CN113849901B - Improved self-adaptive optimization method and system for contact heat exchange coefficient identification - Google Patents

Abstract

The invention provides an improved self-adaptive optimization method and system aiming at contact heat exchange coefficient identification, comprising the following steps: step S1: randomly generating an initial population comprising a plurality of individuals; step S2: defining material thermophysical parameters of the grid; establishing a heat conduction calculation model based on a finite volume method; step S3: calculating to obtain temperature response T _cal of each position of the structure changing along with time; jumping to the step S6 when the solution is not available; step S4: selecting a temperature response calculated value T _cal,l of a specified position I in the structure to obtain weights w _i,t of each measuring point i and a calculating time T; step S5: calculating the fitness fit of an individual; step S6: evaluating the fitness of all individuals in the population; step S7: the values of the crossover probability p _c and the mutation probability p _m are calculated, a new generation population is produced, and steps S3 to S6 are repeated. The invention can realize the direct identification of the thermal test parameters under the actual structural condition, improve the engineering precision and reduce the identification time and the economic cost.

Description

Improved self-adaptive optimization method and system for contact heat exchange coefficient identification

Technical Field

The invention relates to the technical field of parameter identification and heat conduction inverse problem analysis, in particular to an improved self-adaptive optimization method and system aiming at contact heat exchange coefficient identification.

Background

With the increasing demands for unmanned aerial vehicles at flying speeds and flying distances, high-speed unmanned aerial vehicles are increasingly subjected to pneumatic heating environments. Under severe pneumatic heating conditions, thermal protection design has become an important and indispensable link in the development process of high-speed unmanned aerial vehicles. In the design process of the thermal protection scheme, thermal conduction analysis needs to be carried out on the elastomer structure and equipment, and the thermal protection scheme meeting the requirements is given according to the obtained temperature response. In practice, the object for conducting the thermal conductivity analysis is composed of a plurality of components, and heat exchange occurs between the components through contact interfaces. The contact interface heat exchange directly influences the accuracy of temperature calculation of equipment inside the structure. The heat exchange process is affected by the temperature and the contact heat exchange coefficient at the two sides of the contact interface, and it is necessary to obtain an accurate contact heat exchange coefficient if the calculation accuracy is to be improved.

Research shows that the contact heat exchange coefficient is influenced by various factors such as materials, pressure, interface temperature, roughness and the like. The existing method for determining the contact heat exchange coefficient mainly comprises a steady state method and a transient state method. The steady state method heats two materials needing to measure the contact heat exchange coefficient in a special device designed for the materials, so that the temperature response of a plurality of measuring points is obtained after the materials reach a steady state, and then the contact heat exchange coefficient is obtained through linear extrapolation and solving. The steady state method has lower solving difficulty, but in engineering practice, parameters such as pressure, roughness and the like are affected by various aspects such as assembly precision, machining precision and the like, if the contact heat exchange coefficient is measured by the steady state method, a large number of sample pieces are required to be machined, tests are respectively carried out aiming at a large number of working conditions, high economic cost and time cost are generated, and certain limitations are realized. Heating a material needing to be measured for the contact heat exchange coefficient according to a certain boundary condition by a transient method, and carrying out inversion calculation on the thermophysical parameters through temperature response of measuring points to obtain the contact heat exchange coefficient. The transient method is flexible in test method, but the inversion calculation difficulty is high. The existing inversion method mainly aims at carrying out inversion on thermophysical parameters or boundary conditions, and is not improved on the characteristics of contact heat exchange coefficients. When the loading heat flow condition is complex or errors exist in measurement, the inversion precision is not enough. In order to facilitate engineering designers to estimate the contact heat exchange coefficient between solid interfaces, it is necessary to provide a new optimization algorithm improved for contact heat exchange coefficient identification.

The invention patent with publication number CN103616406A discloses a device and a method for measuring the heat exchange coefficient of a solid-solid contact interface. The invention provides a device for measuring a heat exchange coefficient of a solid-solid contact interface and a measuring method thereof, wherein the device comprises the following components: the system comprises a workbench, a heating device, a guide heat preservation device, a loading device and an information processing system, and provides a response measurement method. The invention can be used for researching the influence of solid-solid contact temperature, contact pressure and contact surface roughness on interface heat exchange coefficient.

The invention patent with publication number CN102507636A discloses a method for measuring the interface heat exchange coefficient of a rapid cooling process of steel, which comprises the steps of welding a thermocouple on the surface of a workpiece, reading surface temperature change data through a temperature acquisition module, and obtaining the interface heat exchange coefficient of the cooling process by utilizing the interface heat exchange coefficient verification function of heat treatment software. The method can be used for solving the problem that the interface heat exchange coefficient is difficult to accurately determine in the rapid cooling process of the workpiece.

However, the following limitations exist in the prior art: (1) Special devices or test pieces are required to be manufactured for the contact heat exchange coefficient identification production, or measurement is required to be carried out under specific conditions, so that the economic cost is obviously increased; (2) A series of tests under different parameter conditions need to be carried out for identifying the contact heat exchange coefficient, and the time cost is high. Therefore, in order to overcome the above-mentioned shortcomings, a method is needed to be designed to realize the rapid identification of the heat test parameters directly under the actual structural conditions, and to reduce the identification time and the economic cost.

Disclosure of Invention

Aiming at the defects in the prior art, the invention provides an improved self-adaptive optimization method and system aiming at contact heat exchange coefficient identification.

According to the improved self-adaptive optimization method and system for identifying the contact heat exchange coefficient, the scheme is as follows:

in one aspect, the present invention provides an improved adaptive optimization method for contact heat exchange coefficient identification, comprising:

step S1: according to the number of the contact heat exchange coefficients to be identified, generating an initial population containing a plurality of individuals at random, wherein each individual contains a group of contact heat exchange coefficients [ h ₁,h₂,...,h_N ];

Step S2: dividing grids based on a structural geometric model needing to be subjected to contact heat exchange coefficient identification, and defining material thermophysical parameters of the grids;

setting initial conditions and boundary conditions for a thermal conduction calculation model, and establishing a thermal conduction calculation model based on a finite volume method;

Step S3: for each individual in the population in the step S1, judging whether the thermal conduction calculation model in the step S2 is solvable, if so, calculating the thermal conduction calculation model to obtain a temperature response T _cal of each position of the structure, which changes with time;

For individuals in which the thermal conduction calculation model is insoluble, assigning the fitness fit to a default value, and jumping to step S6;

Step S4: for each individual in the population in the step S1, selecting a temperature response calculated value T _cal,l of a designated position in the structure, and calculating the temperature change rate of a measuring point at each calculation moment to obtain a weight w _i,t based on the temperature change rate at each calculation moment;

Step S5: for each individual in the population in the step S1, based on the weight w _i,t and the temperature response calculated value T _cal,l,t obtained in the step S4, carrying out heat conduction analysis by taking the finally obtained new generation population as a final contact heat exchange coefficient, and calculating the fitness fit of the individual according to the corresponding position measured value T _real,l,t given by the identification data;

step S6: evaluating the fitness of all individuals in the population, and if the difference value between the average fitness and the minimum fitness of the individuals in the population is smaller than a threshold value, completing optimization;

Taking the contact heat exchange coefficient contained in the individual with the minimum fitness in the current population as the solution of the identification, otherwise, continuing;

step S7: and calculating the values of the crossover probability p _c and the mutation probability p _m according to the iteration times of the current population and the self-adaptive parameters, selecting individuals from the current population to reserve, crossover and mutation, producing a new generation population, and repeating the steps S3-S6.

