosing control instruction, one or more of the coagulant, the flocculant or the disinfectant is added to the sewage treatment system through the dosing pump and the flow regulating valve, and the addition amount is adjusted in real time;

[0043] Based on the real-time feedback data, the dosing control instruction is iteratively optimized by an adaptive algorithm to form a closed-loop control.

[0044] Beneficial effects: the monitoring system and method for the sewage treatment process of the application realize dynamic and accurate monitoring of the sewage treatment process through the synergistic effect of the multi-parameter real-time monitoring module, the data fusion and analysis module, the automatic dosing control module and the closed-loop feedback optimization module, solve the problems of single parameter, control lag and high energy consumption in the prior art, increase the water standard compliance rate to ≥95%, and reduce the reagent cost and equipment maintenance cost. BRIEF DESCRIPTION OF DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0046] Figure 1 The structural block diagram of the monitoring system for the sewage treatment process provided by the embodiments of the application is shown in the figure. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be clearly and completely described below. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0048] In this document, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or further includes elements inherent to such a process, method, article or device. Without more limitations, the elements defined by the statement "comprising" do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0049] The embodiments disclose a monitoring system for a sewage treatment process in the first aspect, which aims to solve the technical problems in the prior art by means of multi-parameter real-time monitoring, dynamic machine learning model, closed-loop feedback optimization and other technical solutions. Figure 1 The system includes a multi-parameter real-time monitoring module, a data fusion and analysis module, an automatic dosing control module and a feedback optimization module. The multi-parameter real-time monitoring module, the data fusion and analysis module and the automatic dosing control module are sequentially connected in communication. The feedback optimization module is connected in communication with the multi-parameter real-time monitoring module and the data fusion and analysis module.

[0050] The multi-parameter real-time monitoring module includes a water quality sensor group and a flow monitoring device. The water quality sensor group is used to collect real-time chemical oxygen demand (COD), biochemical oxygen demand (BOD), pH value, heavy metal concentration and suspended matter concentration data of sewage. The flow monitoring device is used to collect real-time flow, flow rate and temperature data of sewage. These data are transmitted to the data fusion and analysis module through wireless transmission or wired connection.

[0051] The data fusion and analysis module is used to receive feedback data of the multi-parameter real-time monitoring module and perform dynamic correlation analysis on water quality parameters through a machine learning algorithm to generate a dosing control instruction.

[0052] The automatic dosing control module is used to add one or more of a coagulant, a flocculant or a disinfectant to the sewage treatment system according to the dosing control instruction through a dosing pump and a flow regulating valve, and to adjust the addition amount in real time.

[0053] The automatic dosing control module is used to add one or more of a coagulant, a flocculant or a disinfectant to the sewage treatment system according to the dosing control instruction through a dosing pump and a flow regulating valve, and to adjust the addition amount in real time.

[0054] a feedback optimization module, configured to iteratively optimize the dosing control instruction through an adaptive algorithm based on the feedback data of the multi-parameter real-time monitoring module to form a closed-loop control.

[0055] Specifically, the machine learning algorithm includes a space-time coupling regression model, a dosing amount optimization model and a fuzzy PID control algorithm.

[0056] The space-time coupling regression model is configured to extract features of the space-time distribution of the chemical oxygen demand, biochemical oxygen demand and pH value in the water quality parameters through a convolutional neural network, and predict the water quality change trend in the future time by combining a long short-term memory network. For example, assuming that we have a historical database containing water quality data of the past 24 hours, we can train the model with these data to predict the water quality change in the next 10 minutes. In simple terms, the model combines the spatial distribution features (such as the spatial correlation of COD, BOD, pH and turbidity at different monitoring points) and time series features (such as the change rule of water quality parameters over time) of water quality parameters through joint modeling of space-time features, to predict the water quality state at the future time, and its expression is:

[0057]

[0058] wherein X t is the input data tensor of time step t, including the feature data extracted from the space-time distribution of the chemical oxygen demand, biochemical oxygen demand and pH value in the water quality parameters; f CNN is a convolutional neural network for extracting local features in the spatial dimension; f LSTM is a long short-term memory network for modeling dynamic changes in the time dimension; is the predicted water quality parameter value at the future time t+Δt. The space-time coupling regression model combines CNN-LSTM to predict the water quality change trend in the next 10 minutes, and the dosing error is reduced to ≤3%, which is significantly better than the traditional PID control (error ≥10%).

