An intelligent carbon source dosing system for sewage treatment plants

Abstract

The present invention relates to a carbon source intelligent dosing system for a sewage treatment plant, comprising: a monitoring and acquisition module for collecting influent COD, TN, NH4 + -N, nitrification effluent DO, total aeration volume of nitrification tank, denitrification tank inlet flow and hydraulic retention time, sodium acetate accumulation, effluent TN and temperature; carbon source addition prediction module is used to collect influent COD, TN, NH4 + The carbon source dosage prediction model uses the following inputs: NH, nitrification effluent DO, total aeration volume of the nitrification tank, inlet flow rate and hydraulic retention time of the denitrification tank, cumulative sodium acetate, effluent TN, and temperature to obtain a predicted carbon source dosage. A feedback adjustment module is used to adjust the predicted carbon source dosage based on the collected effluent TN to obtain a carbon source dosage control value. A dosing device module is used to control the carbon source dosing device to add the carbon source based on the carbon source dosage control value. This invention makes the dosage more accurate, the effluent TN compliance rate higher, and the dosage lower.

Description

Technical Field

[0001] The present invention relates to the technical field of sewage treatment, and in particular to an intelligent carbon source dosing system for a sewage treatment plant. Background Art

[0002] Biological denitrification is the most commonly used method for nitrogen removal in urban wastewater treatment in my country. The process consists of two main steps: nitrification in an aerobic tank converts ammonia nitrogen into nitrate nitrogen, and denitrification in an anoxic tank converts nitrate nitrogen into nitrogen gas, which escapes from the water. Denitrification in anoxic tanks requires a carbon source. Currently, the amount of carbon source added is primarily based on the experience of sewage plant staff, which can easily lead to over- or under-addition, resulting in risks such as excessive effluent total nitrogen levels, wasted energy, and increased costs. Summary of the Invention

[0003] The present invention provides an intelligent carbon source dosing system for a sewage treatment plant, which solves the problems of inaccurate carbon source dosage, easily exceeding the TN standard of effluent, and the need for continuous manual regulation when carbon source addition is performed based on experience.

[0004] The technical solution adopted by the present invention to solve the technical problem is to provide a carbon source intelligent dosing system for a sewage treatment plant, comprising:

[0005] Monitoring and collection module, used to collect influent COD, TN, NH4 + -N, nitrification effluent DO, total aeration volume of nitrification tank, influent flow rate and hydraulic retention time of denitrification tank, cumulative amount of sodium acetate, effluent TN and temperature;

[0006] Carbon source dosing prediction module is used to collect the influent COD, TN, NH4 + -N, nitrification effluent DO, total aeration volume of nitrification tank, influent flow rate and hydraulic retention time of denitrification tank, cumulative amount of sodium acetate, effluent TN and temperature are used as inputs of carbon source dosage prediction model to obtain the predicted amount of carbon source dosage;

[0007] A feedback adjustment module is used to adjust the predicted amount of carbon source addition based on the collected effluent TN to obtain a carbon source addition control value;

[0008] The dosing device module is used to control the carbon source dosing device to add the carbon source based on the carbon source dosage control value.

[0009] The carbon source dosage prediction model is constructed based on a BP neural network.

[0010] The feedback regulation module includes:

[0011] a determination unit, configured to determine a current state of the outlet water TN;

[0012] The first setting unit is configured to set the feedback adjustment value to be greater than zero when the outlet water TN is higher than the upper limit value, or the continuous growth amplitude exceeds the growth threshold and approaches the upper limit value;

[0013] The second setting unit is used to make the feedback adjustment value less than zero when the outlet water TN is lower than the lower limit value, or the continuous decreasing amplitude exceeds the reduction threshold and is lower than the excellent rate threshold range;

[0014] a third setting unit, configured to set the feedback adjustment value to zero when the outlet water TN is within an excellent rate threshold range;

[0015] The calculation unit is used to add the feedback adjustment value to the carbon source dosage prediction amount to obtain the carbon source dosage control value.

