Intelligent dosing system of distributed sewage station

Abstract

The invention provides an intelligent dosing system of a distributed sewage station. The intelligent dosing system comprises a dosing device and a control system, then the control system is used for executing the following steps: acquiring water quality monitoring information and the current medicine material content; determining a dosing curve according to the water quality monitoring information, the current medicine material content and a pre-established deep learning model; and finally, controlling the dosing device to add chemicals into the sewage reaction tank according to the dosing curve. By applying a real-time control concept, a numerical simulation analysis technology, a deep learning technology, dynamic control and on-demand dosing, the invention realizes stable and up-to-standard effluent quality, optimizes the operation of a sewage treatment plant, and further realizes sustainable stable up-to-standard, energy saving and consumption reduction.

Description

Technical Field

[0001] The invention belongs to the technical field of sewage treatment, and in particular relates to an intelligent dosing system for a decentralized sewage station. Background Art

[0002] The pollutants in tunnel construction wastewater are mainly suspended solids SS, some oils, and excessive pH values. The other indicators such as COD, ammonia nitrogen, and total phosphorus are all low. Among them, the SS and oil contents in highway tunnel construction wastewater are relatively high, while the SS content in subway and railway tunnel construction wastewater is relatively high. In the application of concrete during tunnel construction, materials such as mixing mortar and concrete will enter the water environment with tunnel water. These substances entering the water environment will be brought back into the surface water, resulting in an increase in surface water values, especially the concentration of suspended solids.

[0003] In the sewage treatment system, in order to remove or reduce suspended matter in the water, produce large-particle agglomerates, and accelerate the sedimentation rate of impurities and sludge in the water, a certain concentration of flocculant is generally added to the sedimentation part of the sewage treatment. Flocculants used for oily sewage treatment include three major categories: inorganic flocculants, organic flocculants, and composite flocculants. In order to ensure the dosing concentration, reduce the labor intensity of workers, and reduce the cost of chemicals, the automatic dosing technology of flocculants is applied. Using the traditional manual dosing method, there is a significant lag in the concentration control of the outlet PH and SS, and the outlet SS and outlet PH are unstable, with too high and too low points. Summary of the invention

[0004] In view of this, the present invention provides an intelligent dosing system for a decentralized sewage station, aiming to solve the problem of obvious lag in the control of the concentration of outlet PH and SS in the traditional manual dosing method, and the outlet SS and outlet PH are unstable, with too high and too low points.

[0005] A first aspect of an embodiment of the present invention provides an intelligent dosing method for a decentralized sewage station, comprising: Dosing device and control system; The control system is used to perform the following steps: Obtain water quality monitoring information and current drug content; Determine the dosing curve based on water quality monitoring information, current drug content and pre-established deep learning model; The dosing device is controlled to add drugs into the sewage reaction tank according to the dosing curve.

[0006] In a possible implementation, the dosing device includes an automatic feeding device and a stirring device; The automatic feeding device is provided with a medicine bin; the control system is used to control the medicine bin to feed medicine into the stirring device according to the medicine adding curve; The stirring device is used to monitor the drug content and report it to the control system.

[0007] In a possible implementation, the medicine chamber includes an oxalic acid medicine adding chamber body and a PAC medicine adding chamber body; The four bottom corners of the oxalic acid dosing bin and the PAC dosing bin are equipped with weighing modules, which are used to monitor the remaining weight and material level of oxalic acid or PAC medicine, and feed back the weight information and material level information to the control system.

[0008] In a possible implementation, a vibration device is provided at the bottom of the automatic feeding device; The control system is used to send a control signal to the vibration device when it detects that the amount of medicine fed is insufficient, so that the vibration device drives the medicine bin to vibrate.

[0009] In a possible implementation, the intelligent dosing system of the decentralized sewage station is also provided with an image acquisition device; The image acquisition device is used to monitor SS suspended matter in the sewage reaction tank and report it to the control system.