[0059] In this embodiment, the core principle of spatiotemporal coupling modeling is: (1) spatial dimension modeling: CNN automatically captures the spatial correlation of water quality parameters (such as whether the COD concentration of adjacent monitoring points changes synchronously) through a local receptive field, solving the problem of manually defining spatial weights in traditional methods; (2) time dimension modeling: LSTM dynamically retains or discards historical information through the gating mechanism (input gate, forget gate), effectively handling long-term dependencies in time series (such as the delayed rise of BOD concentration after rainfall); (3) joint modeling: combine spatial features (CNN output) and temporal dynamics (LSTM output) to form spatiotemporal joint features, avoiding the limitations of single dimension modeling (such as ignoring spatial heterogeneity by only using time series). Compared with traditional technology, the traditional method relies on manual definition of spatial weights, while the model automatically learns spatial patterns through CNN; LSTM solves the problem of traditional regression models that cannot capture multi-time scale dependencies; and directly from the original spatiotemporal data to the prediction result, without complex preprocessing. Using the spatiotemporal coupling regression model of the present application, although the calculation time is slightly higher, the improvement of prediction accuracy (such as a 49% reduction in COD prediction error) significantly improves the real-time and reliability of water quality management.

[0060] The dosing amount optimization model is used to calculate the dosing amount based on the multi-parameter nonlinear relationship, and the calculation formula of the dosing amount is:

[0061]

[0062] wherein Q t is the dosing amount at time t; a i , b i and g i are time-varying coefficients obtained by training historical data (such as ridge regression), and correspond to the weights of chemical oxygen demand, biochemical oxygen demand and pH value, respectively; COD i , BOD i and pH i correspond to chemical oxygen demand, biochemical oxygen demand and pH value, respectively, and l is a decay coefficient 0.05 i -7) sensitivity. e -λ·t The time-varying nature of water quality parameters (such as the decay of pollutant concentration over time) is considered to avoid hysteresis effects.

[0063] The proportional coefficient K pIntegral time T i Derivative time T d Adaptive adjustment according to real-time flow Q 流量 and flow rate v; the specific adjustment formula is:

[0064]

[0065] T d = T d0 ·(1+μ·ΔQ)

[0066] Wherein, K p0 , T i0 and T d0 are initial proportional coefficient, initial integral time and initial derivative time, Q 阈值 and v 阈值 are preset flow threshold and flow rate threshold, μ is proportional coefficient gain factor, and ΔQ is the amount of change in the amount of drug. For example, when the flow increases, the proportional coefficient K p will also increase accordingly, so as to respond to water quality changes faster and respond quickly; when the flow is stable, the integral time T i is extended, and the integral accumulation is reduced. This adaptive adjustment mechanism is more suitable for complex and variable working conditions than the traditional PID control, and the control error is reduced to within 5% when the flow fluctuates ±20%.

[0067] Further, in order to ensure the accuracy of the spatiotemporal feature extraction, the spatiotemporal coupling regression model is also used to compensate for the time delay and noise of the sensor data through a Kalman filtering algorithm. This step is crucial to ensure the reliability of the subsequent analysis results.

[0068] Secondly, in order to ensure the reliability of the processing process, in the embodiment, the spatiotemporal coupling regression model further comprises a dynamic weight distribution mechanism, which dynamically adjusts the weight of each parameter in the regression model according to the real-time change rate of the water quality parameter, and the weight calculation formula is: Wherein, θ i is the weight of the i-th parameter, θ i ∈(α i ,β i ,γ i ); ΔX i is the current change rate of the i-th parameter; X i is the current value of the i-th parameter, X 目标 is the preset target value; η is the sensitivity coefficient, which is used to adjust the sensitivity of the weight to the deviation of the parameter from the target value, and is set to η=0.3. The dynamic weight distribution mechanism adjusts the weight according to the parameter change rate, so that the response time of the system to COD mutation (such as instantaneous increase of 20%) is ≤2 seconds, avoiding processing lag.

[0069] As a preferred embodiment of the present embodiment, the automatic dosing control module further comprises a multi-stage dosing distributor and a medicament concentration correction unit.

