A method for arranging a sensor for sensing greenhouse gases in a municipal wastewater treatment process
By establishing a multi-objective optimization model for the deployment of greenhouse gas sensors in the urban wastewater treatment process and using a multi-objective genetic algorithm to solve the problem, the measurement deviation caused by the differences in greenhouse gas concentrations within the wastewater treatment plant was solved, achieving accurate monitoring and cost optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING DRAINAGE GRP CO LTD
- Filing Date
- 2026-03-10
- Publication Date
- 2026-06-12
AI Technical Summary
Differences in greenhouse gas concentrations at different locations within urban wastewater treatment plants can lead to measurement errors, resulting in substandard effluent quality and excessive energy consumption, making accurate monitoring difficult.
A multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes was established. The optimal deployment scheme of the sensors was determined by solving the multi-objective genetic algorithm to reduce data acquisition errors and deployment costs.
It enables accurate monitoring of greenhouse gas data from wastewater treatment plants, reduces sensor deployment costs, and minimizes data acquisition errors.
Smart Images
Figure CN122197579A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of water treatment, and more specifically, to a method for deploying greenhouse gas sensors in urban wastewater treatment processes. Background Technology
[0002] With the continuous improvement of urban wastewater treatment optimization and control technologies, organic pollutants in wastewater have been effectively controlled. However, due to the possible differences in gas concentrations at different locations within the same water treatment zone, measurement deviations occur, making it difficult to achieve accurate monitoring of greenhouse gas data at wastewater treatment plants. This results in effluent quality exceeding standards and excessive operating energy consumption, placing a heavy burden on both the urban environment and wastewater treatment plants.
[0003] Therefore, it is necessary to develop a method for deploying greenhouse gas sensors in urban wastewater treatment processes.
[0004] The information disclosed in the background section of this invention is intended only to enhance the understanding of the general background of this invention and should not be construed as an admission or in any way implying that such information constitutes prior art known to those skilled in the art. Summary of the Invention
[0005] This invention proposes a method for deploying greenhouse gas sensors in urban wastewater treatment processes. Based on the gas generation characteristics and zoning conditions, a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes is established. A multi-objective evolutionary algorithm is designed to solve the model, obtaining the number of collection points and the deployment scheme. This method ensures that the collection error is minimized while reducing the sensor deployment cost, and achieves accurate monitoring of greenhouse gas data from wastewater treatment plants.
[0006] This disclosure provides a method for deploying greenhouse gas sensors in an urban wastewater treatment process, including: The deployment of multiple candidate sensor locations is determined, and then the greenhouse gas emission concentration vectors of the candidate locations and the center location are obtained respectively. Establish a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes; A multi-objective optimization model is solved by a multi-objective genetic algorithm to obtain the output vector of the multi-objective optimization model, and the output vector is used as the optimization deployment scheme for greenhouse gas sensors in urban sewage treatment process.
[0007] Preferably, obtaining the greenhouse gas emission concentration vectors for the candidate sites and the center site respectively includes: The layout of multiple candidate points is used as the key vector:
[0008] Where x is the vector of key variables, xl For the first l The layout of each candidate point is indicated by "1" for placement and "0" for no placement. l ∈[1, L]; Greenhouse gas emission concentrations at multiple candidate sites were monitored, and the greenhouse gas emission concentration vectors at the candidate sites were obtained as follows:
[0009] Wherein, H(x) ,t )for t Greenhouse gas emission concentration vectors at candidate sites at given time points. H ( x l , t )for t Time of the first l Greenhouse gas emission concentrations at each candidate site; Therefore, the greenhouse gas emission concentration vector at the center point is obtained as follows:
[0010] Wherein, H(y) ,t )for t Greenhouse gas emission concentration vector at the center point at any given time. H ( y l , t ) is the first l and the l+ The center point between 1 candidate point is t Greenhouse gas emission concentration at any given time.
