Intelligent high-density pool water quality control method and system

Abstract

The invention discloses an intelligent high-density pool water quality control method and system, and relates to the technical field of intelligent water quality treatment.The intelligent high-density pool water quality control method comprises the steps that newest data and local data in a high-density pool are collected, the newest data are preprocessed, the preprocessed data are input into a three-dimensional digital twinborn model to update the state of the three-dimensional digital twinborn model, and the state of the three-dimensional digital twinborn model is obtained; and meanwhile, exchanging the local data between adjacent agents to obtain local interaction data of the agents. Extreme working conditions are simulated in the digital twinborn model, a water quality change prediction result is obtained, a bionic swarm intelligence algorithm is started for distributed collaborative optimization, an optimized control instruction is obtained and distributed to each agent, a water quality control task is executed, and a high-density pool water quality state report is generated. According to the method, the ant colony algorithm is used for optimization, the requirement that the water quality reaches the standard is considered, the requirement of saving energy consumption is also considered, the economic adaptability is greatly improved, and it is ensured that efficient water quality management can be achieved through control instructions under various environment conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent water quality treatment, and particularly to an intelligent high-density sedimentation tank water quality control method and system. Background Art

[0002] As the core process in the field of water treatment, the water quality control technology of the high-density sedimentation tank has evolved from mechanical control to intelligent development. With the popularization of the Internet of Things technology, the existing high-density sedimentation tank water quality control technology has gradually introduced a multi-parameter sensor network (such as turbidity, dissolved oxygen), realizing the real-time acquisition and remote control of key water quality indicators.

[0003] There are still many deficiencies in the existing high-density sedimentation tank water quality control methods. When sudden load shocks occur, the existing optimization algorithms (such as genetic algorithms, particle swarm algorithms) have a very slow convergence speed and are easily trapped in local optimal solutions. For example, when the chemical oxygen demand of the influent exceeds the standard, multiple iterations are required to generate a feasible solution, during which the water quality of the high-density sedimentation tank may have deteriorated to an irreversible state. The correlation analysis between existing distributed control instructions is very weak, often causing problems such as over-dosing or flow field disorder. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides an intelligent high-density sedimentation tank water quality control method to solve the problems of insufficient accuracy in working condition prediction and poor coordination of distributed control instructions in the existing high-density sedimentation tank water quality control methods.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides an intelligent high-density sedimentation tank water quality control method, which includes constructing a three-dimensional digital twin model, defining the core positions of the high-density sedimentation tank as independent agents at the same time, setting control rules for each agent, and outputting the initial configuration information of the agent;

[0008] According to the initial configuration information of the agent, collect the latest data and local data in the high-density sedimentation tank, preprocess the latest data, input the preprocessed data into the three-dimensional digital twin model to update the state of the three-dimensional digital twin model, and exchange the local data between adjacent agents at the same time to obtain the local interaction data of the agent;

[0009] According to the local interaction data of the agent, simulate extreme working conditions in the digital twin model, obtain the water quality change prediction result and start the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instructions;

[0010] Assign the optimized control instructions to each agent, execute the water quality control task, and generate a high-density tank water quality status report.

[0011] As a preferred embodiment of the intelligent high-density tank water quality control method described in the present invention, the method includes the following steps of constructing a three-dimensional digital twin model:

[0012] Collect multi-source water quality data and upload it to the central server, select ANSYS Fluent as the core modeling tool, and construct a three-dimensional digital twin model.

[0013] As a preferred embodiment of the intelligent high-density tank water quality control method described in the present invention, the dosing device of the metering pump, the agitator, and the sludge pump are used as agents. Initialize the agents and set control rules, and at the same time verify the agents after setting the control rules to obtain the initial configuration information of the agents.

[0014] As a preferred embodiment of the intelligent high-density tank water quality control method described in the present invention, the latest data in the high-density tank includes the latest dissolved oxygen, sludge concentration, water flow velocity, and water temperature, and the local data includes COD concentration, flow field uniformity, and sludge concentration;

[0015] The preprocessing is performed by applying the five-point moving average method.

[0016] As a preferred embodiment of the intelligent high-density tank water quality control method described in the present invention, select the rainstorm scenario as the extreme working condition, configure the extreme working condition parameters and integrate them into a rainstorm working condition configuration file, and load the rainstorm working condition configuration file into the three-dimensional digital twin model for simulation to output the predicted results of water quality changes.

[0017] As a preferred embodiment of the intelligent high-density tank water quality control method described in the present invention, select the ant colony algorithm as the bionic swarm intelligence algorithm, set the core parameters of the ant colony algorithm, generate an ant exploration space according to the core parameters of the ant colony algorithm, and generate the path result of the control instruction when the ants complete the exploration in the ant exploration space;

[0018] According to the path result of the control instruction, set the multi-objective function and water quality standard, calculate the score of each control instruction path, and output the optimized control instruction.

[0019] As a preferred embodiment of the intelligent high-density tank water quality control method described in the present invention, integrate the optimized control instructions into the edge computing node, use the priority queue mechanism to perform hierarchical scheduling on the optimized control instructions, and after the scheduling is completed, each agent executes an independent water quality control task to generate the final high-density tank water quality status report.