Preferably, the step S1 includes:

For the calculated one-dimensional structure, assuming that the contact interfaces of N parts are shared, N contact heat exchange coefficients H to be identified are shared, an initial population is generated through a random function, each individual j in the population comprises N contact heat exchange coefficients H, and the individual available contact heat exchange coefficient array H _j is expressed as:

H_j＝[h_j,1 h_j,2 ... h_j,N](j＝1,2,...,J)

wherein j is the number of the individual in the population;

J is the total number of individuals in the population;

p _j represents an array of all contact heat exchange coefficients in the jth individual;

h _j,n represents the contact heat transfer coefficient of the jth individual nth (n=1, 2,., N) part contact interface;

the entire population of individuals can be represented as a two-dimensional array:

Preferably, the step S2 includes:

for the geometric structure needing to be subjected to contact heat exchange coefficient identification, a limited volume method is adopted to discrete a geometric model into a series of units, and explicit solving is adopted; the general form of the one-dimensional unsteady-state heat conduction differential equation under the Cartesian coordinate system is as follows:

Wherein ρ, c, λ, The density, specific heat capacity, heat conductivity coefficient and internal heat source items of the material are respectively;

T and x are temperature and coordinates, respectively;

τ represents time;

Representing a first partial derivative of the time tau of the temperature T;

Representing that the unknown number x of the function T is subjected to one-time partial derivative, then the product of the unknown number x and lambda is obtained, and then the unknown number x is subjected to one-time partial derivative;

The one-dimensional structure is discretized by adopting a finite volume method, and for each control body, the following equation is obtained according to the energy balance principle:

wherein, T _P、T_W、T_E is the temperature of the current control body, the left control body of the current control body and the right control body of the current control body respectively;

_P Representing the current control body;

_W Representing a control body adjacent to the left side of the current control body;

_E Representing a control body adjacent to the right side of the current control body;

Δτ represents the calculated time step after the discretization;

Δx is the control body width;

A temperature value of the current control body representing the current time τ ₁;

A temperature value of the current control body indicating the previous time τ ₀;

(^-)^0-1 shows that the value takes the mean of the current time τ ₁ and the previous time τ ₀;

Representing the average value of the control body W adjacent to the left side of the current control body at the current time tau ₁ and the previous time tau ₀;

representing the average value of the control body E adjacent to the right side of the current control body at the current time tau ₁ and the previous time tau ₀;

When the time display format is adopted, the value of the previous time tau ₀ is adopted to replace the mean value, and the differential is adopted to replace the differential, so that a discrete equation is obtained:

Wherein, A temperature value representing the control body E adjacent to the right side of the current control body at the current time τ ₀;

A temperature value representing the control body W adjacent to the left side of the current control body at the current time τ ₀;

The distances from the center point of the current control point to the center points of the left control point and the right control point of δx _w、δx_e are calculated by the following formulas respectively;

δx_w＝|x_p-x_w|

δx_e＝|x_p-x_e|

Wherein x _p represents the position of the center point of the current control point;

x _w denotes the position of the center point W of the control W adjacent to the left of the current control point;

_w A center point of the control W adjacent to the left side of the current control point;

x _e denotes the position of the center point E of the control E adjacent to the right of the current control point;

_e A center point representing the control E adjacent to the right of the current control point;

for internal boundaries between materials, a third class of boundary conditions can be considered; calculating a boundary of one side of the control body includes:

Calculating the left boundary of the control body, wherein the boundary is obtained by equal heat flows on the interface, and the boundary is provided with:

Wherein h is a convective heat transfer coefficient;

T _s represents the structural surface temperature in contact with the fluid;

_s Representing a structured surface;

t _f is the fluid temperature;

_f Representing a fluid;

and obtaining a discrete equation by adopting a time display format:

Wherein, Representing the fluid temperature at the previous calculation time τ ₀;

Sorting the discrete equations to obtain an expression adopted by the explicit calculation program:

preferably, the step S3 includes:

For the contact heat exchange coefficient h _j,n of each interface n in each individual j in the population, judging whether the following conditions are met:

h_j,n＞0

If the contact heat exchange coefficient generated by the individual does not have the actual physical meaning, the contact heat exchange coefficient is regarded as insoluble;

if the contact heat exchange coefficient generated by the individual has the actual physical meaning, the contact heat exchange coefficient is considered to be solvable;

for the resolvable individual j, the contact heat exchange coefficient of the individual is transmitted into the heat conduction positive problem calculation model established in the step S2, and the method comprises the following steps:

h₁＝h_j,1

h₂＝h_j,2

h_N＝h_j,N

for solvable individuals, the thermal conduction positive problem calculation model established in the step S2 meets all input conditions, and the thermal conduction positive problem calculation model is operated to obtain temperature response values T _cal of all nodes at each calculation time:

Wherein L is the total number of the measuring point nodes;

v is the total number of the calculated time points;

T _cal,l,t represents a temperature response calculated value of the measuring point node l at the time point T; t=1, 2,; l=1, 2,..l.

Preferably, the step S4 includes:

Extracting temperature response T _cal,l of corresponding positions from a temperature response value matrix T _cal of all nodes given in the step S3 according to the measuring point positions corresponding to the temperature response measured values T _real,l,t given in the identification data;

And respectively calculating the weight W _i,t of each measuring point in the adaptive value function for each selected measuring point.

The specific steps of calculating the weight W _i,t include:

step S4.1: for the temperature response data of each measuring point, calculating a temperature change rate dT _cal,i/dτ from the second moment;

wherein i is the extracted ith measuring point;

t is the number of the calculated time after the discretization;

In actual calculation, the following formula is adopted instead:

Wherein τ _t represents the time corresponding to the t-th calculation time after the discretization;

τ _t-1 represents the time corresponding to the t-1 th calculation time after the discretization;

step S4.2: summing the absolute values of the temperature change rates at each moment of each measuring point;

Step S4.3: for each moment of each measuring point, the weight of each moment of each measuring point in the process of calculating the adaptive value is given as follows:

Wherein w _i,t represents a weight;

If the substitution formula given in step S4.1 above is adopted, the weights are expressed as:

Wherein, T _cal,i,t-1 represents the temperature response calculated value of the i-th measuring point selected at the T-1-th calculating moment.

