A method for tracing the source of a river pollution emergency based on an improved FA algorithm

Abstract

The present application relates to a kind of sudden pollution event of river based on the method of improved FA algorithm traceability. The method is mainly according to the diffusion law of pollution source in surface water through mathematical method to establish water pollution diffusion model, and through the method of coupling probability density to deduce the traceability method of water pollution diffusion, it is converted into water pollution traceability model;For the solution of model, improved firefly algorithm is used to solve, from the two aspects of firefly algorithm being easy to fall into local extreme value and oscillation convergence speed being slow at extreme point, division firefly subpopulation and adaptive step length method are used to strengthen the global search ability and convergence speed of firefly algorithm, so that more accurately water pollution traceability related parameter derivation is carried out, and the precision of water pollution traceability is improved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of water environment quality monitoring, and particularly relates to a method for tracing the source of a river sudden pollution event based on an improved FA algorithm. BACKGROUND

[0002] Water is the basis for human survival, and the water quality of inland rivers is related to the life, health and safety of the people, so the monitoring and tracing research on river pollution problems has important practical significance.

[0003] In recent years, water pollution events have occurred frequently, and the traditional manual investigation of polluting enterprises consumes many human resources and has a long investigation period, and there may be situations of concealment or non-reporting, so the automatic monitoring of pollution and tracing through an information system is imminent, and information support can be provided for government decision-making through information monitoring and tracing. SUMMARY

[0004] In order to better supervise the pollution discharge of enterprises and quickly locate the polluting enterprises when pollution occurs, the application provides a method for tracing the source of a river sudden pollution event based on an improved FA algorithm, which can quickly and accurately locate the polluting enterprises, thereby providing scientific support for the safety of water resources of the people.

[0005] The technical scheme adopted by the application to achieve the above-mentioned purpose is:

[0006] A method for tracing the source of a river sudden pollution event based on an improved FA algorithm, comprising the following steps:

[0007] Obtaining heavy metal pollutant information collected by a monitoring station and river information at the monitoring station;

[0008] Based on the improved firefly algorithm, the heavy metal pollutant information and the river information are used to obtain the discharge location and discharge time of the pollutant;

[0009] Based on the improved firefly algorithm, the discharge location and discharge time of the pollutant are used to obtain the discharge concentration of the pollutant;

[0010] The discharge location, discharge time and discharge concentration of the pollutant are summarized to generate a list and visualized.

[0011] The heavy metal pollutant information includes: pollutant type type, degradation coefficient of the type type pollutant in river water, pollutant concentration and current time; and the river information at the monitoring station includes: current position river flow velocity, position information of the monitoring station relative to the river and current river dispersion coefficient.

[0012] The improved firefly algorithm comprises the following steps:

[0013] 1) Construct the objective function

[0014] where, ;

[0015] 2) Set the number of fireflies m, the number of firefly subpopulations N, the light absorption coefficient γ, the maximum attraction factor , the step factor α of each firefly subpopulation, set the convergence condition according to the objective function, and randomly generate the initial position of each firefly in the solution space

[0016]

[0017] where: , represents the position of the i-th firefly in the j-th subpopulation, and d represents the dimension of the independent variable.

[0018] 3) Substitute the initial position of the firefly into the objective function , calculate the maximum fluorescence of the i-th firefly in each subpopulation

[0019] =

[0020] where, represents the fluorescence of the firefly at position ;

[0021] Calculate the relative attraction degree between fireflies in the subpopulation

[0022]

[0023] where, represents the position of the i-th firefly in the j-th subpopulation, , represents the spatial distance between the i-th and j-th fireflies in the j-th subpopulation, defined as ;

[0024] The initial position of the firefly in each subpopulation evolves, and the spatial position is updated according to the following formula

[0025]

[0026] where: , is the step attenuation coefficient, taking the value range (0.95, 1); rand is a random factor that is uniformly distributed in [0, 1], and r represents the current iteration number;

[0027] ​​4) Record the current individual data value that minimizes the objective function for each subpopulation as the optimal individual data. If the current data value and the value of the past 4 iterations have a continuous change of less than 1, then the optimal individual data is considered. Return to step 3), learn from the best individuals in other subpopulations, and update;

[0028] 5) According to the formula Adjust the step size;

[0029] 6) Iterate from step 3) to step 5) until the maximum number of iterations is reached or the objective function is minimized ahead of time.

