• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    The present disclosure relates to an ant colony algorithm-based method for optimal design of an artificial ground freezing (AGF) method under seepage conditions, including the steps of: determining the number of freezing pipes, radii of the freezing 5 pipes, a type of a tunnel under construction, a type of a freezing liquid, ground parameters and groundwater conditions, determining a layout equation of the freezing pipes under seepage conditions according to the data, building a coupled thermo-hydraulic finite element model according to the layout equation of the freezing pipes in conjunction with given boundary conditions and initial conditions, and 10 establishing a mapping relationship between the optimal layout of the freezing pipes under seepage conditions and the ant colony algorithm using an ant colony optimization algorithm in conjunction with the coupled thermo-hydraulic finite element model, and obtaining optimized layout results of the freezing pipes under different seepage conditions. Compared with the prior art, the present disclosure may provide the 15 optimal design for the layout of the freezing pipes in the presence of seepage rapidly and efficiently by a combination of the ant colony algorithm and the finite element model, allowing for formation of a frozen wall regular in shape and satisfactory in engineering quality with the least time while not increasing the number of the freezing pipes. 20 [Fig l] 14
    公开号:NL2027278A
    申请号:NL2027278
    申请日:2021-01-06
    公开日:2021-08-30
    发明作者:Li Zeyao；Wang Chuanhe；Tang Yiqun；Zhao Wenqiang；Ren Junjie；Zhou Jie
    申请人:Univ Tongji；
    IPC主号:
    专利说明:

    ANT COLONY ALGORITHM-BASED METHOD FOR OPTIMAL DESIGN OFARTIFICIAL GROUND FREEZING METHOD UNDER SEEPAGECONDITIONS
    TECHNICAL FIELD The present disclosure relates to the field of construction by artificial ground freezing (AGF) method, and in particular, to an ant colony algorithm-based method for optimal design of an AGF method under seepage conditions.
    BACKGROUND During the construction of underground engineering in a soft soil with a high groundwater level or other poor engineering geological conditions, the AGF method has been widely used. It can be used in excavation of underground engineering for its advantages such as reversibility, environmental friendliness, and wide application scope. It is common to adopt a uniform layout of freezing pipes. However, during freezing, underground seepage may occur, which often leads to nonuniform thickness of a frozen wall formed around the uniformly arranged freezing pipes and the freezing time gets extended. In some strata having high-velocity groundwater, it may be even hard to form a high strength frozen closed body around the uniformly arranged freezing pipes. Thus, the engineering cost may be increased, and the engineering efficiency may be reduced. Therefore, optimal design of layout of freezing pipes under seepage conditions is of great engineering significance.
    SUMMARY An objective of the present disclosure is to overcome the above-described defects in the prior art and practice, and provide an ant colony algorithm-based method for optimal design of an AGF method under seepage conditions. The method requires less time for forming a frozen wall by changing the positions of distributed freezing pipes without increasing the total number. Besides, the frozen wall has a more uniform and regular shape, and improved quality. In this way, the engineering benefits is increased, and energy conservation and emission reduction are implemented. The objective of the present disclosure may be achieved by the following technical solution: An ant colony algorithm-based method for optimal design of an AGF method under seepage conditions includes the following steps: S1, determining the number of freezing pipes, radii of the freezing pipes, a type of a tunnel under construction, a desired frozen wall thickness, a type of a freezing liquid, ground parameters and groundwater conditions according to construction engineering 1 requirements and an investigation report; S2, determining a layout equation of the freezing pipes under seepage conditions according to the data in step S1; S3, building a coupled thermo-hydraulic finite element model according to the layout equation of the freezing pipes in conjunction with given boundary conditions and initial conditions, and obtaining temperature fields under different layouts of the freezing pipes and different seepage conditions; and S4, establishing a mapping relationship between the optimal layout of the freezing pipes under seepage conditions and the ant colony algorithm using an ant colony optimization algorithm in conjunction with the coupled thermo-hydraulic finite element model, and obtaining optimized layout results of the freezing pipes under different seepage conditions.
