Heat exchanger optimization design system based on genetic algorithm

Abstract

The invention belongs to the field of heat exchange equipment design, particularly relates to a heat exchanger optimization design system based on a genetic algorithm, and aims at solving the problem that heat exchange equipment designed by an existing design system for designing a shell-and-tube heat exchanger is low in performance. The heat exchanger optimization design system based on the genetic algorithm comprises a design index input module and a heat exchanger parameter optimization module. According to the system, seven key design parameters including the tube pass number, the tube outer diameter, the heat exchange tube length, the tube inner flow speed, the shell side flow speed, the tube wall thickness and the heat exchange tube number serve as optimization variables, and a fitness function with the heat transfer coefficient and pressure drop ratio as the target is established. A hybrid coding strategy is introduced to carry out differentiation processing on continuous parameters and integer parameters, an improved genetic algorithm is adopted to realize iterative evolution of a population through intelligent optimization operations such as coding, selection, crossover and variation, and finally, a global optimal parameter combination meeting engineering constraint conditions is converged.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of heat exchanger design, and more particularly to a heat exchanger optimization design system based on a genetic algorithm. BACKGROUND

[0002] As one of the most widely used devices in industrial heat transfer systems, the optimization degree of the design parameters of the shell-and-tube heat exchanger directly determines the heat transfer efficiency, energy consumption and operation economy of the entire system. Under the background of the rapid development of current industrial 4.0 and intelligent manufacturing, higher requirements are put forward for the precision, efficiency and reliability of heat exchanger design, and the traditional design method based on empirical formula and manual calculation has been difficult to meet the technical needs of modern industry. The existing technology analysis shows that the current heat exchanger design method has the following technical defects:

[0003] (1) The traditional heat exchanger design method mainly relies on the experience judgment and manual calculation of the design engineer, and the entire design process needs to be iteratively optimized in multiple links such as heat transfer coefficient calculation, pressure drop checking and structural parameter determination. A single design task usually takes several weeks or even months, and the design efficiency is extremely low.

[0004] (2) In order to simplify the calculation complexity, the traditional design method usually adopts many idealized assumptions and simplified thermodynamic models, such as ignoring the three-dimensional flow characteristics of the shell-side fluid, the mutual influence effect between the tube bundles, and the flow distribution unevenness caused by the baffle plate. These simplifications result in a significant deviation of 15%-25% between the theoretical calculation results and the actual operating performance, which is difficult to meet the technical requirements of high-precision engineering design, especially in the fields of nuclear energy, aerospace and other fields with extremely high safety and reliability requirements.

[0005] (3) In terms of optimization algorithm, the traditional method usually adopts single-parameter optimization or simple parameter scanning strategy. This method has obvious insufficient ability when dealing with the complex optimization problems such as multi-variable strong coupling, strong nonlinearity and multiple constraints which are common in heat exchanger design. When facing the combination optimization of multiple design variables such as tube number, outer diameter, length, flow rate and wall thickness, the traditional algorithm is easy to converge to a local optimal solution and cannot obtain a globally optimal design scheme, thereby limiting the full excavation of the performance potential of the heat exchanger.

[0006] The above-mentioned technical problems seriously affect the efficiency and quality of heat exchanger design, resulting in the performance of the existing design shell-and-tube heat exchanger design system being poor SUMMARY

[0007] The purpose of the present application is to solve the problem of low performance of the existing design shell-and-tube heat exchanger design system. A heat exchanger optimization design system based on a genetic algorithm is provided, which comprises:

[0008] The design index input module and the heat exchanger parameter optimization module are designed.

[0009] The design index input module is used for inputting heat exchanger design necessary parameters.

[0010] The heat exchanger design necessary parameters include thermal design necessary parameter values, optimization parameters, and optimization parameter ranges.

[0011] The design index input module is also used for generating an initial population according to the optimization parameters.

[0012] The heat exchanger thermal calculation module is used for obtaining optimal heat exchanger parameters by using a genetic algorithm optimization according to the initial population and the thermal design necessary parameter values.