[0020] Second aspect, the present invention provides an intelligent high-density pond water quality control system, including an output configuration module, which constructs a three-dimensional digital twin model, and at the same time defines the core positions of the high-density ponds as independent agents, sets control rules for each agent, and outputs the initial configuration information of the agents;

[0021] A data interaction module, according to the initial configuration information of the agents, collects the latest data and local data in the high-density ponds, preprocesses the latest data, inputs the preprocessed data into the three-dimensional digital twin model to update the state of the three-dimensional digital twin model, and at the same time exchanges the local data between adjacent agents to obtain the local interaction data of the agents;

[0022] An optimization instruction module, according to the local interaction data of the agents, simulates extreme working conditions in the digital twin model, obtains the water quality change prediction result and starts the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instructions;

[0023] A report generation module, distributes the optimized control instructions to each agent, executes the water quality control task, and generates a high-density pond water quality status report.

[0024] Third aspect, the present invention provides a computer device, including a memory and a processor, the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the intelligent high-density pond water quality control method described in the first aspect of the present invention is implemented.

[0025] Fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, wherein: when the computer program is executed by the processor, any step of the intelligent high-density pond water quality control method described in the first aspect of the present invention is implemented.

[0026] The beneficial effects of the present invention are as follows: By simulating extreme working conditions in the digital twin model, obtaining the water quality change prediction result and starting the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instructions, distributing the optimized control instructions to each agent, and executing the water quality control task to generate a high-density pond water quality status report. Using the ant colony algorithm for optimization not only considers the requirement of water quality compliance but also takes into account the requirement of energy consumption saving, greatly improving the adaptability and economy, and ensuring that the control instructions can achieve efficient water quality management under various environmental conditions. The use of the priority queue mechanism ensures the priority execution of key tasks such as COD control, and the application of the high-speed GPIO interface reduces the risk of delay and data loss, improving the execution accuracy of distributed control instructions. Description of the Drawings

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0028] Figure 1 Schematic diagram of the high-density pond water quality control architecture and three-dimensional digital twin model construction in Embodiment 1.

[0029] Figure 2 Flowchart of data collection and local interaction of agents in Embodiment 1.

[0030] Figure 3 Extreme condition simulation and ant colony algorithm optimization path diagram in Embodiment 1. Specific implementation manners

[0031] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific implementation manners of the present invention in detail with reference to the drawings in the specification.

[0032] In the following description, many specific details are set forth to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0033] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that can be included in at least one implementation manner of the present invention. The "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that mutually excludes other embodiments.

[0034] Embodiment 1, referring to Figures 1 to 3 , this embodiment provides an intelligent high-density pond water quality control method, including the following steps:

[0035] S1. Construct a three-dimensional digital twin model, and at the same time define the core positions of the high-density pond as independent agents, load preset rules for each agent, and output the initial configuration information of the agent.

[0036] Including the following steps,

[0037] S1.1. First, deploy a high-precision sensor network to collect multi-source water quality data in the high-density pond. Specifically, select a dissolved oxygen probe (measurement range: 0 - 20 mg / L, accuracy: ±0.1 mg / L), a turbidimeter (measurement range: 0 - 1000 NTU, accuracy: ±1 NTU), an ultrasonic flowmeter (measurement range: 0 - 5 m / s, accuracy: ±0.01 m / s), and a temperature sensor (measurement range: 0 - 50 °C, accuracy: ±0.1 °C). These four sensors form a four-in-one sensor group, and each group is fixedly installed at three main positions in the high-density pond: the water inlet (monitoring the input water quality characteristics), the central area (reflecting the average state in the high-density pond), and the sludge outlet (detecting the bottom sediment situation). During installation, the sensor probe is inserted 0.5 meters underwater and fixed on a stainless steel bracket to avoid interference from floating objects. The sampling frequency is set to once every five minutes. After startup, an original data set is collected and generated, that is, dissolved oxygen 6 mg / L, sludge concentration 700 mg / L, water flow velocity 0.4 m / s, and water body temperature 25 °C. The four types of data are uploaded to the central server in the form of encrypted data packets through an Internet of Things gateway (transmission rate: 10 Mbps). The Internet of Things gateway is installed in the rainproof box on the border of the high-density pond, and it should be noted that the distance to the nearest sensor should not exceed 20 meters to ensure stable signals.

[0038] S1.2. After the central server receives the four types of data uploaded by the sensors, it is necessary to construct a three-dimensional digital twin model. Specifically, the central server needs to be configured as a high-performance workstation and then run ANSYS Fluent software (as the core modeling tool) to process the four types of uploaded data. First, open the ANSYS Fluent software interface. Enter the specific physical dimensions of the high-density tank in the geometry module, such as a length of 20 m, a width of 12 m, and a depth of 5 m. The specific physical dimensions are provided by the design drawings of the high-density tank body, and it is necessary to ensure consistency with the reality. The ANSYS Fluent software will automatically generate a basic three-dimensional rectangular framework (used to represent the outer boundary of the high-density tank body, that is, the three-dimensional geometric network). Next, set the mesh division parameters in the mesh module of the ANSYS Fluent software, specify the number of mesh cells as 500,000 (the side length of each cell is about 0.1 m), select the hexahedral mesh type to improve the calculation accuracy. The division process takes about 2 minutes. After the division is completed, a fine mesh structure is generated, covering the entire interior of the high-density tank body, which can capture the details of the local flow field and material distribution, such as the high-flow velocity area near the inlet or the low-velocity deposition area at the bottom of the sludge outlet. After the mesh structure is generated, the central server receives the four types of data uploaded by the sensors (i.e., dissolved oxygen of 6 mg / L, sludge concentration of 700 mg / L, water flow velocity of 0.4 m / s, and temperature of 25 °C) and imports them into the settings module of the ANSYS Fluent software, and distributes them to the corresponding areas of the mesh structure according to the positions (i.e., the inlet, the central area, and the sludge outlet). For example, the dissolved oxygen at the inlet grid point is assigned 7 mg / L, the central area is 6 mg / L, and the sludge outlet is 4 mg / L.