[0011] Preferably, based on the greenhouse gas emission concentration vectors of candidate sites and the central site, a multi-objective optimization model for the deployment of greenhouse gas sensors in the urban wastewater treatment process is established, including: Based on the greenhouse gas emission concentration vectors of candidate sites and the center site, an optimization model for acquisition error and a sensor deployment cost optimization model are established. Based on the data acquisition error optimization model and the sensor deployment cost optimization model, the multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes is obtained as follows:
[0012] in, E (x) represents the objective function value for optimizing the acquisition error of decision variable x. C (x) represents the objective function value for optimizing the sensor deployment cost of decision variable x.
[0013] Preferably, the acquisition error optimization model is as follows:
[0014] Where E(x) is the time interval T The total acquisition error optimization model within the system, E (x ,t )for t Optimization model for acquisition error at any given time. x is t time E (x ,t The input vector of ) H ( x l , t ) is the first l Candidate points t Greenhouse gas emission concentration at any given time ( x l , t ) is the first l Candidate points t Estimated greenhouse gas emission concentrations at time t. , H ( y lk , t ) is the first l Each central point t Greenhouse gas emission concentration at any given time ( y l , t ) is the first l Each central point t Estimated greenhouse gas emission concentrations at time t. d ( x l-1 ,x l ) is the first l- 1 and l The distance between candidate points.
[0015] Preferably, the sensor deployment cost optimization model is as follows:
[0016] in, C (x) represents the sensor deployment cost optimization model. λ 1 and λ 2 is the weighting coefficient. I com ( l ) represents the complexity coefficient of the candidate points.
[0017] Preferably, the output vector of the multi-objective optimization model is obtained by solving the multi-objective optimization model using a multi-objective genetic algorithm, including: Set the total number of iterations and population size for the multi-objective genetic algorithm, randomly initialize the population, and obtain the... g Population vector; The next generation of key variable vectors is generated through crossover mutation, and then the population is updated through environmental selection to obtain the next generation. g+ First generation population; Iteratively update the population until the total number of iterations is reached, and output the last generation of the population as the output vector.
[0018] Preferably, the first g The population vector is:
[0019] in, g Let P be the number of iterations. g ) is the first g Population vector, x1( g ) is the first g The first key variable vector in the algebra, x2( g ) is the first g The second key variable vector in the algebra, x3 (g ) is the first g The third key variable vector in the algebra, x N ( g ) is the first g The middle generation N A vector of key variables.
[0020] Preferably, the next-generation key variable vector is generated through crossover mutation as follows:
[0021] Where, x l ( g+ 1) is the first g+ 1st generation l A vector of key variables, x SBX For the first g The middle generation l The key variable vectors are generated by simulating binary crossover. , x PM For the first g The middle generation l The key variable vector is generated by multinomial mutation of the key variable vector. , , e The cross-distribution index, m It is the distribution index of variation. u and r These are uniformly distributed random numbers.
[0022] Preferably, the first g+ The first generation population is:
[0023] Wherein, P( g+ 1) is the first g+ Generation 1 population, x i ( g+ 1) Calculate the key variable vector. , The dominance relationship is defined as follows: , E (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing the acquisition error is calculated. E (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of the acquisition error; C (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing sensor deployment costs; C (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of sensor deployment cost. dis (·) represents the distance of congestion level. , E (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing the acquisition error is calculated. E (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing the acquisition error is calculated. C (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing sensor deployment costs; C (x i-1 ( g )) represents the key variable vector x i-1 (g The objective function value for optimizing sensor deployment costs is determined by...
[0024] Preferably, it further includes: From the last generation of the population g max The solution set P( g max )=[x1( g max ), x2( g max ), x3( g max ), …, x N ( g max A key variable vector is randomly selected from the data to serve as the optimal deployment scheme for greenhouse gas sensors in the urban wastewater treatment process.