[0070] The multi-stage dosing distributor is used to realize segmented dosing of medicaments through pressure sensors and electromagnetic valves, adjust the dosing position of medicaments according to the depth of sewage, and optimize the flocculation effect through dynamic mixing ratio adjustment, so as to realize precise segmented dosing and multi-medicament collaborative control of medicaments, improve the flocculation effect, and reduce the waste of medicaments. The technical principle is: according to the depth stratification of the aeration tank (such as the surface layer, the middle layer, and the bottom layer), the flow rate and pressure of the sewage in different areas are monitored through the pressure sensor, and the dosing position of the medicament is controlled by the electromagnetic valve to ensure uniform dispersion of the medicament. The flow control accuracy of the segmented dosing is ±0.5%, and this accuracy parameter is realized through high-precision electromagnetic valves (such as stepper motor driven valves) and closed-loop PID control. The segmented dosing logic is: according to the depth division of the aeration tank (such as 0-2m, 2-4m, and 4-6m), the pressure sensor feedbacks the pressure of each layer in real time, and the electromagnetic valve opening is dynamically adjusted to ensure uniform distribution of the medicament in the target area.

[0071] When treating wastewater containing multiple pollutants, the optimal treatment effect can be achieved by adjusting the proportion of different medicaments. The dynamic mixing ratio formula is: P r =1:(1+k·Conc 悬浮物 ·sin(ω·t)), where P r is the calculated mixing ratio, k is an empirical coefficient, Conc 悬浮物 is the suspended solids concentration, and ω is the adjustment frequency, and ω=2π·0.5Hz. By periodically adjusting the mixing ratio through the sine function, the medicament forms alternating high / low concentration zones during the flocculation process, enhancing particle collision efficiency. The implementation steps are: the sensor monitors turbidity and pH value in real time; according to the current water quality data, dynamically select the ω value (such as increasing the frequency when the water quality fluctuates greatly); adjust the electromagnetic valve opening through PLC control, and proportionally distribute the medicament.

[0072] Based on the design of the above multi-stage dosing distributor, the invalid loss of medicaments in the surface layer is reduced, the flocculation efficiency of the bottom layer is improved, and the utilization rate of medicaments is increased by 15%-20%. And through periodic proportion adjustment, the flocculation particles form tighter alum flowers, and the suspended solids removal rate is increased to more than 95%.

[0073] The medicament concentration correction unit is used to correct the dosing ratio of medicament in real time by ion selective electrode according to the heavy metal concentration data, to avoid overdosing or insufficient treatment. For example, if the heavy metal concentration is detected to increase, the medicament dosing amount is appropriately increased to ensure the treatment effect. The ion selective electrode (ISE) monitors the concentration of heavy metal ions (such as lead, cadmium) in real time, adjusts the medicament concentration through a nonlinear correction formula, and suppresses the antagonistic effect of heavy metals on flocculants. The correction formula used by the medicament concentration correction unit is: wherein 0.5 is a linear gain coefficient, controlling the correction amplitude; C 校正 is the corrected medicament concentration; C 初始 is the initial medicament concentration; Conc 重金属 is the heavy metal concentration data; Conc 重金属阈值 is the preset heavy metal concentration threshold, set to 10 mg / L, which is a preset safety concentration threshold, and when it is exceeded, the medicament dosing needs to be strengthened. The implementation steps are: the ISE detects the heavy metal concentration every 5 seconds; the data is input into the correction formula to calculate C 校正 ; the medicament pump frequency is adjusted to make the actual concentration close to C 校正 .

[0074] Through the design of the medicament concentration correction unit, the heavy metal removal rate is increased to 99%, and the overdosing of medicament is reduced (saving cost by about 10%-15%). Moreover, the competitive adsorption of heavy metals and flocculants is suppressed, ensuring stable flocculation effect.

[0075] As a preferred embodiment of the present embodiment, the feedback optimization module includes an abnormal working condition recognition unit and an energy consumption optimization sub-module. The abnormal working condition recognition unit detects mutations or abnormal fluctuations of water quality parameters through cluster analysis, and triggers an emergency dosing mode when the COD instantaneous increase exceeds 20%. The energy consumption optimization sub-module is used to optimize the start-stop frequency of the dosing pump.