[0010] In a possible implementation, an MBR water purification device is provided at the outlet of the sewage reaction tank; The MBR water purification device is used to deeply purify part of the sewage flowing through the outlet of the sewage treatment tank, and transport the purified water to the medicine barrel used to dissolve oxalic acid through a suction pump.

[0011] In a possible implementation, the control system is used to: The water quality monitoring information and the current drug content are input into the pre-established deep learning model to obtain the dosing type and target dosing amount at each target moment, and the dosing curve is fitted.

[0012] In a possible implementation, the deep learning model includes a reward value calculation module and a parameter update module; The reward value calculation module is used to determine the reward value based on water quality monitoring information, current drug content and drug addition curve; The parameter update module is used to update the parameters of the deep learning model according to the reward value.

[0013] In a possible implementation, the reward value calculation module is used to: Under the constraints, the reward function is determined with the goal of minimizing the dosing cost and maximizing the water quality stability; The water quality monitoring information and the current drug content are input into the reward function to obtain the reward value.

[0014] In a possible implementation, the parameter updating module is used to: Calculate the policy gradient based on the reward value; Update the parameters of the deep learning model according to the policy gradient.

[0015] The intelligent dosing system for a decentralized sewage treatment plant provided by the embodiment of the present invention first includes a dosing device and a control system; then the control system is used to perform the following steps: then obtain water quality monitoring information and current drug content; then determine a dosing curve based on water quality monitoring information, current drug content and a pre-established deep learning model; finally, control the dosing device to add drugs to the sewage reaction tank based on the dosing curve. The present invention applies real-time control concepts, numerical simulation analysis technology, deep learning technology and dynamic control to add drugs on demand, achieve stable effluent quality compliance, optimize sewage treatment plant operation, and further achieve sustainable stable compliance and energy saving and consumption reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0017] Figure 1 It is a structural schematic diagram of an intelligent dosing system for a decentralized sewage station provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0019] Figure 1 FIG. 1 is a schematic diagram of the structure of the intelligent dosing system of the distributed sewage station provided by the embodiment of the present invention. Figure 1 As shown, in some embodiments, the intelligent dosing system of the decentralized sewage station includes: a dosing device 11 and a control system 12; The control system 12 is used to perform the following steps: obtain water quality monitoring information and current drug content; determine a dosing curve based on the water quality monitoring information, current drug content and a pre-established deep learning model; and control the dosing device 11 to add drugs into the sewage reaction tank according to the dosing curve.

[0020] In the embodiment of the present invention, the water quality monitoring information is the water quality information collected by the sensor installed in the control system, such as pH, ss content, etc. The current drug content is the sensor detection value of the drug content currently put into the sewage reaction tank.

[0021] In the embodiment of the present invention, the control system collects the inlet and outlet water flow and water quality information in the sewage reaction tank in real time; The control system outputs dosing parameters to the dosing device according to the collected water quality information and water quality qualified standard parameters in the sewage reaction tank, and controls the dosing device to dosing the medicine to the sewage reaction tank; When the control system monitors the water quality pH value of the sewage reaction tank inlet > 9, the first dosing is started; and the dosing parameters are transmitted to the automatic feeding device; The automatic feeding device conveys the medicine into the stirring device according to the instruction information output by the control system. The control system controls the stirring device to stir the medicine and monitors the medicine content inside the stirring device until the medicine concentration is within the range of 3.2%-4.3%; The control system controls the dosing device to add the stirred medicine to the sewage reaction tank until the pH value of the sewage reaction tank outlet is between 8.0 and 8.6; The control system controls the dosing device to start the second dosing to feed the sewage reaction tank.

[0022] The first dosing agent and the second dosing agent are different types of water purification agents.

