Intelligent scum degassing device and process based on image online identification

Abstract

The invention discloses a scum degassing intelligent device and process based on image online identification, the scum degassing intelligent device comprises a scum collection module, a scum degassing module, a scum precipitation module and a sludge treatment module, the scum collection module is located at the tail end of an air flotation-precipitation tank, the scum degassing module is connected with the scum collection module through a scum discharge pipe, and the scum discharge pipe is connected with the scum precipitation module through a sludge discharge pipe. The scum degassing module is located below the air floatation-sedimentation tank and connected with the scum sedimentation module through a pipeline, and the scum sedimentation module is connected with the sludge treatment module through a pipeline. The invention provides an intelligent scum degassing device and process based on image on-line identification, which are used for degassing scum, removing gas in the scum, mixing the gas into lower-layer liquid and finally discharging the scum along with the liquid, so that the problem that the scum is difficult to settle and discharge due to the fact that the scum contains a large number of bubbles is solved, and the service life of the scum is prolonged. The scaling phenomenon of the scum treatment system is avoided, harmful gas is prevented from being generated, and the sludge pump is prevented from being eroded by bubbles.

Description

Technical Field

[0001] The present invention relates to the technical field of urban water supply treatment, and in particular to an intelligent device and process for scum degassing based on online image recognition. Background Art

[0002] In the flotation-sedimentation treatment process or the single flotation process, the raw water reacts with the added coagulant in the coagulation zone and enters the flocculation zone. Micro flocs are produced by flocculation and meet the pressurized dissolved air water in the contact chamber at the end of the flocculation zone. Microbubbles attach to the flocs and gradually increase, eventually causing the flocs to float to the water surface to form a scum layer. The sludge is discharged by mechanical scraping or hydraulic pushing to achieve solid-liquid separation. The clarified effluent flows from the bottom of the tank into the clarified water collection area and enters the subsequent treatment stage. A part of the produced water is refluxed for the preparation of pressurized dissolved air water. The scum is discharged from the surface of the tank into the scum collection tank, degassed and enters the scum sedimentation tank, and the sludge formed by the sedimentation is pumped to the sludge treatment system.

[0003] When the flotation process is running, pollutants are removed by forming flocs attached to microbubbles and then floating to the water surface. Therefore, the discharged scum still contains a large amount of gas and is difficult to dissipate on its own. At the same time, the scum will also contain residual flocculants, resulting in its viscous texture and difficulty in dissolving in water. The above two factors will cause the scum discharged from the flotation tank to float on the surface of the waste liquid for a long time, unable to be discharged with the waste liquid, and eventually accumulate in the scum tank of the flotation-sedimentation tank.

[0004] According to the feedback from the operation of the flotation process, the scum waste liquid in the scum sedimentation tank still has obvious stratification, with the lower layer mainly liquid and the upper layer gathering foamy scum, which seriously affects the sedimentation efficiency. At this time, whether it is through gravity flow or pumping, the first thing to be discharged is often the liquid in the lower layer. As the liquid level in the lower layer drops, the scum in the upper layer containing a large amount of pollutants will remain at the bottom of the pool or stick to the pool wall and cannot be discharged, which greatly limits the scum discharge efficiency and eventually causes scaling of the scum pool, which affects the appearance and even produces toxic and harmful gases, threatening human health. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] In view of the deficiencies of the prior art, the present invention provides an intelligent scum degassing device and process based on online image recognition, which is used to degas the scum, remove the gas in the scum and mix it into the lower layer of liquid, and finally be discharged together with the liquid.

[0007] (II) Technical solution

[0008] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: an intelligent scum degassing device based on online image recognition, comprising a scum collection module, a scum degassing module, a scum sedimentation module and a sludge treatment module, wherein the scum collection module is located at the end of a flotation-sedimentation tank, the scum degassing module is connected to the scum collection module through a slag discharge pipe, and is located below the flotation-sedimentation tank, the scum degassing module is connected to the scum sedimentation module through a pipeline, and the scum sedimentation module is connected to the sludge treatment module through a pipeline.