[0025] The method of the present invention has other features and advantages that will be apparent from or will be set forth in detail in the accompanying drawings and following detailed description, which together serve to explain the particular principles of the invention. Attached Figure Description
[0026] The above and other objects, features and advantages of the present invention will become more apparent from the more detailed description of exemplary embodiments of the invention in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same parts.
[0027] Figure 1 A flowchart illustrating the steps of a method for deploying greenhouse gas sensors in an urban wastewater treatment process according to an embodiment of the present invention is shown. Detailed Implementation
[0028] Preferred embodiments of the invention will now be described in more detail. While preferred embodiments of the invention are described below, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.
[0029] Figure 1 A flowchart illustrating the steps of a method for deploying greenhouse gas sensors in an urban wastewater treatment process according to an embodiment of the present invention is shown.
[0030] like Figure 1 As shown, the method for deploying greenhouse gas sensors in the urban wastewater treatment process includes: Step 101: Determine the layout of multiple candidate sensor locations, and then obtain the greenhouse gas emission concentration vectors of the candidate locations and the center location respectively. Step 102: Establish a multi-objective optimization model for the deployment of greenhouse gas sensors in the urban wastewater treatment process; Step 103: Solve the multi-objective optimization model using a multi-objective genetic algorithm to obtain the output vector of the multi-objective optimization model, and use the output vector as the optimization deployment scheme for greenhouse gas sensors in the urban sewage treatment process.
[0031] In one example, obtaining the greenhouse gas emission concentration vectors for the candidate sites and the center site respectively includes: The layout of multiple candidate points is used as the key vector:
[0032] Where x is the vector of key variables, x l For the first l The layout of each candidate point is indicated by "1" for placement and "0" for no placement. l ∈[1, L]; Greenhouse gas emission concentrations at multiple candidate sites were monitored, and the greenhouse gas emission concentration vectors at the candidate sites were obtained as follows:
[0033] Wherein, H(x) ,t )for t Greenhouse gas emission concentration vectors at candidate sites at given time points. H ( x l , t )for t Time of the first l Greenhouse gas emission concentrations at each candidate site; Therefore, the greenhouse gas emission concentration vector at the center point is obtained as follows:
[0034] Wherein, H(y) ,t )for t Greenhouse gas emission concentration vector at the center point at any given time. H ( y l , t ) is the first l and the l+ The center point between 1 candidate point is t Greenhouse gas emission concentration at any given time.
[0035] In one example, a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes is established based on the greenhouse gas emission concentration vectors of candidate sites and the center site, including: Based on the greenhouse gas emission concentration vectors of candidate sites and the center site, an optimization model for acquisition error and a sensor deployment cost optimization model are established. Based on the data acquisition error optimization model and the sensor deployment cost optimization model, the multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes is obtained as follows:
[0036] in, E (x) represents the objective function value for optimizing the acquisition error of decision variable x. C (x) represents the objective function value for optimizing the sensor deployment cost of decision variable x.
[0037] In one example, the acquisition error optimization model is as follows:
[0038] Where E(x) is the time interval T The total acquisition error optimization model within the system, E (x ,t )for t Optimization model for acquisition error at any given time. x is t time E (x ,t The input vector of ) H ( x l , t ) is the first l Candidate points t Greenhouse gas emission concentration at any given time ( x l , t ) is the first l Candidate points t Estimated greenhouse gas emission concentrations at time t. , H ( y lk , t ) is the first l Each central point t Greenhouse gas emission concentration at any given time ( y l , t ) is the first l Each central point t Estimated greenhouse gas emission concentrations at time t. d ( x l-1 ,x l ) is the first l- 1 andl The distance between candidate points.
[0039] In one example, the sensor deployment cost optimization model is as follows:
[0040] in, C (x) represents the sensor deployment cost optimization model. λ 1 and λ 2 is the weighting coefficient. I com ( l ) represents the complexity coefficient of the candidate points.