[0023] Specifically, the control system includes a data acquisition module, a data processing module, a deep learning module, a data storage module, a human-computer interaction module and a dosing control module. The data acquisition module includes: a weighing level meter, a suction pump for MBR water purification equipment, a radar level meter, an RSI oxalic acid concentration meter, a PAC dosing pump, an oxalic acid dosing pump, a water quality collection device, an electromagnetic flowmeter, a distance meter, and an electromagnetic flowmeter. Among them, the weighing level meter and the suction pump for MBR water purification equipment are connected to the control system by wired communication. The water quality collection device, the electromagnetic flowmeter, the distance meter, the electromagnetic flowmeter, the radar level meter, the oxalic acid dosing pump, the RSI oxalic acid concentration meter, and the PAC dosing pump transmit signals to the data processing module through a self-organizing network, thereby reducing the amount of wiring in the system, and the sensor node can automatically adjust the communication path according to the signal strength and network topology. When a sensor node fails or the communication is interrupted, other nodes can automatically reroute the data to ensure stable data transmission and improve the reliability of the system.

[0024] Specifically, when the data collected by the sensor is transmitted to the data processing module, it is realized through the low-power wide area network (LPWAN) technology, and specifically through LoRaWAN to realize the control and communication between various sensors. The node devices of the self-organizing network created by this system are integrated with WIFI modules and wireless Bluetooth modules to realize signal communication between various sensors and the self-organizing network.

[0025] The deep learning module calls the collected monitoring information from the data storage module for machine learning. The deep learning module outputs different dosing parameters for different sewage quality conditions according to the deep learning recognition model, and the dosing parameters form corresponding dosing curves. Specifically, the deep learning model is Each The corresponding target value is the amount of oxalic acid added to the medicine barrel .

[0026] in, The water inlet flow rate of the medicine barrel is the water outlet flow rate of the medicine barrel, is the oxalic acid concentration in the medicine barrel, S is the volume of the sewage pool, and PH is the pH value of the sewage pool.

[0027] By formula The data is normalized, where , are the mean and standard deviation of the input data, and are the mean and standard deviation of the target data.

[0028] When the input data passes through the convolution layer, for a specific convolution kernel At layer l, its output It can be expressed as in, is the input of layer l, I and J are the sizes of convolution kernels is the bias term is the activation function (such as ReLU, The ReLU function sets all negative values ​​to 0 and leaves positive values ​​unchanged).

[0029] Pooling layer, the output of the maximum pooling layer is expressed as: , where s is the size of the pooling window.

[0030] The fully connected layer performs the following operations: The fully connected layer maps the features learned by the convolution and pooling layers to the final output. In this system, the fully connected layer has 30 neurons. For the input feature vector , output It can be expressed as: ,in, is a The weight matrix of (M is the dimension of the input feature vector) is the bias vector.

[0031] The mean square error (MSE) is used as the loss function during training: ,in, is the predicted value of the model. The parameters of the model are updated by an optimization algorithm (such as the Adam optimizer) to minimize the loss function. “T” represents the total number of time steps or samples in the time series data.

[0032] Evaluating the model Use the test set data to evaluate the performance of the model and calculate the root mean square error (RMSE) metric: Understand the degree of deviation between the predicted dosage and the actual dosage, and adjust the parameters of the model to improve the accuracy and reliability of the model.

[0033] The dosing curve formed by the deep learning module based on historical dosing information can further predict the dosage of the next time period. At the same time, the data of the dosing curve can be updated in real time according to the actual dosage, so as to adjust the dosage of the intelligent dosing system. The system is also equipped with a front monitoring component, which includes a water quality monitor, a rangefinder and an electromagnetic flowmeter. The front water quality monitor transmits water quality information, flow rate information and position information to the control system in real time. The control system calculates the time when the sewage at the front monitoring component reaches the water inlet of the sewage pool, and issues an early warning to remind the staff to prepare medicine.

[0034] In some embodiments, the dosing device includes an automatic feeding device and a stirring device; the automatic feeding device is provided with a medicine bin; the control system is used to control the medicine bin to feed medicine into the stirring device according to the dosing curve; the stirring device is used to monitor the medicine content and report it to the control system.