    Further, in the step S4, the mapping relationship between the optimal layout of the freezing pipes under seepage conditions and the ant colony algorithm is as follows: with each ant in an ant colony representing a layout style of the freezing pipes and different parameters in the layout equation representing different nodes, when an ant travels from one node to another node, it is representative of a random combination of different parameters to give a different specific layout equation of the freezing pipes; each combination of styles is a different layout equation determined by a random combination of different parameters in the layout equation; a frozen wall quality function constitutes an objective function of the ant colony algorithm; a better solution is chosen as an optimal solution for the ant colony from a population based on the objective function, and therefore, the optimal ant in the ant colony characterizes an optimized configuration result.
    Further, in the step S3, the coupled thermo-hydraulic finite element model may be expressed as: rl T.(R).d, 2, Ay, Aas Pi, Poh | 0 pL CCC (DP KAP, 1, where #, denotes a temperature in soil, while /, an outer wall temperature of the freezing pipe, fo an initial ground temperature, 7 freezing time, (R) layout parameters for the freezing pipes, ‚7 a diameter of the freezing pipe, 4, a thermal conductivity coefficient of unfrozen soil, A, a thermal conductivity coefficient of frozen soil, A, a thermal conductivity coefficient of water, p, a density of unfrozen soil, Pr 4 density of frozen soil, p, a density of water, }; a frozen wall thickness, (; specific heat of unfrozen soil, C, specific heat of frozen soil, C; specific heat of 2 water, y latent heat released by soil mass per unit volume during freezing of the soil mass, (D) parameters of a freezing material, V, a seepage velocity, x a permeability coefficient, AP a seepage water pressure, and x a dynamic viscosity coefficient of water.
    Further, the freezing material may be a brine, and the parameters of the freezing material may include: a brine flow velocity V,, a brine pressure AP, a dynamic viscosity coefficient £, of the brine, a brine density 9,, and a coefficient of convective heat transfer #, between the brine and the freezing pipe wall.
    Further, the step 4 may specifically include: S41, initializing parameters of the ant colony optimization algorithm, and reading temperature field quality assessment data, as inputs to the ant colony algorithm, into a heuristic value matrix, where the parameters of the algorithm may include the number n of ants, a pheromone evaporation rate # , a standard deviation © , etc.; and algorithm termination criterion parameters may include a maximum number I of iterations, the longest running time T, and an expected objective convergence value G; S42, initializing ant memory, setting 7; to an arbitrary small constant C, setting Az, to 0, and placing n ants on several nodes; S43, for each ant k (k=1, 2, ..., n), calculating a selection probability P, of traveling from the current node / to next node j: S44, calculating an objective function value corresponding to each ant in conjunction with the coupled thermo-hydraulic finite element model, recording the current best solution, and ranking such values from worst to best; S45, updating pheromone: calculating A7, and 7," tor each ant and each route ij, where AT, denotes the strength of a pheromone trail left by the ant when passing by; and Ti” denotes the updated pheromone trail strength in the connecting line of the route ij after the ant passing by; S46, comparing a solution constructed by the current ant with the best solution, updating the solution, if better, to the best solution, and determining whether all ants complete solution construction; if no, selecting next ant as the current ant; otherwise, going to step S48; S47, determining whether a termination criterion is met; if yes, going to step S48; otherwise, going to step S41; and S48, outputting a result, and obtaining an optimal layout style of the freezing pipes under the layout equation of the freezing pipes.
    Further, in the step S43, the selection probability Z, may be calculated according to a formula expressed as: 2 Ti where 7, denotes the pheromone trail strength in the connecting line of the route ij; the pheromone trail strength 7, may be initially the same in different routes and set to an arbitrary small constant; and size (DD, ) denotes the number of nodes obtained by discretizing each parameter in the layout equation of the freezing pipes at certain intervals within a certain value range.
    Further, in the step S45, the updated pheromone may be calculated according to a formula expressed as: (RJ Len on 553) ’ | r= (1-p)1, + par, where n denotes the number of ants; R, denotes the objective function value corresponding to each ant in the current iteration obtained in step S44, and all objective function values are successively assigned to n ants from worst to best, from 1 ton; © denotes a standard deviation, which, as a scale parameter, is an input parameter of the ant colony algorithm; mean is a position parameter of the node; and © denotes a pheromone evaporation rate.
    Further, in the step S47, the termination criterion is one or more of a maximum number I of iterations, the longest running time T, and an expected objective convergence value G; and when any termination criterion is met, the looping of the algorithm ends.
