AU2021100130A4 - Intelligent control method and system for sewage treatment aeration based on deep learning - Google Patents
Intelligent control method and system for sewage treatment aeration based on deep learning Download PDFInfo
- Publication number
- AU2021100130A4 AU2021100130A4 AU2021100130A AU2021100130A AU2021100130A4 AU 2021100130 A4 AU2021100130 A4 AU 2021100130A4 AU 2021100130 A AU2021100130 A AU 2021100130A AU 2021100130 A AU2021100130 A AU 2021100130A AU 2021100130 A4 AU2021100130 A4 AU 2021100130A4
- Authority
- AU
- Australia
- Prior art keywords
- aeration
- sewage treatment
- water
- optimal
- blower
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005273 aeration Methods 0.000 title claims abstract description 121
- 239000010865 sewage Substances 0.000 title claims abstract description 88
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000013135 deep learning Methods 0.000 title claims abstract description 24
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 85
- 238000012549 training Methods 0.000 claims abstract description 45
- 239000013598 vector Substances 0.000 claims abstract description 39
- 238000003062 neural network model Methods 0.000 claims abstract description 35
- 239000010802 sludge Substances 0.000 claims abstract description 27
- 238000006243 chemical reaction Methods 0.000 claims abstract description 17
- 230000000306 recurrent effect Effects 0.000 claims abstract description 9
- 238000012544 monitoring process Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 10
- 238000005265 energy consumption Methods 0.000 abstract description 8
- 238000004364 calculation method Methods 0.000 abstract description 2
- 239000007787 solid Substances 0.000 description 9
- XKMRRTOUMJRJIA-UHFFFAOYSA-N ammonia nh3 Chemical compound N.N XKMRRTOUMJRJIA-UHFFFAOYSA-N 0.000 description 8
- 238000013528 artificial neural network Methods 0.000 description 4
- 230000004913 activation Effects 0.000 description 3
- 238000007664 blowing Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 3
- 239000007788 liquid Substances 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 239000000126 substance Substances 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 230000014759 maintenance of location Effects 0.000 description 2
- 229910052760 oxygen Inorganic materials 0.000 description 2
- 239000001301 oxygen Substances 0.000 description 2
- IJGRMHOSHXDMSA-UHFFFAOYSA-N Atomic nitrogen Chemical compound N#N IJGRMHOSHXDMSA-UHFFFAOYSA-N 0.000 description 1
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- CKUAXEQHGKSLHN-UHFFFAOYSA-N [C].[N] Chemical compound [C].[N] CKUAXEQHGKSLHN-UHFFFAOYSA-N 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000033116 oxidation-reduction process Effects 0.000 description 1
- 239000002957 persistent organic pollutant Substances 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 239000011574 phosphorus Substances 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F7/00—Aeration of stretches of water
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W10/00—Technologies for wastewater treatment
- Y02W10/10—Biological treatment of water, waste water, or sewage
Landscapes
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Automation & Control Theory (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Hydrology & Water Resources (AREA)
- Environmental & Geological Engineering (AREA)
- Water Supply & Treatment (AREA)
- Chemical & Material Sciences (AREA)
- Organic Chemistry (AREA)
- Activated Sludge Processes (AREA)
Abstract
The present disclosure provides an intelligent control method and system for sewage
treatment aeration based on deep learning (DL). The present disclosure uses a gated recurrent unit
(GRU) neural network model to determine an optimal aeration rate under a current working
condition (a minimum aeration rate for effluent compliance), and performs aeration control
according to the optimal aeration rate. The present disclosure reduces the energy consumption in
the sewage treatment process and stabilizes the quality of the effluent. In addition, the present
disclosure does not require complicated calculations, and improves the sewage treatment
efficiency.