[0041] In one example, a multi-objective optimization model is solved using a multi-objective genetic algorithm, and the output vector of the multi-objective optimization model is obtained as follows: Set the total number of iterations and population size for the multi-objective genetic algorithm, randomly initialize the population, and obtain the... g Population vector; The next generation of key variable vectors is generated through crossover mutation, and then the population is updated through environmental selection to obtain the next generation. g+ First generation population; Iteratively update the population until the total number of iterations is reached, and output the last generation of the population as the output vector.
[0042] In one example, the first g The population vector is:
[0043] in, g Let P be the number of iterations. g ) is the first g Population vector, x1( g ) is the first g The first key variable vector in the algebra, x2( g ) is the first g The second key variable vector in the algebra, x3 (g ) is the first g The third key variable vector in the algebra, x N ( g ) is the first g The middle generation N A vector of key variables.
[0044] In one example, the next generation key variable vector generated through crossover mutation is:
[0045] Where, x l ( g+ 1) is the first g+1st generation l A vector of key variables, x SBX For the first g The middle generation l The key variable vectors are generated by simulating binary crossover. , x PM For the first g The middle generation l The key variable vector is generated by multinomial mutation of the key variable vector. , , e The cross-distribution index, m It is the distribution index of variation. u and r These are uniformly distributed random numbers.
[0046] In one example, the first g+ The first generation population is:
[0047] Wherein, P( g+ 1) is the first g+ Generation 1 population, x i ( g+ 1) Calculate the key variable vector. , The dominance relationship is defined as follows: , E (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing the acquisition error is calculated. E (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of the acquisition error; C (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing sensor deployment costs; C (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of sensor deployment cost. dis (·) represents the distance of congestion level. , E (xi+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing the acquisition error is calculated. E (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing the acquisition error is calculated. C (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing sensor deployment costs; C (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing sensor deployment costs is determined by...
[0048] In one example, it also includes: From the last generation of the population g max The solution set P( g max )=[x1( g max ), x2( g max ), x3( g max ), …, x N ( g max A key variable vector is randomly selected from the data to serve as the optimal deployment scheme for greenhouse gas sensors in the urban wastewater treatment process.
[0049] Specifically, collect greenhouse gas emission data from urban wastewater treatment: Select L The layout of each candidate point is taken as a key variable and expressed as follows: (1) Where x is the vector of key variables, x 1 This indicates the placement of the first candidate point, where "1" indicates placement and "0" indicates no placement. x 2 This indicates the placement of the second candidate point; "1" indicates placement, and "0" indicates no placement. x l For the first l The layout of each candidate point is indicated by "1" for placement and "0" for no placement. xL For the first L The deployment status of each candidate point is indicated by "1" for deployment and "0" for no deployment; temporary monitoring is conducted using mobile sensor deployment. L Candidate locations and L- The greenhouse gas emission concentration at a single centroid is expressed as: (2) (3) Wherein, H(x) ,t )for t Greenhouse gas emission concentration vector at time t. H ( x 1 , t ) is the first candidate point t Greenhouse gas emission concentration at any given time, in mg / L. H ( x 2 , t ) is the second candidate point. t Greenhouse gas emission concentration at any given time, in mg / L. H ( x l , t ) is the first l Candidate points t Greenhouse gas emission concentration at any given time, in mg / L. H ( x L , t ) is the first L Candidate points t Greenhouse gas emission concentration at time t, in mg / L; H(y ,t )for t Greenhouse gas emission concentration vector at the center point at any given time. H ( y 1 , t () is the center point between the first and second candidate points. t Greenhouse gas emission concentration at any given time, in mg / L. H ( y 2 , t () is the center point between the second and third candidate points. t Greenhouse gas emission concentration at any given time, in mg / L. H ( y l , t ) is the first l and the l+Center point between 1 candidate point t Greenhouse