[0035] In an embodiment of the present invention, the control system controls the medicine bin to feed medicine into the stirring device, and the stirring device monitors the medicine content inside it and transmits the monitoring information to the control system to keep the oxalic acid medicine concentration in the stirring device within the range of 3.2%-4.3%, and the PAC medicine concentration within the range of 20-50 mg / L.

[0036] In some embodiments, the medicine bin includes an oxalic acid dosing bin body and a PAC dosing bin body; weighing modules are configured at the four corners of the bottom of the oxalic acid dosing bin body and the PAC dosing bin body, and the weighing modules are used to monitor the remaining weight and material level of oxalic acid or PAC medicine, and feed back the weight information and material level information to the control system.

[0037] In an embodiment of the present invention, the side wall structures of the two silos form a sliding connection, and weighing modules are respectively arranged at the bottom. The weighing modules and the radar level meter are respectively connected to the control system through a self-organizing network and the control system, and a screen is installed on the top of the silo.

[0038] The dosing system also includes a stirring structure arranged in the sewage reaction tank, on which is provided an image acquisition device for monitoring real-time images of suspended matter and floating with the liquid surface of the sewage reaction tank. The acquisition device transmits the acquisition signal to the control system, and the acquisition device is installed in an isolation cover.

[0039] In some embodiments, a vibration device is provided at the bottom of the automatic feeding device; when the control system detects that the amount of medicine fed is insufficient, it sends a control signal to the vibration device, so that the vibration device drives the medicine bin to vibrate.

[0040] In the embodiment of the present invention, the weighing module specifically adopts a weighing level meter, and the oxalic acid dosing bin and the PAC dosing bin are weighed independently, and the remaining weight and material level of oxalic acid and PAC medicine materials are monitored in real time, and the weight information and material level information are fed back to the control system. A vibration device is arranged at the bottom of the feeder connected to the medicine bin. When the control system detects that the amount of medicine is insufficient, a control signal is sent to the vibration device. After the vibration device is turned on, it drives the medicine bin to vibrate, so that oxalic acid and PAC flocculants can smoothly enter the dosing pipeline, thereby ensuring the amount of medicine added in the oxalic acid medicine barrel and the PAC medicine barrel. By respectively configuring a weighing level meter at the four corners of the bin body, it is effective to prevent errors in the collection of residual information of medicine materials, accurately monitor the increase or decrease of oxalic acid in the bin, and combine the real-time data of the oxalic acid concentration meter to calculate the amount of medicine to be added to the bin, thereby realizing accurate delivery of medicine materials. By monitoring the weight and material level signal of oxalic acid and PAC bins in the bin, the amount of medicine added to the medicine bin is controlled, and an early warning is given to remind the operation and maintenance personnel to interrupt the unreasonable addition of medicines to the medicine bin. A screen is slidably installed on the top of the bin. When medicine is added to the bin, the screen is pushed to the top of the bin to filter the medicine materials that have aggregated into blocks to prevent large blocks of medicine materials from entering the bin and causing blockage.

[0041] In an embodiment of the present invention, the stirring device includes: a medicine barrel for dissolving oxalic acid and a medicine barrel for dissolving PAC flocculant and stirring mechanisms respectively installed on the top of the medicine barrel, an RSI oxalic acid concentration meter is provided in the oxalic acid medicine barrel, a spectrophotometer is provided in the PAC medicine barrel for real-time monitoring of the concentration of the PAC flocculant, and the RSI oxalic acid concentration meter is used to collect the oxalic acid concentration in the medicine barrel in real time. The medicine barrel is configured according to the dosage, and a large-capacity medicine barrel can be configured in large-scale production to ensure the tonnage of sewage purified by this system.

[0042] In an embodiment of the present invention, after the PAC flocculant is added, it is transported to the PAC medicine barrel by the PAC dosing pump and adjusted to a predetermined ratio concentration, and then quantitatively input into the sewage reaction tank by the feeder. A fixed radar level meter is installed on the top of the sewage reaction tank. The radar level meter feeds back the detected depth information in the sewage reaction tank to the deep learning module.