    Compared with the prior art, the present disclosure may have the following advantages: Due to the presence of seepage, a frozen wall formed in existing uniform layout style of freezing pipes is irregular and may take longer time than in the absence of seepage, and a closed frozen wall cannot be formed even after a long freezing time.
    Consequently, the progress and benefits of construction are greatly reduced.
    In consideration of the influence of seepage on the formation of the frozen wall, the present disclosure may provide the optimal design for the layout of the freezing pipes in the presence of seepage rapidly and efficiently by a combination of the ant colony algorithm and the finite element model.
    With the optimal design for the layout of the freezing pipes provided in the present disclosure, a frozen wall regular in shape and satisfactory in engineering quality can be formed with the least time while not 4 increasing the number of the freezing pipes, thus effectively reducing the construction cost and improving the benefits of construction.
    BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a flow chart according to an embodiment of the present disclosure.
    FIG. 2 is a schematic diagram of a layout equation of freezing pipes according to an embodiment of the present disclosure.
    FIG. 3 is a schematic diagram of an ant colony algorithm according to an embodiment of the present disclosure.
    DETAILED DESCRIPTION The present disclosure will be described in detail below in conjunction with the accompanying drawings and a specific embodiment. The embodiment is carried out on the premise of the technical solution of the present disclosure. Details of the embodiment and a specific operation process are given, but the protection scope of the present disclosure is not limited to the following embodiment.
    As shown in FIG. 1, this embodiment provides an ant colony algorithm-based method for optimal design of an AGF method under seepage conditions. According to the method, a coupled thermo-hydraulic finite element model is built by a combination of a layout equation of freezing pipes, a freezing temperature field system, a seepage system and a freezing pipe heat exchange system. With variable parameters in the layout equation of freezing pipes as inputs to the ant colony algorithm, temperature field outputs and frozen wall formation time under different combinations of parameters are analyzed using the coupled thermo-hydraulic finite element model, and an optimization degree of each combination of parameters is assessed on this basis. Finally, a relatively optimal design under the layout equation of freezing pipes obtained after multiple iterations is output according to the positive feedback mechanism of the ant colony algorithm.
    Firstly, the critical point of this embodiment lies in quantitative analysis of effects of seepage on temperature field and frozen wall formation under uniform layout of freezing pipes. For this purpose, this embodiment permits building of the coupled thermo-hydraulic finite element model to obtain, under given boundary and geometry conditions, temperature fields and morphologies of the formed frozen wall under different seepage conditions. Secondly, on the basis of the known effects, a layout optimization equation of freezing pipes is built, so that specific positions of a group of freezing pipes with a fixed number can be controlled. Then, variable parameters in the equation are input to the ant colony algorithm. A value range and a value interval of 5 each parameter are determined, and the number of nodes of each different parameter is obtained.
    As shown in FIG. 3, a1-an are different value points of one parameter, and then N (N=m;*m;*ms...*m,) combinations of different parameters can be obtained, as “different routes of ants searching for food”. Subsequently, the temperature field under this combination of parameters and the time needed for forming the qualified frozen wall are analyzed according to the previously used coupled thermo-hydraulic finite element model, and the optimization effect under this combination of parameters is assessed on this basis.
    The core of building of the ant colony algorithm lies in building and updating of a “pheromone trail strength” equation.
    At the initial step of the algorithm, the pheromone trail strength is the same in each route, and therefore, an “ant” may select different routes with the same probability.
    After calculating once, the “pheromone trail strength” equation 1s updated according to assessment results obtained by the coupled thermo-hydraulic finite element model, so that the “pheromone” becomes stronger in a “route”, i.e., a combination of parameters with a better optimization result, and weaker in a “route”, 1.e., a combination of parameters with a poor optimization result.
    As a result, when calculating next time, “ants” will search for food along the “route” with the relatively better optimization effect, that is, the combination of parameters with better optimization is chosen, forming a “positive feedback mechanism” of the ant colony algorithm.
    In the ant colony algorithm, the maximum number of loops and no degradation (i.e., all found solutions are the same) are taken as criteria for determining whether looping continues, and finally, the relatively optimal combination of parameters, namely the relatively optimal layout of freezing pipes under the layout equation of freezing pipes is output.
