Sewage treatment device, system and method based on AI decision control and storage medium

Abstract

The invention relates to a sewage treatment technology, and discloses a sewage treatment device, system and method based on AI decision control and a storage medium, and the method comprises the following steps: data collection and real-time treatment: collecting sewage treatment data in real time by using an online instrument; sewage treatment load determination: carrying out load sensing and allocation on the acquired data through AI, and determining real-time sewage treatment load; and controlling the sewage, wherein the sewage is controlled according to the real-time sewage treatment load. According to the method, the actual load condition of the incoming water is calculated in real time by monitoring the incoming water indexes, the actual cost of each agent and electric quantity is combined, the treatment load level between related process sections is effectively adjusted from operation requirements, and the stable and economical optimal operation state is achieved.

Description

Technical Field

[0001] The present invention relates to the treatment technology of sewage, and particularly to a sewage treatment device, system, method and storage medium based on AI decision control. Background Art

[0002] Sewage treatment plants are mainly responsible for treating domestic sewage of urban residents, and the sewage treated in sewage treatment plants may also contain a small amount of industrial wastewater. A sewage treatment plant is a complex system composed of multiple unit processes. The costs and efficiencies of each unit process are interrelated and interact with each other, and ultimately determine the cost and efficiency of the entire system. As the core link in the sewage treatment process, the sewage biological treatment system plays an important role in water environmental pollution control. The efficient and stable operation of the sewage biological treatment system is of great significance for the sustainable development of the economy and society. However, the frequent occurrence of fluctuations in the influent load, changes in operating conditions and mechanical equipment failures in the sewage biological treatment system during the sewage treatment process, and the transience and uncertainty of the process operation pose severe challenges to the stable operation and energy conservation and consumption reduction of the sewage biological treatment system.

[0003] For the prior art such as Prior Art 1: CN202311191814.3; its system is independently controlled and cannot achieve overall optimal regulation; for the process simulation and simulation of all processes, effective execution cannot be achieved. Summary of the Invention

[0004] In view of the problems in the prior art that its system is independently controlled and cannot achieve overall optimal regulation, and for the process simulation and simulation of all processes, effective execution cannot be achieved, the present invention provides a sewage treatment device, system, method and storage medium based on AI decision control.

[0005] To solve the above technical problems, the present invention is solved by the following technical solutions:

[0006] A sewage treatment method based on AI decision control, including

[0007] Data collection and real-time processing, using on-line instruments to collect data of sewage treatment in real time;

[0008] Determination of sewage treatment load, perceiving and allocating the load of the collected data through AI to determine the real-time sewage treatment load;

[0009] Control of sewage, controlling the sewage according to the real-time sewage treatment load.

[0010] Preferably: perceiving and allocating the load of the collected data to determine the sewage treatment load includes:

[0011] Monitoring the real-time total nitrogen (TN) removal rate Q1 of the first system and the real-time total nitrogen (TN) removal rate Q2 of the second system, and monitoring the real-time total nitrogen (TN) removal rate Q1 of the first system and the real-time total nitrogen (TN) removal rate Q2 of the second system in real time;

[0012] The real-time total nitrogen (TN) removal rate Q1 of the first system is:

[0013]

[0014] The real-time total nitrogen (TN) removal rate Q2 of the second system is:

[0015]

[0016] Setting the dynamic target value. According to the monitoring of the real-time total nitrogen (TN) removal rate Q1 of the first system and the real-time total nitrogen (TN) removal rate Q2 of the second system, determine the dynamic target value;

[0017] Determining the sewage treatment load. According to the dynamic target value, determine the real-time sewage treatment load.

[0018] Preferably: The setting and calculation of the dynamic target value include:

[0019] Comparing the real-time total nitrogen (TN) removal rate Q1 of the first system and the real-time total nitrogen (TN) removal rate Q2 of the second system; when the real-time total nitrogen (TN) removal rate Q1 of the first system is greater than the real-time total nitrogen (TN) removal rate Q2 of the second system, then determine the dynamic target value through the first system; otherwise, determine the dynamic target value through the second system.