1/3
Establish a gated recurrent unit (GRU) neural network model
Acquire quality parameters of water at a water inlet and a water outlet of a sewage treatment 102 system under different working conditions, quality parameters of water and sludge in each reaction
tank, operating parameters of the sewage treatment system and operating parameters of a blower as
training input vectors
Determine optimal aeration rates under different working conditions as optimal aeration rates
corresponding to different training input vectors
Establish a training set including the training input vectors and the optimal aeration rates
corresponding to the training input vectors
Train the GRU neural network model by using the training set to obtain a trained GRU neural
network model
Acquire the quality parameters of the water at the water inlet and the water outlet of the sewage 106
treatment system under a current working condition, the quality parameters of the water and the
sludge in each reaction tank, the operating parameters of the sewage treatment system and the
operating parameters of the blower, and establish a monitored data vector
107
Input the monitored data vector into the trained GRU neural network model to obtain an optimal
aeration rate under the current working condition
108
Perform aeration control according to the optimal aeration rate under the current working condition
FIG. 1
Description
1/3
Establish a gated recurrent unit (GRU) neural network model
Acquire quality parameters of water at a water inlet and a water outlet of a sewage treatment 102 system under different working conditions, quality parameters of water and sludge in each reaction tank, operating parameters of the sewage treatment system and operating parameters of a blower as training input vectors
Determine optimal aeration rates under different working conditions as optimal aeration rates corresponding to different training input vectors
Establish a training set including the training input vectors and the optimal aeration rates corresponding to the training input vectors
Train the GRU neural network model by using the training set to obtain a trained GRU neural network model
Acquire the quality parameters of the water at the water inlet and the water outlet of the sewage 106 treatment system under a current working condition, the quality parameters of the water and the sludge in each reaction tank, the operating parameters of the sewage treatment system and the operating parameters of the blower, and establish a monitored data vector
107 Input the monitored data vector into the trained GRU neural network model to obtain an optimal aeration rate under the current working condition
108 Perform aeration control according to the optimal aeration rate under the current working condition
FIG. 1
INTELLIGENT CONTROL METHOD AND SYSTEM FOR SEWAGE TREATMENT AERATION BASED ON DEEP LEARNING TECHNICAL FIELD The present disclosure relates to the technical field of sewage treatment, in particular to an intelligent control method and system for sewage treatment aeration based on deep learning (DL). BACKGROUND The sewage treatment industry is an energy-intensive industry (ElI), which mainly consumes electricity and chemicals. Biological treatment method is the most widely used method in modem sewage treatment. As the core unit of the sewage treatment plant, the biological treatment system undertakes the task of removing organic pollutants in sewage. The power consumption required for the operation of the aeration system of the sewage treatment plant accounts for about 50% to % of the energy consumption of the whole plant. Therefore, the optimization and transformation of the aeration system in the biological treatment process of the sewage treatment plant to improve its stability and energy resource utilization is the key to the sewage treatment plant's discharge compliance and energy saving. Most of the current aeration control methods for sewage treatment adopt manual proportional-integral-derivative (PID) control or manual empirical control, which has low control accuracy. The sewage biological treatment system carries out biological, chemical and physical reactions simultaneously. It is a complex system in which multiple variables interact with each other and show strong nonlinearity. In the sewage biological treatment system, the aeration control system has the characteristics of strong multi-variable interaction, strong nonlinear correlation, difficulty in online monitoring, uncertainty, time variability, time lag and large inertia. The traditional sewage treatment aeration control system is restricted by hardware parameters, which makes it difficult for the controller to process a large number of complex sewage treatment data acquired in real time on site. Therefore, it is an urgent technical problem to reduce energy consumption in the sewage treatment process, stabilize the quality of the effluent and improve the sewage treatment efficiency. SUMMARY The present disclosure aims to provide an intelligent control method and system for sewage treatment aeration based on deep learning (DL). The present disclosure reduces the energy consumption in the sewage treatment process, stabilizes the quality of the effluent and improves the sewage treatment efficiency. To achieve the above objective, the present disclosure provides the following solutions:
An intelligent control method for sewage treatment aeration based on DL, where the intelligent control method includes the following steps: establishing a gated recurrent unit (GRU) neural network model; acquiring quality parameters of water at a water inlet and a water outlet of a sewage treatment system under different working conditions, quality parameters of water and sludge in each reaction tank, operating parameters of the sewage treatment system and operating parameters of a blower as training input vectors; determining optimal aeration rates under different working conditions as optimal aeration rates corresponding to different training input vectors, where the optimal aeration rates are minimum aeration rates that make the quality parameters of the water at the water outlet meet a sewage treatment discharge requirement; establishing a training set including the training input vectors and the optimal aeration rates corresponding to the training input vectors; training the GRU neural network model by using the training set to obtain a trained GRU neural network model; acquiring the quality parameters of the water at the water inlet and the water outlet of the sewage treatment system under a current working condition, the quality parameters of the water and the sludge in each reaction tank, the operating parameters of the sewage treatment system and the operating parameters of the blower, and establishing a monitored data vector; inputting the monitored data vector into the trained GRU neural network model to obtain an optimal aeration rate under the current working condition; and performing aeration control according to the optimal aeration rate under the current working condition. Optionally, the determining optimal aeration rates under different working conditions specifically includes: adjusting a blower speed under an i-th working condition, monitoring blower aeration rates at different speeds and the quality parameters of the water at the water outlet, and determining a minimum aeration rate that makes the quality parameters of the water at the water outlet meet the sewage treatment discharge requirement as the optimal aeration rate under the i-th working condition. Optionally, the performing aeration control according to the optimal aeration rate under the current working condition specifically includes: determining an adjusted value of the blower speed corresponding to a difference between the optimal aeration rate under the current working condition and an actual aeration rate under the current working condition according to a relationship between the blower speed, air pressure and the blower aeration rate; and inputting the adjusted value of the blower speed into a proportional-integral-derivative (PID) controller for blower speed control in a blower control system to control the aeration rate. Optionally, the GRU neural network model includes an update gate and a reset gate. According to the specific embodiments of the present disclosure, the present disclosure has the following technical effects: The present disclosure provides an intelligent control method and system for sewage treatment aeration based on DL. The intelligent control method includes the following steps: establishing a gated recurrent unit (GRU) neural network model; acquiring quality parameters of water at a water inlet and a water outlet of a sewage treatment system under different working conditions, quality parameters of water and sludge in each reaction tank, operating parameters of the sewage treatment system and operating parameters of a blower as training input vectors; determining optimal aeration rates under different working conditions as optimal aeration rates corresponding to different training input vectors; establishing a training set including the training input vectors and the optimal aeration rates corresponding to the training input vectors; training the GRU neural network model by using the training set to obtain a trained GRU neural network model; acquiring the quality parameters of the water at the water inlet and the water outlet of the sewage treatment system under a current working condition, the quality parameters of the water and the sludge in each reaction tank, the operating parameters of the sewage treatment system and the operating parameters of the blower, and establishing a monitored data vector; inputting the monitored data vector into the trained GRU neural network model to obtain an optimal aeration rate under the current working condition; and performing aeration control according to the optimal aeration rate under the current working condition. The present disclosure uses a GRU neural network model to determine the optimal aeration rate under the current working condition (the minimum aeration rate for effluent compliance), and performs aeration control according to the optimal aeration rate. The present disclosure reduces the energy consumption in the sewage treatment process and stabilizes the quality of the effluent. In addition, the present disclosure does not require complicated calculations, and improves the sewage treatment efficiency. BRIEF DESCRIPTION OF THE DRAWINGS To describe the technical solutions in embodiments of the present disclosure or in the prior art more clearly, the accompanying drawings needed in the embodiments are introduced below briefly. Apparently, the accompanying drawings in the following description show merely some embodiments of the present disclosure, and other drawings can be derived from these accompanying drawings by those of ordinary skill in the art without creative efforts. FIG. 1 is a flowchart of an intelligent control method for sewage treatment aeration based on deep learning (DL) according to the present disclosure. FIG. 2 is a schematic diagram of the intelligent control method for sewage treatment aeration based on DL according to the present disclosure. FIG. 3 is a structural diagram of a gated recurrent unit (GRU) neural network according to the present disclosure. DETAILED DESCRIPTION The technical solutions of the embodiments of the present disclosure are clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are merely a part rather than all of the embodiments of the present disclosure. All other embodiments derived from the embodiments of the present disclosure by a person of ordinary skill in the art without creative efforts should fall within the protection scope of the present disclosure. An objective of the present disclosure is to provide an intelligent control method and system for sewage treatment aeration based on deep learning (DL). The present disclosure reduces the energy consumption in the sewage treatment process, stabilizes the quality of the effluent and improves the sewage treatment efficiency. To make the above objectives, features and advantages of the present disclosure clearer and more comprehensible, the present disclosure is described in further detail below with