CN111290275A - Sewage treatment optimization control method based on reinforcement learning particle swarm algorithm - Google Patents
Sewage treatment optimization control method based on reinforcement learning particle swarm algorithm Download PDFInfo
- Publication number
- CN111290275A CN111290275A CN202010110081.6A CN202010110081A CN111290275A CN 111290275 A CN111290275 A CN 111290275A CN 202010110081 A CN202010110081 A CN 202010110081A CN 111290275 A CN111290275 A CN 111290275A
- Authority
- CN
- China
- Prior art keywords
- sewage treatment
- neural network
- particle swarm
- energy consumption
- concentration set
- 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.)
- Granted
Links
Images
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
- C02F1/00—Treatment of water, waste water, or sewage
- C02F1/008—Control or steering systems not provided for elsewhere in subclass C02F
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F2209/00—Controlling or monitoring parameters in water treatment
- C02F2209/15—N03-N
-
- C—CHEMISTRY; METALLURGY
- C02—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F—TREATMENT OF WATER, WASTE WATER, SEWAGE, OR SLUDGE
- C02F2209/00—Controlling or monitoring parameters in water treatment
- C02F2209/22—O2
Abstract
The invention relates to a sewage treatment optimization control method based on a reinforcement learning particle swarm algorithm, which comprises the following steps: (1) constructing four elements of an intelligent body of a sewage treatment process based on reinforcement learning: status, context, rewards, and actions; (2) establishing a sewage treatment optimization control flow based on a reinforcement learning particle swarm algorithm: firstly, predicting the adjustment trend of a concentration set value by a neural network model, weighting to a position and speed updating formula of a standard particle swarm algorithm, carrying out iterative updating, and taking global optimum as the concentration set values of nitrate nitrogen and dissolved oxygen; then obtaining progressive particles, recording concentration set values and adjustment trends of the progressive particles, and training a neural network model; and finally, evaluating and updating the neural network model. The method is repeatedly carried out according to the optimization period, and is beneficial to optimizing the sewage treatment process through strategy evaluation and continuous improvement.
Description
Technical Field
The invention belongs to the technical field of sewage treatment, relates to an intelligent optimization algorithm for optimal control of a sewage treatment process, and particularly relates to a sewage treatment optimal control method based on a reinforcement learning particle swarm algorithm.
Background
Water pollution is one of the most troublesome problems in the world and will, if left untreated, exhibit a tendency to continue to worsen in the coming decades. At present, an activated sludge process is one of the most effective methods for removing organic pollutants, and the key problem of research is how to minimize energy consumption under the condition of ensuring that water quality reaches the standard. One of the important measures of this method is to maintain the concentration of dissolved oxygen in the aerobic zone by feeding suitable oxygen into the aeration tank by means of a blower and to maintain the concentration of nitrate nitrogen in the anoxic zone by returning the sewage water by means of a return pump, which requires a large supply of electrical energy. However, the operation of the blower and the reflux pump requires a large energy supply, which inevitably increases the operating cost. Meanwhile, from the biochemical reaction mechanism, the nitrification and denitrification can be ensured to be smoothly carried out only by proper dissolved oxygen concentration and nitrate nitrogen concentration. Therefore, dynamic optimization of the concentration set values of the dissolved oxygen and the nitrate nitrogen is required, and an optimal control strategy aiming at reducing energy consumption and effluent quality is constructed, so that the sewage treatment effect is improved.
The optimization control process of the sewage treatment process can be divided into four methods: firstly, the PID-based dissolved oxygen and nitrate nitrogen concentration control method is adopted, but the traditional control strategy is difficult to realize the online adjustment of control parameters and is only effective to specific working conditions. And secondly, a concentration control method based on predictive control, but the method needs a high-precision predictive model to realize a better control effect. The concentration control method based on neural network control, the concentration control method based on intelligent optimization, however, in the two algorithms, optimization information is not fully utilized, except sewage data, the optimization information is not continued, namely, the later suboptimization does not draw useful information from the former suboptimization process, and the former suboptimization does not play a guiding role for the later suboptimization, thereby causing the defects of low control precision, poor real-time performance, high energy consumption, unsatisfactory water quality control and the like.