[0016] The first given unit includes:

[0017] The first given subunit is used to calculate the feedback adjustment value n according to n=(TN-13.5)×0.15×denitrification tank inlet flow / 3000 when the effluent TN is greater than 14 mg / L and less than or equal to 15 mg / L;

[0018] The second given subunit is used to calculate the feedback adjustment value n according to n = (TN-13.5) × 0.2 × denitrification tank inlet flow / 3000 when the effluent TN is greater than 15 mg / L and less than or equal to 16 mg / L;

[0019] The third given subunit is used to calculate the feedback adjustment value n according to n=(TN-14.5)×0.25×denitrification tank inlet flow / 3000 when the effluent TN is greater than 16 mg / L;

[0020] The fourth given subunit is used to calculate the feedback adjustment value n according to n=0.1×denitrification tank inlet flow / 3000 when the continuous growth amplitude of the effluent TN exceeds the growth threshold and the current effluent TN is greater than 13.3 mg / L and less than 14 mg / L.

[0021] The first given unit further includes:

[0022] A judging subunit, configured to judge the number of times the outlet water TN exceeds the standard;

[0023] The adjustment subunit is used to adjust the feedback adjustment value n to 1.5 times the original value when the outlet water TN exceeds the standard twice in a row, and to adjust the feedback adjustment value n to 2.3 times the original value when the outlet water TN exceeds the standard three times in a row.

[0024] The second given unit includes:

[0025] The fifth given subunit is used to calculate the feedback adjustment value n according to n=(TN-11)×0.1×denitrification tank inlet flow / 6000 when the effluent TN is less than or equal to 8 mg / L;

[0026] The sixth given subunit is used to calculate the feedback adjustment value n according to n=(TN-11)×0.06×denitrification tank inlet flow / 6000 when the effluent TN is greater than 8 mg / L and less than or equal to 10 mg / L;

[0027] The seventh given subunit is used to calculate the feedback adjustment value n according to n=(TN-12)×0.06×denitrification tank inlet flow / 6000 when the effluent TN is greater than 10 mg / L and less than or equal to 11.5 mg / L;

[0028] The eighth given subunit is used to calculate the feedback adjustment value n according to n=-0.1×denitrification tank inlet flow / 6000 when the effluent TN continuously decreases and exceeds the reduction threshold and the current effluent TN is less than 12 mg / L.

[0029] Beneficial effects

[0030] Due to the adoption of the above-mentioned technical scheme, the present invention has the following advantages and positive effects compared with the existing technology: the present invention constructs a carbon source addition prediction model through historical data, and then predicts the carbon source addition amount based on the monitoring data by the carbon source addition prediction model, and adjusts the predicted carbon source addition amount in combination with the real-time data of effluent TN to ensure that the effluent TN is stably up to standard. During the test, the effluent TN compliance rate reached 94.7%, and the excellent rate (TN is 12-14 mg / L) accounted for 45%, which was 15% higher than the existing technology. At the same time, 14% of sodium acetate consumption was saved, about 4.2t / d, which reduced the daily operating costs of the water plant and achieved the goal of energy saving and consumption reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a structural block diagram of the intelligent carbon source dosing system for a sewage treatment plant according to an embodiment of the present invention;

[0032] Figure 2 It is a structural schematic diagram of an embodiment of the present invention applied to a sewage treatment plant. DETAILED DESCRIPTION

[0033] Below in conjunction with specific embodiment, further set forth the present invention.Should be understood that these embodiments are only used to illustrate the present invention and are not used in limiting the scope of the present invention.In addition, should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms fall equally within the scope limited by the appended claims of the application.

[0034] The embodiment of the present invention relates to a carbon source intelligent dosing system for a sewage treatment plant, such as Figure 1 Shown, including:

[0035] Monitoring and collection module, used to collect influent COD, TN, NH4 + -N, nitrification effluent DO, total aeration volume of nitrification tank, influent flow rate and hydraulic retention time of denitrification tank, cumulative amount of sodium acetate, effluent TN and temperature;

[0036] Carbon source dosing prediction module is used to collect the influent COD, TN, NH4 + -N, nitrification effluent DO, total aeration volume of nitrification tank, influent flow rate and hydraulic retention time of denitrification tank, cumulative amount of sodium acetate, effluent TN and temperature are used as inputs of carbon source dosage prediction model to obtain the predicted amount of carbon source dosage;

[0037] A feedback adjustment module is used to adjust the predicted amount of carbon source addition based on the collected effluent TN to obtain a carbon source addition control value;

[0038] The dosing device module is used to control the carbon source dosing device to add the carbon source based on the carbon source dosage control value.

[0039] like Figure 2 As shown, the monitoring and acquisition module includes a COD monitor 1 installed in the main water inlet pipeline, a first TN monitor 2, and a NH4 + -N monitor 3, blower aeration rate reader 4 installed in the blower room, DO monitor 5 installed at the outlet of the nitrification tank, sodium acetate cumulative amount reader 8 installed in the dosing room, flow monitor 9 and temperature measuring instrument 10 installed at the inlet of the denitrification tank, and a second TN monitor 11 installed at the outlet.