[0013] The thermal design necessary parameters include shell-side working medium flow Ms, shell-side fluid inlet temperature Ts,i, shell-side fluid outlet temperature Ts,o, shell-side fluid inlet pressure Ps,i, tube-side fluid inlet temperature Tt,i, tube-side fluid outlet temperature Tw,o, tube-side fluid inlet pressure Pw,i, and heat transfer tube thermal conductivity λ w These input variables are known data in subsequent calculations, and are fixed values according to actual design requirements.

[0014] The optimization parameters include tube pass number Z t , tube outer diameter d o , heat transfer tube length L, tube-side flow velocity w t , shell-side flow velocity w s , tube wall thickness δ w , and heat transfer tube number N.

[0015] The initial population is generated according to the heat exchanger design necessary parameters, and the specific process is as follows.

[0016] Step A1: An encoding matrix is generated for each optimization parameter. The specific process of generating an encoding matrix for one of the optimization parameters is as follows.

[0017] T parameter values are randomly generated within the optimization parameter range of the optimization parameter, and the T parameter values are binary coded to obtain T binary sequences of the optimization parameter. The T binary sequences are combined in rows to form an encoding matrix, where T is a positive integer.

[0018] Each optimization parameter is composed of a binary number with a specified number of coding bits, and the number of coding bits is positively correlated with the parameter accuracy (that is, the higher the accuracy of the parameter, the greater the number of coding bits). The generated binary sequence is stored as a data matrix, and the matrix rows correspond to individuals and the columns correspond to parameter coding, as shown in Figure 2 .

[0019] Step A2: Latin hypercube sampling method is used to process the encoding matrix of each optimization parameter to generate M initial individuals, and the gene structure of each individual is composed of the binary coding of each optimization parameter, wherein M is a positive integer;

[0020] An initial population is generated based on the M initial individuals, the population size is set to M, and the maximum number of iterations is set to Ngmax

[0021] The gene structure of the individual is represented by the formula: [X=[x1, x2, x3, x4, x5, x6, x7]],

[0022] x1 represents the binary coding of the number of tube passes Z t x2 represents the binary coding of the outer diameter of the tube d o x3 represents the binary coding of the length of the heat exchange tube L; x4 represents the binary coding of the flow rate w t x5 represents the binary coding of the shell side flow rate w s x6 represents the binary coding of the tube wall thickness δ w x7 represents the binary coding of the number of heat exchange tubes N;

[0023] Each individual is a series of binary coding of the parameters. The parameter settings include the population size M and the maximum number of iterations, each of which represents a set of optimization parameters. The variables in the gene structure of the individual are treated as known quantities in the subsequent optimization algorithm, and the values are obtained by extraction and processing for calculation

[0024] The heat exchanger thermal calculation module uses genetic algorithm optimization to obtain the optimal heat exchanger parameters according to the initial population and the necessary parameter values of thermal design, and the specific process is as follows:

[0025] S1: Decode each individual in the initial population into real parameter values, and calculate the fitness value of each individual in the initial population according to the real parameter values and the necessary parameter values of thermal design;

[0026] S2: Arrange each individual in the initial population in ascending order according to the fitness value;

[0027] S3: According to the arrangement order of each individual in the initial population, perform mixed selection processing on the initial population to select M1 elite individuals; form an elite population;

[0028] S4: Perform cross genetic processing and mutation genetic processing on the elite population in turn to obtain M2 offspring individuals; form an offspring population;

[0029] S5: Calculate the fitness value of each offspring individual in the offspring population; the offspring individual with the highest fitness value is taken as the optimal offspring individual;

[0030] S6: Set the stopping condition and judge whether the stopping condition is met according to the fitness value of the optimal offspring individual.

[0031] When the stopping condition is met, the optimal offspring individual is used as the output individual; enter S7;

[0032] When the stopping condition is not met, the offspring population is used as the initial population for the next iteration and returns to S2;

[0033] S7: Decode the output individual into the optimal optimization parameters, and calculate the optimal heat exchanger parameters based on the optimal optimization parameters. The beneficial effects of the present invention are:

[0034] 1. This invention overcomes the limitations of traditional design methods, which rely on empirical formulas and manual calculations, by deeply coupling a genetic algorithm with the heat exchanger's thermal model, significantly shortening the design cycle. Furthermore, the genetic algorithm's global search capability ensures multi-parameter collaborative optimization, effectively avoiding the inherent drawback of traditional optimization methods, which are prone to falling into local optimal solutions. This significantly enhances the global optimization capabilities of heat exchanger design.