Preferably, the step S5 includes:

According to the temperature response calculated value T _cal,l extracted in the step S4, the temperature response measured value T _real,l given by the identification data and the weight w _i,t of each measuring point at each calculating moment calculated in the step S4, and the fitness fit of the individual is calculated by carrying out square weighted summation on all the data of each measuring point and each moment, wherein the specific formula is as follows:

wherein i is the extracted ith measuring point;

t is the calculation time number, namely the calculation time point;

w _i,t represents the weight of the ith measuring point at the time t;

v is the number of times of calculation output.

Preferably, the step S6 includes:

For all individuals in the population, calculating the average fitness

Wherein j is the number of the individual in the population;

J is the total number of individuals in the population;

fit _j represents fitness of the jth individual;

For all individuals in the population, searching for its minimum fitness and comparing with the average fitness, when:

then the optimization is considered to have been completed; otherwise, the optimization is considered to be incomplete;

Wherein fit _min represents the minimum fitness among all individuals within the population;

Δα _thres represents a threshold value for ending optimization;

And giving an individual corresponding to the fit _min for the situation that the optimization is completed, wherein the contact heat exchange coefficient h ₁,h₂,...,h_N corresponding to the individual is the contact heat exchange coefficient obtained by the optimization aiming at the current input condition.

For the case where the optimization is not completed, the process proceeds to step S7.

Preferably, the step S7 includes:

For the current population, a crossover probability p _c is calculated:

Wherein p _max is the upper cross probability limit;

p _min is the crossover probability lower bound;

r _max is the corresponding percentage when the upper crossover probability limit is reached;

r _min is the corresponding percentage when the lower crossover probability limit is reached;

Δα ₀ is the difference between the minimum and average population fitness values at the first iteration;

Δα is the difference between the minimum and average population fitness values at the current iteration number, expressed as:

for the current population, a variation probability p _m is calculated:

p_m＝1-p_e-p_c

wherein p _e is the elite individual proportion, i.e. the elite individual ratio in the population, for elite individuals, no crossing and variant individuals are involved in the iteration.

Preferably, for the current population, multiplying the probability by the number of individuals in the population to obtain the number of individuals remaining, crossing and varying in the next generation;

The method for determining elite individuals comprises the following steps: the individuals in the population are arranged according to the ascending order of fitness, and the individuals are sequentially selected from front to back until the number of reserved individuals is reached;

The method for determining the crossed individuals comprises the following steps: randomly extracting two individuals from the population each time to cross to obtain a crossed individual until the number of crossed individuals is reached;

The method for determining variant individuals comprises the following steps: randomly extracting individuals from the population to randomly mutate the individuals until the number of mutated individuals is reached;

And (3) forming a matrix by newly generated reserved individuals, crossed individuals and contact heat exchange coefficients corresponding to variant individuals, namely forming a new generation population, and repeating the steps S3-S6.

In another aspect, the present invention also provides an improved adaptive optimization system for contact heat exchange coefficient identification, comprising:

Module M1: according to the number of the contact heat exchange coefficients to be identified, generating an initial population containing a plurality of individuals at random, wherein each individual contains a group of contact heat exchange coefficients [ h ₁,h₂,...,h_N ];

Module M2: dividing grids based on a structural geometric model needing to be subjected to contact heat exchange coefficient identification, and defining material thermophysical parameters of the grids;

setting initial conditions and boundary conditions for a thermal conduction calculation model, and establishing a thermal conduction calculation model based on a finite volume method;

module M3: for each individual in the population in the module M1, judging whether the thermal conduction calculation model in the module M2 is solvable, if so, calculating the thermal conduction calculation model to obtain a temperature response T _cal of each position of the structure changing along with time;

For individuals in which the thermal conduction calculation model is insoluble, assign fitness fit to a default value and jump to module M6;

Module M4: for each individual in the population in the module M1, selecting a temperature response calculated value T _cal,l of a designated position in the structure, and calculating the temperature change rate of a measuring point at each calculation moment to obtain a weight w _i,t based on the temperature change rate at each calculation moment;

Module M5: for each individual in the population in the module M1, based on the weight w _i,t and the temperature response calculated value T _cal,l,t obtained by the module M4, carrying out heat conduction analysis by taking the finally obtained new-generation population as a final contact heat exchange coefficient, and calculating the fitness fit of the individual according to the corresponding position measured value T _real,l,t given by the identification data;

Module M6: evaluating the fitness of all individuals in the population, and if the difference value between the average fitness and the minimum fitness of the individuals in the population is smaller than a threshold value, completing optimization;

Taking the contact heat exchange coefficient contained in the individual with the minimum fitness in the current population as the solution of the identification, otherwise, continuing;

Module M7: and calculating the values of the crossover probability p _c and the mutation probability p _m according to the iteration times of the current population and the self-adaptive parameters, selecting individuals from the current population to reserve, crossover and mutation, producing a new generation population, and repeating the modules M3-M6.

Compared with the prior art, the invention has the following beneficial effects:

1. the invention realizes the identification of the contact heat exchange coefficient in the common heating test, can synchronously complete the identification of the contact heat exchange coefficient in the tests such as the sample high-temperature test, the static heat combined test and the like, has high convenience and flexibility, saves the related economic cost and time cost of a special device and the test, and improves the research and development efficiency;

2. By pertinently improving the fitness function, the sensitivity of the algorithm to the contact heat exchange coefficient is improved, and the inversion precision of the contact heat exchange coefficient under the conditions of complex loading heat flow conditions and measurement errors of identification data is improved;

3. By pertinently improving the optimization algorithm, the invention reduces the calculated amount required by the inversion process and reduces the calculation force requirement on inversion.

Drawings

Other features, objects and advantages of the present invention will become more apparent upon reading of the detailed description of non-limiting embodiments, given with reference to the accompanying drawings in which:

FIG. 1 is a schematic flow diagram of an improved adaptive optimization algorithm for contact heat transfer coefficient identification;

FIG. 2 is a graph of the loading heat flow in an embodiment of the present invention;

FIG. 3 is a diagram illustrating the identification data according to an embodiment of the present invention;

FIG. 4 is an example initial population distribution in an embodiment of the invention;

FIG. 5 is a schematic diagram of an example geometric model and boundary conditions in an embodiment of the present invention;

FIG. 6 is a graph showing fitness functions of an exemplary first generation population in accordance with an embodiment of the present invention;

FIG. 7 is an example optimized population distribution in an embodiment of the invention;

FIG. 8 is an example optimized population fitness distribution in an embodiment of the present invention;

FIG. 9 is a comparison of the identification result and the input data in the embodiment of the invention.

Detailed Description

The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the present invention, but are not intended to limit the invention in any way. It should be noted that variations and modifications could be made by those skilled in the art without departing from the inventive concept. These are all within the scope of the present invention.