    This embodiment may specifically include the following steps: Step 1, determine a layout equation of freezing pipes under seepage conditions, taking circular tunnel construction for example: R =R,+6R(6),i=12,.,N A vd) 4,9 <7 SR(8,) = 7 2 d,.0 >= 2 0 - hy =1 | 20 0, +060 _,i>1 6
    30-00“)
    2.G(j.m0) i . 1 za) G(x, u,0)= Shee 20 where as shown in FIG. 2, R denotes a radius of a tunnel formed by freezing pipes, while R, an original uniform distribution radius, 0R(6,) a correction to the original uniform distribution radius, &, a distribution angle of freezing pipes, and d,, d,, * and 9 variable parameters used in the correction. The plane positions and relative position relationship of freezing pipes are adjusted by controlling the parameters d,, d,, ¥ and © . In this example, the original uniform distribution radius R, of freezing pipes is 2.75 m.
    Step 2, build a coupled thermo-hydraulic finite element model in the presence of seepage with a combination of 4 parameters d,, d,, ' and 9 in the layout equation and given conditions in the construction example, which is specifically expressed as: le A | 0 Pi CC WD) VK AP kt, | where t, denotes a temperature in soil, °C, while #, an outer wall temperature of the freezing pipe, °C, f, initial ground temperature, °C, T freezing time, s, (R) layout parameters for the freezing pipes, determined by parameters required in a position control equation in step 1, specifically including d,, d,, * and 9, 4 a diameter of the freezing pipe, m, A, a thermal conductivity coefficient of unfrozen soil, W/(m-K), A, a thermal conductivity coefficient of frozen soil, W/(m-K), A, a thermal conductivity coefficient of water, W/mK), p‚ a density of unfrozen soil, kg/m’, p, a density of frozen soil, kg/m’, ©, a density of water, kg/m’, h a frozen wall thickness, m, C, specific heat of unfrozen soil, kJ (kg K), C, specific heat of frozen soil, kJ (kg'K), C specific heat of water, kJ (kg'K), w latent heat released by soil mass per unit volume during freezing of the soil 7 mass, J/m’, (D) parameters of a freezing material, where a brine is used as a freezing liquid in this example, and then specific parameters range from a brine flow velocity V,, m/s, a brine pressure AP, pa, a dynamic viscosity coefficient 4, of the brine, kg/(m's), a brine density p,, kg/m’, a coefficient #, of convective heat transfer between the brine and the freezing pipe wall, W/(m2-K).
    V‚ a seepage velocity, m/s, K a permeability coefficient, m/s, AP, a seepage water pressure, pa, 44, a dynamic viscosity coefficient of water, kg/(m-s).
    Step 3, perform optimization using an ant colony optimization algorithm based on step 1 and step 2, and establish a mapping relationship between the optimal layout of the freezing pipes under seepage conditions and the ant colony algorithm to obtain optimized layout results of the freezing pipes under different seepage conditions. In this embodiment, there are 4 parameters in the layout equation. Let the value range of # be [2,8] with a step length of 0.25, 25 nodes are obtained; let the value range of © be [2.75,10] with a step length of 0.25, 30 nodes are obtained; and let the value range of d, and d, be [0,0.5] with a step length of 0.025, 21 nodes are obtained for each. Then, there are 21x21x25x30=330750 combinations of different parameters in the ant colony algorithm in this example, hence 330750 different routes (layout styles of freezing pipes).
    The step 3 may specifically include the following steps: Step 3.1, initialize different algorithm parameters, including the number n of ants, a pheromone evaporation rate © , a standard deviation © , etc, and algorithm termination criterion parameters, including a maximum number I of iterations, the longest running time T, and an expected objective convergence value G; and read temperature field quality assessment data, as inputs to the ant colony algorithm, into a heuristic value matrix.
    Step 3.2, initialize ant memory, set 7, to an arbitrary small constant C, set A7, to 0, and place n ants on several nodes.
    Step 3.3, for each ant k (k=1, 2, ..., n), calculate a selection probability £, of traveling from the current node / to next node J , the probability being calculated according to a formula expressed as: Do Tx where 7, denotes the pheromone trail strength in the connecting line of the route 8 ij; the pheromone trail strength 71s initially the same in different routes and set to an arbitrary small constant C. As shown in FIG. 3, each node corresponds to a value of a different parameter in the layout equation, and parameters of different values are combined to obtain different routes, namely different layout equations composed of different parameters. Let (rj) denote the ith node in the rth column and (r+1,j) the jth node in the (r+1)th layer, the connecting line of the route ij is a line connecting the node (r,i) and the node (r+1,)).
    size (1), ) denotes the number of nodes obtained by discretizing each parameter in the layout equation of the freezing pipes at certain intervals within a certain value range.