[0009] Preferably, the scum collection module comprises a scum scraper, a scum collection tank and a scum discharge pipe;

[0010] The scraper of the scum collection module adopts a chain scraper, the scraper is less than 60mm away from the pool wall, the scum collection tank is inclined 45 to 60 degrees toward the water surface, and the scum discharge pipe is vertically arranged at the bottom of the scum collection tank and connected to the scum degassing channel.

[0011] Preferably, the scum degassing module includes a scum degassing channel, a scum stirring device, an inert gas aeration device, an ultrasonic emitting device, an intelligent image online recognition system, a scum pump and a water spraying device:

[0012] The scum degassing channel is provided with a scum stirring device, an inert gas aeration device, an ultrasonic emitting device and a water sprinkling device, and a high-speed camera movable along the degassing channel is provided at the upper end of the degassing channel, and a scum pump is provided at the end, and the scum pump outlet pipe is connected to the scum precipitation module.

[0013] Preferably, the scum stirring device adopts a high-speed submersible stirrer with an adjustable speed range of 0-980 rpm, a power of 4.0 kW, and the speed is determined by a stirring shear degassing test;

[0014] The inert gas aeration device is provided with a perforated aeration pipe in the scum degassing channel. The perforated aeration pipe slowly releases helium through the mobile phase to drive away the air dissolved in the scum. The aeration flow rate range is 0.25-2.5Nm 3 / h adjustable; gas input pressure 0.4~0.6MPa;

[0015] The ultrasonic emitting device includes an ultrasonic generator and a transducer, which are located above the degassing channel and can generate ultrasonic waves of different intensities. The ultrasonic waves are perpendicular to the scum degassing channel and can generate ultrasonic waves of 25-40 Hz, using cavitation benefits to destroy bubbles in the shallow layer under the water surface.

[0016] The water spraying device is arranged at the top of the scum degassing channel, and the water source comes from the upper clear water of the sedimentation tank of the sedimentation module. The sedimentation tank supplies water from top to bottom to dissipate the bubbles on the water surface of the scum degassing channel.

[0017] Preferably, the scum degassing module innovatively uses the random forest method in machine learning to construct a multi-parameter degassing control model based on image recognition, which can identify the relationship between the rotation speed of the stirring device, the inert gas aeration time, the ultrasonic power and the bubble removal, so as to optimize the operating parameters of the degassing device based on the model; the model first performs data preprocessing and training based on the operating data and degassing conditions over a period of time, and then predicts the structure and evaluates the performance of the model; at the same time, the scum degassing module develops a microbubble content test system in the scum (composed of a test device, a constant temperature water bath and a data acquisition system) based on the measurement principle of the vapor pressure method, and measures the microbubble data in the scum under different operating conditions through multiple experiments, and uses the data to evaluate the model and correct the multi-parameter degassing control model.

[0018] Preferably, the sludge treatment module includes a sludge thickening tank, a sludge agitator, a decanter, a sludge pump, a dosing pump and a sludge dewatering machine; the sludge thickening tank is provided with a float-type decanter, a submersible sludge pump and a high-speed submersible agitator, and is equipped with a polyaluminium chloride dosing device, and the sludge dewatering machine adopts a centrifugal dewatering machine.

[0019] A scum degassing method for a flotation-sedimentation tank, based on the scum degassing device, comprises the following steps:

[0020] S1. The scum generated when the flotation-sedimentation tank is running the flotation process, the scum discharge mode is started, the outlet pipe of the flotation-sedimentation tank is closed, the water level rises, and the scraper of the chain scraper scrapes the scum on the water surface into the slag discharge trough at the end of the tank. The water-facing surface of the slag discharge trough is inclined at 45° to 60°, so that the scum can flow smoothly into the slag discharge trough. The scum flows into the scum degassing channel through the vertical scum outlet pipe at the bottom of the slag discharge trough, and some bubbles in the scum are released by the waterfall effect;