[0035] 2. This invention adopts a high-precision thermal model, fully considering the dynamic effects of the shell-side fluid (such as leakage, bypass and other complex flow characteristics), avoiding the design deviation caused by the traditional method due to model simplification. The optimization results strictly meet the industrial constraints (tube side pressure drop ≤ 50kPa, shell side pressure drop ≤ 100kPa), and improve the overall heat transfer coefficient by up to 1000W / (m 2 ·℃) and above, ensuring the reliability of the design results and the feasibility of practical application.

[0036] 3. This invention introduces dynamic adjustment mechanisms (such as adaptive adjustment of the elite retention ratio and mutation probability) to adjust algorithm parameters in real time based on the convergence state during the optimization process, significantly improving the algorithm's convergence efficiency and enhancing its ability to escape local optimal solutions. Furthermore, by constructing a fitness function based on the comprehensive efficiency ratio (the ratio of the sum of the heat transfer coefficient and the pressure drop), it maximizes the combined effectiveness of heat transfer efficiency and flow resistance, providing solid technical support for efficient thermal design in complex industrial scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a flow chart of a thermal design method for a shell and tube heat exchanger based on genetic algorithm optimization according to an embodiment of the present invention;

[0038] Figure 2 is a schematic diagram of coding individual parameters of a heat exchanger according to an embodiment of the present invention;

[0039] Figure 3 is a calculation process of the individual adaptation value of the heat exchanger according to an embodiment of the present invention;

[0040] Figure 4 is a heat exchanger individual sorting and selection schematic diagram according to an embodiment of the present application;

[0041] Figure 5 is a crossover genetic schematic diagram of a genetic algorithm according to an embodiment of the present application;

[0042] Figure 6 is a mutation genetic schematic diagram of a genetic algorithm according to an embodiment of the present application. DETAILED DESCRIPTION

[0043] Specific implementation one: combined with Figures 1-6 To illustrate the present application, the present application comprises:

[0044] a design index input module and a heat exchanger parameter optimization module;

[0045] The design index input module is used for inputting heat exchanger design necessary parameters, and generating an initial population according to the heat exchanger design necessary parameters;

[0046] The heat exchanger thermal calculation module is used for obtaining optimal heat exchanger parameters by using a genetic algorithm optimization according to the initial population.

[0047] Specific implementation two: the difference between the present embodiment and specific implementation one is that,

[0048] The heat exchanger design necessary parameters include thermal design necessary parameters and optimization parameter ranges;

[0049] The thermal design necessary parameters include shell side working medium flow Ms, shell side fluid inlet temperature Ts, i, shell side fluid outlet temperature Ts, o, shell side fluid inlet pressure Ps, i, tube side fluid inlet temperature Tt, i, tube side fluid outlet temperature Tw, o, tube side fluid inlet pressure Pw, i, heat transfer tube thermal conductivity λ w ; these input variables are known data in subsequent calculation; are fixed values according to actual design requirements;

[0050] The optimization parameters include tube pass number Z t , tube outer diameter d o , heat transfer tube length L, tube internal flow rate w t , shell side flow rate w s , tube wall thickness δ w , and heat transfer tube number N.

[0051] The optimization parameter range includes tube pass number range, Z t ∈ [Z t，min , Z t，max ]; Z t，min represents the minimum value of tube pass, and Z t，max represents the maximum value of tube pass.