Example 1:

The embodiment 1 of the invention provides an improved self-adaptive optimization method for identifying a contact heat exchange coefficient, which is shown by referring to fig. 1, and specifically comprises the following steps:

step S1: according to the number of the contact heat exchange coefficients to be identified, generating an initial population containing a plurality of individuals at random, wherein each individual contains a group of contact heat exchange coefficients [ h ₁,h₂,...,h_N ];

Step S2: dividing grids based on a structural geometric model needing to be subjected to contact heat exchange coefficient identification, and defining material thermophysical parameters of the grids;

setting initial conditions and boundary conditions for a thermal conduction calculation model, and establishing a thermal conduction calculation model based on a finite volume method;

Step S3: for each individual in the population in the step S1, judging whether the thermal conduction calculation model in the step S2 is solvable, if so, calculating the thermal conduction calculation model to obtain a temperature response T _cal of each position of the structure, which changes with time;

For individuals in which the thermal conduction calculation model is insoluble, assigning the fitness fit to a default value, and jumping to step S6;

Wherein, the insoluble individual means that when any one of the set of contact heat exchange coefficients [ h ₁,h₂,...,h_N ] randomly generated by the step S1 is less than 0, the individual has no physical meaning, or the equation set can not be solved.

Step S4: for each individual in the population in the step S1, selecting a temperature response calculated value T _cal,l of a specified position I in the structure, and calculating the temperature change rate of the measuring point at each calculation moment to obtain a weight w _i,t of each measuring point i and each calculation moment T based on the temperature change rate;

Step S5: for each individual in the population in the step S1, based on the weight w _i,t of each calculation time T obtained in the step S4 and the temperature response calculated value T _cal,l,t corresponding to the time, carrying out heat conduction analysis by taking the finally obtained new generation population as a final contact heat exchange coefficient, and calculating the fitness fit of the individual according to the corresponding position measured value T _real,l,t given by the identification data;

step S6: evaluating the fitness of all individuals in the population, and if the difference value between the average fitness and the minimum fitness of the individuals in the population is smaller than a threshold value, completing optimization;

Taking the contact heat exchange coefficient contained in the individual with the minimum fitness in the current population as the solution of the identification, otherwise, continuing;

step S7: and calculating the values of the crossover probability p _c and the mutation probability p _m according to the iteration times of the current population and the self-adaptive parameters, selecting individuals from the current population to reserve, crossover and mutation, producing a new generation population, and repeating the steps S3-S6.

The step S1 includes:

For a structure with N interfaces needing to identify contact heat exchange coefficients, generating an initial population through a random function, wherein each individual P in the population comprises N contact heat exchange coefficients, which are expressed as:

P_j＝[h_j,1 h_j,2 ... h_j,N](j＝1,2,...,J)

wherein j is the number of the individual in the population;

J is the total number of individuals in the population;

p _j represents an array of all contact heat exchange coefficients in the jth individual;

h _j,n represents the contact heat transfer coefficient of the jth individual nth (n=1, 2,., N) part contact interface;

the total number of individuals in the population is determined according to the number N of the contact heat exchange coefficients, and the generated population is expressed as follows by adopting a matrix:

The step S2 includes:

for the geometric structure needing to be subjected to contact heat exchange coefficient identification, a limited volume method is adopted to discrete a geometric model into a series of units, and explicit solving is adopted; the general form of the one-dimensional unsteady-state heat conduction differential equation under the Cartesian coordinate system is as follows:

Wherein ρ, c, λ, The density, specific heat capacity, heat conductivity coefficient and internal heat source items of the material are respectively;

T and x are temperature and coordinates, respectively;

τ represents time;

Representing a first partial derivative of the time tau of the temperature T;

Representing that the unknown number x of the function T is subjected to one-time partial derivative, then the product of the unknown number x and lambda is obtained, and then the unknown number x is subjected to one-time partial derivative;

The one-dimensional structure is discretized by adopting a finite volume method, and for each control body, the following equation is obtained according to the energy balance principle:

Wherein T _P、T_W、T_E is the temperature of the current control body, the left control body and the right control body respectively;

_P Representing the current control body;

_W Representing a control body adjacent to the left side of the current control body;

_E Representing a control body adjacent to the right side of the current control body;

Δτ represents the calculated time step after the discretization;

Δx is the control body width;

A temperature value of the current control body representing the current time τ ₁;

A temperature value of the current control body indicating the previous time τ ₀;

(^-)^0-1 shows that the value takes the mean of the current time τ ₁ and the previous time τ ₀;

Representing the average value of the control body W adjacent to the left side of the current control body at the current time tau ₁ and the previous time tau ₀;

representing the average value of the control body E adjacent to the right side of the current control body at the current time tau ₁ and the previous time tau ₀;

When the time display format is adopted, the value of the previous time tau ₀ is adopted to replace the mean value, and the differential is adopted to replace the differential, so that a discrete equation is obtained:

Wherein, A temperature value representing the control body E adjacent to the right side of the current control body at the current time τ ₀;

A temperature value representing the control body W adjacent to the left side of the current control body at the current time τ ₀;

The distances from the center point of the current control point to the center points of the left control point and the right control point of δx _w、δx_e are calculated by the following formulas respectively;

δx_w＝|x_p-x_w|

δx_e＝|x_p-x_e|

Wherein x _p represents the position of the center point of the current control point;

x _w denotes the position of the center point W of the control W adjacent to the left of the current control point;

_w A center point of the control W adjacent to the left side of the current control point;

x _e denotes the position of the center point E of the control E adjacent to the right of the current control point;

_e A center point representing the control E adjacent to the right of the current control point;

for internal boundaries between materials, a third class of boundary conditions can be considered; calculating a boundary of one side of the control body includes:

Calculating the left boundary of the control body, wherein the boundary is obtained by equal heat flows on the interface, and the boundary is provided with:

Wherein h is a convective heat transfer coefficient;

T _s represents the structural surface temperature in contact with the fluid;

_s Representing a structured surface;

t _f is the fluid temperature;

_f Representing a fluid;

and obtaining a discrete equation by adopting a time display format:

Wherein, Representing the fluid temperature at the previous calculation time τ ₀;

Δτ represents the calculation step size of the transient heat conduction problem;

Sorting the discrete equations to obtain an expression adopted by the explicit calculation program:

the step S3 specifically comprises the following steps:

For the contact heat exchange coefficient in each individual P in the population, judging whether the contact heat exchange coefficient meets the following conditions:

h_j,n＞0

If the contact heat exchange coefficient generated by the individual does not have the actual physical meaning, the contact heat exchange coefficient is regarded as insoluble;

if the contact heat exchange coefficient generated by the individual has the actual physical meaning, the contact heat exchange coefficient is considered to be solvable;

for the resolvable individual j, the contact heat exchange coefficient of the individual is transmitted into the heat conduction positive problem calculation model established in the step S2, and the method comprises the following steps:

h₁＝h_j,1

h₂＝h_j,2

h_N＝h_j,N

for solvable individuals, the thermal conduction positive problem calculation model established in the step S2 meets all input conditions, and the thermal conduction positive problem calculation model is operated to obtain temperature response values T _cal of all nodes at each calculation time:

Wherein L is the total number of the measuring point nodes;

v is the total number of the calculated time points;

T _cal,l,t represents a temperature response calculated value of the measuring point node l at the time point T; wherein t=1, 2,. -%, V; l=1, 2,..l.