[0021] S2. When the water level in the scum degassing channel reaches the specified level, the scum is broken up by the scum agitator in the channel to release some bubbles;

[0022] S3, the inert gas aeration section through which the scum passes again. The bottom of this section of the channel is arranged with a perforated aeration pipe, and the inert gas is released to drive out some bubbles in the scum; the scum then passes through the ultrasonic emission device, and the ultrasound can produce cavitation and chemical action to further remove the bubbles in the scum;

[0023] S4. Finally, the water spraying section is used to spray the water surface with clean water pumped from the decanter reflux pump of the sedimentation tank to eliminate the bubbles on the surface; a high-speed camera that can move along the scum degassing channel is set at the top of the scum degassing channel to observe the bubbles contained in the scum, and a multi-parameter degassing control model based on image recognition is constructed using the random forest method in machine learning, and the operating parameters of the degassing device are optimized based on the multi-parameter degassing control model, which includes:

[0024] Data preparation: Separate the data set by features and labels to ensure that the features and labels of each sample correspond;

[0025] Parameter setting: set the parameters of random forest as needed;

[0026] Train the model and input the prepared data into the random forest model for training;

[0027] Prediction results: Use the trained random forest model to predict new data;

[0028] Model evaluation: evaluate the model through evaluation indicators to judge performance and effect;

[0029] The bubble photos in the scum are analyzed according to the image recognition technology, and the spatial frequency distribution characteristics and bubble content of the bubble group are identified. Based on this, the destruction effects of stirring, ultrasound, inert gas and other measures on the bubbles are analyzed, and they are correlated with the speed of the stirring device, the inert gas aeration time and the ultrasonic power. According to the results of bubble photo identification, the operating parameters of each process unit are further optimized to give full play to the removal effect of the process unit on the bubbles in the scum.

[0030] S5. The scum at the end of the scum degassing channel, which has been freed of most bubbles, is pumped to the scum sedimentation tank by a sludge pump. After the scum is pumped out, the scum degassing channel can receive scum again.

[0031] S6. When the scum sedimentation tank is full, static sedimentation begins. When the sedimentation is completed, the supernatant is sucked back to the raw water pipe and the water sprinkling device through the decanter, the supernatant pipe and the water pump for reuse;

[0032] S7. The accumulated mud at the bottom of the scum sedimentation tank is scraped into the mud pit by a wire rope traction type scraper;

[0033] S8. After the supernatant is pumped out, the sludge is pumped into the sludge thickening tank through the sludge pump for gravity concentration or chemical concentration. After the sludge is pumped out, the slag and sludge sedimentation tank can receive the slag again.

[0034] S9. After the sludge in the sludge thickening tank is concentrated, it is pumped to the sludge storage tank. When the sludge in the sludge storage tank accumulates to a certain amount, the sludge centrifugal dewatering machine is turned on for dehydration treatment, and the dehydrated sludge is transported out for treatment.

[0035] (III) Beneficial effects

[0036] Compared with the prior art, the present invention provides an intelligent device and process for scum degassing based on online image recognition, which has the following beneficial effects:

[0037] 1. Solve the problem that scum is difficult to settle and remove because it contains a large number of bubbles, avoid scaling in the scum treatment system, and avoid the generation of harmful gases;

[0038] 2. After the scum is degassed, when the water pump is used for transfer, the bubbles in the scum are prevented from eroding the pump blades;

[0039] 3. The flotation sedimentation tank and the scum and sludge treatment module adopt an upper and lower stacking tank design to save floor space, which saves 30-50% of floor space compared with the traditional flat-laying process;

[0040] 4. The process tanks in the pretreatment module, sludge treatment module and deep treatment module are adjacent to each other and connected by channels, so the head loss is small. The head loss between the tanks is only 0.1-0.4m, which saves 30% to 50% of the head loss compared with the traditional process in which the tanks are connected by pipes, and saves 8% to 12% of the energy consumption of the raw water lifting pump. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The structure of the slag degassing device of the present invention is shown in FIG. Figure 1

[0042] Figure 2 The structure of the slag degassing device of the present invention is shown in FIG. Figure 2 ;