    In this embodiment, size ( D, ) has the values of 21, 25, and 30.
    Step 3.4, calculate an objective function value corresponding to each ant in conjunction with the finite element model built in step 2, record the current best solution, and rank such values from worst to best.
    Step 3.5, update pheromone: calculate A7, and 7 for each ant and each route ij: /R 1 Ln ti =(1-p)7, + par, where n denotes the number of ants; A, denotes the objective function value corresponding to each ant in the current iteration obtained in step 3.4, and all objective function values are successively assigned to n ants from worst to best, from 1 ton. The ant colony algorithm used in the present disclosure can be extended to continuous thresholds. Based on Gaussian distribution, © denotes a standard deviation, which, as a scale parameter, is an input parameter of the ant colony algorithm; mean is a position parameter of the node; and # denotes a pheromone evaporation rate.
    Step 3.6, compare a solution constructed by the current ant with the best solution, update the solution, if better, to the best solution, and determine whether all ants complete solution construction; if no, select next ant as the current ant; otherwise, go to step 3.7.
    Step 3.7, determine whether a termination criterion is met. The termination criterion may be selected from one or more of the algorithm termination criteria in step
    3.1: the maximum number I of iterations, the longest running time T, the expected objective convergence value G, etc.; and when any termination criterion is met, the looping of the algorithm ends. Otherwise, go to step 3.1.
    Step 3.8, output a result, and obtain an optimal layout style of the freezing pipes 9 under the layout equation of the freezing pipes.
    Checking calculations are carried out by the above method, and the optimization results of the freezing time for forming a 1.5 m thick frozen wall under different seepage velocities are obtained based on given construction conditions, as shown in table 1. Table 1 Optimization Results of Freezing Time
    1.2 Seepage Velocity (m/d) 05 075 1 5 Uniform layout of freezing 21 27 53 — Frozen wall formation pipes time (d) Optimal layout of freezing 9.6 10.5 103 pipes 7 0 3 5 In the table, for the convenience of expressing the seepage velocity, “d” in the table denotes a day. ‘—” denotes that no frozen wall can be formed.
    As can be seen from the above table, under the same seepage velocity, the freezing time needed under the optimal layout of freezing pipes obtained by the method in the present disclosure is significantly reduced over the traditional uniform layout. Under an increased seepage velocity, the freezing time for the traditional uniform layout is significantly increased, and the increasing of the freezing time under the optimal layout of freezing pipes obtained by the method in the present disclosure is obviously slowed down. Apparently, the position optimization effect of freezing pipes in the presence of seepage achieved by this method is better.
    The foregoing are detailed descriptions of the preferred specific embodiment of the present disclosure. It will be understood that a person of ordinary skill in the art can make various modifications and variations according to the concept of the present disclosure without creative efforts. Therefore, all technical solutions that a person skilled in the art can arrive at based on the prior art through logical analysis, deduction, or limited experiments according to the concept of the present disclosure shall fall within the protection scope defined by the appended claims.
    10
    权利要求:
    Claims (8)
    [1]
    An ant colony algorithm based method for an optimal design of an artificial soil freezing (AGF) process under seepage conditions, comprising the following steps: S1, determining the number of freezing tubes, radii of the freezing tubes, a type of a tunnel under construction, a desired frozen wall thickness, a type of a freezing liquid, soil parameters and groundwater conditions according to construction-technical requirements and a research report; S2, determining a layout comparison of the freezing tubes under seepage conditions according to the data in step S1; S3, building a coupled thermo-hydraulic finite element model according to the freezing tube layout equation together with given boundary conditions and initial conditions, and obtaining temperature fields under different freezing tube layouts and different seepage conditions; and S4, building an assignment relationship between the optimal layout of the freezing tubes under seepage conditions and the ant colony algorithm using an ant colony optimization algorithm together with the coupled thermo-hydraulic finite element model, and obtaining optimized layout results of the freezing tubes under different seepage conditions.