[0039] S1.3. The next operation is to activate the computational fluid dynamics module in the options of the physical model of the ANSYS Fluent software, select the turbulence model (k-ε model) to simulate the nature of water flow turbulence, and enable the multiphase flow function to simulate the interaction between sludge particles and water bodies. The set boundary conditions include an inlet water flow velocity of 0.6 m / s, the pressure at the sludge outlet is the atmospheric pressure, and the pool wall is a no-slip wall surface (all three boundary conditions are determined based on the specific operating parameters of the high-density tank body). Then adjust the solution parameters, set the time step as 0.1 s, simulate the steady-state distribution of the initial state, and after clicking the calculation button, the ANSYS Fluent software completes the simulation and generates detailed results of the flow field and material distribution in the pool (i.e., the preliminary simulation results), such as the sludge concentration gradually increasing from 500 mg / L at the inlet to 900 mg / L at the sludge outlet, the water flow velocity decreasing from 0.5 m / s in the center to 0.2 m / s at the edge, and the dissolved oxygen decreasing from 6 mg / L on the surface to 4 mg / L at the bottom (reflecting the spatial heterogeneity in the high-density tank). Finally, the constructed three-dimensional digital twin model is stored in the local hard disk of the central server in the fluent file format.

[0040] S1.4. To achieve distributed control, it is necessary to initialize the agents at the core position of the high-density pond. Specifically, select the chemical dosing device of the metering pump, with the dosing range of 0 - 10 g / m 3 , the agitator (with a power of 500 W and a rotation speed range of 0 - 500 revolutions per minute), and the sludge discharge pump (with a flow rate range of 0 - 100 L per time) as the agent carriers. Each agent carrier is equipped with an embedded microcontroller of ARM Cortex-M4 and a supporting local sensor (i.e., the chemical dosing device of the metering pump is equipped with a COD sensor, with a range of 0 - 500 mg / L and an accuracy of ±2 mg / L; the agitator is equipped with a rotation speed sensor; the sludge discharge pump is equipped with a flow meter).

[0041] When installing the microcontroller, it needs to be fixed in the equipment control box. All sensors are adjacent to the execution components (for example, the COD sensor is placed 5 cm away from the chemical dosing outlet). Each agent forms a star-shaped local communication network through the ZigBee communication network, and the connection distance of the farthest agent should not exceed 80 meters. At the same time, set control rules for each agent, that is, the chemical dosing device of the metering pump is set to increase the sodium hypochlorite dosing amount at a rate of 0.5 g / m per minute when the COD (i.e., chemical oxygen demand) exceeds 120 mg / L until the COD drops below 100 mg / L. The agitator is set to increase the rotation speed by 50 revolutions per minute when the flow rate is lower than 0.3 m / s. The sludge discharge pump is set to discharge sludge once every 6 hours, with each discharge of 50 L (all three preset control rules are set according to the preliminary simulation results). 3

[0042] S1.5. After all agents are initialized, conduct a self-check communication through the ZigBee communication network to let each agent report its current status. For example, the chemical dosing device of the metering pump detects that the COD is 100 mg / L, the rotation speed of the agitator is 200 revolutions per minute, and the sludge discharge pump is on standby. The communication takes about 5 seconds. Upload the current status detection results of the agents to the central server and synchronize and verify them with the four types of data in the 3D digital twin model. For example, confirm that the COD (100 mg / L) displayed in the central area of the 3D digital twin model is consistent with the measured value of the chemical dosing device of the metering pump, and the deviation is controlled within 2%. If it exceeds the standard, re-collect the data and adjust the 3D digital twin model. After passing the verification, output the initial configuration information of the agents (including the current status of the agents, such as the COD of 100 mg / L of the chemical dosing device of the metering pump and the preset rules, such as dosing 0.5 g / m when the COD exceeds the standard 3 ).

[0043] ​S2. According to the initial configuration information of the agent, collect the latest data and local data in the high-density pool, preprocess the latest data, input the preprocessed data into the three-dimensional digital twin model to update the state of the three-dimensional digital twin model, and at the same time exchange the local data among adjacent agents to obtain the local interaction data of the agents.

[0044] It includes the following steps.

[0045] S2.1. The latest data in the high-density pool here refers to the latest dissolved oxygen, sludge concentration, water flow velocity, and water temperature. For the specific collection process, a dissolved oxygen probe, a turbidimeter, an ultrasonic flowmeter, and a temperature sensor are also installed at three main positions in the high-density pool to collect the latest data.

[0046] S2.2. Send the collected latest data to the industrial-grade edge computing node, and the industrial-grade edge computing node preprocesses the latest data. Specifically, apply the five-point moving average method to each group of data, that is, take the average of the current data point and the two time points before and after (a total of 5 minutes) to smooth out abnormal fluctuations. For example, assuming the dissolved oxygen sequence is [5.2, 6.0, 5.8, 5.0, 5.1] and the sludge concentration sequence is [780, 820, 790, 800, 810], after being processed by the five-point moving average method, the dissolved oxygen becomes 5.42 mg / L and the sludge concentration becomes 800 mg / L.