[0020] Preferably: Determining the dynamic target value according to the first system includes:

[0021] Based on the TN removal rate of the first system, calculate the real-time total nitrogen (TN) removal amount of the first system under the maximum carbon source dosage;

[0022] Calculate the difference between the total influent real-time total nitrogen (TN) amount of the first system and the maximum real-time total nitrogen (TN) removal amount of the first system;

[0023] If the difference between the total influent real-time total nitrogen (TN) amount of the first system and the maximum real-time total nitrogen (TN) removal amount of the first system is less than the total effluent TN target value of the first system, then the dynamic target value is the total effluent TN target value; otherwise, the dynamic target value is the sum of the total effluent TN target value of the first system and the minimum frequency TN removal amount of the second system.

[0024] Preferably: Determining the dynamic target value according to the second system includes: Based on the TN removal rate of the second system, calculate the minimum frequency TN removal amount of the second system under the maximum carbon source dosage;

[0025] The dynamic target value is the sum of the total effluent TN target value of the first system and the TN removal amount at the lowest frequency of the second system.

[0026] Preferably: The sewage is controlled according to the real-time sewage treatment load: The dynamic target value is allocated to the first system and the second system in real time according to the real-time total nitrogen (TN) removal rate Q1 of the first system and the real-time total nitrogen (TN) removal rate Q2 of the second system.

[0027] To solve the above technical problems, the present invention also provides a sewage treatment device based on AI decision control, which includes:

[0028] A data acquisition and real-time processing module, which uses on-line instruments to collect sewage treatment data in real time;

[0029] A sewage treatment load determination module, which senses and apportions the load of the collected data to determine the real-time sewage treatment load;

[0030] A sewage control module, which controls the sewage according to the real-time sewage treatment load.

[0031] To solve the above technical problems, the present invention also provides a computer program product, which is characterized in that when running on a computer, it executes the above method.

[0032] To solve the above technical problems, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and the program is configured to implement the above method when executed.

[0033] Due to the adoption of the above technical solutions, the present invention has significant technical effects:

[0034] Real-time monitoring and dynamic adjustment of the present invention: By using on-line instruments to collect sewage treatment data in real time and dynamically adjusting the treatment process according to the real-time data, the efficiency of sewage treatment can be significantly improved. For example, the setting of the dynamic target value and the real-time load allocation can ensure that the sewage treatment system operates in the best state and reduce the treatment time. Optimization of resource allocation: The intelligent control device can dynamically allocate the sewage treatment load according to the real-time data, optimize the resource allocation, improve the equipment utilization rate, reduce the equipment idle time, and thus improve the overall treatment efficiency.

[0035] This invention reduces energy consumption: By dynamically adjusting the treatment process, the intelligent control device can optimize energy usage and reduce unnecessary energy consumption. For example, by reasonably allocating loads and avoiding overoperation of equipment, energy costs can be reduced. Reduces labor costs: The automated and intelligent operation of the intelligent control device reduces the dependence on manual operation and lowers labor costs. At the same time, it reduces the additional costs caused by human operation errors. Improves equipment lifespan: By optimizing the operating state of equipment, reducing overuse and wear of equipment, the service life of equipment is extended, and the costs of equipment replacement and maintenance are reduced.

[0036] This invention improves treatment quality: The intelligent control device can ensure the stability and reliability of the sewage treatment process, improve the quality of sewage treatment, reduce pollutant emissions, which is of great significance for environmental protection. Reduces the use of chemical agents: By dynamically adjusting the treatment process, the intelligent control device can optimize the use of chemical agents, reduce the waste of chemical agents and secondary pollution to the environment.