[0030] When calculating the emission location and time of pollutants, the objective function of the improved firefly algorithm is:

[0031] min[abs(1-r)]

[0032] in:

[0033]

[0034] in, For the observed concentration series i=1,2,3,…,n, The average concentration is the observed concentration, and n represents the number of observed concentration data. For the inverse probability density series i=1,2,3,…,n; The inverse probability density average; r represents the correlation coefficient between the two series; To substitute the monitoring time of different monitoring stations into the following formula Monitoring station monitoring location A series of values ​​corresponding to the observed concentrations were obtained by combining the results.

[0035]

[0036] in, For the cross-section observed from the location of the monitoring station Judgment Pollution sources are always present The inverse probability density at the location; u is the average flow velocity of the river / canal cross section; E is the longitudinal dispersion coefficient of the river / canal; k is the degradation coefficient of pollutants.

[0037] When calculating the emission concentration of pollutants, the objective function of the improved firefly algorithm is:

[0038] in, This represents the probability density of the positive location of the pollutant.

[0039]

[0040] wherein, is the initial source strength of the instantaneous emission of the pollutant,

[0041]

[0042] wherein, is the coordinate of the pollution source position, is the time of the pollution burst, u is the average flow velocity of the river channel section, E is the longitudinal dispersion coefficient of the river channel, and k is the degradation coefficient of the pollutant.

[0043] A river burst pollution event tracing system based on an improved FA algorithm, comprising:

[0044] A data acquisition module is configured to acquire heavy metal pollutant information collected by a monitoring station and river channel information at the monitoring station.

[0045] A first firefly operation module is configured to obtain a discharge position and a discharge time of the pollutant based on an improved firefly algorithm and by using the heavy metal pollutant information and the river channel information.

[0046] A second firefly operation module is configured to obtain a discharge concentration of the pollutant based on the improved firefly algorithm and by using the discharge position and the discharge time of the pollutant.

[0047] An output module is configured to generate a list by summarizing the discharge position, the discharge time and the discharge concentration of the pollutant and to visualize the list.

[0048] A river burst pollution event tracing system based on an improved FA algorithm, comprising a memory and a processor; the memory is configured to store a computer program; the processor is configured to implement the method for tracing a river burst pollution event based on an improved FA algorithm when the computer program is executed.

[0049] A computer readable storage medium, the storage medium stores a computer program, when the computer program is executed by a processor, the method for tracing a river burst pollution event based on an improved FA algorithm is implemented.

[0050] The present application has the following advantages and benefits:

[0051] 1. Visualization. By collecting and reporting the data of the fixed monitoring system to the pollutant tracing system, the specific possible polluting enterprises can be located on the computer screen, and the entire transmission process of the pollutant can be simulated.

[0052] 2. Automation. Not limited by time and region, without manual intervention, the most possible polluting enterprises can be determined through on-site investigation, and the supervision department can be reported in the first time, which provides valuable time for preventing and controlling pollution expansion.

[0053] 3. Accuracy. By improving the glowworm algorithm, the global search ability is enhanced, and the solution speed is accelerated, and the ability to avoid local extreme value is improved. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 For the schematic diagram of the traceability system operation;

[0055] Figure 2 For the schematic diagram of the improved glowworm algorithm simulation operation. DETAILED DESCRIPTION

[0056] The application will be further described in detail below in combination with the drawings and examples.

[0057] A method for tracing the source of a river pollution emergency based on an improved FA algorithm, comprising the following steps:

[0058] Step 1: Collecting river pollutant data. The fixed pollutant monitoring platform deployed in the river channel monitors the river pollutant data. When the monitoring platform detects that the pollutant concentration exceeds the preset range, an alarm is sent. The historical concentration information of the monitoring platform related to the alarm information is summarized to obtain the data.

[0059] Step 2: River generalization. The natural river model is modeled as a series of small straight segments, and the river is converted into a one-dimensional water flow and water quality problem.

[0060] Step 3: River modeling. According to the simplified river established in step 2, a pollutant transfer model is constructed, which is a one-dimensional convection-diffusion equation model.

[0061] Step 4: Traceability modeling. According to the coupling model in step 3, a coupling probability model is established by using the forward density function and the reverse density function to construct a pollutant traceability model.

[0062] Step 5: Optimization of the original glowworm algorithm. The original glowworm algorithm is optimized to improve its tendency to fall into local minimum points and slow convergence at extreme points.

[0063] Step 6: Using the improved glowworm algorithm to iteratively solve the water pollution traceability model, the time, location, and mass concentration information related to the traceability can be obtained.

[0064] Step 7: Using the parameter information solved in step 6, the nearby enterprises are investigated to obtain the final list of suspicious enterprises.