[0043] Figure 3 It is a schematic diagram of the system module of the present invention;

[0044] Figure 4 This is a schematic diagram of the process flow of the slag degassing module of the present invention;

[0045] Figure 5 It is a schematic diagram of the intelligent scum bubble image online recognition system of the present invention;

[0046] Figure 6 Schematic diagram of the microbubble content testing system of the present invention. DETAILED DESCRIPTION

[0047] In order to better understand the purpose, structure and function of the present invention, the following is a further detailed description of an intelligent slag degassing device and process based on image online recognition of the present invention in conjunction with the accompanying drawings.

[0048] Example 1

[0049] See also Figure 1-6The present invention discloses an intelligent scum degassing device based on online image recognition, comprising a scum collecting module, a scum degassing module, a scum sedimentation module and a sludge treatment module, wherein the scum collecting module is located at the end of a flotation-sedimentation tank, the scum degassing module is connected to the scum collecting module through a slag discharge pipe and is located below the flotation-sedimentation tank, the scum degassing module is connected to the scum sedimentation module through a pipeline, and the scum sedimentation module is connected to the sludge treatment module through a pipeline.

[0050] Specifically, in the present invention, in a scum degassing device for a flotation-sedimentation tank, the scum collection module includes a scraper, a scum collection tank and a scum discharge pipe, wherein the scraper of the scum collection module adopts a chain scraper, the scraper is less than 60 mm away from the tank wall, the scum collection tank is made of stainless steel 316, inclined 45 to 60 degrees toward the water surface, and the scum discharge pipe is vertically arranged at the bottom of the scum collection tank and connected to the scum degassing channel;

[0051] refer to Figure 2 and 4 As shown, the scum degassing module includes a scum degassing channel, a scum stirring device, an inert gas aeration device, an ultrasonic transmitting device, an intelligent image online recognition system, a scum pump and a water sprinkling device, wherein the scum degassing channel is provided with a scum stirring device, an inert gas aeration device, an ultrasonic transmitting device and a water sprinkling device, and a high-speed camera that can move along the degassing channel is provided at the upper end of the degassing channel, and a scum pump is provided at the end, and the scum pump outlet pipe is connected to the scum sedimentation module, and the scum stirring device adopts a high-speed submersible agitator with an adjustable speed range of 0-980 rpm, a power of 4.0kw, and the speed is determined by a stirring shear degassing test. The inert gas aeration device is provided with a perforated aeration pipe in the scum degassing channel, generally using helium, and the aeration flow rate range is 0.25~2.5Nm 3 / h is adjustable; the gas input pressure is 0.4-0.6MPa, and the perforated aeration pipe slowly releases helium through the mobile phase to drive away the air dissolved in the scum. The ultrasonic emitting device includes an ultrasonic generator and a transducer, which are arranged above the scum degassing channel and can generate ultrasonic waves of different intensities. The ultrasonic waves are perpendicular to the scum degassing channel and are used to break up shallow underwater bubbles; the water sprinkling device is arranged on the top of the scum degassing channel, and the water source comes from the upper clear water of the sedimentation tank of the sedimentation module. The sedimentation tank supplies water from top to bottom to eliminate the bubbles in the scum tank.

[0052] Image recognition technology can identify the bubble group in the scum, define the average gradient based on the image gradient, and obtain the grayscale image of the bubble and the average gradient of the grayscale image of the bubble, thereby showing the difference in the shape and scale of the bubble group. The image point gradient value of the image area with high spatial frequency is larger, and the image point gradient value of the image area with low spatial frequency is smaller. Let the image be g(x, y), and define the gradient vector of g(x, y) at point (x, y) as:

[0053]

[0054] The average gradient t of the digital image is as follows:

[0055]

[0056] For a discrete digital image g(i,j), i=1,2,…,m; j=1,2,…,n.