reference to the accompanying drawings and specific implementations. As shown in FIGS. 1 and 2, the present disclosure provides an intelligent control method for sewage treatment aeration based on DL. The intelligent control method includes the following steps: Step 101: Establish a gated recurrent unit (GRU) neural network model. The present disclosure uses a GRU as a deep neural network to determine an optimal aeration rate. A long short-term memory (LSTM) solves the problem that a recurrent neural network (RNN) cannot handle long-distance dependence well. The GRU is a variant of the LSTM, which maintains the effect of the LSTM while making the architecture simpler. The GRU only keeps two gates, an update gate and a reset gate. The update gate is used to control how much state information to pass from a previous moment into the current state. A larger value of the update gate indicates more state information to pass from the previous moment. The reset gate is used to control how much of the state information of the previous moment to forget. A smaller value of the reset gate indicates more state information to forget. The update gate zt at time step t is calculated as follows: zt=a(W(z)xt+U(z)ht-1) In the equation, xt represents an input vector of a t-th time step, that is, a t-th component of an input sequence X, which will undergo a linear transformation (multiplied by a weight matrix W(z)); hi 1 stores information of a previous time step t-1, which will also undergo a linear transformation (multiplied by a weight matrix U(z)). The update gate adds these two pieces of information and puts it into a Sigmoid activation function a, thus compressing an activation result to between 0 and 1. The reset gate mainly determines how much past information needs to be forgotten. Like the update gate, it is calculated using the following expression: rt=aT(W~rxt+U~rht-1) It =tanh(W- [rt * ht_ , xt]) ht (1 - 7) *ht-1 + zt *hi yt = cr(W) - ht)
In the equation, zt represents an update gate output; rt represents a reset gate output; h represents a first intermediate vector; ht represents the second intermediate vector; Ye represents an output of the GRU neural network model; a() represents an activation function; W4 and W respectively represent a weight matrix. Step 102: Acquire quality parameters of water at a water inlet and a water outlet of a sewage treatment system under different working conditions, quality parameters of water and sludge in each reaction tank, operating parameters of the sewage treatment system and operating parameters of a blower as training input vectors. The present disclosure uses multi-channel multi-parameter online sensors to monitor the quality parameters of the water at the water inlet and the water outlet of the sewage treatment system and the quality parameters of the water and the sludge in each reaction tank on line. The online sensors include a water quality sensor and a sludge quality sensor, which are used to monitor dynamic changes in the water quality and sludge quality in real time. The water quality sensor is a multi-channel multi-parameter fast water quality sensor, including a flow meter, an online thermometer, an online level gauge, an online pH meter, an online oxidation-reduction potential (ORP) tester, an online suspended solids (SS) concentration meter, an online dissolved oxygen (DO) meter, an online chemical oxygen demand (COD) meter, an online total nitrogen (TN) meter, an online ammonia nitrogen meter and an online total phosphorus (TP) meter. The
AS quality sensor is an online sludge concentration meter. The water quality parameters and the sludge quality parameters vary depending on sewage treatment locations. The parameters of the water inlet include flow rate, temperature, pH, ORP, SS, COD, DO, TN, ammonia nitrogen and TP. The parameters of an anaerobic/anoxic reactor (reaction tank) include flow rate, temperature, pH, ORP, SS, COD, DO, TN, ammonia nitrogen, TP and mixed liquor suspended solids (MLSS). The parameters of an aerobic reactor (reaction tank) include flow rate, temperature, liquid level, pH, ORP, SS, COD, DO, TN, ammonia nitrogen, TP and MLSS. The parameters of a secondary settling tank include flow rate, temperature, liquid level, pH, ORP, SS, COD, TN, ammonia nitrogen and TP. The parameters of the water outlet include flow rate, pH, ORP, SS, COD, TN, ammonia nitrogen and TP. The acquired water quality parameters and sludge quality parameters are transmitted to a blower room control system. The blower room control system is used to control a blower, and the blower is used to provide air to a biochemical reactor (reaction tank). A main outlet duct of the blower is provided with a pressure meter and a flow meter to monitor operating parameters of the blower in real time, including the blower speed, blowing rate (aeration rate) and air pressure. The blower room control system sends the received water quality parameters, sludge quality parameters and operating parameters of the blower to a main control workstation. The main control workstation is able to acquire other operating data of the sewage treatment plant, including an influent carbon nitrogen (C/N) ratio, a hydraulic retention time (HRT), a sludge retention time (SRT), a sludge recycle ratio (SRR) and a mixed liquor recycle ratio (MLRR). Step 103: Determine optimal aeration rates under different working conditions as optimal aeration rates corresponding to different training input vectors. Step 103 "Determine optimal aeration rates" specifically includes: adjust a blower speed under an i-th working condition, monitor blower aeration rates at different speeds and the quality parameters of the water at the water outlet, and determine a minimum aeration rate that makes the quality parameters of the water at the water outlet meet a sewage treatment discharge requirement as the optimal aeration rate under the i-th working condition. Step 104: Establish a training set including the training input vectors and the optimal aeration rates corresponding to the training input vectors. Step 105: Train the GRU neural network model by using the training set to obtain a trained GRU