[0043] In some embodiments, the intelligent dosing system of the decentralized sewage station is also provided with an image acquisition device; the image acquisition device is used to monitor SS suspended matter in the sewage reaction tank and report it to the control system.

[0044] In an embodiment of the present invention, an image acquisition device is provided in the sewage reaction tank. The image acquisition device is installed in the slide groove of the fixed rod. The fixed rod is connected to the bottom of the sewage reaction tank. The acquisition device includes a float arranged at the bottom. When the liquid level of the sewage reaction tank rises, the acquisition device slides upward along the path of the track of the fixed rod under the action of the buoyancy of the pool water by setting a float structure, and vice versa, it slides downward under the action of gravity, thereby more accurately monitoring the situation of SS suspended matter in the sewage reaction tank through visual recognition technology. A stirring mechanism is also provided at the bottom of the sewage reaction tank to achieve the purpose of rapid dissolution of SS suspended matter.

[0045] In some embodiments, an MBR water purification device is provided at the outlet of the sewage reaction tank; the MBR water purification device is used to deeply purify part of the sewage flowing through the outlet of the sewage treatment tank, and transport the purified water to the medicine barrel used to dissolve oxalic acid through a suction pump.

[0046] In the embodiment of the present invention, the MBR water purification device deeply purifies part of the sewage flowing through the outlet of the sewage treatment pool, and transports the purified clean water to the medicine barrel for dissolving oxalic acid through a suction pump, thereby realizing the recycling of the purified sewage, making the system self-sufficient in water for dissolving oxalic acid and saving water resources.

[0047] In some embodiments, the control system is used to: input water quality monitoring information and current drug content into a pre-established deep learning model, obtain the drug addition type and target drug addition amount at each target moment, and obtain a drug addition curve by fitting.

[0048] In the embodiment of the present invention, a dosing curve is fitted for each drug, the horizontal axis of the curve is time, and the vertical axis is the target dosing amount at the time. According to these curves, dosing can be indicated.

[0049] In some embodiments, the deep learning model includes a reward value calculation module and a parameter update module; the reward value calculation module is used to determine the reward value based on water quality monitoring information, current drug content and dosing curve; the parameter update module is used to update the parameters of the deep learning model according to the reward value.

[0050] In some embodiments, the reward value calculation module is used to: determine the reward function under constraints with the goal of minimizing the cost of adding medicines and maximizing the stability of water quality; input the water quality monitoring information and the current drug content into the reward function to obtain the reward value.

[0051] In an embodiment of the present invention, firstly, suspended solids (SS) stability and pH stability indices are constructed, and cost indices of flocculants, oxalic acid and other agents are calculated. According to the water quality monitoring information at each moment and the current drug content, the drug addition cost and water quality stability can be calculated according to the above indicators (all are normalized values, the higher the cost and the lower the stability, the smaller the indicator value). The drug addition cost index value and the water quality stability index value are summarized according to the preset weights to obtain the reward value.

[0052] In some embodiments, the parameter updating module is used to: calculate the policy gradient according to the reward value; and update the parameters of the deep learning model according to the policy gradient.

[0053] In an embodiment of the present invention, based on the reward value, the advantage function in the policy gradient algorithm is calculated, and then the policy gradient is calculated. The gradient ascent algorithm is used to update the parameters of the policy network because our goal is to maximize the loss function (actually minimize the negative loss function). Continuously collect data and update the parameters of the policy network through the policy gradient algorithm. During the training process, since the parameters of the policy network are constantly changing, the action probability distribution of its output will also change accordingly, so that the model can adapt to different wastewater quality conditions.

[0054] For example, when the SS content in wastewater is high for a long time, through continuous learning, the policy network may gradually adjust the action probability distribution, so that the probability of increasing the flocculant dosage increases, so as to better treat wastewater with high SS content. At the same time, with the stabilization of water quality and the control of costs, the model will also learn the appropriate action combination to achieve the best treatment effect.