Disclosure of Invention
In order to overcome the defects and shortcomings of the prior art, the invention provides a sewage treatment optimization control method based on a reinforcement learning particle swarm algorithm, which is used for dynamically optimizing the concentration set values of dissolved oxygen and nitrate nitrogen in the process of sewage treatment, so that the control accuracy is effectively improved, and the sewage treatment effect is optimized.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a sewage treatment optimization control method based on a reinforcement learning particle swarm algorithm comprises the following steps:
(1) constructing four elements of an intelligent body of a sewage treatment process based on reinforcement learning: status, context, rewards, and actions;
(2) establishing a sewage treatment optimization control flow based on a reinforcement learning particle swarm algorithm: firstly, predicting the adjustment trend of a concentration set value by a neural network model, weighting to a position and speed updating formula of a standard particle swarm algorithm, carrying out iterative updating, and taking global optimum as the concentration set values of nitrate nitrogen and dissolved oxygen; then obtaining progressive particles, recording concentration set values and adjustment trends of the progressive particles, and training a neural network model; and finally, evaluating and updating the neural network model.
Preferably, the state of the four elements of the agent is the sewage component concentration.
Preferably, the four-element environment of the intelligent body is a sewage treatment process.
Preferably, the four-element reward of the agent consists of water quality and total energy consumption; the total energy consumption comprises aeration energy consumption and pumping energy consumption.
Preferably, the actions of the four elements of the agent are nitrate nitrogen and dissolved oxygen concentration set value adjustment strategies.
Preferably, the nitrate nitrogen and dissolved oxygen concentration set values are described by particle candidate solutions, and the reward is described by a fitness value, which is expressed as weighted values of effluent quality, aeration energy consumption and pumping energy consumption and is expressed by the following formula:
f＝c(Ea+Ep)+EQ (1)
in the formula, EQ is the effluent quality and is represented by the fine paid for discharging effluent pollutants to a receiving water body, Ea is the aeration energy consumption, Ep is the pumping energy consumption, and c is the aeration energy consumption and pumping energy consumption coefficient.
Preferably, the step (2) specifically comprises:
(a) initializing neural network parameters, concentration set values and adjustment trends;
(b) the neural network predicts the adjustment trend of the concentration set value and updates the position and the speed of the particles;
(c) globally optimal as the concentration set values of nitrate nitrogen and dissolved oxygen, calculating the effluent quality and the energy consumption adaptability value, and guiding the operation of the sewage treatment process;
(d) acquiring progressive particles, recording concentration set values and adjustment trends of the progressive particles, and training a neural network model;
(e) evaluating and updating a neural network model;
(f) and (c) judging whether the end condition is met, if not, returning to the step (b) for continuing.
Preferably, step (b) is: predicting the adjustment trend of the concentration set values of the nitrate nitrogen and the dissolved oxygen by a neural network function, and weighting to a standard particle swarm algorithm position and speed updating formula:
xid(t+1)＝xid(t)+vid(t+1) (3)
in the formula: xi＝(xi1,xi2,…,xid,…,xiD) Is the position of particle i; vi＝(vi1,vi2,…,vid,…,viD) Is the velocity component of particle i; pi＝(pi1,pi2,…,pid,…,piD) Is optimal for the individual of particle i; and P isg＝(pg1,pg2,…,pgd,…,pgD) Is composed ofA global optimum value; omega is the inertial weight; c. C1And c2Is a learning factor, r1And r2Is a random probability value; vi_y＝(vi1_y,vi2_y,…,vid_y,…,viD_y) Adjusting the trend for the concentration set value predicted by the neural network; k is a neural network model prediction coefficient; ω vid(t) is the inertial component;
is the predicted component.
Preferably, step (d) is: in the optimization process, if the effluent quality and the energy consumption are reduced, the particles are called progressive particles, and the current nitrate nitrogen and dissolved oxygen concentration set values and the adjustment trend are stored and used as training data to train a neural network model.
Preferably, step (e) is: after training is finished, evaluating the neural network, only keeping an inertia component and a neural network prediction component in the evaluation process, and operating by neural network particles to obtain a concentration set value; and if the water quality and the energy consumption mean value obtained by the new network are superior to those of the previous generation network, reserving parameters of the new network for guiding the prediction optimization step.