[0076] In order to quickly respond to water quality mutations and avoid treatment failure, in the emergency dosing mode, the dosing amount is adjusted through a dosing amount adjustment formula. In the daily monitoring process, water quality data (such as COD, etc.) are analyzed in real time through the DBSCAN clustering algorithm, and abnormal fluctuations are detected. When the COD instantaneous increase exceeds the threshold, the emergency dosing is started. The dosing amount adjustment formula is specifically:

[0077]

[0078] wherein Q 当前 is the dosing amount before triggering the emergency dosing mode, ΔParm is the change amount of any one of chemical oxygen demand, biochemical oxygen demand and pH value in the water quality parameters, Parm 阈值This is a trigger threshold for any one of the preset water quality parameters: chemical oxygen demand (COD), biochemical oxygen demand (BOD), and pH. 阈值 =100mg / L, σ is the nonlinear gain coefficient, controlling the response slope (the larger the σ, the more sensitive the response), σ = 0.75 is selected. The implementation steps are: the DBSCAN algorithm clusters the real-time COD data to identify abrupt change points; when ΔCOD exceeds the 20% threshold, Q is calculated. 加急 The emergency mode is triggered, and the dosing pump dispenses chemicals (such as flocculants and coagulants) at maximum flow rate.

[0079] The emergency dosing mode design enables rapid response; for example, in the event of a 30% surge in COD, the concentration can be restored to below the threshold within 5 minutes. Furthermore, by using nonlinear gain to avoid overdosing, energy consumption only increases by 15% in emergency mode.

[0080] It is known that each time the dosing pump is started, the motor needs to overcome static friction and acceleration inertia, resulting in a significant increase in instantaneous energy consumption (especially in high-power pumps). Frequent start-stop cycles accumulate additional energy consumption. Reducing the number of start-stop cycles can lower start-up energy consumption, but may lead to fluctuations in the dosing dosage (due to continuous pump operation or prolonged shutdown). Correspondingly, the pump can adjust the dosing dosage more frequently (e.g., to quickly respond to changes in water quality), making the actual dosing dosage at time t closer to the preset dosage, but may lead to increased energy consumption due to frequent start-stop cycles. Conversely, a more stable pump operation may result in a larger deviation between the actual dosing dosage at time t and the preset dosage due to the inability to adjust the dosing dosage in a timely manner (e.g., insufficient dosing when influent turbidity suddenly increases). Therefore, in this embodiment, the energy consumption optimization submodule optimizes the start-stop frequency of the dosing pump using a dynamic programming algorithm to achieve a balance between energy consumption and dosing accuracy. The technical principle is as follows: establish a weighted objective function of energy consumption and dosing accuracy, combine it with dynamic constraints (start-stop frequency limits, dosage range), and use a dynamic programming algorithm to solve for the optimal start-stop strategy. The objective function of the energy consumption optimization submodule is: Where E(t) is the energy consumption (kW·h) at time t, including start-up energy consumption (e.g., 0.5 kW·h / cycle) and operating energy consumption (e.g., 0.1 kW·h / min), τ is a penalty coefficient used to balance energy consumption and dosing accuracy, balancing energy consumption (low priority) and accuracy (high priority), and is taken as τ = 15, W(t) is the actual amount of pesticide added at time t, W 目标 This is the preset dosage.

[0081] Corresponding to the objective function, the energy consumption optimization submodule is configured with a dynamic constraint algorithm, and the constraints include dynamic limits on the start and stop frequency of the dosing pump and dynamic constraints on the dosage of the reagent.

[0082] The specific dynamic limit on the start / stop frequency of the dosing pump is as follows: Where, ΔS tN is the number of start-stop of the dosing pump at time t max (t) is the dynamic start-stop number threshold, when the flow or flow rate increases, N max (t) increases linearly, allowing more frequent start-stop to cope with load changes, Q 流量 (t) and v(t) are the real-time flow and flow rate, Q 阈值 and v 阈值 are the preset flow threshold and flow rate threshold, N 基准 is the preset baseline start-stop number, N 基准 = 5 times / hour.