For an insoluble individual, since it does not have a practical physical meaning, it is not calculated to save calculation power. In order to avoid uncontrollable influence of individuals on population scale, a larger default adaptation value is directly given to the individuals so as to eliminate the individuals in iteration, thereby realizing the purpose of screening out ineffective contact heat exchange coefficients. The larger adaptation value is determined according to the actual problem. The determining method comprises the following steps: and (3) transferring the upper and lower boundaries of the contact heat exchange coefficient into a heat conduction positive problem calculation model established in the step (S2) for calculation, converting the calculated temperature response value into an adaptation value according to the method provided in the step (S5), and taking the maximum value of the adaptation value obtained in the calculation as a default adaptation value.

For the insoluble individual, go to step S6.

The step S4 includes:

Extracting temperature response T _cal,l of corresponding positions from a temperature response value matrix T _cal of all nodes given in the step S3 according to the measuring point positions corresponding to the temperature response measured values T _real,l,t given in the identification data;

And respectively calculating the weight W _i,t of each measuring point in the adaptive value function for each selected measuring point. The specific steps of calculating the weight W _i,t include:

step S4.1: for the temperature response data of each measuring point, calculating a temperature change rate dT _cal,i/dτ from the second moment;

wherein i is the extracted ith measuring point;

t is the number of the calculated time after the discretization;

generally, in actual calculation, the following formula is adopted instead:

Wherein τ _t represents the time corresponding to the t-th calculation time after the discretization;

τ _t-1 represents the time corresponding to the t-1 th calculation time after the discretization;

step S4.2: summing the absolute values of the temperature change rates at each moment of each measuring point;

Step S4.3: for each moment of each measuring point, the weight of each moment of each measuring point in the process of calculating the adaptive value is given as follows:

Wherein w _i,t represents a weight;

If the substitution formula given in step S4.1 above is used, the weights can be expressed as:

Wherein, T _cal,i,t-1 represents the temperature response calculated value of the i-th measuring point selected at the T-1-th calculating moment.

The step S5 specifically includes:

According to the temperature response calculated value T _cal,l extracted in the step S4, the temperature response measured value T _real,l given by the identification data and the weight w _i,t of each measuring point at each calculating moment calculated in the step S4, and the fitness fit of the individual is calculated by carrying out square weighted summation on all the data of each measuring point and each moment, wherein the specific formula is as follows:

wherein i is the extracted ith measuring point;

t is the calculation time number, namely the calculation time point;

v is the number of times of calculation output.

The step S6 specifically includes:

For all individuals in the population, calculating the average fitness

Wherein j is the number of the individual in the population;

J is the total number of individuals in the population;

fit _j represents fitness of the jth individual;

For all individuals in the population, searching for its minimum fitness and comparing with the average fitness, when:

then the optimization is considered to have been completed; otherwise, the optimization is considered to be incomplete;

Wherein fit _min represents the minimum fitness among all individuals within the population;

Δα _thres represents a threshold value for ending optimization;

And giving an individual corresponding to the fit _min for the situation that the optimization is completed, wherein the contact heat exchange coefficient h ₁,h₂,...,h_N corresponding to the individual is the contact heat exchange coefficient obtained by the optimization aiming at the current input condition.

For the case where the optimization is not completed, the process proceeds to step S7.

The step S7 specifically includes:

For the current population, a crossover probability p _c is calculated:

Wherein p _max is the upper cross probability limit;

p _min is the crossover probability lower bound;

r _max is the corresponding percentage when the upper crossover probability limit is reached;

r _min is the corresponding percentage when the lower crossover probability limit is reached;

Δα ₀ is the difference between the minimum and average population fitness values at the first iteration;

Δα is the difference between the minimum and average population fitness values at the current iteration number, and can be expressed as:

for the current population, a variation probability p _m is calculated:

p_m＝1-p_e-p_c

wherein p _e is the elite individual proportion, i.e. the elite individual ratio in the population, for elite individuals, no crossing and variant individuals are involved in the iteration.

For the current population, multiplying the probability by the number of individuals in the population to obtain the number of individuals reserved, crossed and mutated in the next generation;

The method for determining elite individuals comprises the following steps: the individuals in the population are arranged according to the ascending order of fitness, and the individuals are sequentially selected from front to back until the number of reserved individuals is reached;

The method for determining the crossed individuals comprises the following steps: randomly extracting two individuals from the population each time to cross to obtain a crossed individual until the number of crossed individuals is reached;

The method for determining variant individuals comprises the following steps: randomly extracting individuals from the population to randomly mutate the individuals until the number of mutated individuals is reached;

And (3) forming a matrix by newly generated reserved individuals, crossed individuals and contact heat exchange coefficients corresponding to variant individuals, namely forming a new generation population, and repeating the steps S3-S6.

The embodiment 1 of the invention also provides an improved self-adaptive optimization system aiming at the contact heat exchange coefficient identification, which specifically comprises the following steps:

Module M1: according to the number of the contact heat exchange coefficients to be identified, generating an initial population containing a plurality of individuals at random, wherein each individual contains a group of contact heat exchange coefficients [ h ₁,h₂,...,h_N ];

Module M2: dividing grids based on a structural geometric model needing to be subjected to contact heat exchange coefficient identification, and defining material thermophysical parameters of the grids;

setting initial conditions and boundary conditions for a thermal conduction calculation model, and establishing a thermal conduction calculation model based on a finite volume method;

module M3: for each individual in the population in the module M1, judging whether the thermal conduction calculation model in the module M2 is solvable, if so, calculating the thermal conduction calculation model to obtain a temperature response T _cal of each position of the structure changing along with time;

For individuals in which the thermal conduction calculation model is insoluble, assign fitness fit to a default value and jump to module M6;

Module M4: for each individual in the population in the module M1, selecting a temperature response calculated value T _cal,l of a specified position I in the structure, and calculating the temperature change rate of the measuring point at each calculation moment to obtain a weight w _i,t of each measuring point i and each calculation moment T based on the temperature change rate;

Module M5: for each individual in the population in the module M1, based on the weight w _i,t of each calculation time T obtained by the module M4 and the temperature response calculation value T _cal,l,t corresponding to the time, carrying out heat conduction analysis by taking the finally obtained new generation population as a final contact heat exchange coefficient, and calculating the fitness fit of the individual according to the corresponding position actual measurement value T _real,l,t given by the identification data;

Module M6: evaluating the fitness of all individuals in the population, and if the difference value between the average fitness and the minimum fitness of the individuals in the population is smaller than a threshold value, completing optimization;

Taking the contact heat exchange coefficient contained in the individual with the minimum fitness in the current population as the solution of the identification, otherwise, continuing;

Module M7: and calculating the values of the crossover probability p _c and the mutation probability p _m according to the iteration times of the current population and the self-adaptive parameters, selecting individuals from the current population to reserve, crossover and mutation, producing a new generation population, and repeating the modules M3-M6.