[0040] Carbon source addition prediction module (i.e. Figure 2 The prediction system in the embodiment (1) predicts the carbon source dosage through a carbon source dosage prediction model. The carbon source dosage prediction model can be obtained by the following method:

[0041] First, we collected water quality reports from the past three years and three months of real-time operating data. We cleaned the data, added value using Bessel interpolation, and normalized the data. We then determined input parameters based on correlation analysis and generated a sample training set. This shortened the model building cycle and ensured the algorithm could be quickly applied. We then compared and analyzed the three algorithms: BP, XGBoost, and Decision Tree. We selected the algorithm with the smallest average relative error in simulation results and the strongest anti-interference ability to build a carbon source dosage prediction model.

[0042] Through comparative analysis, this implementation method built a 3-layer BP neural network model with 10 neurons and a model loss value of 0.0378. Subsequently, the model was optimized by adjusting the data latency and the weights of each network. The final carbon source dosage prediction model can use real-time data to predict the carbon source dosage, and at the same time add the operating data to the training sample set to realize the system's adaptive learning and iterative optimization.

[0043] The feedback adjustment module (i.e. Figure 2 Feedback regulation in the

[0044] a determination unit, configured to determine a current state of the outlet water TN;

[0045] The first setting unit is configured to set the feedback adjustment value to be greater than zero when the outlet water TN is higher than the upper limit value or the continuous growth amplitude exceeds the growth threshold and approaches the upper limit value; the first setting unit also includes:

[0046] The first given subunit is used to calculate the feedback adjustment value n according to n=(TN-13.5)×0.15×denitrification tank inlet flow / 3000 when the effluent TN is greater than 14 mg / L and less than or equal to 15 mg / L;

[0047] The second given subunit is used to calculate the feedback adjustment value n according to n = (TN-13.5) × 0.2 × denitrification tank inlet flow / 3000 when the effluent TN is greater than 15 mg / L and less than or equal to 16 mg / L;

[0048] The third given subunit is used to calculate the feedback adjustment value n according to n=(TN-14.5)×0.25×denitrification tank inlet flow / 3000 when the effluent TN is greater than 16 mg / L;

[0049] The fourth given subunit is used to calculate the feedback adjustment value n according to n=0.1×denitrification tank inlet flow / 3000 when the effluent TN continuously increases beyond the growth threshold and the current effluent TN is greater than 13.3 mg / L and less than 14 mg / L;

[0050] A judging subunit, configured to judge the number of times the outlet water TN exceeds the standard;

[0051] The adjustment subunit is used to adjust the feedback adjustment value n to 1.5 times the original value when the outlet water TN exceeds the standard twice in a row, and to adjust the feedback adjustment value n to 2.3 times the original value when the outlet water TN exceeds the standard three times in a row.

[0052] The second given unit is configured to set the feedback adjustment value to be less than zero when the outlet water TN is lower than the lower limit value, or the continuous decreasing amplitude exceeds the reduction threshold and is lower than the excellent rate threshold range; wherein the second given unit includes:

[0053] The fifth given subunit is used to calculate the feedback adjustment value n according to n=(TN-11)×0.1×denitrification tank inlet flow / 6000 when the effluent TN is less than or equal to 8 mg / L;

[0054] The sixth given subunit is used to calculate the feedback adjustment value n according to n=(TN-11)×0.06×denitrification tank inlet flow / 6000 when the effluent TN is greater than 8 mg / L and less than or equal to 10 mg / L;

[0055] The seventh given subunit is used to calculate the feedback adjustment value n according to n=(TN-12)×0.06×denitrification tank inlet flow / 6000 when the effluent TN is greater than 10 mg / L and less than or equal to 11.5 mg / L;

[0056] The eighth given subunit is used to calculate the feedback adjustment value n according to n=-0.1×denitrification tank inlet flow / 6000 when the continuous decrease amplitude of the effluent TN exceeds the reduction threshold and the current effluent TN is less than 12 mg / L;

[0057] a third setting unit, configured to set the feedback adjustment value to zero when the outlet water TN is within an excellent rate threshold range;

[0058] The calculation unit is used to add the feedback adjustment value to the carbon source dosage prediction amount to obtain the carbon source dosage control value.