[0065] The monitoring data of the pollutant monitoring platform is a series of real-time concentration data of pollutants monitored by the fixed monitor.

[0066] The water pollution alarm is a water pollution alarm issued when a certain water pollutant of the monitor exceeds a preset concentration range.

[0067] The historical concentration information of the monitoring platform related to the alarm information is historical concentration record information of the monitoring platform related to the alarm monitoring platform upstream and downstream.

[0068] The natural river is modeled as a series of tiny straight segments. In general, the water depth and width of an inland river are small relative to the length of the river. After the pollutant is discharged, the pollutant concentration in the water depth and width direction of the river can be uniformly distributed in a short time, so the depth and width of the river can be ignored, the river can be approximated as a straight segment, and the river pollution problem can be converted into a one-dimensional water flow quality problem.

[0069] The established river simplification construction pollutant transfer model uses a one-dimensional convection-diffusion equation to describe the migration law of the pollutant in the one-dimensional river

[0070]

[0071] C (x, t) is the average concentration of the river channel x section at time t, mg / L.

[0072] u is the average flow velocity of the river channel section, m / s.

[0073] E is the longitudinal dispersion coefficient of the river channel, / s, the dispersion coefficient is constant for the same river section; k is the degradation coefficient of the pollutant, s-1.

[0074] The Fourier transform is used to obtain the expression of C (x, t)

[0075] The pollution source position coordinate is m.

[0076] The time of the sudden pollution is s.

[0077] The initial surface source strength of the instantaneous discharge of the pollutant is g / s. .

[0078] The coupling probability model is established by using the forward density function and the reverse density function

[0079]

[0080]

[0081] The observation section is determination probability of the pollution source at time

[0082] —reversely calculate the time point;

[0083] —pollutant concentration observation time point

[0084] Fourier transform is performed on the same

[0085]

[0086] The concentration relationship of can be obtained

[0087] The optimization of the original glowworm algorithm comprises the following steps:

[0088] Step 1: Improve the population mechanism and divide it into N sub-populations.

[0089] Different initial parameters are set for each population, and iteration is performed respectively, so as to realize more comprehensive global search capability, and the position of the optimal individual in each sub-population is recorded in each iteration, and when it is found that the optimal individual of a sub-population is not updated for many times, it is indicated that the sub-population falls into a local optimal solution, at this time, the individual learns from the optimal individual of other sub-populations, and jumps out of the local optimal solution.

[0090] Step 2: Improve the step length to an adaptive variable step length mechanism.

[0091] The step length of the original algorithm is improved to be larger in the initial stage of the algorithm and smaller in the later stage, so as to achieve fast initial search speed and better global search, and gradually reduce the step length with the increase of the number of iterations, so as to avoid oscillation of the algorithm at extreme points and faster convergence. The adaptive step length adjustment formula is

[0092]

[0093] —step length attenuation coefficient, value range (0.95, 1)

[0094] The improved glowworm algorithm is used to iteratively solve the water pollution tracing model, comprising the following steps:

[0095] Step 1: According to the alarm data of the fixed monitoring platform, the initial value of the measured concentration of the pollutant concentration at time t is set , time , which is substituted into the expression of the correlation coefficient r

[0096]

[0097] In the formula: - observed concentration average

[0098] - inverse calculated concentration average

[0099] The closer the correlation coefficient is to 1, the closer it is to the location of the pollutant discharge , discharge time

[0100] Step 2: Use the improved firefly algorithm to find the minimum value of the objective function 1-r, and find the objective function ,

[0101] Step 3: Set the maximum number of generations of m firefly subpopulations, population size, dimension, attraction degree, perturbation step factor, medium light absorption coefficient, step attenuation factor parameters.

[0102] Step 4: After every 4 iterations, judge whether the numerical value of all subpopulation optimal individuals has changed, if the optimal individual of a subpopulation has not changed in the past 4 times, it means that the subpopulation has fallen into a local extremum, adjust the optimal individual of the population to learn from the optimal individual of other subpopulations, jump out of the local extremum, and continue iteration.

[0103] Step 5: When the iteration reaches the maximum iteration number or the objective function obtains the optimal solution in advance, the algorithm ends.

[0104] The method for tracing the source of a river pollution event based on an improved FA algorithm according to claim 1, characterized in that,

[0105] The suspicious enterprise pollution source investigation is to investigate the enterprises related to the emission of the relevant pollutants in the area with R as the radius (R<5000m) according to the pollution source position, time and concentration size obtained by inversion, combined with the posterior probability obtained from the historical emission of the enterprise, to find the most possible pollution enterprise.