Example 2:

Example 2 is a modification of example 1:

taking a three-layer material, a model of two contact interfaces as an example, wherein the three-layer material is ceramic, fiber board and steel respectively, and the thermal physical parameters adopted in calculation are shown in table 1:

TABLE 1

Material 1, material 2 and material 3 are arranged in sequence from front to back, and the thicknesses of the materials are respectively 2mm, 1mm and 1.5mm. A time-varying heat flow is applied to the outer surface of the material 1, the heat flow profile of which over time is shown in fig. 2. Taking the front surface of the second layer material and the rear surface of the third layer material as measuring points, extracting temperature response data and adding noise, wherein the expression is as follows:

Wherein T _exact represents the true value of the measurement point, in this example the extracted temperature response data;

gamma is a random error, in this example a value in the range 0 to 1 is generated by a random function;

ζ is the measurement error, in this example taken ζ=1%;

T _real is the measured point temperature response value as the identification data after adding noise, and the identification data in this example is shown in FIG. 3.

1. The identification range of the contact heat exchange coefficient is set to be 1< h <3000, 40 contact heat exchange coefficients are randomly generated in the range as individuals, and an initial population P containing 40 individuals is generated, and the distribution of the initial population P is shown in fig. 4.

2. A one-dimensional calculation model of the calculation example is established, the first layer, the second layer and the third layer are respectively divided into 20 units, 10 units and 15 units, the thermophysical properties of materials given in table 1 are transferred into the calculation model, and the heat flow of fig. 2 is applied to the corresponding node of the front surface of the first layer. The initial temperature is set to 310K, the solving time is set to be 214 seconds which is the same as the loading time of the heat flow, one step of calculation is carried out every 0.5 seconds, and the establishment of a heat conduction calculation model based on a finite volume method is completed, wherein the geometric schematic diagram is shown in fig. 5.

3. For each individual within P, it is determined whether each contact heat transfer coefficient thereof is within the range (1,3000). And for the individuals within the range, the individuals are considered to be solvable, and the contact heat exchange coefficient is transmitted into a calculation model for calculation, so that a temperature response matrix of each position of the structure with time change is obtained. For individuals not within this range, the individual is considered to be insoluble, and for this example, the adaptation value default value fit _default =110 is taken and jumps to 6.

4. For each individual in P, selecting temperature response data of a front surface node of the second layer and a rear surface node of the third layer corresponding to the identification data, and calculating the temperature change rates of the two measuring points at 427 times except the first calculating time:

Further calculating weights in calculating the fitness value for each of the 427 points:

5. for each individual within P, calculate the fitness fit of the individual:

taking the first generation population as an example, the fitness distribution calculated for randomly generated individuals is shown in fig. 6.

6. Calculating the average fitness of the population according to the fitness distribution obtained by 5

Searching for the minimum fitness in the population and comparing with the average fitness, if the absolute value of the difference is smaller than the threshold, then the optimization is considered to be completed, otherwise, continuing to 7. In this embodiment, the threshold is 1×10 ^-4.

For this example, the population distribution and the fitness distribution at the completion of the optimization are shown in fig. 7 and 8. And finally identifying and obtaining the contact heat exchange coefficient h ₁＝307.97(W/m²/K),h₂＝1014.22(W/m²/K of the interface). The comparison between the identification input data T _real and the temperature response T _cal calculated by using the interface contact heat exchange coefficient obtained by identification is shown in fig. 9.

7. For the present embodiment, the upper limit p _max =80% of the crossover probability, the lower limit p _min =20% of the crossover probability, the upper limit of the crossover probability corresponds to the percentage r _max =100%, and the lower limit of the crossover probability corresponds to the percentage r _min =30%. For the first generation population, Δα ₀ was calculated:

/>

Wherein, Representing the average value of fitness fit of all individuals in the generation 1 population;

fit _min,1 represents the minimum of fitness of all individuals in the generation 1 population;

For the first generation population, the crossover probability is the crossover probability upper limit p _max =80%, and for the rest of the various generation populations, the crossover function p _c:

for this example, the elite individual proportion p _e =5% is taken, yielding the probability of variation p _m:

p_m＝0.95-p_c

For the first generation population, the number of reserved, crossed and mutated individuals is respectively 2, 32 and 6 according to the total number of population individuals 40 and the elite individual proportion, the cross probability and the mutation probability. For the remaining generation populations, the number is determined according to the above equation. And sequentially generating new generation individuals after the determination to form a next generation population.

The newly generated contact heat exchange coefficients corresponding to reserved individuals, crossed individuals and variant individuals form a matrix, namely a new generation population is formed, and the process is repeated for 3-6 times.

The embodiment of the invention provides an improved self-adaptive optimization method for identifying the contact heat exchange coefficient, which realizes the identification of the contact heat exchange coefficient in a general heating test, can synchronously complete the identification of the contact heat exchange coefficient in a sample high-temperature test, a static heat combined test and other tests, has high convenience and flexibility, saves the related economic cost and time cost of a special device and the test, and improves the research and development efficiency; by pertinently improving the fitness function, the sensitivity of the algorithm to the contact heat exchange coefficient is improved, and the inversion precision of the contact heat exchange coefficient under the conditions of complex loading heat flow conditions and measurement errors of identification data is improved; by pertinently improving the optimization algorithm, the calculation amount required by the inversion process is reduced, and the calculation force requirement on inversion is reduced.

Those skilled in the art will appreciate that the invention provides a system and its individual devices, modules, units, etc. that can be implemented entirely by logic programming of method steps, in addition to being implemented as pure computer readable program code, in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers, etc. Therefore, the system and various devices, modules and units thereof provided by the invention can be regarded as a hardware component, and the devices, modules and units for realizing various functions included in the system can also be regarded as structures in the hardware component; means, modules, and units for implementing the various functions may also be considered as either software modules for implementing the methods or structures within hardware components.

The foregoing describes specific embodiments of the present application. It is to be understood that the application is not limited to the particular embodiments described above, and that various changes or modifications may be made by those skilled in the art within the scope of the appended claims without affecting the spirit of the application. The embodiments of the application and the features of the embodiments may be combined with each other arbitrarily without conflict.