[0083] Second, the dosage constraint is:

[0084] W min (t)≤W(t)≤W max (t)

[0085] Reduce reagent dosage at low COD

[0086] Increase the upper limit at high COD to avoid under-treatment

[0087] Where, Conc COD is the concentration value of COD, W min (t) is the minimum dosage, W max (t) is the maximum dosage, W 基准 is the baseline dosage, COD 阈值 is the preset COD concentration threshold.

[0088] The implementation steps are: the dynamic programming algorithm predicts the energy consumption and error for the next T = 60 minutes with a time step of 1 minute; at each time point, select the start / stop decision to meet the constraint conditions and minimize the objective function; output the optimal start-stop sequence to control the pump to execute.

[0089] Through the design of the above dynamic constraint algorithm, by adjusting the upper limit of the start-stop number of the dosing pump (N max (t)) according to the real-time flow and flow rate, compared with the fixed threshold (such as 5 times / hour), the energy consumption is reduced by 15%-20%. The dynamic constraint of reagent dosage (W min (t) and W max (t)) avoids over-dosing, and the reagent cost is reduced by 20%-25%.

[0090] In the embodiment, the water quality sensor group includes an electrochemical sensor array for real-time monitoring of pH, dissolved oxygen and oxidation-reduction potential, an electrochemical sensor sampling frequency of 10 Hz, a resolution of 0.01 pH / 0.1 mg / L / mV, a spectral analysis unit for online qualitative and quantitative analysis of organic pollutants by Raman spectroscopy or fluorescence spectroscopy technology, identification of organic pollutant types, auxiliary selection of reagent types (such as selective flocculants), a detection limit of ≤3 mg / L, and simultaneous analysis of TOC, benzene series, polycyclic aromatic hydrocarbons, etc., and a multi-sensor data fusion unit for fusion of sensor data by a Bayesian filtering algorithm to eliminate noise interference. The Bayesian filtering algorithm can use any one of the existing technologies. By eliminating sensor drift and environmental noise, the data confidence is improved by 30%. The implementation steps are: sensor data is uploaded to the central controller every second; the Bayesian filter calculates the weight and fuses the data; the fusion result drives the reagent addition control and abnormal alarm.

[0091] Through the configuration of the water quality sensor group, the following is achieved: pH error ≤±0.1, heavy metal detection limit reaches ppb level; multi-component identification: simultaneous quantitative analysis of 5-10 kinds of organic pollutants; anti-interference ability: data availability is improved by 50% in a noisy environment.

[0092] In summary, the present application solves the core problems of single parameter, control lag, high energy consumption, high maintenance cost, etc. in the existing sewage treatment monitoring system by integrating multi-parameter monitoring, dynamic machine learning model, closed-loop feedback optimization, etc. It realizes precise, energy-saving and stable sewage treatment process control, and has significant technical progress and practicality.

[0093] The embodiment provides a monitoring method for a sewage treatment process in a second aspect. The method is applied to the monitoring system for the sewage treatment process as described above, and includes the following steps:

[0094] Through the multi-parameter sensor group and the flow monitoring device, the chemical oxygen demand, biochemical oxygen demand, pH value, heavy metal concentration, suspended solids concentration, flow, flow rate and temperature data of the sewage are collected;

[0095] According to the collected data, the water quality parameters are dynamically correlated and analyzed by a machine learning algorithm to generate a reagent addition control instruction;

[0096] According to the reagent addition control instruction, one or more of the coagulant, flocculant or disinfectant is added to the sewage treatment system through the reagent addition pump and the flow regulating valve, and the addition amount is adjusted in real time;

[0097] Based on the real-time feedback data, the reagent addition control instruction is iteratively optimized by an adaptive algorithm to form a closed-loop control.

[0098] It should be noted that the monitoring method for the sewage treatment process of the embodiment corresponds to the monitoring system for the sewage treatment process described above, and therefore, the parts not disclosed in the monitoring method for the sewage treatment process of the embodiment (specific technical means and corresponding technical effects) can be referred to the description in the monitoring system for the sewage treatment process described above, which will not be described herein.