[0059] In this embodiment, a feedback regulation system is constructed, which can fine-tune the predicted carbon source dosage given by the prediction system according to the value of the effluent TN. The specific regulation mechanism is as follows: Assume the predicted dosage is m and the feedback regulation amount is n. When 14 < effluent TN ≤ 15 mg / L, n = (TN - 13.5) × 0.15 × flow rate / 3000; when 15 < effluent TN ≤ 16 mg / L, n = (TN - 13.5) × 0.2 × flow rate / 3000; when effluent TN > 16 mg / L, n = (TN - 14.5) × 0.25 × flow rate / 3000; when the effluent TN exceeds the standard continuously for two times, n = 1.5 × n, and when it exceeds the standard continuously for three times, n = 2.3 × n; when the continuous growth rate of the effluent TN exceeds 3 mg / L and the current effluent TN is greater than 13.3 mg / L, n = 0.1 × flow rate / 3000; when effluent TN ≤ 8 mg / L, n = (TN - 11) × 0.1 × flow rate / 6000, when 8 < TN ≤ 10 mg / L, n = (TN - 11) × 0.06 × flow rate / 6000, when 10 < TN ≤ 11.5 mg / L, n = (TN - 12) × 0.06 × flow rate / 6000, when the effluent TN decreases continuously by more than 2 mg / L and the current effluent TN < 12 mg / L, n = -0.1 × flow rate / 6000; when the effluent TN is between 12 and 14 mg / L, n = 0; the finally determined dosage control value k of the system = m + n.

[0060] The dosing device module of this embodiment includes a chemical storage tank, a dosing pump 6 and a water injector 7. After determining the carbon source dosing amount control value, the control system issues an instruction to the water injector and the dosing pump, adjusts the valve opening of the water injector and the frequency of the dosing pump to adjust the carbon source dosing amount to the control value, and realizes the closed-loop of carbon source dosing. Among them, the water injector is mainly responsible for the situation where the carbon source dosing amount is low. It has more precise control over the flow rate and can prevent the dosing pump from malfunctioning and being unable to dose. When the required dosing amount is high, the dosing pump can be turned on. The dosing pump 6 and the water injector 7 together constitute the carbon source dosing device.

[0061] Relying on the water plant Internet of Things and the existing operation process control system to build an intelligent management platform, the process operation, prediction process, and dosing results are monitored in real time. The carbon source addition system has multiple control modes, including automatic and manual. At the same time, it is combined with the existing water plant operation process control module to facilitate staff to manage and inspect the process. The intelligent carbon source dosing system for sewage treatment plants in this embodiment features five control modes: a. Manual Control Mode: Staff control the on / off and frequency settings of the dosing pump via a computer in the central control room. b. Automatic Control Mode: An algorithm calculates predicted sodium acetate dosage values. After feedback adjustment, the control system issues commands to the carbon source dosing device to adjust the ejector valve opening and dosing pump frequency. Commands are issued four times per hour. c. Inertial Dosing Mode: When a small amount of abnormal data is detected, there is no need to switch to manual mode; the dosage can be predicted based on other parameters and historical data. d. Water Diversion Dosing Mode: This is specifically designed for situations where effluent quality is good during diversion, reducing dosage and saving drug consumption. e. Edge Computing Mode: In the event of system failures, network or server disconnections, carbon source dosing control commands are issued via an edge gateway. The platform uses a BP neural network algorithm to continuously adjust the existing prediction model based on new water quality data, optimizing and iterating the prediction model to ensure the accuracy of carbon source dosage predictions. Staff can directly modify the carbon source dosage on the management platform to respond to emergencies.

[0062] Table 1 Comparison results of the trial operation of this embodiment and the existing carbon source addition system

[0063]

[0064] Table 2: Formal operation of this implementation method

[0065]

[0066] As can be seen from Tables 1 and 2, under the control of the intelligent carbon source dosing system for sewage treatment plants based on the BP neural network, after 13 months of formal operation, the qualified rate of TN (TN<15 mg / L) of the effluent of the water plant in this embodiment was basically above 97%, the excellent rate (TN was 12-14 mg / L) increased from 26% to 50%, the sodium acetate consumption rate decreased by about 14% (4.2 t / d), the denitrification tank operated normally, and the sodium acetate consumption rate was reduced while ensuring that the effluent TN met the standards.