[0106] As shown in Figure 1 , the river pollution tracing system is built. When river pollution occurs, the fixed monitoring station deployed in the river channel monitors that the pollutant concentration exceeds the threshold value, reports the alarm data to the tracing system platform to start the processing process including the following steps:

[0107] Step 1: A sudden heavy metal pollution is monitored at a certain place in the river channel, the monitoring station of the river channel monitors the pollution and reports to the tracing system.

[0108] Step 2: The tracing system starts to collect heavy metal pollution related information and river channel related information at the monitoring station, and records the information to the corresponding database.

[0109] Step 3: Start calling the tracing model to calculate the emission location x and emission time t of the pollutant by improving the firefly algorithm.

[0110] Step 4: According to the x and t calculated in step 3, use the objective function constructed by the pollutant emission model to solve the pollutant emission concentration.

[0111] Step 5: Summarize the simulation results to obtain a list of all information related to the pollutant, and report the information to the relevant person in charge.

[0112] As shown in Figure 2 , the simulation algorithm is run. The pollutant concentration value m of the current river section, the emission location x of the pollutant, and the emission time t are calculated. The following steps are included:

[0113] Step 1: Construct the objective function for solving the emission location and emission distance = min [abs (1-r)], where

[0114]

[0115] Observe the average concentration

[0116] Calculate the average concentration in reverse

[0117]

[0118] The probability of the pollution source at when the observation section is determined

[0119] The simulation algorithm data is initialized. For example, the information of the sewage outlet, the basic information of the river, and the basic information of the map, etc.

[0120] Step 2: Construct the objective function for solving the emission concentration , where

[0121] Step 3: Substitute the objective functions , into the improved firefly algorithm respectively to start solving, which is divided into the following steps:

[0122] 1). Set the number of fireflies m and the number of firefly subgroups n. Set the light absorption coefficient γ, the maximum attraction factor , and the step factor α for each firefly subgroup. According to the objective function, set the convergence condition (the value of the objective function tends to a fixed value after multiple iterations), and randomly generate the initial position ​

[0123]

[0124] wherein: represents the position of the i-th firefly in the j-th sub-population, and d represents the dimension of the independent variable.

[0125] 2). Substitute the initial position of the firefly into the objective function, and calculate the maximum fluorescence of the i-th firefly in each sub-population

[0126] =

[0127] wherein: represents the fluorescence of the firefly at position

[0128] 3). Calculate the relative attraction degree between the fireflies in the sub-population

[0129]

[0130] wherein: represents the position of the i-th firefly in the j-th sub-population, , represents the spatial distance between the i-th and the j-th firefly in the j-th sub-population, and is defined as .

[0131] The initial position of the firefly in each sub-population evolves, and the spatial position is updated according to the following formula

[0132]

[0133] wherein: , is a step attenuation coefficient, and the value range is (0.95, 1); rand is a random factor obeying uniform distribution in [0, 1], and t represents the current iteration number;

[0134] 4) Record the current individual data value of each sub-population that minimizes the objective function as the optimal individual data, and if the continuous change amount of the current data value and the past four iteration values is less than , return to step 3) to learn from the optimal individual of the other sub-population and update;

[0135] 5) Adjust the step size according to the formula

[0136] 6). If the maximum iteration number is reached or the optimal solution of the objective function is obtained (the minimum value is obtained and convergence is achieved), the algorithm ends.​​​​​

Claims

1. A method for tracing the source of a river pollution emergency event based on an improved FA algorithm, characterized in that, The method comprises the following steps: obtain heavy metal pollutant information collected by a monitoring station and river information at the monitoring station; based on the improved firefly algorithm, use the heavy metal pollutant information and the river information to obtain the discharge location and discharge time of the pollutant; based on the improved firefly algorithm, use the discharge location and discharge time of the pollutant to obtain the discharge concentration of the pollutant; aggregate the discharge location, discharge time and discharge concentration of the pollutant to generate a list and visualize it; improve the original firefly algorithm, including the following steps: Step 1: Improve the population mechanism and divide it into N sub-populations: Set different initial parameters for each population, iterate separately, and record the optimal individual position in each sub-population in each iteration. When it is found that the optimal individual of a sub-population has not been updated for a long time, it means that the sub-population has fallen into a local optimal solution. At this time, the individual learns from the optimal individual of other sub-populations to jump out of the local optimal solution. Step 2: Improve the step length to an adaptive variable step length mechanism: Original algorithm step size The improvement is to use larger step size in the initial stage and smaller step size in the later stage. The adaptive step size adjustment formula is: ； wherein is a step attenuation coefficient, taking a value in the range (0.95, 1); When calculating the discharge location and discharge time of the pollutant, the objective function of the improved firefly algorithm is: min[abs(1-r)]; Wherein: ； wherein, is a series of observed concentrations i = 1, 2, 3,..., n, is the mean of the observed concentrations, n represents the number of observed concentration data; is a series of inverse probability densities i = 1, 2, 3,..., n; is the mean of the inverse probability densities; r represents the correlation coefficient between the two series; is a series of values corresponding to the observed concentrations obtained by combining the different monitoring station monitoring times , the monitoring station monitoring positions ​ ； wherein is observed by the monitoring station determination the instantaneous backward probability density of the pollution source at time; u is the average flow velocity of the river section, E is the longitudinal dispersion coefficient of the river, and k is the degradation coefficient of the pollution material.