[0047] S2.3. Open the real-time update module of ANSYS Fluent software, import the preprocessed data into the three-dimensional digital twin model through the real-time update module, and allocate the preprocessed data by position using 500,000 cells. For example, update the inlet to dissolved oxygen 5.5 mg / L, the central area 5 mg / L, and the sludge outlet 4.8 mg / L, and update the sludge concentrations to 750 mg / L, 800 mg / L, and 850 mg / L respectively. Then run a quick simulation to generate the updated state of the three-dimensional digital twin model. For example, the area where the dissolved oxygen decreases is concentrated at the bottom of the high-density pool (4.8 mg / L), and the sludge concentration rises to 850 mg / L near the sludge outlet. The updated three-dimensional digital twin model is displayed through a visualization interface (i.e., dynamic curves and heat maps). The dynamic curves show the parameter change trends within 5 minutes (such as the curve of dissolved oxygen decreasing from 6 mg / L to 5 mg / L), and the heat map marks the low dissolved oxygen area (blue, bottom of the pool) and the high sludge area (red, sludge outlet). The heat map supports dynamic adjustment, such as dragging the mouse to rotate the view to view the side of the pool body, or using the cross-section tool to cut the central area to check the vertical changes in flow velocity and dissolved oxygen. The update frequency is maintained at once every 5 minutes.

[0048] S2.4. The local data here refers to the COD concentration, flow field uniformity, and sludge concentration, which are specifically collected using a COD sensor, a rotational speed sensor, and a flow meter. After the local data is collected, the ZigBee communication network (with a data rate of 250 kbps) is started to perform data exchange, and the exchange frequency is once every 10 seconds. The specific exchange process is as follows: The dosing device of the metering pump broadcasts the data packet of the COD concentration through the ZigBee communication network to the star-shaped local communication network (the specific meaning of this broadcast here is that the agent simultaneously sends any one of the local data to other devices within the star-shaped local communication network through the ZigBee communication network, rather than the point-to-point transmission communication method). The broadcast signal is first sent to the central communication tower (at a distance of 50 m). When the central communication tower receives it, it forwards it to the stirrer (at a distance of 30 m) and the sludge discharge pump (at a distance of 40 m). After the stirrer receives it, it analyzes the data packet and records the COD concentration value. At the same time, it broadcasts 70% of the data on the flow field uniformity, which is forwarded by the central communication tower to the dosing device of the metering pump and the sludge discharge pump. The sludge discharge pump also broadcasts the sludge concentration, which is forwarded by the central communication tower to the other two parties. It should be noted that the entire exchange process should be controlled within 50 milliseconds. When all agents confirm successful reception, the data between each agent's exchanges (for example, the dosing device of the metering pump records that 70% of the data on the flow field uniformity may affect the chemical agent diffusion) is recorded and stored in the temporary buffer (with a capacity of 16 KB) to form the local interaction data of the agent.

[0049] S3. According to the local interaction data of the agent, simulate the extreme working conditions in the digital twin model, obtain the water quality change prediction results, and start the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instructions.

[0050] including the following steps

[0051] S3.1. Based on the interaction data of the agent, it is necessary to configure extreme operating condition parameters through adaptive boundary conditions to simulate the future water quality changes. The extreme operating condition parameters include the increase in water body flow rate, dissolved oxygen, and sludge concentration. Specifically, considering the problem of a sharp increase in flow rate during the rainy season that the high-density sedimentation tank often faces, the rainstorm scenario is selected as the extreme operating condition. First, determine the current inlet state of the high-density sedimentation tank (assuming the water body flow rate is 100 cubic meters per hour, the dissolved oxygen is 5.5 mg / L, and the sludge concentration is 750 mg / L). Use the meteorological disturbance database to statistically analyze the probability of rainstorm events in the high-density sedimentation tank area (this meteorological disturbance database stores the rainfall data in the area where the high-density sedimentation tank is located in the past 5 years, including the intensity, duration, rainfall type, and rainfall frequency of each rainfall, and the rainstorm event is defined as the rainfall exceeding 50 millimeters per hour. Filter out the data that meet the rainstorm event from the rainfall data). According to the probability of rainstorm events in the high-density sedimentation tank area, start real-time water body flow-capacity coupling analysis to determine the specific increase in water body flow rate. Specifically, first check the instantaneous load-bearing limit of the high-density sedimentation tank body (assuming it is 200 cubic meters per hour, which can be understood as the highest flow rate that the high-density sedimentation tank body can safely handle without triggering overflow). Combine the current water body flow rate at the inlet to calculate the remaining capacity of the high-density sedimentation tank, that is, subtract the current water body flow rate of 100 cubic meters per hour at the inlet from the instantaneous load-bearing limit of 200 cubic meters per hour of the high-density sedimentation tank body, and the remaining capacity of the high-density sedimentation tank is obtained as 100 cubic meters per hour. Considering that the actual impact of the rainstorm will not reach the instantaneous load-bearing limit, initially estimate the increase in water body flow rate to be 35%, that is, increase the water body flow rate to 135 cubic meters per hour (the water body flow rate of 100 cubic meters per hour at the inlet plus 35% of the remaining capacity of 100 cubic meters per hour of the high-density sedimentation tank, that is, 35 cubic meters per hour).

[0052] Based on the initially estimated increase in water body flow rate, combined with the specific volume of the high-density sedimentation tank body (assuming it is 1000 cubic meters) and the rainstorm duration (assuming it is 6 hours), estimate the total water input volume of the high-density sedimentation tank to be 135 cubic meters per hour multiplied by 6 hours, which is equal to 810 cubic meters, occupying 81% of the high-density sedimentation tank body capacity, close to but not exceeding 90% (i.e., the safe range). For further optimization, compare the historical rainstorm cases of the high-density sedimentation tank and find that an increase of 40% (i.e., 140 cubic meters per hour, and the total water input volume is 840 cubic meters) is more common under the condition of an 80% rainstorm probability, and the total water input volume is still controlled within the safe range (84% capacity). Finally, verify whether 140 cubic meters per hour exceeds the instantaneous load-bearing limit, and confirm that there is still a margin of 60 cubic meters per hour, meeting the design constraints. Therefore, the specific increase in water body flow rate is determined to be 40%, adjusted from 100 cubic meters per hour to 140 cubic meters per hour.