[0067] It is not difficult to find that the present invention adopts an automatically adjustable and remotely controlled carbon source dosing system. Compared with the existing homemade system based on empirical formulas in water plants, the dosage prediction is more accurate, the TN compliance rate of the effluent is higher, and the control effect is better. At the same time, it also reduces the carbon source dosage, reduces the daily operating costs of the water plant, and achieves the goal of energy saving and consumption reduction.

Claims

1. A carbon source intelligent dosing system for a sewage treatment plant, characterized in that: include: Monitoring and collection module, used to collect influent COD, TN, NH4 + -N, nitrification effluent DO, total aeration volume of nitrification tank, influent flow rate and hydraulic retention time of denitrification tank, cumulative amount of sodium acetate, effluent TN and temperature; Carbon source dosing prediction module is used to collect the influent COD, TN, NH4 + -N, nitrification effluent DO, total aeration volume of nitrification tank, influent flow rate and hydraulic retention time of denitrification tank, cumulative amount of sodium acetate, effluent TN and temperature are used as inputs of carbon source dosage prediction model to obtain the predicted amount of carbon source dosage; A feedback adjustment module is used to adjust the predicted amount of carbon source addition based on the collected effluent TN to obtain a carbon source addition control value; The feedback adjustment module includes: a determination unit, configured to determine a current state of the outlet water TN; The first given unit is configured to set the feedback adjustment value to be greater than zero when the outlet water TN is higher than the upper limit value or the continuous growth amplitude exceeds the growth threshold and approaches the upper limit value. The first given unit includes: The first given subunit is used to calculate the feedback adjustment value n according to n=(TN-13.5)×0.15×denitrification tank inlet flow / 3000 when the effluent TN is greater than 14 mg / L and less than or equal to 15 mg / L; The second given subunit is used to calculate the feedback adjustment value n according to n = (TN-13.5) × 0.2 × denitrification tank inlet flow / 3000 when the effluent TN is greater than 15 mg / L and less than or equal to 16 mg / L; The third given subunit is used to calculate the feedback adjustment value n according to n=(TN-14.5)×0.25×denitrification tank inlet flow / 3000 when the effluent TN is greater than 16 mg / L; The fourth given subunit is used to calculate the feedback adjustment value n according to n = 0.1 × denitrification tank inlet flow / 3000 when the continuous growth rate of effluent TN exceeds the growth threshold and the current effluent TN is greater than 13.3 mg / L and less than 14 mg / L The second setting unit is configured to set the feedback adjustment value to be less than zero when the outlet water TN is lower than the lower limit value, or the continuous decreasing amplitude exceeds the reduction threshold and approaches the lower limit value; a third setting unit, configured to set the feedback adjustment value to zero when the outlet water TN is at an excellent rate; A calculation unit, configured to add the feedback adjustment value to the predicted carbon source dosage to obtain a carbon source dosage control value; The dosing device module is used to control the carbon source dosing device to add the carbon source based on the carbon source dosage control value.

2. The intelligent carbon source dosing system for sewage treatment plants according to claim 1, characterized in that: The carbon source dosage prediction model is constructed based on a BP neural network.

3. The intelligent carbon source dosing system for sewage treatment plants according to claim 1, characterized in that: The first given unit further includes: A judging subunit, configured to judge the number of times the outlet water TN exceeds the standard; The adjustment subunit is used to adjust the feedback adjustment value n to 1.5 times the original value when the outlet water TN exceeds the standard twice in a row, and to adjust the feedback adjustment value n to 2.3 times the original value when the outlet water TN exceeds the standard three times in a row.

4. The intelligent carbon source dosing system for sewage treatment plants according to claim 1, characterized in that: The second given unit includes: The fifth given subunit is used to calculate the feedback adjustment value n according to n=(TN-11)×0.1×denitrification tank inlet flow / 6000 when the effluent TN is less than or equal to 8 mg / L; The sixth given subunit is used to calculate the feedback adjustment value n according to n=(TN-11)×0.06×denitrification tank inlet flow / 6000 when the effluent TN is greater than 8 mg / L and less than or equal to 10 mg / L; The seventh given subunit is used to calculate the feedback adjustment value n according to n=(TN-12)×0.06×denitrification tank inlet flow / 6000 when the effluent TN is greater than 10 mg / L and less than or equal to 11.5 mg / L; The eighth given subunit is used to calculate the feedback adjustment value n according to n=-0.1×denitrification tank inlet flow / 6000 when the effluent TN continuously decreases and exceeds the reduction threshold and the current effluent TN is less than 12 mg / L.

Images