[0052] Tube outer diameter range, d o ∈[d o，min , d o，max ];d o，min Indicates the minimum outer diameter of the tube, d o，max Indicates the maximum outer diameter of the tube

[0053] Heat exchange tube length range, L∈[L min , L max ]; L min Indicates the minimum length of the heat exchange tube, L max Indicates the maximum length of the heat exchange tube

[0054] Flow rate range in the tube, w t ∈[w t，min , w t，max ];w t，min Indicates the minimum flow velocity in the pipe, w t，max Indicates the maximum flow velocity in the pipe

[0055] Shell side velocity range, w s ∈[w s，min , w s，max ];w s，min Indicates the minimum shell side velocity, w s，max Indicates the maximum shell side velocity

[0056] Pipe wall thickness range, δ w ∈[δ w，min , δ w，max ];δ w，min Indicates the minimum wall thickness, δ w，max Indicates the maximum wall thickness

[0057] The range of the number of heat exchange tubes, N∈[N min , N max ]; N min Indicates the minimum number of heat exchange tubes, N max Indicates the maximum number of heat exchange tubes

[0058] Expressed in a table:

[0059]

[0060]

[0061] Other steps and parameters are the same as those in the first embodiment.

[0062] Specific embodiment three: This embodiment differs from specific embodiment one in that:

[0063] The initial population is generated according to the necessary design parameters of the heat exchanger. The specific process is as follows:

[0064] Step A1: Generate a coding matrix for each optimized parameter. The specific process of generating a coding matrix for one of the optimized parameters is as follows:

[0065] Randomly generating T parameter values ​​within an optimization parameter range of the optimization parameter, and performing binary encoding on the T parameter values ​​to obtain T binary sequences of the optimization parameter; combining the T binary sequences row by row to form a coding matrix, where T is a positive integer;

[0066] Each optimized parameter is composed of a binary number with a specified number of encoding bits. The number of encoding bits is positively correlated with the parameter accuracy (i.e., the higher the accuracy of the parameter, the larger the number of encoding bits). The generated binary sequence is stored as a data matrix, where the rows correspond to individuals and the columns correspond to parameter encodings, such as Figure 2 shown.

[0067] Step A2: Use the Latin hypercube sampling method to process the encoding matrix of each optimization parameter to generate M initial individuals. The genetic structure of each individual is composed of the binary encoding of each optimization parameter in series, where M is a positive integer.

[0068] Generate an initial population based on the M initial individuals, set the population size to M, and the maximum number of iterations to Ngmax

[0069] The individual gene structure is expressed as: [X = [x1, x2, x3, x4, x5, x6, x7]],

[0070] x1 represents the number of tube passes Z t The binary code of x2 represents the outer diameter of the tube d o x3 represents the binary code of the heat exchange tube length L; x4 represents the flow velocity w in the tube t Binary code; x5 represents the shell side flow velocity w s The binary code of x6 represents the wall thickness δ w The binary code of x7 represents the binary code of the number of heat exchange tubes N;

[0071] Each individual is a binary-encoded concatenation of various parameters. Parameter settings include the population size M and the maximum number of iterations, each representing a set of optimization parameters. The variables in the individual's genetic structure are treated as known quantities in the subsequent optimization algorithm, and the values ​​obtained through extraction and processing are used for calculation. The other steps and parameters are the same as those in Specific Embodiments 1-2.

[0072] Specific embodiment 4: This embodiment differs from specific embodiments 1 to 4 in that:

[0073] In the heat exchanger thermal calculation module, the optimal heat exchanger parameters are obtained by using a genetic algorithm based on the initial population. The specific process is as follows:

[0074] S1: decode each individual in the initial population into real parameter values, calculate the fitness value of each individual in the initial population according to the real parameter values,

[0075] S2: arrange each individual in the initial population in ascending order according to the fitness value;

[0076] S3: according to the arrangement order of each individual in the initial population, perform mixed selection processing on the initial population to select M1 elite individuals; form an elite population;

[0077] S4: perform cross genetic processing and mutation genetic processing on the elite population in turn to obtain M2 offspring individuals; form an offspring population;

[0078] S5: calculate the fitness value of each offspring individual in the offspring population; the offspring individual with the highest fitness value is taken as the optimal offspring individual;

[0079] S6: determine whether the stop condition is met according to the fitness value of the optimal offspring individual,

[0080] When the stop condition is met, the optimal offspring individual is taken as the output individual; enter S7;

[0081] When the stop condition is not met, the offspring population is taken as the initial population of the next round of iteration, and returns to S2;

[0082] S7: decode the output individual into optimal optimization parameters, and calculate the optimal heat exchanger parameters according to the optimal optimization parameters;

[0083] The other steps and parameters are the same as one of the first to third embodiments.