[0053] S3.2. Next, the operating parameters of dissolved oxygen need to be configured. Specifically, since the dissolved oxygen content at the current inlet of the high-density tank is 5.5 mg / L, and considering that the water flow rate has increased from 100 cubic meters per hour to 140 cubic meters per hour due to heavy rain, first calculate the change in water velocity. Assuming the initial water velocity at the inlet is 0.6 m / s, then the cross-sectional area of the inlet is the current water flow rate of 100 cubic meters per hour at the inlet of the high-density tank divided by 3600 seconds and then multiplied by the initial water velocity of 0.6 m / s at the inlet, which is approximately 0.167 square meters. When the water flow rate increases to 140 cubic meters per hour, the water velocity at the inlet is 0.84 m / s (obtained by dividing 140 cubic meters per hour by 3600 seconds and then dividing by the cross-sectional area of the inlet, 0.167 square meters). In this way, the change in water velocity can be compared, which is approximately 40%.

[0054] S3.3. Use the meteorological disturbance database to analyze the impact of the change in water velocity on oxygen mass transfer. Specifically, since the high water velocity of the water body will accelerate the contact between the water body and the air, but heavy rain will also bring low-oxygen rainwater (typical value is 4 - 6 mg / L), so the dilution effect dominates. Take the reference value of low-oxygen rainwater as 5 mg / L (which is the average value of low-oxygen rainwater), and compare it with the dissolved oxygen content of 5.5 mg / L at the current inlet of the high-density tank to estimate the mixed value after the dissolved oxygen content drops. The specific estimation process is as follows: Assume that the new dissolved oxygen is the value obtained by weighted averaging the volumes of the two. That is, the current water flow rate of 100 cubic meters per hour in the high-density tank contributes 5.5 mg / L, and the newly added water flow rate of 40 cubic meters per hour contributes 5 mg / L. Under the condition of the adjusted total water flow rate of 140 cubic meters per hour, calculate the mixed value after the dissolved oxygen content drops as (100 multiplied by 5.5 plus 40 multiplied by 5) divided by 140, which is approximately 5.36 mg / L. However, further considering the turbulence effect of the water velocity of 0.84 m / s at the inlet, this will enhance the attenuation of oxygen mass transfer (due to the turbulence effect causing the low-oxygen water at the bottom to surge). According to the water velocity attenuation coefficient table (preset in the meteorological disturbance database, with a decrease of 0.05 mg / L for every 0.1 m / s increase in water velocity), a 0.24 m / s increase in water velocity corresponds to a decrease of 0.12 mg / L. Subtract 0.12 mg / L from the mixed value of 5.36 mg / L after the dissolved oxygen content drops to get 5.24 mg / L. To reflect the most unfavorable low-oxygen characteristics of heavy rain, take the integer of the dissolved oxygen to 4 mg / L (i.e., the lower limit), and this adjustment is consistent with the low-oxygen characteristics of rainwater (starting from 4 mg / L).

[0055] S3.4. Finally, configure the operating parameters of the sludge concentration. When adjusting the sludge concentration, call the deposition-scouring equilibrium algorithm to analyze the impact of rainstorms based on the sludge concentration of 750 mg / L at the inlet of the current high-density tank. The water flow rate increases from 100 cubic meters per hour to 140 cubic meters per hour, with an increase of 40%. First, evaluate the scouring and deposition effects of this change on the sludge. Since the increased rainstorm flow will wash in new sediments, assume that every 10 cubic meters per hour increase in flow rate brings 5 mg / L of new sludge (based on the regional soil particle content). The additional 40 cubic meters per hour corresponds to an increase of 50 mg / L. The initial estimated sludge concentration rises to 800 mg / L (i.e., the current sludge concentration of 750 mg / L at the inlet of the high-density tank plus the corresponding increase of 50 mg / L per hour). However, the rainstorm will simultaneously dilute the water in the high-density tank. According to the current water volume in the high-density tank (assumed to be 800 cubic meters, close to 80% of the volume of 1000 cubic meters) and the additional water volume (40 cubic meters per hour, 240 cubic meters in 6 hours), the total water volume increases to 1040 cubic meters (overflow treatment). Calculate the dilution ratio as the initial 800 cubic meters divided by the total water volume of 1040 cubic meters, approximately 0.77. Multiply the original sludge concentration of 750 mg / L by the dilution ratio of 0.77 to get 577 mg / L, which means the sludge concentration decreases by approximately 173 mg / L after dilution. Considering the combined effects of sediment inflow and dilution, calculate the net change in the sludge concentration. Subtract the 173 mg / L reduced by dilution from 800 mg / L (including the additional 50 mg / L) to get 627 mg / L. To conform to the actual law of rainstorm scouring, refer to historical rainstorm data (the influent sludge concentration is between 550 and 650 mg / L). Therefore, the final value of the sludge concentration is rounded to 600 mg / L to ensure a conservative estimate.