2. The method for tracing the source of a river pollution emergency event based on the improved FA algorithm according to claim 1, characterized in that, The heavy metal pollutant information includes: pollutant type type, degradation coefficient of type type pollutant in river water, pollutant concentration, current time; The river information at the monitoring station includes: current position river flow velocity, monitoring station relative to the river position information, current river dispersion coefficient.

3. The method for tracing the source of a river pollution emergency event based on the improved FA algorithm according to claim 1, characterized in that, The improved firefly algorithm includes the following steps: 1) Constructing the objective function : wherein ; 2) Set the number of fireflies as m, the number of sub-population of fireflies as N, set the light intensity absorption coefficient γ, the maximum attraction factor , the step factor α for each sub-population of fireflies, set the convergence condition according to the objective function, and randomly generate the initial position of each firefly in the solution space: ； wherein: , denotes the position of the i-th firefly in the subpopulation j, d denotes the dimension of the argument; 3) Set the initial position of the firefly Substitute the objective function Calculate the maximum fluorescence of the i-th firefly in each sub-population: = ； wherein representing the position of the firefly's fluorescence; Calculate the relative attraction degree between fireflies in the sub-population: ； wherein, represents the position of the i-th firefly in the sub-population j, only fireflies, , represents the spatial distance between the i-th and the j-th firefly in the sub-population j, defined as only fireflies, ; Evolve the starting position of the firefly individual in each sub-population, and update the spatial position according to the following formula: ； Where: rand is a random factor uniformly distributed in [0, 1], and r represents the current iteration number; 4) record the current individual data value of each sub-population that minimizes the objective function as the optimal individual data, if the current data value and the past 4 iteration values change continuously less than , return to step 3) to learn from the optimal individual of other sub-populations and update; 5) according to the formula adjustment step; 6) Iterate steps 3) to 5) until the maximum iteration number is reached or the objective function is minimized in advance.

4. The method for tracing the source of a river pollution emergency event based on an improved FA algorithm according to claim 1, characterized in that, When calculating the discharge concentration of the pollutant, the objective function of the improved firefly algorithm is: ； wherein, Pcontam+is the contaminant forward position probability density; ； wherein is the initial source strength of the instantaneous emission pollutant, ； wherein, is the pollution source location coordinate, is the pollution burst occurrence time, is the river x section t time average concentration.

5. A river pollution emergency event tracing system based on an improved FA algorithm, used to implement the method of tracing a river pollution emergency event based on an improved FA algorithm in claim 1, characterized in that, including: A data acquisition module for obtaining heavy metal pollutant information collected by a monitoring station and river information at the monitoring station; A first firefly calculation module for obtaining the discharge location and discharge time of the pollutant based on the improved firefly algorithm and using the heavy metal pollutant information and the river information; A second firefly calculation module for obtaining the discharge concentration of the pollutant based on the improved firefly algorithm and using the discharge location and discharge time of the pollutant; An output module for aggregating the discharge location, discharge time and discharge concentration of the pollutant to generate a list and visualize it.

6. A river pollution emergency event tracing system based on an improved FA algorithm, characterized in that, A memory and a processor; the memory is used to store a computer program; the processor is used to realize the method of tracing the source of a river sudden pollution event based on the improved FA algorithm as claimed in any one of claims 1-4 when executing the computer program.

7. A computer-readable storage medium, characterized in that, The storage medium has a computer program stored thereon, and when the computer program is executed by the processor, the method of tracing the source of a river sudden pollution event based on the improved FA algorithm as claimed in any one of claims 1-4 is realized.

Images