    [2]
    An ant colony algorithm based method for an optimal design of a fresh produce method under seepage conditions according to claim 1, wherein in the step S4 the assignment relationship between the optimal layout of the freezing tubes under seepage conditions and the ant colony algorithm is as follows: when each ant in an ant colony represents a layout style of the freezing tubes and different parameters in the layout equation represent different nodes, when an ant moves from one node to another node, this is representative of a random combination of different parameters to create a different specific layout equation of to give the freezing tubes; each combination of styles is a different layout comparison determined by a random combination of different parameters in the layout comparison; a frozen wall quality function constitutes an objective function of the ant colony algorithm; a better solution is chosen as an optimal solution for the ant colony from a population based on the objective function, and therefore the optimal ant in the ant colony characterizes an optimized configuration result.
    [3]
    An ant colony algorithm based method for an optimal design of an AGF method under seepage conditions according to claim 1, wherein in the step S3 the coupled thermo-hydraulic finite element model is expressed as: 40 a © RA 2 Bo, Bo po 0 Lp, > Cs C,, C, 2 Y, (0), V, 2 K, AP, HH, denoting herein: # a temperature in the soil, £, an outer wall temperature of the freezing tube, /, an initial ground temperature, . freezing time, (R) layout parameters for the freezing tubes, ‚7 a diameter of the freezing tube, A a thermal conductivity coefficient of unfrozen soil, A, a thermal conductivity coefficient of frozen soil, A a thermal conductivity coefficient of water, ©, a density of unfrozen soil, ©, a density of frozen ground, ©, a density of water, J; a frozen wall thickness, specific heat of unfrozen soil, C, specific heat of frozen soil, C, specific heat of water, y latent heat released by ground mass per unit volume during freezing of the ground mass, (D) parameters of a freezing material , V, a seepage rate, z a permeability coefficient, A, a seepage water pressure, and £4, a dynamic viscosity coefficient of water.
    [4]
    An ant colony algorithm based method for an optimal design of a fresh produce process under seepage conditions according to claim 3, wherein the freezing material is a brine, and the parameters of the freezing material vary from a brine flow rate to a brine pressure AP, a dynamic viscosity coefficient 4 , of the brine, a brine density P, , and a convective heat transfer coefficient h, between the brine and the freezing tube wall.
    [5]
    The ant colony algorithm based method for an optimal design of a fresh produce method under seepage conditions according to claim 1, wherein the step S4 specifically comprises: S41, initializing parameters of the ant colony optimization algorithm; S42, initializing the memory of ants, and placing n ants at various nodes; S43, for each ant k (k=1, 2, ..., n), calculating a selection probability F 1 of moving from the current node I to the next node J ; S44, calculating an objective function value corresponding to each ant along with the coupled thermo-hydraulic finite element model, registering the current best solution, and classifying such values from worst to best;
    $45, updating pheromone: calculating AT, and 7. for each ant and each route ij, where AT, denotes the strength of a pheromone trail left by the ant as it passes; and 7," denotes the updated pheromone trace strength in the connecting line of the route ij after the ant has passed; S486, comparing a solution constructed by the actual ant with the best solution, updating the solution, if better, to the best solution, and determining whether all ants complete a solution construct; if not, then selecting the next ant as the current ant; otherwise go to step S48; S47, determining whether a termination criterion is met; if so, to go to step S48, otherwise go to step S41: and S48, outputting a result, and obtaining an optimal layout style of the freezing tubes under the layout comparison of the freezing tubes.
    [6]
    An ant colony algorithm based method for an optimal design of a fresh produce method under seepage conditions according to claim 5, wherein in step S43 the selection probability F 1 is calculated according to a formula expressed as: _ Ti size( £) ;) Do Ta where 7, denotes the pheromone trace strength in the connecting line of the route ij; the pheromone trace strength 7, is initially the same in different routes and is set to an arbitrary small constant; and size (D) denotes the number of nodes obtained by discretizing each parameter in the freeze tube layout equation over certain intervals within a certain range of values.
    [7]
    An ant colony algorithm based method for an optimal design of a fresh produce method under seepage conditions according to claim 5, wherein in step S45 the updated pheromone is calculated according to a formula expressed as: {i-mean 2 | 1 ey At, = (2 —=—e 1st v 2 m—1) oN27 1 7, =(1-p)7; +PAT,
    where n denotes the number of ants; R, denotes the objective function value corresponding to each ant in the current iteration obtained in step S44, and all objective function values are sequentially assigned to n ants from worst to best, from 1 to n; © denotes a standard deviation which, as a scale parameter, is an input parameter of the ant colony algorithm; mean is a position parameter of the node; and p denotes a pheromone evaporation rate.