[0084] Embodiment five: the difference between this embodiment and the first to fourth embodiments is that,

[0085] In S1, the x i th variable of each individual is decoded into the real parameter value of the ith optimization parameter, and different decoding formulas are used according to the parameter type:

[0086] For continuous parameters:

[0087]

[0088] For integer parameters:

[0089]

[0090] Where, x i represents the real value decoded from the ith optimization parameter; a i is the minimum value of the ith optimization parameter, that is, the lower boundary of the parameter value range; bi represents the maximum value of the i-th optimization parameter, i.e. the upper boundary of the parameter value range; n i represents the number of coding bits of the i-th optimization parameter; v i represents the decimal value of the binary gene segment corresponding to the i-th optimization parameter; represents a floor function; wherein, continuous parameters include: pipe outer diameter d o , heat exchange pipe length L, pipe internal flow rate w t , shell side flow rate w s , pipe wall thickness δ w ; integer parameters include: pipe pass number Z t , heat exchange pipe number N;

[0091] The fitness value of a certain individual in the initial population in S1 is calculated by the formula:

[0092]

[0093] In the formula, K represents the overall heat exchange coefficient of the heat exchanger; ΔP t represents the pipe side pressure drop, ΔP s represents the shell side pressure drop;

[0094] The other steps and parameters are the same as one of the first four embodiments.

[0095] Embodiment six: the difference between this embodiment and the first to fifth embodiments is that,

[0096] The formula for calculating the overall heat exchange coefficient K of the heat exchanger is:

[0097]

[0098] In the formula, α t represents the pipe internal flow heat exchange coefficient; α s represents the pipe external shell internal heat exchange coefficient; r s，i represents the pipe internal fouling thermal resistance, r s，o represents the pipe external fouling thermal resistance, r s，i The fouling thermal resistance value is obtained by querying the fouling coefficient table, and the thermal resistance r s，o corresponding to seawater is obtained by querying the fouling coefficient table. Here, the calculation process is considered as a known quantity; λ w represents the thermal conductivity coefficient of the pipe wall material, and the thermal conductivity coefficient λ w of the pipe wall material is an input variable and belongs to a known quantity; d o represents the pipe outer diameter, δ w represents the pipe wall thickness; d o , δ w is obtained according to individual decoding, d i represents the pipe inner diameter; d iThe value of the tube inside diameter is based on the tube outside diameter d o Subtracting twice the tube wall thickness δ w Yields; this is a well known calculation;

[0099] The tube side heat transfer coefficient α t Is calculated by:

[0100] Nu t = 0.023Re t 0.8 Pr t n

[0101]

[0102] Where Nu t Is the tube side Nusselt number; Re t Is the tube side Reynolds number; Pr t Is the tube side Prandtl number; λ t Is the tube side fluid thermal conductivity; and n is an intermediate variable,

[0103] n = 0.4 when the fluid is being heated and n = 0.3 when the fluid is being cooled;

[0104] The shell side heat transfer coefficient α s Is calculated by:

[0105]

[0106] Where j H Is the baffle window correction factor, j c Is the bypass correction factor, j l Is the leakage correction factor, j b Is the tube bundle stagger correction factor, these correction factors are based on experimental data and engineering experience and are found in the TEMA standards; Pr s Is the shell side Prandtl number; λ s Is the shell side fluid thermal conductivity; μ s Is the shell side fluid dynamic viscosity, μ w Is the shell side fluid dynamic viscosity at the shell wall; the fluid dynamic viscosity can be found by looking up the fluid dynamic viscosity at the fluid temperature and pressure.