[0099] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be realized by hardware, software, firmware, middleware, codes or any proper combination thereof. For hardware implementation, the processor can be realized in one or more of the following: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, micro controller, microprocessor, other electronic unit designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the functions of the embodiments can be instructed by a computer program to relevant hardware. In implementation, the above program can be stored in a computer readable storage medium or transmitted as one or more instructions or codes on a computer readable storage medium. The computer readable storage medium includes computer storage medium and communication medium, wherein the communication medium includes any medium that facilitates the transfer of computer program from one place to another. The storage medium can be any available medium that can be accessed by a computer. The computer readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage medium or other magnetic storage devices, or any other medium that can be used to carry or store desired program codes in the form of instructions or data structures and can be accessed by a computer.

[0100] Finally, it should be noted that: the above only describes the preferred embodiments of the present application and is not used to limit the present application, although the present application has been described in detail with reference to the foregoing embodiments, for those skilled in the art, the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalent, any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A supervisory system for a sewage treatment process, characterized in that The method comprises the following steps: A multi-parameter real-time monitoring module, comprising a water quality sensor group and a flow monitoring device, the water quality sensor group is used to collect chemical oxygen demand, biochemical oxygen demand, pH value, heavy metal concentration and suspended solids concentration data of sewage in real time; the flow monitoring device is used to collect flow, flow rate and temperature data of sewage in real time; A data fusion and analysis module is used to receive feedback data of the multi-parameter real-time monitoring module, and to perform dynamic correlation analysis on water quality parameters through a machine learning algorithm to generate a dosing control instruction; An automatic dosing control module is used to add one or more of coagulants, flocculants or disinfectants to the sewage treatment system according to the dosing control instruction through a dosing pump and a flow regulating valve, and to adjust the dosage in real time; A feedback optimization module is used to form a closed-loop control by iteratively optimizing the dosing control instruction through an adaptive algorithm based on the feedback data of the multi-parameter real-time monitoring module; The machine learning algorithm comprises a space-time coupling regression model, a dosing amount optimization model and a fuzzy PID control algorithm; The space-time coupling regression model is used to extract features of the space-time distribution of chemical oxygen demand, biochemical oxygen demand and pH value in water quality parameters through a convolutional neural network, and to predict water quality trends in future time by combining a long short-term memory network, and its expression is: wherein X t is the input data tensor of time step t, including the feature data extracted from the spatiotemporal distribution of the water quality parameters, such as chemical oxygen demand, biochemical oxygen demand, and pH value; f CNN is a convolutional neural network used to extract local features in the spatial dimension; f LSTM is a long short-term memory network used to model the dynamic changes in the time dimension; is the predicted water quality parameter value at the future time step t+△t. The dosing amount optimization model is used to calculate the dosing amount based on the multi-parameter nonlinear relationship, and the calculation formula of the dosing amount is: wherein Q t is the dosage at time t; α i , β i and γ i are time-varying coefficients trained by historical data, and correspond to the weights of chemical oxygen demand, biochemical oxygen demand and pH value, respectively; COD i , BOD i and pH i correspond to chemical oxygen demand, biochemical oxygen demand and pH value, respectively, λ is a decay coefficient 0.05≤λ≤1, used to reflect the decay trend of water quality parameters over time; n is the number of water quality parameters involved in the calculation. The proportional coefficient K of the fuzzy PID control algorithm p , the integral time T i , and the differential time T d is adaptively adjusted according to the real-time flow Q 流量 and the flow rate v; the specific adjustment formula is: T d = T d0 ·(1 + μ · ΔQ) Wherein, K p0 , T i0 and T d0 are initial proportional coefficient, initial integral time and initial derivative time respectively, Q 阈值 and v 阈值 are preset flow threshold and flow rate threshold, μ is proportional coefficient gain factor, and △Q is the amount of change of the reagent.

2. The supervisory system for a wastewater treatment process according to claim 1, characterized in that, The space-time coupling regression model is also used to compensate for the time delay and noise of sensor data through a Kalman filtering algorithm.