neural network model. The main control workstation performs structured and normalized processing by the received water quality parameters, sludge quality parameters, operating parameters of the blower and other operating parameters of the sewage treatment plant, and constructs a GRU neural network model for training to obtain a trained GRU neural network model for intelligent control of sewage treatment aeration. Before the model training, the present disclosure performs structured and normalized processing on the water quality parameters, the sludge quality parameters, the operating parameters of the blower and other operating parameter of the sewage treatment plant in the training set to obtain a high-quality data set. The data are divided into a training set, a validation set and a test set according to a ratio of 8:1:1 to build a deep neural network for training to obtain a correlation model between the monitored data and controllable parameters, specifically an interaction model of the water quality parameters, the sludge quality parameters, the operating parameters of the sewage treatment plant and the operating parameters of the blower. Thus, a trained GRU neural network model for aeration rate prediction and blowing intelligent control of sewage treatment aeration is further obtained. The model can calculate the most efficient and energy-saving aeration rate in real time based on the current monitored data. During the model training process, the present disclosure performs regularization processing by dropout to obtain the trained GRU neural network model to predict the optimal aeration rate. The trained GRU neural network model is installed into an algorithm controller. The algorithm controller calculates the optimized aeration rate of the sewage treatment aeration control system according to the parameters monitored in real time, further calculates the optimized blower speed, and feeds back the calculated results to the blower. The blower adjusts the speed to the optimized speed according to the feedback, thereby realizing the intelligent control of the aeration rate. Step 106: Acquire the quality parameters of the water at the water inlet and the water outlet of the sewage treatment system under a current working condition, the quality parameters of the water and the sludge in each reaction tank, the operating parameters of the sewage treatment system and the operating parameters of the blower, and establish a monitored data vector. Step 107: Input the monitored data vector into the trained GRU neural network model to obtain an optimal aeration rate under the current working condition. Step 108: Perform aeration control according to the optimal aeration rate under the current working condition. Step 108 "Perform aeration control according to the optimal aeration rate under the current working condition" specifically includes: determine an adjusted value of the blower speed corresponding to a difference between the optimal aeration rate under the current working condition and an actual aeration rate under the current working condition according to a relationship between the blower speed, air pressure and the blower aeration rate; and input the adjusted value of the blower speed into a proportional-integral-derivative (PID) controller for blower speed control in a blower control system to control the aeration rate. The present disclosure installs the trained GRU neural network model into the algorithm controller located in the sewage treatment plant. The algorithm controller calculates the optimal aeration rate according to the water quality parameters, the sludge quality parameters and the operating parameters of the sewage treatment plant, then calculates the blower speed, and feeds the calculated results back to the blower room control system located in a blower room. The blower room control system adjusts the speed based on the feedback and the PID control algorithm to realize the control of the aeration rate. In the intelligent control for sewage treatment aeration based on DL provided by the present disclosure, the water quality parameters include flow rate, temperature, liquid level, pH, ORP, SS, COD, DO, TN, ammonia nitrogen and TP. The sludge quality parameter is MLSS. The operating parameters of the blower include blower speed, blowing rate and air pressure. Other operating parameters of the sewage treatment plant include influent C/N ratio, HRT, SRT, SRR and MLRR. In the intelligent control for sewage treatment aeration based on DL provided by the present disclosure, the algorithm controller calculates the optimal aeration rate of the sewage treatment aeration control system according to water quality parameters, compares the actual aeration rate with the optima aeration rate to obtain the blower speed that actually needs to be regulated, and generates a regulation command for a control module. According to the specific embodiments of the present disclosure, the present disclosure has the following technical effects: 1. The present disclosure performs real-time detection and transmission of the quality parameters of water and sludge in the reactor and the blower operating parameters, and integrates the DL algorithm into the intelligent control of sewage treatment aeration. In this way, the present disclosure realizes the precise aeration prediction and control in the sewage biochemical treatment process, and overcomes the shortcomings of sewage treatment systems such as strong multi-variable interaction, strong nonlinear correlation, strong time variability and time lag. 2. The present disclosure effectively improves the control accuracy of the aeration rate, and solves the problems of large energy consumption and high cost caused by the inability of precise aeration in the existing sewage treatment aeration technology. 3. The present disclosure ensures the efficient and stable operation of the sewage biochemical treatment process, and ensures that the effluent BOD, COD, ammonia nitrogen and other water quality indicators are stable and compliant. 4. The present disclosure realizes the high efficiency, low energy consumption and low cost of sewage treatment.