[0057] Furthermore, the scum degassing module innovatively uses the random forest method in machine learning to construct a multi-parameter degassing control model based on image recognition. The random forest method has the advantages of high accuracy, fast training speed, not easy to fall into overfitting, and can process nonlinear data. It has advantages in fast image recognition. The established multi-parameter degassing control model can identify the relationship between the speed of the stirring device, the inert gas aeration time, the ultrasonic power and the bubble removal, so as to optimize the operating parameters of the degassing device based on the model; at the same time, the scum degassing module develops a micro-bubble content test system in the scum (composed of a test device, a constant temperature water bath and a data acquisition system) based on the measurement principle of the vapor pressure method, and measures the micro-bubble data in the scum under different operating conditions through multiple experiments, and uses the data to correct the multi-parameter degassing control model. Afterwards, according to the results of image recognition of the bubble photos in the scum, they are compared with the multi-parameter degassing control model, so that the multi-parameter degassing control model calculates the optimal process parameters (speed of the stirring device, inert gas aeration time, ultrasonic power), and controls the process device to adjust to the optimal process parameters, so as to give full play to the removal effect of the process device on the bubbles in the scum.

[0058] Specifically, the sludge treatment module includes a sludge thickening tank, a sludge agitator, a decanter, a sludge pump, a dosing pump and a sludge dewatering machine; the sludge thickening tank is provided with a float-type decanter, a submersible sludge pump and a high-speed submersible agitator, and is equipped with a polyaluminium chloride dosing device, and the sludge dewatering machine adopts a centrifugal dewatering machine.

[0059] Example 2

[0060] The present invention proposes a scum degassing method for a flotation tank, based on the scum degassing device, comprising the following steps:

[0061] S1. The scum generated when the flotation tank runs the flotation process, starts the slag discharge mode, closes the outlet pipe of the flotation-sedimentation tank, and the water level rises. The scraper of the chain scraper scrapes the scum on the water surface into the SUS304 slag discharge trough at the end of the tank. The water-facing surface of the slag discharge trough is inclined at 45 degrees, so that the scum can flow smoothly into the slag discharge trough. The scum flows into the slag degassing channel through the vertical slag outlet pipe at the bottom of the slag discharge trough, and some bubbles in the scum are released by the waterfall effect;

[0062] S2. When the water level in the scum degassing channel reaches the specified level, the scum is broken up by the scum agitator in the channel to release some bubbles;

[0063] S3, the inert gas aeration section through which the scum passes again. The bottom of this section of the channel is arranged with a perforated aeration pipe, and inert gas (helium, neon, argon, etc.) is released to drive out some bubbles in the scum; the scum then passes through an ultrasonic transmitter, and ultrasound can produce cavitation to further remove bubbles in the scum;

[0064] S4, finally through the water spraying section, use the clean water pumped from the sedimentation tank decanter return pump to spray water to the water surface to eliminate the bubbles on the surface. The flow range is 0~4m 3 / h is adjustable; a high-speed camera that can move along the scum degassing channel is set at the top of the scum degassing channel to observe the bubbles contained in the scum, and a multi-parameter degassing control model based on image recognition is constructed by using the random forest method in machine learning. After that, the multi-parameter degassing control model is compared with the result of image recognition of the bubble photos in the scum, so that the multi-parameter degassing control model calculates the optimal process parameters (speed of the stirring device, inert gas aeration time, ultrasonic power), and controls the process device to adjust to the optimal process parameters, so as to give full play to the removal effect of the process device on the bubbles in the scum. The steps of constructing the model include:

[0065] Data preparation: Separate the data set by features and labels to ensure that the features and labels of each sample correspond;

[0066] Parameter setting: set the parameters of random forest as needed;

[0067] Train the model and input the prepared data into the random forest model for training;

[0068] Prediction results: Use the trained random forest model to predict new data;

[0069] Model evaluation: evaluate the model through evaluation indicators (such as accuracy, recall, etc.) to judge performance and effect;

[0070] S5. The scum at the end of the scum degassing channel, which has been freed of most bubbles, is pumped to the scum sedimentation tank by a sludge pump. After the scum is pumped out, the scum degassing channel can receive scum again.