[0037] This invention features automated operation: The automated and intelligent operation of the intelligent control device reduces manual intervention, improves the convenience and accuracy of operation. Operators can easily understand the status of the sewage treatment process through the monitoring system and make necessary adjustments. Easy to maintain and upgrade: The modular design makes the maintenance and upgrade of the intelligent control device more convenient. When the system needs to be updated or modules need to be replaced, it can be completed quickly, reducing downtime.

[0038] After introducing the intelligent control device designed according to this invention, through real-time data collection and dynamic adjustment, the treatment efficiency has increased by 30%, the energy consumption has decreased by 20%, the treatment quality has been significantly improved, and the effluent water quality has stably reached the standard. Description of the Drawings

[0039] Figure 1 is the schematic diagram of the architecture of this invention.

[0040] Figure 2 is the schematic diagram of load sharing of this invention.

[0041] Figure 3 is the flowchart of obtaining the sewage treatment load of this invention.

[0042] Figure 4 is the flowchart of the influent load of this invention. Detailed Description of the Invention

[0043] The following further describes this invention in detail in conjunction with the drawings and embodiments.

[0044] Embodiment 1

[0045] A sewage treatment method based on AI decision control, combined with Figure 1 and Figure 2 which includes

[0046] Data collection and real-time processing, using on-line instruments to collect sewage treatment data in real time;

[0047] Determination of sewage treatment load, perceiving and sharing the load of the collected data through AI to determine the real-time sewage treatment load;

[0048] Sewage control, controlling sewage according to the real-time sewage treatment load.

[0049] Figure 3 Among them, perceiving and sharing the load of the collected data to determine the sewage treatment load includes:

[0050] Monitoring the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system, and monitoring the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system in real time;

[0051] The real-time total nitrogen TN removal rate Q1 of the first system is:

[0052]

[0053] The real-time total nitrogen TN removal rate Q2 of the second system is:

[0054]

[0055] Setting of dynamic target value, determining the dynamic target value according to monitoring the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system;

[0056] Determination of sewage treatment load, determining the real-time sewage treatment load according to the dynamic target value.

[0057] Setting and calculation of dynamic target value include:

[0058] Comparing the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system; when the real-time total nitrogen TN removal rate Q1 of the first system is greater than the real-time total nitrogen TN removal rate Q2 of the second system, the dynamic target value is determined through the first system; otherwise, the dynamic target value is determined through the second system.

[0059] Determining the dynamic target value according to the first system includes:

[0060] Calculating the real-time total nitrogen TN removal amount of the first system under the maximum carbon source dosage according to the TN removal rate of the first system;

[0061] Calculating the difference between the total inlet real-time total nitrogen TN amount of the first system and the maximum real-time total nitrogen TN removal amount of the first system;

[0062] If the difference between the real-time total nitrogen (TN) amount of the total influent water of the first system and the maximum real-time TN removal amount of the first system is less than the TN target value of the total effluent water of the first system, the dynamic target value is the TN target value of the total effluent water; otherwise, the dynamic target value is the sum of the TN target value of the total effluent water of the first system and the TN removal amount at the lowest frequency of the second system.

[0063] Determining the dynamic target value according to the second system includes: calculating the TN removal amount at the lowest frequency of the second system under the maximum carbon source dosage based on the TN removal rate of the second system;

[0064] The dynamic target value is the sum of the TN target value of the total effluent water of the first system and the TN removal amount at the lowest frequency of the second system.

[0065] Figure 4 Among them, the sewage is controlled according to the real-time sewage treatment load: the dynamic target value is allocated to the first system and the second system in real time according to the real-time total nitrogen (TN) removal rate Q1 of the first system and the real-time total nitrogen (TN) removal rate Q2 of the second system.

[0066] Example 2

[0067] Based on Example 1, this example is a sewage treatment device based on AI decision control, which includes:

[0068] A data acquisition and real-time processing module, which uses on-line instruments to collect sewage treatment data in real time;

[0069] A sewage treatment load determination module, which senses and apportions the load for the collected data to determine the real-time sewage treatment load;

[0070] A sewage control module, which controls the sewage according to the real-time sewage treatment load.