[0107] Are well known to those skilled in the art;

[0108] The tube side pressure drop ΔP t Is calculated by:

[0109]

[0110] Where Δpi is the frictional resistance along the flow path; Δp r is the return bend resistance; is the viscosity correction factor, f i represents the friction factor of the tube, ρ t represents the fluid density on the tube side, all of which are known quantities;

[0111] w t represents the flow velocity of the fluid on the tube side; L represents the length of the heat exchange tube, Z t represents the temperature of the tube side, which is obtained according to individual decoding;

[0112] when Re≥2100, Re represents the Reynolds number of the fluid in the tube; the definition and calculation of the Reynolds number are well known to those skilled in the art

[0113] when Re<2100, μ f represents the dynamic viscosity of the fluid on the tube side, μ w represents the dynamic viscosity of the fluid on the tube side wall

[0114] The dynamic viscosity can be obtained according to the temperature and pressure of the fluid, which is well known to those skilled in the art;

[0115] The shell side pressure drop ΔP s is calculated by the following formula:

[0116] When no baffle is used and the smooth heat exchange tube is a parallel tube bundle, the shell side resistance is calculated by the following formula:

[0117]

[0118] In the formula, N represents the number of heat exchange tubes; w s represents the flow velocity on the shell side; N represents the number of heat exchange tubes; μ represents the dynamic viscosity of the fluid on the shell side; ρ represents the fluid density; and the number of heat exchange tubes N is obtained according to individual decoding; w represents the dynamic viscosity of the fluid on the shell side wall; and ρ represents the fluid density; the number of heat exchange tubes N is obtained according to individual decoding;

[0119] When no baffle is used and the smooth heat exchange tube is a parallel tube bundle, the shell side resistance is calculated by the following formula:

[0120]

[0121] In the formula, w s represents the flow velocity on the shell side, m / s; N represents the number of heat exchange tubes; μ represents the dynamic viscosity of the fluid on the shell side; and ρ represents the fluid density; the other steps and parameters are the same as one of the first to fifth embodiments.

[0122] Embodiment seven: The difference between this embodiment and the first to sixth embodiments is that,

[0123] In S3, the initial population is subjected to mixed selection according to the arrangement order of each individual in the initial population, and M1 elite individuals are selected to form an elite population. The specific process is as follows:

[0124] S3.1. According to the ranking order of each individual in the initial population, select the top m1 individuals as elite individuals;

[0125] The number of m1 selections is set to 5% of the total by default. If the optimal fitness does not improve for 20 consecutive generations, the elite retention ratio will be increased to 15%.

[0126] S3.2. Perform roulette wheel selection on the individuals not selected in S3.1, and select m2 individuals as elite individuals, ultimately obtaining M1 elite individuals; forming an elite population; M1 = m1 + m2;

[0127] The roulette wheel selection uses the fitness value of the individual as the roulette wheel share. The larger the fitness value of the individual, the larger its roulette wheel share and the greater the probability of being selected as the next generation. When the population size is M-m1, the fitness of individual i is F i , the probability of being selected is calculated as follows:

[0128]

[0129] Where, F i is the fitness of individual i, M-m1 is the remaining population size, the higher the fitness, the greater the probability of selection.

[0130] The other steps and parameters are the same as those in the first to sixth embodiments.

[0131] Specific embodiment eight: This embodiment differs from specific embodiments one to seven in that:

[0132] In S4, crossover genetic processing and variation genetic processing are sequentially performed on the elite population to obtain M2 offspring individuals, which form an offspring population. The specific process is as follows:

[0133] S4.1: Set the crossover probability p c , according to the crossover probability p c Perform cross-genetic operations on the elite population to obtain individuals after cross-genetic operation;

[0134] In S4.1, two elite individuals in the elite population undergo a crossover genetic operation to obtain two individuals after crossover genetics. The specific process is as follows:

[0135] S4.1: Randomly exchange the coding of the structure parameters of two elite individuals at a random cross point with a random length, to generate two new individuals with new corresponding heat exchanger structure parameters, as two new heat exchanger individuals

[0136] The two elite individuals subjected to the cross genetic process are separated by one position, i.e. the latter individual 1 and individual 3.