[0071] S6. When the scum sedimentation tank is full, static sedimentation begins. When the sedimentation is completed, the supernatant is sucked back to the raw water pipe and the water sprinkling device through the decanter, the supernatant pipe and the water pump for reuse;

[0072] S7. The accumulated mud at the bottom of the scum sedimentation tank is scraped into the mud pit by a wire rope traction type scraper;

[0073] S8. After the supernatant is pumped out, the sludge is pumped into the sludge thickening tank through the sludge pump for gravity concentration or chemical concentration. After the sludge is pumped out, the slag and sludge sedimentation tank can receive the slag again.

[0074] S9. After the sludge in the sludge thickening tank is concentrated, it is pumped to the sludge storage tank. When the sludge in the sludge storage tank accumulates to a certain amount, the sludge centrifugal dewatering machine is turned on for dehydration treatment, and the dehydrated sludge is transported out for treatment.

[0075] The random forest method can handle multi-category problems without additional transformation or processing. By counting the usage of features in the random forest, the importance ranking of each feature can be obtained.

[0076] It is understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. An intelligent scum degassing device based on online image recognition, characterized in that: It includes a scum collection module, a scum degassing module, a scum sedimentation module and a sludge treatment module. The scum collection module is located at the end of the flotation-sedimentation tank. The scum degassing module is connected to the scum collection module through a slag discharge pipe and is located below the flotation-sedimentation tank. The scum degassing module is connected to the scum sedimentation module through a pipeline, and the scum sedimentation module is connected to the sludge treatment module through a pipeline.

2. The intelligent scum degassing device based on online image recognition according to claim 1 is characterized by: The scum collection module includes a scum scraper, a scum collection tank and a scum discharge pipe; The scraper of the scum collection module adopts a chain scraper, the scraper is less than 60mm away from the pool wall, the scum collection tank is inclined 45 degrees to 60 degrees toward the water surface, and the scum discharge pipe is vertically arranged at the bottom of the scum collection tank and connected to the scum degassing channel.

3. The intelligent scum degassing device based on online image recognition according to claim 1 is characterized by: The scum degassing module includes a scum degassing channel, a scum stirring device, an inert gas aeration device, an ultrasonic transmitting device, an intelligent image online recognition system, a scum pump and a water spraying device; The scum degassing channel is provided with a scum stirring device, an inert gas aeration device, an ultrasonic emitting device and a water sprinkling device, and a high-speed camera movable along the degassing channel is provided at the upper end of the degassing channel, and a scum pump is provided at the end, and the scum pump outlet pipe is connected to the scum precipitation module.

4. The intelligent scum degassing device based on online image recognition according to claim 3 is characterized by: The scum stirring device adopts a high-speed submersible stirrer with an adjustable speed range of 0-980 rpm and a power of 4.0 kW, and the speed is determined by a stirring shear degassing test; The inert gas aeration device is provided with a perforated aeration pipe in the slag degassing channel. The inert gas is usually helium, and the aeration flow rate range is 0.25~2.5Nm 3 / h adjustable; gas input pressure 0.4~0.6MPa, perforated aeration tube slowly releases helium through the mobile phase to drive out the air dissolved in the slag; The ultrasonic emitting device includes an ultrasonic generator and a transducer, which can generate ultrasonic waves of 25-40 Hz, and the ultrasonic waves are perpendicular to the slag degassing channel; The water spraying device is arranged at the top of the slag degassing channel, and the flow rate range is 0~4m 3 / h is adjustable. The water source comes from the upper clear water of the sedimentation tank of the sedimentation module. The sedimentation tank supplies water from top to bottom to dissipate the bubbles in the scum tank.