[0056] S3.5. Integrate the three extreme condition parameters of water body flow rate increase, dissolved oxygen, and sludge concentration into a rainstorm condition configuration file. Load the rainstorm condition configuration file into the 3D digital twin model, and at the same time configure the simulation parameters to simulate extreme conditions (the simulation parameters are a time step of 1 minute, a total simulation duration of 12 hours, and an iteration number of 20 time steps). Specifically, the 3D digital twin model extracts the rainstorm condition configuration file from memory, initializes 700,000 grid cells at the same time, adjusts the inlet flow velocity to 0.84 m / s, and sets the dissolved oxygen to 4 mg / L. After the simulation starts, by allocating grid tasks, with a time step of 1 minute, gradually deduce the whole process of water body flow distribution, dissolved oxygen migration, and sludge deposition. During the simulation, the 3D digital twin model will update the grid state in real time (that is, update the progress bar and the residual convergence curve). The residual is used as a measure of the physical consistency of the simulation results. When the residual convergence curve converges to 0.001, the calculation is confirmed to be stable. In the simulation, the input of an influent flow rate of 140 cubic meters per hour causes the bottom layer flow velocity to drop to 0.1 m / s. The dilution of dissolved oxygen and the deposition of sludge concentration gradually appear in the grid. The prediction data for 12 hours is generated minute by minute to obtain the final predicted result of water quality change (that is, the dissolved oxygen drops from 5 mg / L to 4 mg / L, the dilution effect is significantly improved due to the increase in water body flow rate, the deposition thickness of sludge concentration increases by 2 cm, the deposition is aggravated due to the reduction of the bottom layer water body flow velocity in the high-density tank to 0.1 m / s, and the COD rises to 160 mg / L).

[0057] S3.6. According to the predicted result of water quality change, select the ant colony algorithm as the bionic swarm intelligence algorithm (that is, the core of distributed collaborative optimization, which can be understood as running distributively among each intelligent agent to jointly determine the global optimal control instruction). Specifically, first, the core parameters of the ant colony algorithm need to be set (including the number of ants, heuristic factor, pheromone concentration, pheromone evaporation coefficient). The number of ants is set to 30. Each ant represents a combination of control instructions of an intelligent agent. The specific number is selected based on the balance between the number of intelligent agents (3) and the calculation efficiency (that is, 10 times the number of intelligent agents). The heuristic factor (that is, the intuitive tendency to guide ants to select the control instruction path) is set to 2. The usual value range is between 1-3. The median value is taken to balance the exploration and convergence speed, and it is easier to reflect the priority of the control effect (such as the reduction amplitude of COD). The pheromone concentration is set to 1, covering all initial control instruction paths. For example, the dosing of the chemical dosing device of the metering pump is 6 g / m 3 、the rotation speed of the stirrer is 300 revolutions per minute, and the frequency of the sludge discharge pump is every 4 hours. The pheromone evaporation coefficient is set to 0.1, that is, the pheromone concentration decreases by 10% after each iteration, which not only prevents ants from forgetting high-quality control instruction paths too quickly but also prevents falling into local optima.

[0058] After the core parameters of the ant colony algorithm are set, each agent generates an initial control range of an ant as the exploration space of the ant according to the local data of the high-density pond and the prediction result of water quality change. For example, the dosing amount of the dosing device of the metering pump is 5-8 g / m 3 , the rotation speed of the stirrer is 250-350 revolutions per minute, and the sludge discharge pump frequency is every 3-5 hours.

[0059] S3.7. Command 30 ants to disperse and explore within the exploration space of the ants, and each ant randomly selects a combination. For example, the 1st ant selects a dosing amount of 5 g / m 3 , a rotation speed of 250 revolutions per minute, and a frequency of every 5 hours. The 2nd ant selects 6 g / m 3 , 300 revolutions per minute, and every 4 hours, and so on. When the ants are exploring, each agent needs to estimate the exploration effect of the ants. For example, for the dosing device of the metering pump, every 1 g / m 3 reduces COD by 5 mg / L. 5 g / m 3 reduces COD from 150 mg / L to 125 mg / L. The stirrer increases the rotational speed uniformity by 5% per 50 revolutions per minute, and increases from 70% to 80% when it reaches 300 revolutions per minute. For every 1-hour reduction in the frequency of the sludge discharge pump, the sludge thickness decreases by 0.25 cm, and decreases from 2 cm to 1.5 cm every 4 hours. After each exploration by the ants, each agent records the path result of the control instruction through the ZigBee communication network. For example, the combination of the 2nd ant (6 g / m 3 , 300 revolutions per minute, and every 4 hours) corresponds to COD of 110 mg / L, uniformity of 80%, and sludge thickness of 1.5 cm.

[0060] S3.8. According to the path results of the control instructions, set the multi-objective function (the multi-objectives include the water quality compliance rate and the energy consumption saving rate, and the weights of the two objectives are respectively assigned as 80% and 20%, which are set according to the specific operation requirements of the high-density pond) and the water quality standard (set according to the water quality specification requirements of the high-density pond effluent, that is, COD < 100 mg / L, dissolved oxygen > 4.5 mg / L, and the energy consumption limit is dosing < 8 g / m 3 , stirrer rotation speed < 350 revolutions per minute) to evaluate the advantages and disadvantages of the 30 control instruction paths traversed by the ants. Specifically, to calculate the score of each control instruction path. For example, for ant A, the COD of 110 mg / L does not meet the water quality standard compared with the water quality standard, and the dissolved oxygen of 4 mg / L is also lower than the water quality standard. Then, use the compliance ratio, that is, 110 divided by 100 is 1.1. Since it exceeds the standard by 10%, 0.9 is calculated. Then subtract the dissolved oxygen deviation (4.5 - 4 divided by 4.5 is 0.11) and multiply by the weight of 80% to get the water quality compliance rate score of 0.632. And the energy consumption limit of 6 g / m 3 accounts for 8 g / m of the water quality standard 375%, 300 revolutions per minute accounts for 86% of the agitator speed of 350 in the water quality standard. Taking the average value of 80% and multiplying it by the weight of 20%, the score of the energy consumption saving rate is 0.16. Adding the score of the water quality compliance rate and the score of the energy consumption saving rate, the total score is 0.632 + 0.16 = 0.79 (i.e., the score of the control instruction path).