[0071] Example 3

[0072] Based on Example 1, this example is a computer program product, which is characterized in that when it runs on a computer, it executes the described method.

[0073] Example 4

[0074] Based on Example 1, this example is a computer-readable storage medium, on which a computer program is stored, and the program is configured to implement the described method when executed.

[0075] Example 5

[0076] Based on Example 1, the loads in this example are hydraulic load, pollutant load, volume load, sludge load, and temperature. All generalized loads that have an impact on production control are referred to as load Y; load perception = continuously cycle through the process of hypothesis verification and expectation adjustment until cognitive convergence.

[0077] The load perception process is as follows:

[0078] Hypothesis: AI believes that the current load level in the biochemical pool is Y = 1460;

[0079] Prediction: After performing the optimal action, the DO will drop from 2.8 mg / L to 2.0 mg / L after 2 minutes;

[0080] Action: Fan power "-2%", air valve opening "-3.6°";

[0081] Observation: After 2 minutes, DO = 2.3 mg / L;

[0082] Reality - expectation gap: DO +0.3 mg / L;

[0083] New hypothesis: The DO drops slower than expected, the load is lower than expected, and AI believes that the current load level is Y = 1330;

[0084] Prediction: After performing the optimal action, the DO will drop from 2.3 mg / L to 2.0 mg / L after 2 minutes;

[0085] Action: Fan power "-1%", air valve opening "-1.5°";

[0086] Observation: After 2 minutes, DO = 2.0 mg / L; Reality - expectation gap: DO 0.0 mg / L

[0087] At this time, the load Y is the same as the expected Y = 1330.

[0088] For load sharing: Automatically calculate the feasibility of load transfer according to the load perceived by AI;

[0089] Based on the calculation results, achieve dynamic load sharing while ensuring safety and economy.

[0090] In this example, the first system is the anoxic tank, and the second system is the denitrification filter; by comparing the TN removal rate Q1 of the anoxic tank and the TN removal rate Q2 of the denitrification filter, determine which system has a higher TN removal rate; if Q1 > Q2, calculate the TN removal amount R1 at the maximum dosing amount based on Q1 of the anoxic tank;

[0091] The difference between the influent TN amount R and the effluent target TN amount R2 is R3

[0092] If the anoxic tank R1 is compared with the difference R3 and R1 > R3, the operation and maintenance decision center module decides to add medicine only in the anoxic tank, and the TN removal is completed in the anoxic tank

[0093] If R1 ≤ R3, the operation and maintenance decision center module decides to add medicine at the minimum power of the denitrification filter, and the remaining TN to be removed is completed in the anoxic tank.

[0094] If Q1 ≤ Q2, calculate the maximum chemical dosage TN removal amount T1 based on Q2;

[0095] The difference T3 between the influent TN amount T and the effluent target TN amount T2; Compare T1 with T3. If T1 > T3, the operation and maintenance decision center module decides to add medicine only in the denitrification filter, and the TN removal is completed in the denitrification filter

[0096] If T1 ≤ T3, the operation and maintenance decision center module decides to add medicine at the minimum power of the anoxic tank, and the remaining TN to be removed is completed in the denitrification filter. If adding medicine at the minimum power of the anoxic tank and adding medicine at the maximum power of the denitrification filter still do not meet the effluent requirements, then add medicine at the maximum power of the denitrification filter, and the remaining TN is removed in the anoxic tank. Finally, through the perception and sharing of the load, combined with the floating DO control effect, the self-optimization of the denitrification process control target is realized, that is, the "TN floating target value" in the biochemical section is self-set.