Claims (10)

1. An improved adaptive optimization method for contact heat exchange coefficient identification, comprising the steps of:

step S1: according to the number of the contact heat exchange coefficients to be identified, generating an initial population containing a plurality of individuals at random, wherein each individual contains a group of contact heat exchange coefficients [ h ₁,h₂,...,h_N ];

Step S2: dividing grids based on a structural geometric model needing to be subjected to contact heat exchange coefficient identification, and defining material thermophysical parameters of the grids;

setting initial conditions and boundary conditions for a thermal conduction calculation model, and establishing a thermal conduction calculation model based on a finite volume method;

Step S3: for each individual in the population in the step S1, judging whether the thermal conduction calculation model in the step S2 is solvable, if so, calculating the thermal conduction calculation model to obtain a temperature response T _cal of each position of the structure, which changes with time;

For individuals in which the thermal conduction calculation model is insoluble, assigning the fitness fit to a default value, and jumping to step S6;

Step S4: for each individual in the population in the step S1, selecting a temperature response calculated value T _cal,l of a specified position I in the structure, and calculating the temperature change rate of the measuring point at each calculation moment to obtain a weight w _i,t of each measuring point i and each calculation moment T based on the temperature change rate;

step S5: for each individual in the population in the step S1, based on the weight w _i,t of each calculation time T obtained in the step S4 and the temperature response calculation value T _cal,l,t corresponding to the time, carrying out heat conduction analysis by taking the finally obtained new generation population as a final contact heat exchange coefficient, and calculating the fitness fit of the individual according to the corresponding position actual measurement value T _real,l,t given by the identification data;

step S6: evaluating the fitness of all individuals in the population, and if the difference value between the average fitness and the minimum fitness of the individuals in the population is smaller than a threshold value, completing optimization;

Taking the contact heat exchange coefficient contained in the individual with the minimum fitness in the current population as the solution of the identification, otherwise, continuing;

Step S7: and calculating the values of the crossover probability p _c and the mutation probability p _m according to the iteration times of the current population and the self-adaptive parameters, selecting individuals from the current population to reserve, crossover and mutation, producing a new generation population, and repeating the steps S3-S6.

2. The improved adaptive optimization method for contact heat transfer coefficient identification according to claim 1, wherein the step S1 comprises:

For the calculated one-dimensional structure, assuming that the contact interfaces of N parts are shared, N contact heat exchange coefficients H to be identified are shared, an initial population is generated through a random function, each individual j in the population comprises N contact heat exchange coefficients H, and the individual available contact heat exchange coefficient array H _j is expressed as:

H_j＝[h_j,1 h_j,2 ... h_j,N](j＝1,2,...,J)

wherein j is the number of the individual in the population;

J is the total number of individuals in the population;

p _j represents an array of all contact heat exchange coefficients in the jth individual;

h _j,n represents the contact heat transfer coefficient of the contact interface of the nth part of the jth individual, n=1, 2,.;

the entire population of individuals can be represented as a two-dimensional array:

the total number of rows J of the contact heat exchange coefficient array H _j of each behavior unit is the total number of units, and the total number of units is determined according to the computing capability and the number N of interfaces to be identified.

3. The improved adaptive optimization method for contact heat transfer coefficient identification according to claim 1, wherein the step S2 comprises:

for the geometric structure which needs to be identified by the contact heat exchange coefficient, the geometric structure is discretized into a series of units by adopting a finite volume method, and explicit solving is adopted; the general form of the one-dimensional unsteady-state heat conduction differential equation under the Cartesian coordinate system is as follows:

Wherein ρ, c, λ, The density, specific heat capacity, heat conductivity coefficient and internal heat source items of the material are respectively;

T and x are temperature and coordinates, respectively;

τ represents time;

Representing a first partial derivative of the time tau of the temperature T;

The one-dimensional structure is discretized by adopting a finite volume method, and for each control body, the following equation is obtained according to the energy balance principle:

wherein, T _P、T_W、T_E is the temperature of the current control body, the left control body of the current control body and the right control body of the current control body respectively;

P represents the current control body;

w represents a control body adjacent to the left side of the current control body;

E represents a control body adjacent to the right side of the current control body;

Δτ represents the calculated time step after the discretization;

Δx is the control body width;

A temperature value of the current control body representing the current time τ ₁;

A temperature value of the current control body indicating the previous time τ ₀;

(^-)^0-1 shows that the value takes the mean of the current time τ ₁ and the previous time τ ₀;

Representing the average value of the control body W adjacent to the left side of the current control body at the current time tau ₁ and the previous time tau ₀;

representing the average value of the control body E adjacent to the right side of the current control body at the current time tau ₁ and the previous time tau ₀;

When the time display format is adopted, the value of the previous time tau ₀ is adopted to replace the mean value, and the differential is adopted to replace the differential, so that a discrete equation is obtained:

Wherein, A temperature value representing the control body E adjacent to the right side of the current control body at the current time τ ₀;

A temperature value representing the control body W adjacent to the left side of the current control body at the current time τ ₀;

The distances from the center point of the current control point to the center points of the left control point and the right control point of δx _w、δx_e are calculated by the following formulas respectively;

δx_w＝|x_p-x_w|

δx_e＝|x_p-x_e|

Wherein x _p represents the position of the center point of the current control point;

x _w denotes the position of the center point W of the control W adjacent to the left of the current control point;

w represents the center point of the control W adjacent to the left side of the current control point;

x _e denotes the position of the center point E of the control E adjacent to the right of the current control point;

E represents the center point of control E adjacent to the right of the current control point;

for internal boundaries between materials, a third class of boundary conditions can be considered; calculating a boundary of one side of the control body includes:

Calculating the left boundary of the control body, wherein the boundary is obtained by equal heat flows on the interface, and the boundary is provided with:

Wherein h is a convective heat transfer coefficient;

T _s represents the structural surface temperature in contact with the fluid;

s represents the surface of the structure;

t _f is the fluid temperature;

f represents a fluid;

and obtaining a discrete equation by adopting a time display format:

Wherein, Representing the fluid temperature at the previous calculation time τ ₀;

Sorting the discrete equations to obtain an expression adopted by the explicit calculation program:

4. The improved adaptive optimization method for contact heat transfer coefficient identification according to claim 1, wherein the step S3 comprises:

For the contact heat exchange coefficient h _j,n of each interface n in each individual j in the population, judging whether the following conditions are met:

h_j,n＞0

if the contact heat exchange coefficient generated by the individual does not have the actual physical meaning, the heat conduction calculation model is considered as insoluble;

If the contact heat exchange coefficient generated by the individual has the actual physical meaning, the heat conduction calculation model is considered to be solvable;

for the resolvable individual j, the contact heat exchange coefficient of the individual is transmitted into the heat conduction positive problem calculation model established in the step S2, and the method comprises the following steps:

for the solvable individuals, the thermal conduction positive problem calculation model established in the step S2 has satisfied all the input conditions, and the thermal conduction positive problem calculation model is operated to obtain the temperature response value T _cal of all the nodes at each calculation time:

Wherein L is the total number of the measuring point nodes;

v is the total number of the calculated time points;

t _cal,l,t represents a temperature response calculated value of the measuring point node l at the time point T; wherein t=1, 2,. -%, V; l=1, 2,..l.