[0061] Update the pheromone concentration according to the score of the control instruction path (i.e., increase the pheromone concentration for the high-score control instruction path and decrease the pheromone concentration for the low-score control instruction path). For example, the pheromone concentration of the traversed control instruction path of ant A increases from the initial setting of 1 to 1.2 (because the score of the control instruction path traversed by ant A is relatively high, that is, 0.79), while the score of the control instruction path traversed by ant B is relatively low (assumed to be 0.42), so the pheromone concentration is reduced from 1 to 0.9.

[0062] S3.9. Repeat the process of updating the pheromone concentration until the preset number of iterations is reached, and then output the optimal control instruction (i.e., the optimized control instruction). Specifically, the ant re-explores according to the updated pheromone (the higher-concentration path has a greater attraction) and the heuristic factor. For example, in the second iteration, the ant tends to 6 g / m 3 、6.5 g / m 3 、320 revolutions per minute, every 4 hours, the COD drops to 98 mg / L, the uniformity is 85%, the sludge thickness is 1 cm, the score of the control instruction path rises to 0.92, and the pheromone concentration increases to 1.3. When all 10 iterations are completed, the pheromone concentration converges to the optimal path, that is, dosing 6 g / m 3 (COD 98 mg / L), rotation speed 320 revolutions per minute (uniformity 85%), frequency every 4 hours (sludge thickness 1 cm), the total score is 0.95, the water quality compliance rate reaches 98%, and the energy consumption saving rate reaches 85%.

[0063] S4. Assign the optimized control instruction to each agent and execute the water quality control task to generate a high-density pond water quality status report.

[0064] Including the following steps,

[0065] S4.1. The central server integrates the optimized control instruction into the edge computing node for distribution processing. Specifically, it is connected to the central server through a wired Ethernet. The central server first encapsulates the optimized control instruction into a structured JSON data frame and transmits it to the edge computing node through the Ethernet. When the edge node receives it, it uses the priority queue mechanism to perform hierarchical scheduling on the optimized control instruction. The priority is sorted according to the degree of influence on water quality, that is, the dosing device of the metering pump (COD control) has the highest priority, set as level 1, the agitator (flow field uniformity) is level 2, and the sludge discharge pump (sludge thickness) is level 3.

[0066] During the scheduling process, the edge node writes the instructions into the queue in order of priority. The instructions for dosing are first assigned to the head of the queue, followed by the agitator and the sludge pump. The edge node directly establishes point-to-point communication with the microcontroller of each intelligent body through the high-speed GPIO interface. The dosing device of the metering pump receives 6g / m 3 Instructions: the agitator is set to 320 rpm, the sludge pump is set to operate once every 4 hours and the sludge discharge volume is 50L.

[0067] S4.2. After the optimized control instructions are assigned, each agent independently performs the water quality control task and monitors the water quality control effect in real time through local sensors. Specifically, the dosing device of the metering pump starts the metering pump and uses the microcontroller to send a pulse signal to the pump driver. The pump driver adds sodium hypochlorite to the high-density pool at a rate of 0.1g per second, and reaches 6g / m in 60 seconds. 3 During the dosing process, the COD sensor samples once every 5 seconds and records the gradual decrease of COD from 150mg / L, for example, 130mg / L after 10 seconds, 110mg / L after 30 seconds, and 98mg / L after 60 seconds, confirming that it drops below 100mg / L. The agitator adjusts the speed to 320 rpm, and the microcontroller drives the motor through the PWM signal (ie, duty cycle). The speed sensor detects once per second, and the flow rate increases from 0.4m / s to 0.5m / s. The sludge pump runs at a frequency of once every 4 hours. The microcontroller sets the timer (cycle is 14400 seconds). Each time the sludge pump runs, it discharges sludge at a flow rate of 50L / minute (a total of 50L, which takes 1 minute). At the same time, the flow meter monitors the sludge discharge volume. The sludge concentration drops from 850mg / L to 700mg / L, and the sludge thickness is reduced to 1 cm.

[0068] S4.3. Upload the water quality monitoring results to the edge computing node for integration to form a water quality status report for the high-density pool (including COD of 98 mg / L (100% compliance rate, target <100 mg / L), flow field uniformity of 85%, sludge concentration of 700 mg / L, sludge deposition thickness of 1 cm (reduced by 50%), with a timestamp and spatial distribution description, such as the COD decrease is mainly concentrated in the central area of the high-density pool).

[0069] The high-density pool water quality report generated in this step utilizes a priority queue mechanism to hierarchically schedule control instructions based on the degree of water quality impact, ensuring that critical tasks (such as dosing) are prioritized. This mechanism improves responsiveness to core water quality indicators and avoids the resource waste associated with conventional uniform distribution. High-speed GPIO interfaces enable point-to-point communication between edge nodes and intelligent agents, reducing latency and the risk of data loss while significantly improving the accuracy of control instruction execution.