Compared with the prior art, the invention has the beneficial effects that:
(1) the method is based on the reinforcement learning theory, does not need to accumulate man-made mark training samples for a long time, continuously and interactively tries through environment and action factors, adjusts the strategy according to feedback information, and finally judges the optimal concentration set value in various states according to the strategy so as to improve the sewage treatment effect;
(2) the method is based on a swarm intelligence algorithm-a particle swarm algorithm, simulates the foraging behavior of birds, has the advantages of high convergence rate, concise concept and easy realization, and is generally applied in the optimization field; in the application process of the sewage treatment process, the method is beneficial to improving the diversity distribution of the solution so as to find the global optimal concentration set value;
(3) the method has a memory function, and the sewage treatment process is optimized to be periodic optimization, namely, optimization calculation is performed at intervals. In order to improve the treatment effect, the influence trend of sewage parameters on the concentration set values of dissolved oxygen and nitrate nitrogen needs to be recorded, information is stored and reused, reference data is provided for the next optimization calculation, and therefore the calculation efficiency is improved.
Drawings
FIG. 1 is a functional block diagram of the present invention;
FIG. 2 is a schematic block diagram of four elements of an intelligent agent for wastewater treatment according to the present invention;
FIG. 3 is a flow chart of the present invention based on the reinforced learning particle swarm optimization for optimizing the control of sewage treatment.
Detailed Description
The invention is further illustrated with reference to specific examples, without however being limited thereto. Those skilled in the art can and should understand that any simple changes or substitutions based on the spirit of the present invention should fall within the protection scope of the present invention.
Referring to fig. 1 to 3, a sewage treatment optimization control method based on reinforcement learning particle swarm optimization comprises the following steps:
(1) four factors for constructing intelligent sewage treatment process based on reinforcement learning
The intelligent four elements of the sewage treatment process based on reinforcement learning are as follows: state, environment, reward and action, as shown in fig. 2, the agent returns state and reward according to environment, gives action strategy, thereby improving environment process; the state is set as the concentration of components such as ammonia nitrogen, nitrate nitrogen, COD and the like in the sewage biochemical tank, and the indexes can reflect the current state of the system; the environment is a sewage treatment process; the reward is composed of two parts, namely, the penalty required to be paid for discharging the effluent pollutants to the receiving water body represents the effluent quality EQ; second, total energy consumption, including aeration energy consumption EaAnd pumping energy consumption Ep(ii) a The intelligent action, namely the adjustment strategy of the concentration set values of nitrate nitrogen and dissolved oxygen, is realized by a reinforced learning particle swarm algorithm; in the whole process: the intelligent agent gives out a nitrate nitrogen and dissolved oxygen concentration set value adjusting strategy according to the sewage component concentration, effluent quality and energy consumption fed back by the environment, thereby improving the sewage treatment process.
The reinforcement learning facing the sewage treatment process is realized by a particle swarm optimization algorithm. In the method, the concentration set values of nitrate nitrogen and dissolved oxygen are described by adopting a particle candidate solution, and the set values regulate the trend, namely the particle speed. The reward is described by a fitness value and is expressed as effluent quality EQ and aeration energy consumption EaAnd pumping energy consumption EpThe weighted value of (c) is shown as follows:
f＝c(Ea+Ep)+EQ (1)
wherein c is the aeration energy consumption and pumping energy consumption coefficient.