gas emission concentration at any given time, in mg / L. H ( x L-1 , t ) is the first L- 1 and the L The center point between the candidate points t Greenhouse gas emission concentration at any given time, in mg / L; A multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes was established, including a data acquisition error optimization model, a sensor deployment cost optimization model, and a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes. This model expresses the relationship between key variables and data acquisition error and sensor deployment cost. ① Establish a model for optimizing data acquisition errors: (4) (5) (6) Where E(x) is the time interval T The total acquisition error optimization model within the system, E (x ,t )for t The data acquisition error optimization model at time x is t time E (x ,t The input vector of ) H ( x l , t ) No. l Candidate points t The greenhouse gas emission concentration at time [0, 2] is given. ( x l , t ) is the first l Candidate points t The estimated greenhouse gas emission concentration at time [0, 2] is given. H ( y lk , t ) is the first l Each central point t The greenhouse gas emission concentration at time [0, 2] is given. ( y l , t ) is the first l Each central point t The estimated greenhouse gas emission concentration at time [0, 2] is given. d (x l-1 , x l ) is the first l- 1 and l The distance between candidate points; ② Establish a sensor deployment cost optimization model: (7) in, C (x) represents the sensor deployment cost optimization model. λ 1 and λ 2 is the weighting coefficient and its value ranges from [0, 1]. I com ( l ) represents the complexity coefficient of the candidate point and its value ranges from [1, 3]. ③ Establish a multi-objective optimization function for the deployment of greenhouse gas sensors in the urban wastewater treatment process: (8) in E (x) represents the objective function value for optimizing the acquisition error of decision variable x. C (x) represents the objective function value for optimizing the sensor deployment cost of decision variable x.
[0050] The deployment locations of greenhouse gas sensors are determined by solving a multi-objective optimization function using a multi-objective genetic algorithm. ① Set the total number of iterations for the multi-objective genetic algorithm to be... g max =500, population size is N =100, randomly initialize the population: (9) in, g The iteration number is [1, ..., ...] and its value range is [1, ...]. g max ], P( g ) is the first g Population vector, x1( g ) is the first g The first key variable vector in the algebra, x2( g ) is the first g The second key variable vector in the algebra, x3 (g ) is the first g The third key variable vector in the algebra, x N ( g ) is the first g The middle generation N A vector of key variables; ② Crossover mutation generates the next generation of key variable vectors: (10) (11) (12) (13) (14) Where, x l ( g+ 1) is the first g+ 1st generation l A vector of key variables, x SBX For the first g The middle generation l The key variable vectors are generated by simulating binary crossover, x PM For the first g The middle generation l The key variable vector is generated by multinomial mutation of the key variable vector. e The cross-distribution index is defined as [20, 100]. m The variance distribution index is defined as [20, 100]. u and r The random number is uniformly distributed and its value ranges from [0, 1]. ③ Environmental selection renews the population: (15) Wherein, P( g+ 1) is the first g+ First-generation population, key variable vector x i ( g+ 1) The calculation is as follows: (16) in, Represents a dominance relationship and is defined as follows: (17) in, E (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing the acquisition error is calculated. E (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of the acquisition error; C (xi ( g )) represents the key variable vector x i ( g The objective function value for optimizing sensor deployment costs; C (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of sensor deployment cost. dis (·) represents the crowding distance and is defined as: (18) in, E (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing the acquisition error is calculated. E (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing the acquisition error is calculated. C (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing sensor deployment costs; C (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing sensor deployment costs; ④ Check if the evolutionary process has reached the stopping condition: if the number of generations... g =500, terminate the evolution process and jump to step ⑤; otherwise, the evolutionary generation... g Increase by 1 and return to step ②; ⑤ From the first g max The solution set P( g max )=[x1( g max ), x2( g max ), x3( g max ), …, x N ( g max Randomly select a key variable vector from )] As an optimized deployment scheme for greenhouse gas sensors in urban wastewater treatment processes.