5. The improved adaptive optimization method for contact heat transfer coefficient identification according to claim 1, wherein the step S4 comprises:

Extracting temperature response T _cal,l of corresponding positions from a temperature response value matrix T _cal of all nodes given in the step S3 according to the measuring point positions corresponding to the temperature response measured values T _real,l,t given in the identification data;

respectively calculating the weight W _i,t of each measuring point in the adaptive value function for each selected measuring point;

The specific steps of calculating the weight W _i,t include:

step S4.1: for the temperature response data of each measuring point, calculating a temperature change rate dT _cal,i/dτ from the second moment;

wherein i is the extracted ith measuring point;

t is the number of the calculated time after the discretization;

In actual calculation, the following formula is adopted instead:

Wherein τ _t represents the time corresponding to the t-th calculation time after the discretization;

τ _t-1 represents the time corresponding to the t-1 th calculation time after the discretization;

step S4.2: summing the absolute values of the temperature change rates at each moment of each measuring point;

Step S4.3: for each moment of each measuring point, the weight of each moment of each measuring point in the process of calculating the adaptive value is given as follows:

Wherein w _i,t represents a weight;

If the substitution formula given in step S4.1 above is adopted, the weights are expressed as:

Wherein, T _cal,i,t-1 represents the temperature response calculated value of the i-th measuring point selected at the T-1-th calculating moment.

6. The improved adaptive optimization method for contact heat transfer coefficient identification according to claim 1, wherein the step S5 comprises:

According to the temperature response calculated value T _cal,l extracted in the step S4, the temperature response measured value T _real,l given by the identification data and the weight w _i,t of each measuring point at each calculating moment calculated in the step S4, and the fitness fit of the individual is calculated by carrying out square weighted summation on all the data of each measuring point and each moment, wherein the specific formula is as follows:

wherein i is the extracted ith measuring point;

t is the calculation time number, namely the calculation time point;

w _i,t represents the weight of the ith measuring point at the time t;

v is the number of times of calculation output.

7. The improved adaptive optimization method for contact heat transfer coefficient identification according to claim 1, wherein the step S6 comprises:

For all individuals in the population, calculating the average fitness

Wherein j is the number of the individual in the population;

J is the total number of individuals in the population;

fit _j represents fitness of the jth individual;

for all individuals in the population, searching for the minimum fitness and comparing with the average fitness, when:

then the optimization is considered to have been completed; otherwise, the optimization is considered to be incomplete;

Wherein fit _min represents the minimum fitness among all individuals within the population;

Δα _thres represents a threshold value for ending optimization;

for the situation that the optimization is completed, giving an individual corresponding to the fit _min, wherein the contact heat exchange coefficient h ₁,h₂,...,h_I corresponding to the individual is the contact heat exchange coefficient obtained by optimization aiming at the current input condition;

For the case where the optimization is not completed, the process proceeds to step S7.

8. The improved adaptive optimization method for contact heat transfer coefficient identification of claim 7, wherein said step S7 comprises:

For the current population, a crossover probability p _c is calculated:

Wherein p _max is the upper cross probability limit;

p _min is the crossover probability lower bound;

r _max is the corresponding percentage when the upper crossover probability limit is reached;

r _min is the corresponding percentage when the lower crossover probability limit is reached;

Δα ₀ is the difference between the minimum and average population fitness values at the first iteration;

Δα is the difference between the minimum and average population fitness values at the current iteration number, expressed as:

for the current population, a variation probability p _m is calculated:

p_m＝1-p_e-p_c

wherein p _e is the elite individual proportion, i.e. the elite individual ratio in the population, for elite individuals, no crossing and variant individuals are involved in the iteration.

9. The improved adaptive optimization method for contact heat transfer coefficient identification of claim 8, wherein for the current population, multiplying the probability by the number of individuals in the population to obtain the number of individuals remaining, crossing, and varying in the next generation;

The method for determining elite individuals comprises the following steps: the individuals in the population are arranged according to the ascending order of fitness, and the individuals are sequentially selected from front to back until the number of reserved individuals is reached;

The method for determining the crossed individuals comprises the following steps: randomly extracting two individuals from the population each time to cross to obtain a crossed individual until the number of crossed individuals is reached;

The method for determining variant individuals comprises the following steps: randomly extracting individuals from the population, and randomly mutating the extracted individuals until the number of mutated individuals is reached;

And (3) forming a matrix by newly generated reserved individuals, crossed individuals and contact heat exchange coefficients corresponding to variant individuals, namely forming a new generation population, and repeating the steps S3-S6.

10. An improved adaptive optimization system for contact heat exchange coefficient identification, comprising:

Module M1: according to the number of the contact heat exchange coefficients to be identified, generating an initial population containing a plurality of individuals at random, wherein each individual contains a group of contact heat exchange coefficients [ h ₁,h₂,...,h_N ];

Module M2: dividing grids based on a structural geometric model needing to be subjected to contact heat exchange coefficient identification, and defining material thermophysical parameters of the grids;

setting initial conditions and boundary conditions for a thermal conduction calculation model, and establishing a thermal conduction calculation model based on a finite volume method;

module M3: for each individual in the population in the module M1, judging whether the thermal conduction calculation model in the module M2 is solvable, if so, calculating the thermal conduction calculation model to obtain a temperature response T _cal of each position of the structure changing along with time;

For individuals in which the thermal conduction calculation model is insoluble, assign fitness fit to a default value and jump to module M6;

Module M4: for each individual in the population in the module M1, selecting a temperature response calculated value T _cal,l of a specified position I in the structure, and calculating the temperature change rate of the measuring point at each calculation moment to obtain a weight w _i,t of each measuring point i and each calculation moment T based on the temperature change rate;

Module M5: for each individual in the population in the module M1, based on the weight w _i,t of each calculation time T obtained by the module M4 and the temperature response calculation value T _cal,l,t corresponding to the time, carrying out heat conduction analysis by taking the finally obtained new generation population as a final contact heat exchange coefficient, and calculating the fitness fit of the individual according to the corresponding position actual measurement value T _real,l,t given by the identification data;

Module M6: evaluating the fitness of all individuals in the population, and if the difference value between the average fitness and the minimum fitness of the individuals in the population is smaller than a threshold value, completing optimization;

Taking the contact heat exchange coefficient contained in the individual with the minimum fitness in the current population as the solution of the identification, otherwise, continuing;

Module M7: and calculating the values of the crossover probability p _c and the mutation probability p _m according to the iteration times of the current population and the self-adaptive parameters, selecting individuals from the current population to reserve, crossover and mutation, producing a new generation population, and repeating the modules M3-M6.