(2) Establishment of optimized control process for sewage treatment based on reinforcement learning particle swarm algorithm
The intelligent action adjustment strategy is how to determine the adjustment trend of the concentration set values of the nitrate nitrogen and the dissolved oxygen, and the specific operation is to give the particle running speed for improving the fitness value according to the current position of the candidate particle solution. In the strategy, the concentration set value is memorized by a neural network function to adjust the trend, namely the optimal running speed of the particles. The second step is divided into the following 6 steps as shown in fig. 3:
(a) initializing neural network parameters, concentration set values and adjustment trends;
(b) the neural network predicts the adjustment trend of the concentration set value and updates the position and the speed of the particles; predicting the adjustment trend of the concentration set values of the nitrate nitrogen and the dissolved oxygen by a neural network function, and weighting to a standard particle swarm algorithm position and speed updating formula:
xid(t+1)＝xid(t)+vid(t+1) (3)
in the formula: xi＝(xi1,xi2,…,xid,…,xiD) Is the position of particle i; vi＝(vi1,vi2,…,vid,…,viD) Is the velocity component of particle i; pi＝(pi1,pi2,…,pid,…,piD) Is optimal for the individual of particle i; and P isg＝(pg1,pg2,…,pgd,…,pgD) Is a global optimum value; omega is the inertial weight; c. C1And c2Is a learning factor, r1And r2Is a random probability value; vi_y＝(vi1_y,vi2_y,…,vid_y,…,viD_y) Adjusting the trend for the concentration set value predicted by the neural network; k is a neural network model prediction coefficient; ω vid(t) is the inertial component;
is a predicted component;
in the optimization process, if the effluent quality and the energy consumption are reduced, namely the particle fitness value is reduced, the particles are called progressive particles, and the current nitrate nitrogen and dissolved oxygen concentration set values and the adjustment trend are stored as training data;
(c) globally optimal as the concentration set values of nitrate nitrogen and dissolved oxygen, calculating the effluent quality and the energy consumption adaptability value, and guiding the operation of the sewage treatment process;
(d) acquiring progressive particles, recording concentration set values and adjustment trends of the progressive particles, and training a neural network model; updating parameters of the neural network model by adopting gradient descent according to the set values of the nitrate nitrogen and dissolved oxygen concentration and the adjustment trend data, and repeatedly iterating until the model converges;
(e) evaluating and updating a network model; after training is finished, evaluating the neural network; in the evaluation process, only the inertia component and the neural network prediction component are reserved, and the neural network particles operate to obtain a concentration set value; if the water quality and the energy consumption mean value obtained by the new network are superior to those of the previous generation network, the new network parameters are reserved for guiding the prediction optimization step;
(f) and (c) judging whether the end condition is met, if not, returning to the step (b) for continuing.
By means of strategy period evaluation and continuous improvement, the optimal set values of nitrate nitrogen and dissolved oxygen concentration are obtained as far as possible under the conditions of different water inflow rates, water quality parameters and component concentrations by taking the effluent quality and energy consumption as targets, so that the sewage treatment efficiency is improved.
Claims (10)
1. A sewage treatment optimization control method based on reinforcement learning particle swarm optimization is characterized by comprising the following steps:
(1) constructing four elements of an intelligent body of a sewage treatment process based on reinforcement learning: status, context, rewards, and actions;
(2) establishing a sewage treatment optimization control flow based on a reinforcement learning particle swarm algorithm: firstly, predicting the adjustment trend of a concentration set value by a neural network model, weighting to a position and speed updating formula of a standard particle swarm algorithm, and carrying out iterative updating; taking the global optimum as the concentration set values of nitrate nitrogen and dissolved oxygen; then obtaining progressive particles, recording concentration set values and adjustment trends of the progressive particles, and training a neural network model; and finally, evaluating and updating the neural network model.
2. The reinforced learning particle swarm algorithm-based sewage treatment optimization control method according to claim 1, wherein the method comprises the following steps: the state of the four elements of the intelligent agent is the sewage component concentration.
3. The reinforced learning particle swarm algorithm-based sewage treatment optimization control method according to claim 1, wherein the method comprises the following steps: the environment of the four elements of the intelligent body is a sewage treatment process.
4. The reinforced learning particle swarm algorithm-based sewage treatment optimization control method according to claim 1, wherein the method comprises the following steps: the reward of the four elements of the intelligent agent consists of water outlet quality and total energy consumption; the total energy consumption comprises aeration energy consumption and pumping energy consumption.
5. The reinforced learning particle swarm algorithm-based sewage treatment optimization control method according to claim 1, wherein the method comprises the following steps: the actions of the four elements of the intelligent agent are the adjustment strategy of the concentration set values of nitrate nitrogen and dissolved oxygen.