Each embodiment in the specification of the present disclosure is described in a progressive manner. Each embodiment focuses on the difference from other embodiments, and the same and similar parts between the embodiments may refer to each other. In this paper, several embodiments are used for illustration of the principles and implementations of the present disclosure. The description of the above embodiments is used to help illustrate the method of the present disclosure and the core principles thereof. In addition, those of ordinary skill in the art may make various modifications in terms of specific implementations and scope of application in accordance with the teachings of the present disclosure. In conclusion, the content of the present specification should not be construed as a limitation to the present disclosure.
Claims (4)
- Claims 1. An intelligent control method for sewage treatment aeration based on deep learning (DL), wherein the intelligent control method comprises the following steps: establishing a gated recurrent unit (GRU) neural network model; acquiring quality parameters of water at a water inlet and a water outlet of a sewage treatment system under different working conditions, quality parameters of water and sludge in each reaction tank, operating parameters of the sewage treatment system and operating parameters of a blower as training input vectors; determining optimal aeration rates under different working conditions as optimal aeration rates corresponding to different training input vectors, wherein the optimal aeration rates are minimum aeration rates that make the quality parameters of the water at the water outlet meet a sewage treatment discharge requirement; establishing a training set comprising the training input vectors and the optimal aeration rates corresponding to the training input vectors; training the GRU neural network model by using the training set to obtain a trained GRU neural network model; acquiring the quality parameters of the water at the water inlet and the water outlet of the sewage treatment system under a current working condition, the quality parameters of the water and the sludge in each reaction tank, the operating parameters of the sewage treatment system and the operating parameters of the blower, and establishing a monitored data vector; inputting the monitored data vector into the trained GRU neural network model to obtain an optimal aeration rate under the current working condition; and performing aeration control according to the optimal aeration rate under the current working condition.
- 2. The intelligent control method for sewage treatment aeration based on DL according to claim 1, wherein the determining optimal aeration rates under different working conditions specifically comprises: adjusting a blower speed under an i-th working condition, monitoring blower aeration rates at different speeds and the quality parameters of the water at the water outlet, and determining a minimum aeration rate that makes the quality parameters of the water at the water outlet meet the sewage treatment discharge requirement as the optimal aeration rate under the i-th working condition.
- 3. The intelligent control method for sewage treatment aeration based on DL according to claim 1, wherein the performing aeration control according to the optimal aeration rate under thein current working condition specifically comprises: determining an adjusted value of the blower speed corresponding to a difference between the optimal aeration rate under the current working condition and an actual aeration rate under the current working condition according to a relationship between the blower speed, air pressure and the blower aeration rate; and inputting the adjusted value of the blower speed into a proportional-integral-derivative (PID) controller for blower speed control in a blower control system to control the aeration rate.