3. The supervisory system for a wastewater treatment process according to claim 1, characterized in that, The automatic dosing control module further comprises a multi-stage dosing distributor and a medicament concentration correction unit; The multi-stage dosing distributor is used for realizing segmented dosing of the medicament through a pressure sensor and a solenoid valve, adjusting a dosing position of the medicament according to sewage depth, and optimizing flocculation effect through dynamic mixing ratio adjustment, and the dynamic mixing ratio formula is: P r =1:(1+k·Conc 悬浮物 ·sin(ω·t)),wherein P r is a calculated mixing ratio, k is an empirical coefficient, Conc 悬浮物 is a suspended matter concentration, and ω is an adjustment frequency, and ω=2π·0.5Hz. The medicament concentration correction unit is used to correct the dosing ratio of the medicament in real time by an ion selective electrode according to the heavy metal concentration data, and the correction formula is: Wherein, C 校正 is the corrected medicament concentration, C 初始 is the initial medicament concentration, Conc 重金属 is the heavy metal concentration data, Conc 重金属阈值 is the preset heavy metal concentration threshold value.

4. The supervisory system for a wastewater treatment process of claim 1, wherein, The feedback optimization module comprises an abnormal working condition identification unit and an energy consumption optimization submodule, the abnormal working condition identification unit detects mutations or abnormal fluctuations of water quality parameters through cluster analysis, and triggers an emergency dosing mode when detecting that COD instantaneously increases by more than 20%, and the energy consumption optimization submodule is used to optimize the start-stop frequency of the dosing pump.

5. The supervisory system for a wastewater treatment process according to claim 4, characterized in that, In the emergency dosing mode, the dosing amount is adjusted through a dosing amount adjustment formula, and the dosing amount adjustment formula is specifically: wherein Q 当前 is the amount of chemical added before triggering the emergency chemical addition mode,△Parm is the variation of any one of chemical oxygen demand, biochemical oxygen demand and pH value in the water quality parameters, Parm 阈值 is the preset trigger threshold of any one of chemical oxygen demand, biochemical oxygen demand and pH value in the water quality parameters, and σ is a nonlinear gain coefficient.

6. The supervisory system for a wastewater treatment process of claim 4, wherein, The objective function of the energy consumption optimization submodule is: Wherein, E(t) is the energy consumption at time t, τ is a penalty coefficient for balancing energy consumption and dosing accuracy, W(t) is the actual dosing amount at time t, W 目标 is the preset dosing amount.

7. The supervisory system for a wastewater treatment process according to claim 6, characterized in that, The energy consumption optimization submodule is configured with a dynamic constraint algorithm, and the constraint conditions include dynamic constraints of the start-stop frequency of the dosing pump and the dynamic constraints of the medicament dosage; The dynamic limit of the start-stop frequency of the dosing pump is specifically: Wherein, △S t is the start-stop frequency of the dosing pump at time t, N max (t) is the dynamic start-stop frequency threshold, Q 流量 (t) and v(t) are the real-time flow and flow rate, Q 阈值 and v 阈值 are the preset flow threshold and flow rate threshold, N 基准 is the preset reference start-stop frequency; The dynamic constraint of the medicament adding amount is specifically: W min (t)≤W(t)≤W max (t), W min (t) is the minimum adding amount, and W max (t) is the maximum adding amount.

8. The supervisory system for a wastewater treatment process of claim 1, wherein, The water quality sensor group comprises an electrochemical sensor array, a spectral analysis unit and a multi-sensor data fusion unit, the electrochemical sensor array is used to monitor pH, dissolved oxygen and oxidation-reduction potential in real time; the spectral analysis unit is used to perform online qualitative and quantitative analysis on organic pollutants through Raman spectroscopy or fluorescence spectroscopy technology; and the multi-sensor data fusion unit is used to fuse sensor data through a Bayesian filtering algorithm to eliminate noise interference.

9. A method for supervising a wastewater treatment process, applied to the supervising system for a wastewater treatment process according to any one of claims 1 to 8, characterized in that, The method comprises the following steps: Through a multi-parameter sensor group and a flow monitoring device, chemical oxygen demand, biochemical oxygen demand, pH value, heavy metal concentration, suspended solids concentration, flow, flow rate and temperature data of sewage are collected; According to the collected data, a machine learning algorithm is used to perform dynamic correlation analysis on water quality parameters to generate a dosing control instruction; According to the dosing control instruction, one or more of a coagulant, a flocculant or a disinfectant is dosed to the sewage treatment system through a dosing pump and a flow regulating valve, and the dosing amount is adjusted in real time; Based on the real-time feedback data, the dosing control instruction is iteratively optimized through an adaptive algorithm to form a closed-loop control.

Images