6. The reinforced learning particle swarm algorithm-based sewage treatment optimization control method according to claim 1, wherein the method comprises the following steps: the nitrate nitrogen and dissolved oxygen concentration set values are described by adopting particle candidate solutions, and the reward is described by adopting a fitness value, which is expressed as weighted values of effluent quality, aeration energy consumption and pumping energy consumption and is expressed by the following formula:
f＝c(Ea+Ep)+EQ
in the formula, EQ is the effluent quality and is represented by the fine paid for discharging effluent pollutants to a receiving water body, Ea is the aeration energy consumption, Ep is the pumping energy consumption, and c is the aeration energy consumption and pumping energy consumption coefficient.
7. The optimized control method for sewage treatment based on the reinforcement learning particle swarm algorithm according to claim 1, wherein the step (2) specifically comprises the following steps:
(a) initializing neural network parameters, concentration set values and adjustment trends;
(b) the neural network predicts the adjustment trend of the concentration set value and updates the position and the speed of the particles;
(c) globally optimal as the concentration set values of nitrate nitrogen and dissolved oxygen, calculating the effluent quality and the energy consumption adaptability value, and guiding the operation of the sewage treatment process;
(d) acquiring progressive particles, recording concentration set values and adjustment trends of the progressive particles, and training a neural network model;
(e) evaluating and updating a neural network model;
(f) and (c) judging whether the end condition is met, if not, returning to the step (b) for continuing.
8. The optimized control method for sewage treatment based on reinforcement learning particle swarm optimization according to claim 7, wherein the step (b) is as follows: predicting the adjustment trend of the concentration set values of the nitrate nitrogen and the dissolved oxygen by a neural network function, and weighting to a standard particle swarm algorithm position and speed updating formula:
xid(t+1)＝xid(t)+vid(t+1)
in the formula: xi＝(xi1,xi2,…,xid,…,xiD) Is the position of particle i; vi＝(vi1,vi2,…,vid,…,viD) Is the velocity component of particle i; pi＝(pi1,pi2,…,pid,…,piD) Is optimal for the individual of particle i; and P isg＝(pg1,pg2,…,pgd,…,pgD) Is a global optimum value; omega is the inertial weight; c. C1And c2Is a learning factor, r1And r2Is a random probability value; vi_y＝(vi1_y,vi2_y,…,vid_y,…,viD_y) Adjusting the trend for the concentration set value predicted by the neural network; k is a neural network model prediction coefficient; ω vid(t) is the inertial component;
is the predicted component.
9. The optimized control method for sewage treatment based on reinforcement learning particle swarm optimization according to claim 7, wherein the step (d) is as follows: in the optimization process, if the effluent quality and the energy consumption are reduced, the particles are called progressive particles, and the current nitrate nitrogen and dissolved oxygen concentration set values and the adjustment trend are stored and used as training data to train a neural network model.
10. The optimized control method for sewage treatment based on reinforcement learning particle swarm optimization according to claim 7, wherein the step (e) is as follows: after training is finished, evaluating the neural network, only keeping an inertia component and a neural network prediction component in the evaluation process, and operating by neural network particles to obtain a concentration set value; and if the water quality and the energy consumption mean value obtained by the new network are superior to those of the previous generation network, reserving parameters of the new network for guiding the prediction optimization step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010110081.6A CN111290275B (en) | 2020-02-23 | 2020-02-23 | Sewage treatment optimization control method based on reinforcement learning particle swarm algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010110081.6A CN111290275B (en) | 2020-02-23 | 2020-02-23 | Sewage treatment optimization control method based on reinforcement learning particle swarm algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111290275A true CN111290275A (en) | 2020-06-16 |