[0055] The dosing system provided by the present invention monitors the water quality at the inlet and outlet ends of the sewage reaction tank, can add different reagents, can perform stable, efficient and safe mixed pretreatment on the medicinal materials, and adjusts the supply amount of the reagent in real time according to the collected medicinal content in the sewage reaction tank, which not only realizes precise adjustment of the dosage adjustment, but also controls the process to respond quickly, can achieve the goal of saving reagents while ensuring the water quality of the outlet, and avoids the delay in detection and waste of reagents and thus pollution of the environment.

[0056] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0057] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0058] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0059] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0060] In the embodiments provided by the present invention, it should be understood that the disclosed devices / controllers and methods can be implemented in other ways. For example, the device / controller embodiments described above are only schematic, for example, the division of modules or units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0061] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0062] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0063] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. Computer-readable media can include: any entity or device that can carry computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0064] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. An intelligent dosing system for a decentralized sewage station, characterized in that: include: Dosing device and control system; The control system is used to perform the following steps: Obtain water quality monitoring information and current drug content; Determine a dosing curve based on the water quality monitoring information, current drug content and a pre-established deep learning model; The dosing device is controlled to add the drug into the sewage reaction tank according to the dosing curve.

2. The intelligent dosing system for a decentralized sewage station according to claim 1 is characterized in that: The dosing device includes an automatic feeding device and a stirring device; The automatic feeding device is provided with a medicine bin; the control system is used to control the medicine bin to feed medicine into the stirring device according to the medicine adding curve; The stirring device is used to monitor the drug content and report it to the control system.

3. The intelligent dosing system for a decentralized sewage station according to claim 2 is characterized in that: The medicine storage chamber includes an oxalic acid dosing chamber body and a PAC dosing chamber body; The four bottom corners of the oxalic acid dosing bin and the PAC dosing bin are each provided with a weighing module, which is used to monitor the remaining weight and material level of the oxalic acid or PAC medicine material, and feed back the weight information and material level information to the control system.

4. The intelligent dosing system for a decentralized sewage station according to claim 2 is characterized in that: A vibration device is provided at the bottom of the automatic feeding device; The control system is used to send a control signal to the vibration device when it detects that the amount of medicine fed is insufficient, so that the vibration device drives the medicine bin to vibrate.

5. The intelligent dosing system for a decentralized sewage station according to claim 3 is characterized in that: The intelligent dosing system of the decentralized sewage station is also provided with an image acquisition device; The image acquisition device is used to monitor SS suspended matter in the sewage reaction tank and report it to the control system.

6. The intelligent dosing system for a decentralized sewage station according to claim 1 is characterized in that: The sewage reaction tank outlet is provided with an MBR water purification device; The MBR water purification device is used to deeply purify part of the sewage flowing through the outlet of the sewage treatment pool, and transport the purified water to the medicine barrel for dissolving oxalic acid through a suction pump.

7. The intelligent dosing system for a decentralized sewage station according to claim 1 is characterized in that: The control system is used to: The water quality monitoring information and the current drug content are input into a pre-established deep learning model to obtain the drug addition type and target drug addition amount at each target moment, and the drug addition curve is obtained by fitting.

8. The intelligent dosing system for a decentralized sewage station according to claim 1 is characterized in that: The deep learning model includes a reward value calculation module and a parameter updating module; The reward value calculation module is used to determine the reward value according to the water quality monitoring information, the current drug content and the drug addition curve; The parameter updating module is used to update the parameters of the deep learning model according to the reward value.

9. The intelligent dosing system for a decentralized sewage station according to claim 1 is characterized in that: The reward value calculation module is used for: Under the constraints, the reward function is determined with the goal of minimizing the dosing cost and maximizing the water quality stability; The water quality monitoring information and the current medicinal material content are input into the reward function to obtain the reward value.

10. The intelligent dosing system for a decentralized sewage station according to claim 1, characterized in that: The parameter updating module is used for: Calculate a policy gradient based on the reward value; According to the policy gradient, the parameters of the deep learning model are updated.

Images