[0097] For example, the total influent TN amount (R) is 100 mg / L; the total effluent TN target value (R2) is 10 mg / L; the removal rate of the AO tank is 70%, and the removal rate of the DNF tank is 30%. The TN removal amount (R1) under the maximum carbon source dosage in the AO tank is 80 mg / L. The TN removal amount at the minimum frequency of the DNF tank is 5 mg / L. Dynamic target value setting and distribution:

[0098] Calculate the difference: R3 = R - R1 = 100 - 80 = 20 mg / L;

[0099] Compare the difference: R3 (20 mg / L) > R2 (10 mg / L), so the dynamic target value = R2 + the TN removal amount at the minimum frequency of the DNF tank = 10 + 5 = 15 mg / L.

[0100] Dynamic distribution:

[0101] The target value of the AO tank = 15 × 70% = 10.5 mg / L;

[0102] The target value of the DNF tank = 15 × 30% = 4.5 mg / L.

[0103] Adjust the carbon source dosage:

[0104] Increase the carbon source dosage in the AO tank to achieve the target value of 10.5 mg / L;

[0105] Adjust the carbon source dosage in the DNF pool to achieve the target value of 4.5 mg / L.

Claims

1. A sewage treatment method based on AI decision control, including Data collection and real-time processing, using on-line instruments to collect sewage treatment data in real time; Determination of sewage treatment load, using AI to sense and allocate the load for the collected data to determine the real-time sewage treatment load; Control of sewage, controlling sewage according to the real-time sewage treatment load.

2. The sewage treatment method based on AI decision control according to claim 1, characterized in that: Sensing and allocating the load for the collected data to determine the sewage treatment load includes: Monitoring the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system, and monitoring the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system in real time; The real-time total nitrogen TN removal rate Q1 of the first system is: The real-time total nitrogen TN removal rate Q2 of the second system is: Setting of dynamic target value, determining the dynamic target value according to monitoring the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system; Determination of sewage treatment load, determining the real-time sewage treatment load according to the dynamic target value.

3. The sewage treatment method based on AI decision-making control according to claim 1, wherein: Setting and calculation of dynamic target value includes: Comparing the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system; when the real-time total nitrogen TN removal rate Q1 of the first system is greater than the real-time total nitrogen TN removal rate Q2 of the second system; then determining the dynamic target value through the first system; otherwise, determining the dynamic target value through the second system.

4. The sewage treatment method based on AI decision-making control according to claim 1, wherein: Determining the dynamic target value according to the first system includes: Calculating the real-time total nitrogen TN removal amount of the first system under the maximum carbon source dosage according to the TN removal rate of the first system; Calculating the difference between the real-time total nitrogen TN amount of the total influent of the first system and the maximum real-time total nitrogen TN removal amount of the first system; If the difference between the real-time total nitrogen TN amount of the total influent of the first system and the maximum real-time total nitrogen TN removal amount of the first system is less than the TN target value of the total effluent of the first system, then the dynamic target value is the TN target value of the total effluent; otherwise, the dynamic target value is the sum of the TN target value of the total effluent of the first system and the minimum frequency TN removal amount of the second system.

5. The sewage treatment method based on AI decision-making control according to claim 1, wherein: Determining the dynamic target value according to the second system includes: calculating the minimum frequency TN removal amount of the second system under the maximum carbon source dosage according to the TN removal rate of the second system; The dynamic target value is the sum of the TN target value of the total effluent of the first system and the minimum frequency TN removal amount of the second system.

6. The sewage treatment method based on AI decision control according to claim 1, characterized in that: Controlling sewage according to the real-time sewage treatment load: allocating the dynamic target value to the first system and the second system in real time according to the real-time total nitrogen TN removal rate Q1 of the first system and the real-time total nitrogen TN removal rate Q2 of the second system.

7. A sewage treatment device based on AI decision control, including: Data collection and real-time processing module, using on-line instruments to collect sewage treatment data in real time; Sewage treatment load determination module, sensing and allocating the load for the collected data to determine the real-time sewage treatment load; Sewage control module, controlling sewage according to the real-time sewage treatment load.

8. A computer program product, characterized in that, When running on a computer, it executes the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, the program being configured to, when executed, implement the method according to any one of claims 1 to 6.

Images