| CN111290275B CN111290275B (en) | 2021-01-05 |
Family
ID=71025385
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010110081.6A Active CN111290275B (en) | 2020-02-23 | 2020-02-23 | Sewage treatment optimization control method based on reinforcement learning particle swarm algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111290275B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418435A (en) * | 2020-11-16 | 2021-02-26 | 软通动力信息技术(集团)股份有限公司 | Reinforcement learning method, reinforcement learning device, electronic equipment and storage medium |
| CN117174198A (en) * | 2023-11-02 | 2023-12-05 | 山东鸿远新材料科技股份有限公司 | Automatic detection cleaning method and system based on zirconium oxychloride production |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102207928A (en) * | 2011-06-02 | 2011-10-05 | 河海大学常州校区 | Reinforcement learning-based multi-Agent sewage treatment decision support system |
| CN102854296A (en) * | 2012-08-30 | 2013-01-02 | 北京工业大学 | Sewage-disposal soft measurement method on basis of integrated neural network |
| CN103942910A (en) * | 2014-04-11 | 2014-07-23 | 北京工业大学 | Machine room fire disaster early warning method and system on basis of IPv6 |
| CN106698642A (en) * | 2016-12-29 | 2017-05-24 | 北京工业大学 | Multi-objective real-time optimization control method for sewage treatment process |
| CN106991493A (en) * | 2017-03-17 | 2017-07-28 | 浙江工商大学 | Sewage disposal water outlet parameter prediction method based on Grey production fuction |
| WO2017181319A1 (en) * | 2016-04-18 | 2017-10-26 | 江南大学 | Particle swarm optimization and reinforcement learning algorithm-based dynamic walking control system for biomimetic biped robot |
| CN108445757A (en) * | 2018-03-12 | 2018-08-24 | 北京工业大学 | A kind of sewage disposal process optimal control method based on dynamic multi-objective particle cluster algorithm |
| CN108762082A (en) * | 2018-06-07 | 2018-11-06 | 北京工业大学 | A kind of sewage disposal process collaboration Optimal Control System |
| US20190049525A1 (en) * | 2017-08-10 | 2019-02-14 | United States Of America As Represented By Secretary Of The Navy | Methods and Systems for Classifying Optically Detected Power Quality Disturbances |
| KR101958903B1 (en) * | 2018-05-31 | 2019-03-15 | 수원대학교산학협력단 | PARTIAL DISCHARGE CLASSIFICATION SYSTEM USING RBFNNs(RADIAL BASIS FUNCTION Neural NETWORKS) MODEL |
| CN110413107A (en) * | 2019-06-21 | 2019-11-05 | 浙江科技学院 | Bionic mechanical hand interaction control method based on electromyography signal pattern-recognition and particle group optimizing |
-
2020
- 2020-02-23 CN CN202010110081.6A patent/CN111290275B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102207928A (en) * | 2011-06-02 | 2011-10-05 | 河海大学常州校区 | Reinforcement learning-based multi-Agent sewage treatment decision support system |
| CN102854296A (en) * | 2012-08-30 | 2013-01-02 | 北京工业大学 | Sewage-disposal soft measurement method on basis of integrated neural network |
| CN103942910A (en) * | 2014-04-11 | 2014-07-23 | 北京工业大学 | Machine room fire disaster early warning method and system on basis of IPv6 |
| WO2017181319A1 (en) * | 2016-04-18 | 2017-10-26 | 江南大学 | Particle swarm optimization and reinforcement learning algorithm-based dynamic walking control system for biomimetic biped robot |
| CN106698642A (en) * | 2016-12-29 | 2017-05-24 | 北京工业大学 | Multi-objective real-time optimization control method for sewage treatment process |
| CN106991493A (en) * | 2017-03-17 | 2017-07-28 | 浙江工商大学 | Sewage disposal water outlet parameter prediction method based on Grey production fuction |
| US20190049525A1 (en) * | 2017-08-10 | 2019-02-14 | United States Of America As Represented By Secretary Of The Navy | Methods and Systems for Classifying Optically Detected Power Quality Disturbances |
| CN108445757A (en) * | 2018-03-12 | 2018-08-24 | 北京工业大学 | A kind of sewage disposal process optimal control method based on dynamic multi-objective particle cluster algorithm |
| KR101958903B1 (en) * | 2018-05-31 | 2019-03-15 | 수원대학교산학협력단 | PARTIAL DISCHARGE CLASSIFICATION SYSTEM USING RBFNNs(RADIAL BASIS FUNCTION Neural NETWORKS) MODEL |
| CN108762082A (en) * | 2018-06-07 | 2018-11-06 | 北京工业大学 | A kind of sewage disposal process collaboration Optimal Control System |
| CN110413107A (en) * | 2019-06-21 | 2019-11-05 | 浙江科技学院 | Bionic mechanical hand interaction control method based on electromyography signal pattern-recognition and particle group optimizing |
Non-Patent Citations (2)