[0051] To facilitate understanding of the solutions and effects of the embodiments of the present invention, a specific application example is given below. Those skilled in the art should understand that this example is merely for the purpose of understanding the present invention, and any specific details therein are not intended to limit the present invention in any way.
[0052] Example 1
[0053] Greenhouse gas emissions data from urban wastewater treatment were collected: the layout of 28 candidate monitoring sites was selected as the key variable, as shown below: (19) Where x is the vector of key variables, x 1 This indicates the placement of the first candidate point, where "1" indicates placement and "0" indicates no placement. x 2 This indicates the placement of the second candidate point; "1" indicates placement, and "0" indicates no placement. x 3 This indicates the placement of the third candidate point; "1" indicates placement, and "0" indicates no placement. x 28 The deployment status of the 28th candidate site is shown, with "1" indicating deployment and "0" indicating no deployment. Greenhouse gas emission concentrations at the 28 candidate sites and 27 intermediate sites are temporarily monitored using mobile sensor deployment, as shown below: (20) (twenty one) Wherein, H(x) ,t )for t Greenhouse gas emission concentration vectors at candidate sites at given time points. H ( x 1, t ) is the first candidate point t Greenhouse gas emission concentration at any given time, in mg / L. H ( x 2, t ) is the second candidate point. t Greenhouse gas emission concentration at any given time, in mg / L. H ( x 3, t ) is the third candidate site t Greenhouse gas emission concentration at any given time, in mg / L. H ( x 28 , t ) is the 28th candidate site tGreenhouse gas emission concentration at any given time, in mg / L; H ( y 1, t The midpoint between the first and second candidate points is... t Greenhouse gas emission concentration at any given time, in mg / L. H ( y 2, t The midpoint between the second and third candidate points is... t Greenhouse gas emission concentration at any given time, in mg / L. H ( y 3, t The midpoint between the 3rd and 4th candidate points is... t Greenhouse gas emission concentration at any given time, in mg / L. H ( y 27 , t The midpoint between the 27th and 28th candidate points is... t Greenhouse gas emission concentration at any given time, expressed in mg / L.
[0054] (2) Establish a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment process: Establish a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment process, including a data acquisition error optimization model, a sensor deployment cost optimization model, and a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment process, to express the relationship between key variables and data acquisition error and sensor deployment cost; ① Establish a model for optimizing data acquisition errors: (twenty two) (twenty three) (twenty four) Where E(x) is the time interval T The total acquisition error optimization model within the system, E (x ,t )for t The data acquisition error optimization model at time x is t time E (x ,t The input vector of ) H ( x l , t ) No. l Candidate points t The greenhouse gas emission concentration at time [0, 2] is given. ( x l , t ) is the first lCandidate points t The estimated greenhouse gas emission concentration at time [0, 2] is given. H ( y lk , t ) is the first l Each central point t The greenhouse gas emission concentration at time [0, 2] is given. ( y l , t ) is the first l Each central point t The estimated greenhouse gas emission concentration at time [0, 2] is given. d ( x l-1 , x l ) is the first l- 1 and l The distance between candidate points; ② Establish a sensor deployment cost optimization model: (25) in, C (x) represents the sensor deployment cost optimization model. λ 1 and λ 2 is the weighting coefficient and its value ranges from [0, 1]. I com ( l ) represents the complexity coefficient of the candidate point and its value ranges from [1, 3]. ③ Establish a multi-objective optimization function for the deployment of greenhouse gas sensors in the urban wastewater treatment process: (26) in E (x) represents the objective function value for optimizing the acquisition error of decision variable x. C (x) represents the objective function value for optimizing the sensor deployment cost of decision variable x.