    [8]
    An ant colony algorithm based method for an optimal design of a fresh produce method under seepage conditions according to claim 5, wherein in the step S47 the termination criterion is one or more of a maximum number | iterations, longest running time T, and an expected objective convergence value G; and when any termination criteria is met, the looping of the algorithm ends.
    类似技术:
    公开号 | 公开日 | 专利标题
    NL2027278A|2021-08-30|Ant colony algorithm-based method for optimal design of artificial ground freezing method under seepage conditions
    WO2017088328A1|2017-06-01|Surface water depth information based ground irrigation control method
    Yuan et al.2020|The performance of six neural-evolutionary classification techniques combined with multi-layer perception in two-layered cohesive slope stability analysis and failure recognition
    Rao et al.2003|Optimal groundwater management in deltaic regions using simulated annealing and neural networks
    CN109117562A|2019-01-01|Intelligent temperature control dam and temperature adjusting method
    Shigidi et al.2003|Parameter estimation in groundwater hydrology using artificial neural networks
    CN106970370A|2017-07-21|Radar Multi Target tracking optimization method based on chaotic neural network
    Benyamin et al.2021|Discrete farmland fertility optimization algorithm with metropolis acceptance criterion for traveling salesman problems
    CN107133398B|2020-09-01|River runoff prediction method based on complex network
    Sokolov et al.2020|Optimization of transmission capacity of energy water pipeline networks with a tree-shaped configuration and multiple sources
    Chen et al.2021|An improved bat algorithm hybridized with extremal optimization and Boltzmann selection
    CN104968047B|2018-06-15|A kind of Forecasting Methodology of mobile network-oriented interior joint network distance
    Chen et al.2018|A meta-optimized hybrid global and local algorithm for well placement optimization
    CN105988145A|2016-10-05|Inversion method of crack parameter
    Mishra et al.2020|Ant colony optimization for slope stability analysis applied to an embankment failure in eastern India
    Vela-Perez et al.2011|Ant foraging and minimal paths in simple graphs
    Li et al.2015|Optimal control for a shield machine subject to multi-point earth pressure balance
    Formentin et al.2018|The new EurOtop Neural Network tool for an improved prediction of wave overtopping
    Singh1976|A distributed converging overland flow model: 3. Application to natural watersheds
    CN110441815B|2021-05-25|Simulated annealing Rayleigh wave inversion method based on differential evolution and block coordinate descent
    WO2021022632A1|2021-02-11|Oil shale continuous exploration drilling position optimization method
    Pan et al.2021|Prediction of the Water Cut with the Hybrid Optimized SVR
    Nematolahi et al.2018|Predicting saturated hydraulic conductivity using particle swarm optimization and genetic algorithm
    CN104270795B|2018-03-02|A kind of method and system in the information potential domain based on backbone node structure sensing network
    Soil1969|Influence of Seepage Stream on the Joining of Frozen Soil Zones in Artificial Soil Freezing
    同族专利:
    公开号 | 公开日
    CN111259468A|2020-06-09|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20050120715A1|1997-12-23|2005-06-09|Christion School Of Technology Charitable Foundation Trust|Heat energy recapture and recycle and its new applications|
    CN102913250B|2012-11-12|2014-07-16|中铁二局股份有限公司|Shield tunnel-entering construction method at stratum rich in water powder and soil powder sand|
    CN107101924B|2017-03-08|2018-05-11|安徽理工大学|The experimental rig and method that monitoring seepage flow influences frozen temperature field|
    CN109239123A|2018-09-04|2019-01-18|北京城建集团有限责任公司|Large-scale freezing test model casing and analogue test platform|
    CN109344547B|2018-11-08|2020-07-28|同济大学|Freezing method model design method and device under combined stratum seepage effect|CN111927463B|2020-07-27|2022-03-01|北京城建集团有限责任公司|Freezing pipe arrangement method in multi-row pipe freezing engineering|
    法律状态:
    优先权:
    申请号 | 申请日 | 专利标题
    CN202010027402.6A|CN111259468A|2020-01-10|2020-01-10|Ant colony algorithm-based optimization design method for freezing method under seepage condition|
    [返回顶部]