5. The intelligent scum degassing device based on online image recognition according to claim 3 is characterized by: The scum degassing module innovatively uses the random forest method in machine learning to construct a multi-parameter degassing control model based on image recognition. The model can identify the relationship between the rotation speed of the stirring device, the inert gas aeration time, the ultrasonic power and the bubble removal, so as to optimize the operating parameters of the degassing device based on the model; at the same time, the scum degassing module develops a microbubble content test system in the scum (composed of a test device, a constant temperature water bath and a data acquisition system) based on the measurement principle of the vapor pressure method, and measures the microbubble data in the scum under different operating conditions through multiple experiments, and uses the data to correct the multi-parameter degassing control model.

6. The intelligent scum degassing device based on online image recognition according to claim 1 is characterized by: The scum sedimentation module comprises a sedimentation tank, a scraper, a decanter and a sludge pump. A traction-type scraper is arranged in the sedimentation tank of the scum sedimentation module to scrape the sludge into a sludge accumulation pit. A sludge pump is arranged in the sludge accumulation pit to pump the sludge to the sludge treatment module. The upper clear water is collected by a float-type decanter, and the supernatant is returned to the raw water pipe, the water sprinkling device, etc. through a supernatant pipe and a water pump for reuse.

7. The intelligent scum degassing device based on online image recognition according to claim 1 is characterized by: The sludge treatment module includes a sludge thickening tank, a sludge agitator, a decanter, a sludge pump, a dosing pump and a sludge dewatering machine; the sludge thickening tank is provided with a buoy-type decanter, a submersible sludge pump and a high-speed submersible agitator, and is equipped with a polyaluminium chloride dosing device, and the sludge dewatering machine adopts a centrifugal dewatering machine.

8. A scum degassing method for a flotation-sedimentation tank, based on the scum degassing device according to claim 1, characterized in that: The steps include: S1. The scum generated when the flotation-sedimentation tank is running the flotation process, the scum discharge mode is started, the outlet pipe of the flotation-sedimentation tank is closed, the water level rises, and the scraper of the chain scraper scrapes the scum on the water surface into the slag discharge trough at the end of the tank. The water-facing surface of the slag discharge trough is inclined at 45 to 60 degrees, so that the scum can flow smoothly into the slag discharge trough. The scum flows into the scum degassing channel through the vertical scum outlet pipe at the bottom of the slag discharge trough, and some bubbles in the scum are released by the waterfall effect; S2. When the water level in the scum degassing channel reaches the specified level, the scum is broken up by the scum agitator in the channel to release some bubbles; S3, the inert gas aeration section through which the scum passes again. The bottom of this section of the channel is arranged with a perforated aeration pipe, and the inert gas is released to drive out some bubbles in the scum; the scum then passes through the ultrasonic emission device, and the ultrasound can produce cavitation and chemical action to further remove the bubbles in the scum; S4. Finally, the water spraying section is used to spray clean water pumped from the sedimentation tank decanter reflux pump to the water surface to eliminate the bubbles on the surface; a high-speed camera that can move along the scum degassing channel is set on the top of the scum degassing channel to observe the bubbles contained in the scum, and the random forest method in machine learning is used to construct a multi-parameter degassing control model based on image recognition, and the operating parameters of the degassing device are optimized based on the multi-parameter degassing control model. S5. The scum at the end of the scum degassing channel, which has been freed of most bubbles, is pumped to the scum sedimentation tank by a sludge pump. After the scum is pumped out, the scum degassing channel can receive scum again. S6. When the scum sedimentation tank is full, static sedimentation begins. When the sedimentation is completed, the supernatant is sucked back to the raw water pipe and the water sprinkling device through the decanter, the supernatant pipe and the water pump for reuse; S7. The accumulated mud at the bottom of the scum sedimentation tank is scraped into the mud pit by a wire rope traction type scraper; S8. After the supernatant is pumped out, the sludge is pumped into the sludge thickening tank through the sludge pump for gravity concentration or chemical concentration. After the sludge is pumped out, the slag and sludge sedimentation tank can receive the slag again. S9. After the sludge in the sludge thickening tank is concentrated, it is pumped to the sludge storage tank. When the sludge in the sludge storage tank accumulates to a certain amount, the sludge centrifugal dewatering machine is turned on for dehydration treatment, and the dehydrated sludge is transported out for treatment.

Images