[0055] The deployment locations of greenhouse gas sensors are determined by solving a multi-objective optimization function using a multi-objective genetic algorithm. ① Set the total number of iterations for the multi-objective genetic algorithm to be... g max =500, population size is N =100, randomly initialize the population: (27) in, g The iteration number is [1, ..., ...] and its value range is [1, ...].g max ], P( g ) is the first g Population vector, x1( g ) is the first g The first key variable vector in the algebra, x2( g ) is the first g The second key variable vector in the algebra, x3 (g ) is the first g The third key variable vector in the algebra, x N ( g ) is the first g The middle generation N A vector of key variables; ② Crossover mutation generates the next generation of key variable vectors: (28) (29) (30) (31) (32) Where, x l ( g+ 1) is the first g+ 1st generation l A vector of key variables, x SBX For the first g The middle generation l The key variable vectors are generated by simulating binary crossover, x PM For the first g The middle generation l The key variable vector is generated by multinomial mutation of the key variable vector. e The cross-distribution index is defined as [20, 100]. m The variance distribution index is defined as [20, 100]. u and r The random number is uniformly distributed and its value ranges from [0, 1]. ③ Environmental selection renews the population: (33) Wherein, P( g+ 1) is the first g+ First-generation population, key variable vector x i ( g+ 1) The calculation is as follows: (34) in, Represents a dominance relationship and is defined as follows: (35) in, E (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing the acquisition error is calculated. E (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of the acquisition error; C (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing sensor deployment costs; C (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of sensor deployment cost. dis (·) represents the crowding distance and is defined as: (36) in, E (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing the acquisition error is calculated. E (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing the acquisition error is calculated. C (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing sensor deployment costs; C (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing sensor deployment costs; ④ Check if the evolutionary process has reached the stopping condition: if the number of generations... g =500, terminate the evolution process and jump to step ⑤; otherwise, the evolutionary generation...g Increase by 1 and return to step ②; ⑤ From the first g max The solution set P( g max )=[x1( g max ), x2( g max ), x3( g max ), …, x N ( g max Randomly select a key variable vector from )] As an optimized deployment scheme for greenhouse gas sensors in urban wastewater treatment processes.
[0056] Those skilled in the art should understand that the above description of the embodiments of the present invention is only intended to illustrate the beneficial effects of the embodiments of the present invention, and is not intended to limit the embodiments of the present invention to any of the examples given.
[0057] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.
Claims
1. A method for deploying greenhouse gas sensors in an urban wastewater treatment process, characterized in that, include: The deployment of multiple candidate sensor locations is determined, and then the greenhouse gas emission concentration vectors of the candidate locations and the center location are obtained respectively. Establish a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes; A multi-objective optimization model is solved by a multi-objective genetic algorithm to obtain the output vector of the multi-objective optimization model, and the output vector is used as the optimization deployment scheme for greenhouse gas sensors in urban sewage treatment process.
2. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 1, wherein, Obtaining the greenhouse gas emission concentration vectors for candidate sites and the center site separately includes: The layout of multiple candidate points is used as the key vector: Where x is the vector of key variables, x l For the first l The layout of each candidate point is indicated by "1" for placement and "0" for no placement. l ∈[1, L]; Greenhouse gas emission concentrations at multiple candidate sites were monitored, and the greenhouse gas emission concentration vectors at the candidate sites were obtained as follows: Wherein, H(x) ,t )for t Greenhouse gas emission concentration vectors at candidate sites at given time points. H ( x l , t )for t Time of the first l Greenhouse gas emission concentrations at each candidate site; Therefore, the greenhouse gas emission concentration vector at the center point is obtained as follows: Wherein, H(y) ,t )for t Greenhouse gas emission concentration vector at the center point at any given time. H ( y l , t ) is the first l and the l+ The center point between 1 candidate point is t Greenhouse gas emission concentration at any given time.
3. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 1, wherein, Based on the greenhouse gas emission concentration vectors of candidate and central locations, a multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes is established, including: Based on the greenhouse gas emission concentration vectors of candidate sites and the center site, an optimization model for acquisition error and a sensor deployment cost optimization model are established. Based on the data acquisition error optimization model and the sensor deployment cost optimization model, the multi-objective optimization model for the deployment of greenhouse gas sensors in urban wastewater treatment processes is obtained as follows: in, E (x) represents the objective function value for optimizing the acquisition error of decision variable x. C (x) represents the objective function value for optimizing the sensor deployment cost of decision variable x.
4. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 3, wherein, The acquisition error optimization model is as follows: Where E(x) is the time interval T The total acquisition error optimization model within the system, E (x ,t )for t Optimization model for acquisition error at any given time. , x is t time E (x ,t The input vector of ) H ( x l , t ) is the first l Candidate points t Greenhouse gas emission concentration at any given time ( x l , t ) is the first l Candidate points t Estimated greenhouse gas emission concentrations at time t. , H ( y lk , t ) is the first l Each central point t Greenhouse gas emission concentration at any given time ( y l , t ) is the first l Each central point t Estimated greenhouse gas emission concentrations at time t. d ( x l-1 ,x l ) is the first l- 1 and l The distance between candidate points.
5. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 3, wherein, The sensor deployment cost optimization model is as follows: in, C (x) represents the sensor deployment cost optimization model. λ 1 and λ 2 represents the weighting coefficient. I com ( l ) represents the complexity coefficient of the candidate point.
6. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 1, wherein, The multi-objective optimization model is solved using a multi-objective genetic algorithm, and the output vector of the multi-objective optimization model is obtained as follows: Set the total number of iterations and population size for the multi-objective genetic algorithm, randomly initialize the population, and obtain the... g Population vector; The next generation of key variable vectors is generated through crossover mutation, and then the population is updated through environmental selection to obtain the next generation. g+ First generation population; Iteratively update the population until the total number of iterations is reached, and output the last generation of the population as the output vector.
7. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 6, wherein, No. g The population vector is: in, g Let P be the number of iterations. g ) is the first g Population vector, x1( g ) is the first g The first key variable vector in the algebra, x2( g ) is the first g The second key variable vector in the algebra, x3 (g ) is the first g The third key variable vector in the algebra, x N ( g ) is the first g The middle generation N A vector of key variables.
8. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 6, wherein, The next generation key variable vector is generated through crossover mutation as follows: Where, x l ( g+ 1) is the first g+ 1st generation l A vector of key variables, x SBX For the first g The middle generation l The key variable vectors are generated by simulating binary crossover. , x PM For the first g The middle generation l The key variable vector is generated by multinomial mutation of the key variable vector. , , e The cross-distribution index, m It is the distribution index of variation. u and r These are uniformly distributed random numbers.
9. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 6, wherein, No. g+ The first generation population is: Wherein, P( g+ 1) is the first g+ Generation 1 population, x i ( g+ 1) Calculate the key variable vector. , The dominance relationship is defined as follows: , E (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing the acquisition error is calculated. E (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of the acquisition error; C (x i ( g )) represents the key variable vector x i ( g The objective function value for optimizing sensor deployment costs; C (x i ( g+ 1)) represents the key variable vector x i ( g+ 1) Optimize the objective function value of sensor deployment cost. dis (·) represents the distance of congestion level. , E (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing the acquisition error is calculated. E (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing the acquisition error is calculated. C (x i+1 ( g )) represents the key variable vector x i+1 ( g The objective function value for optimizing sensor deployment costs; C (x i-1 ( g )) represents the key variable vector x i-1 ( g The objective function value for optimizing sensor deployment costs is determined by...
10. The method for deploying greenhouse gas sensors in urban wastewater treatment processes according to claim 6, wherein, Also includes: From the last generation of the population g max The solution set P( g max )=[x1( g max ), x2( g max ), x3( g max ), …, x N ( g max A key variable vector is randomly selected from the data to serve as the optimal deployment scheme for greenhouse gas sensors in the urban wastewater treatment process.