- 4. The intelligent control method for sewage treatment aeration based on DL according to claim 1, wherein the GRU neural network model comprises an update gate and a reset gate.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011391829.0A CN112782975A (en) | 2020-12-02 | 2020-12-02 | Sewage treatment aeration intelligent control method and system based on deep learning |
| CN202011391829.0 | 2020-12-02 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| AU2021100130A4 true AU2021100130A4 (en) | 2021-04-08 |
Family
ID=75280464
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| AU2021100130A Active AU2021100130A4 (en) | 2020-12-02 | 2021-01-11 | Intelligent control method and system for sewage treatment aeration based on deep learning |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112782975A (en) |
| AU (1) | AU2021100130A4 (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113830896B (en) * | 2021-09-27 | 2022-12-27 | 湖南工商大学 | Aerobic segment oxygen early warning method, system and equipment based on graph neural network |
| CN113845205B (en) * | 2021-10-18 | 2022-08-16 | 江苏科易达环保科技有限公司 | High-salinity high-nitrogen sewage intelligent integrated desalting and denitrification control system |
| CN115259357B (en) * | 2022-07-25 | 2023-05-23 | 河南大学 | Microbial degradation control method and system based on artificial intelligence |
| CN115806344B (en) * | 2022-11-22 | 2023-08-04 | 曲誉环境综合治理有限公司 | ORP (integrated pressure and pressure) control precise aeration system |
| CN115981153B (en) * | 2022-12-30 | 2023-08-04 | 浙江问源环保科技股份有限公司 | Deep learning-based A 2 O process intermittent low-carbon aeration method and control system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107402586A (en) * | 2017-08-29 | 2017-11-28 | 北京易沃特科技有限公司 | Dissolved Oxygen concentration Control method and system based on deep neural network |
| CN109879410B (en) * | 2019-03-27 | 2022-05-13 | 上海西派埃智能化系统有限公司 | Sewage treatment aeration control system |
| CN110378533B (en) * | 2019-07-22 | 2023-04-18 | 中展环能(北京)技术有限公司 | Intelligent aeration management method based on big data analysis |
| CN110577275B (en) * | 2019-08-30 | 2022-03-29 | 绍兴海坤水处理技术有限公司 | Intelligent aeration control system and method for sewage treatment |
| CN111291937A (en) * | 2020-02-25 | 2020-06-16 | 合肥学院 | Method for predicting quality of treated sewage based on combination of support vector classification and GRU neural network |
-
2020
- 2020-12-02 CN CN202011391829.0A patent/CN112782975A/en active Pending
-
2021
- 2021-01-11 AU AU2021100130A patent/AU2021100130A4/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112782975A (en) | 2021-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| AU2021100130A4 (en) | Intelligent control method and system for sewage treatment aeration based on deep learning | |
| US9747544B2 (en) | Method and system for wastewater treatment based on dissolved oxygen control by fuzzy neural network | |
| AU2021101438A4 (en) | Adaptive control method and system for aeration process | |
| Lee et al. | Neural network modeling for on-line estimation of nutrient dynamics in a sequentially-operated batch reactor | |
| US11709463B2 (en) | Control method based on adaptive neural network model for dissolved oxygen of aeration system | |
| CN103744293B (en) | Wastewater treatment monitoring method and system based on fuzzy neural network | |
| Ruano et al. | An advanced control strategy for biological nutrient removal in continuous systems based on pH and ORP sensors | |
| EP2414901B1 (en) | System and method for monitoring an integrated system | |
| Yang et al. | Progress in the development of control strategies for the SBR process | |
| CN103197539A (en) | Wastewater disposal intelligent optimization control aeration quantity method | |
| CN108536106B (en) | Aeration system dissolved oxygen online regulation and control method based on Kalman filtering-extreme learning machine | |
| Luccarini et al. | Soft sensors for control of nitrogen and phosphorus removal from wastewaters by neural networks | |
| Zhao et al. | A novel control strategy for improved nitrogen removal in an alternating activated sludge process—Part I. Process analysis | |
| Chen et al. | Smart energy savings for aeration control in wastewater treatment | |
| Marsili-Libelli et al. | Fuzzy predictive control for nitrogen removal in biological wastewater treatment | |
| Baeza et al. | In-line fast OUR (oxygen uptake rate) measurements for monitoring and control of WWTP | |
| CN111484124A (en) | Intelligent control and biochemical process intensive method and system for sewage treatment | |
| Van Nguyen et al. | Combined ILC and PI regulator for wastewater treatment plants | |
| Huang et al. | Improving nitrogen removal using a fuzzy neural network-based control system in the anoxic/oxic process | |
| Sheik et al. | Fuzzy Logic Control of Active Sludge–Based Wastewater Treatment Plants | |
| Pires et al. | Knowledge-based fuzzy system for diagnosis and control of an integrated biological wastewater treatment process | |
| CN101794117B (en) | Method for optimizing and controlling operation of membrane bioreactor based on mechanism model | |
| Swarnalatha | Simulation and Analysis of Ammonium-Based Aeration Control Strategies to Enhance Efficiency in Wastewater Treatment Plant | |
| Puñal et al. | Automatic start-up of UASB reactors | |
| Zhao et al. | Modeling of activated sludge wastewater treatment processes using integrated neural networks and a first principle model |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FGI | Letters patent sealed or granted (innovation patent) |