[0070] This embodiment also provides an intelligent high-density pond water quality control system, including: an output configuration module, which constructs a three-dimensional digital twin model, defines the core positions of the high-density pond as independent agents at the same time, sets control rules for each agent, and outputs the initial configuration information of the agents;

[0071] A data interaction module, which collects the latest data and local data in the high-density pond according to the initial configuration information of the agents, preprocesses the latest data, inputs the preprocessed data into the three-dimensional digital twin model to update the state of the three-dimensional digital twin model, and exchanges the local data between adjacent agents at the same time to obtain the local interaction data of the agents;

[0072] An optimization instruction module, which simulates extreme working conditions in the digital twin model according to the local interaction data of the agents, obtains the water quality change prediction result and starts the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instructions;

[0073] A report generation module, which distributes the optimized control instructions to each agent, executes the water quality control task, and generates a high-density pond water quality status report.

[0074] This embodiment also provides a computer device, which is applicable to the intelligent high-density pond water quality control method, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to implement the intelligent high-density pond water quality control method proposed in the above embodiment.

[0075] This computer device may be a terminal. This computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of this computer device is used to provide computing and control capabilities. The memory of this computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of this computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a carrier network, NFC (Near Field Communication) or other technologies. The display screen of this computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of this computer device may be a touch layer covered on the display screen, or a button, a trackball or a touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0076] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the intelligent high-density pond water quality control method proposed in the above embodiment. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disc.

[0077] In summary, the present invention: simulates extreme working conditions in the digital twin model to obtain the water quality change prediction results and starts the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instructions, distributes the optimized control instructions to each agent, and executes the water quality control task to generate a high-density pond water quality status report. The use of the ant colony algorithm for optimization not only considers the requirement of water quality compliance but also takes into account the requirement of energy consumption saving, greatly improving the adaptability and economy, and ensuring that the control instructions can achieve efficient water quality management under various environmental conditions. The use of the priority queue mechanism ensures the priority execution of critical tasks such as COD control, and the application of the high-speed GPIO interface reduces the risk of delay and data loss, improving the execution accuracy of distributed control instructions.

[0078] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An intelligent high-density pond water quality control method, characterized in that: including Construct a three-dimensional digital twin model, and at the same time define the core position of the high-density tank as an independent agent, set control rules for each agent, and output the initial configuration information of the agent; According to the initial configuration information of the agent, collect the latest data and local data in the high-density tank, preprocess the latest data, input the preprocessed data into the three-dimensional digital twin model to update the state of the three-dimensional digital twin model, and at the same time exchange the local data between adjacent agents to obtain the local interaction data of the agent; According to the local interaction data of the agent, simulate extreme working conditions in the digital twin model, obtain the predicted result of water quality change and start the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instruction; Allocate the optimized control instructions to each agent, execute the water quality control task, and generate a water quality status report of the high-density tank.

2. The intelligent high-density pond water quality control method according to claim 1, wherein: Construct a three-dimensional digital twin model, including the following steps Collect multi-source water quality data and upload it to the central server, select ANSYS Fluent as the core modeling tool, and construct a three-dimensional digital twin model.

3. The intelligent high-density pond water quality control method according to claim 2, wherein: Take the chemical dosing device, agitator and sludge pump of the metering pump as agents, initialize the agents and set control rules, and at the same time verify the agents after setting the control rules to obtain the initial configuration information of the agents.

4. The intelligent high-density pond water quality control method according to claim 3, characterized in that: The latest data in the high-density tank includes the latest dissolved oxygen, sludge concentration, water flow velocity and water temperature, and the local data includes COD concentration, flow field uniformity and sludge concentration; The preprocessing is carried out by applying the five-point moving average method.

5. The intelligent high-density pond water quality control method according to claim 4, wherein: Select the rainstorm scenario as the extreme working condition, configure the extreme working condition parameters and integrate them into the rainstorm working condition configuration file, load the rainstorm working condition configuration file into the three-dimensional digital twin model for simulation, and output the predicted result of water quality change.

6. The intelligent high-density pond water quality control method according to claim 5, characterized in that: Select the ant colony algorithm as the bionic swarm intelligence algorithm, set the core parameters of the ant colony algorithm, generate the ant exploration space according to the core parameters of the ant colony algorithm, and generate the path result of the control instruction when the ants complete the exploration in the ant exploration space; According to the path result of the control instruction, set the multi-objective function and water quality standard, calculate the score of each control instruction path and output the optimized control instruction.

7. The intelligent high-density pond water quality control method according to claim 6, wherein: Integrate the optimized control instructions into the edge computing node, use the priority queue mechanism to hierarchically schedule the optimized control instructions, and after the scheduling is completed, each agent executes an independent water quality control task to generate the final water quality status report of the high-density tank.

8. An intelligent high-density pond water quality control system, based on the intelligent high-density pond water quality control method according to any one of claims 1 to 7, characterized in that: including Output configuration module, construct a three-dimensional digital twin model, and at the same time define the core position of the high-density tank as an independent agent, set control rules for each agent, and output the initial configuration information of the agent; Data interaction module, according to the initial configuration information of the agent, collect the latest data and local data in the high-density tank, preprocess the latest data, input the preprocessed data into the three-dimensional digital twin model to update the state of the three-dimensional digital twin model, and at the same time exchange the local data between adjacent agents to obtain the local interaction data of the agent; Optimization instruction module, according to the local interaction data of the agent, simulates extreme working conditions in the digital twin model, obtains the water quality change prediction result and starts the bionic swarm intelligence algorithm for distributed collaborative optimization to obtain the optimized control instruction; Report generation module, distributes the optimized control instruction to each agent, executes the water quality control task, and generates the high-density pond water quality status report.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the intelligent high-density pond water quality control method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the intelligent high-density pond water quality control method according to any one of claims 1 to 7.

Images