| Title |
|---|
| HUI ZHENG: "A multi-swarm particle swarm optimization algorithm for tracking multiple targets", 《 2014 9TH IEEE CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS》 * |
| 程志强: "基于粒子群算法的神经网络在污水处理优化中的应用", 《北京工商大学学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418435A (en) * | 2020-11-16 | 2021-02-26 | 软通动力信息技术(集团)股份有限公司 | Reinforcement learning method, reinforcement learning device, electronic equipment and storage medium |
| CN112418435B (en) * | 2020-11-16 | 2023-07-28 | 软通动力信息技术(集团)股份有限公司 | Reinforcement learning method, reinforcement learning device, electronic equipment and storage medium |
| CN117174198A (en) * | 2023-11-02 | 2023-12-05 | 山东鸿远新材料科技股份有限公司 | Automatic detection cleaning method and system based on zirconium oxychloride production |
| CN117174198B (en) * | 2023-11-02 | 2024-01-26 | 山东鸿远新材料科技股份有限公司 | Automatic detection cleaning method and system based on zirconium oxychloride production |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111290275B (en) | 2021-01-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9747544B2 (en) | Method and system for wastewater treatment based on dissolved oxygen control by fuzzy neural network | |
| CN105404151B (en) | Sewage disposal process dynamic multi-objective optimization control method | |
| CN109669352B (en) | Oily sewage treatment process optimization control method based on self-adaptive multi-target particle swarm | |
| CN111290275B (en) | Sewage treatment optimization control method based on reinforcement learning particle swarm algorithm | |
| CN105573115B (en) | Sewage disposal process energy-conserving and optimizing control method based on quantum genetic algorithm | |
| CN108549234A (en) | A kind of multiobjective optimization control method based on dynamic variate | |
| CN108445757B (en) | Sewage treatment process optimization control method based on dynamic multi-target particle swarm algorithm | |
| Holenda et al. | Aeration optimization of a wastewater treatment plant using genetic algorithm | |
| CN109711070A (en) | A kind of dissolved oxygen concentration optimization method based on activated sludge water process | |
| Qiao et al. | Adaptive optimal control for a wastewater treatment plant based on a data-driven method | |
| CN103792844B (en) | A kind of sewage treatment control method based on orderly cluster | |
| CN110647037A (en) | Cooperative control method for sewage treatment process based on two-type fuzzy neural network | |
| CN105843072A (en) | Wastewater treatment energy conservation optimization method based on improved local search immune genetic algorithm | |
| CN108536106A (en) | A kind of aerating system dissolved oxygen based on Kalman filtering-extreme learning machine regulates and controls method online | |
| CN113189881A (en) | Multi-objective optimization control method and system for sewage treatment | |
| Marsili-Libelli et al. | Fuzzy predictive control for nitrogen removal in biological wastewater treatment | |
| CN113568311B (en) | Knowledge information-based intelligent optimal control method for sewage treatment | |
| Dai et al. | Optimal control of sewage treatment process using a dynamic multi-objective particle swarm optimization based on crowding distance | |
| CN114455691A (en) | Composite sewage denitrification process device and operation parameter optimization method | |
| CN116360366B (en) | Sewage treatment process optimization control method | |
| CN116881742A (en) | Task clustering-based multi-working-condition double-layer optimal control method for sewage treatment process | |
| CN114488822B (en) | Intelligent sewage treatment process optimization method based on high-dimensional multi-objective evolutionary algorithm | |
| CN115356930A (en) | Multi-objective optimization control system and method in sewage treatment process | |
| CN114879487A (en) | Sewage treatment process robust model prediction control method based on interval two-type fuzzy neural network | |
| CN111399455B (en) | Urban sewage treatment real-time optimization control method based on operation process information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20220110 Address after: 314214 No. 1, YeDing Road, Caoqiao street, Pinghu City, Jiaxing City, Zhejiang Province Patentee after: ZHEJIANG JINGXING PAPER JOINT STOCK Co.,Ltd. Address before: 310023 No. 318 stay Road, Xihu District, Zhejiang, Hangzhou Patentee before: ZHEJIANG University OF SCIENCE AND TECHNOLOGY |