[0137] According to the cross probability p c = 0.6, two cross genetic individuals are obtained and passed to the next generation of heat exchanger population, as shown in Figure 5

[0138] Further, if the optimal fitness is not improved for 20 consecutive generations, the cross probability is increased to p c = 0.85 to promote population diversity. The individuals are sorted according to the fitness,

[0139] The two elite individuals subjected to the cross genetic process are separated by one position, i.e. the latter individual 1 and individual 3.

[0140] S4.2: Set the mutation probability p m According to the mutation probability p m , the cross genetic individuals are subjected to mutation genetic operation to obtain M2 offspring individuals;

[0141] The specific process of the mutation genetic operation is that the random position of the coding of the structure parameters of the heat exchanger individual is subjected to mutation.

[0142] The initial mutation probability is set to p m = 0.02, and the random position of the coding of the structure parameters of the heat exchanger individual is randomly subjected to mutation (from 1 to 0, or from 0 to 1) to maintain population stability, as shown in Figure 6

[0143] Mutation probability adjustment, if the optimal fitness is not improved for 20 consecutive generations, the mutation probability is dynamically increased to p m = 0.05 to break the local optimum, so that the corresponding structure parameter value is changed to obtain a new heat exchanger individual

[0144] The other steps and parameters are the same as one of embodiments 1 to 7.

[0145] Embodiment 9: The difference between this embodiment and embodiments 1 to 8 is that

[0146] The stop condition in S6 is expressed by a formula: ​​

[0147] |F gen -F gen-100 / F gen |<1%

[0148] Where, F gen represents the optimal fitness; F gen-100 Indicates the optimal fitness within 100 iterations;

[0149] That is, if the optimal fitness changes by less than 1% for 100 consecutive generations, the iteration is terminated and the optimal parameters are output.

[0150] The other steps and parameters are the same as those in the first to eighth embodiments.

[0151] Specific embodiment 10: This embodiment differs from specific embodiments 1 to 9 in that:

[0152] In S7, the output individual is decoded into the optimal optimization parameters, and the optimal heat exchanger parameters are calculated based on the optimal optimization parameters. The specific process is as follows:

[0153] S7.1: Calculate the working fluid flow rate M on the tube side based on the optimal optimization parameters t , heat exchange area A of the heat exchanger, and overall heat transfer coefficient K of the heat exchanger, which can be expressed as follows:

[0154]

[0155] Where: ΔT s Expressed as the shell side inlet and outlet temperature difference (known quantity), ΔT t represents the inlet and outlet temperature difference of the tube side (known quantity), °C; It is expressed as the shell-side and tube-side fluid constant pressure specific heat, kJ / (kgK) (known quantity); K is expressed as the overall heat transfer coefficient of the heat exchanger (unknown quantity, needs to be calculated, the calculation method has been disclosed in S1 and will not be introduced here), W / (m 2 ·K); A represents the heat exchange area of ​​the heat exchanger (unknown quantity, needs to be calculated), m 2 ; ΔT m Expressed as heat transfer logarithmic temperature difference, ℃ (the following formula can be calculated); M s Indicates the shell side working medium flow rate (known quantity), (kg / s); M t Tube side working medium flow rate (unknown quantity, can be calculated using the above formula)

[0156] The heat exchange logarithmic temperature difference ΔT m The calculation formula is expressed as:

[0157]

[0158] Where: T s，aveIndicated as shell side fluid average temperature (known quantity), °C; T s，i Indicated as shell side fluid inlet temperature (known quantity); T s，o Indicated as shell side fluid outlet temperature (known quantity); T t，i Indicated as tube side fluid inlet temperature (known quantity); T t，o Indicated as tube side fluid outlet temperature (known quantity) T t，ave Indicated as tube side fluid average temperature, °C (known quantity); ΔT m Indicated as qualitative temperature, °C (here can be calculated quantity, give above use);

[0159] S7.2: According to tube side working medium flow M t , heat exchanger heat transfer area A to calculate heat exchanger structure parameters;

[0160] The heat exchanger structure parameters include: tube side structure parameters and shell side structure parameters;

[0161] S7.2.1. Calculate tube side structure parameters, the tube side structure parameters include: tube flow area A t , tube number Z t , total number of heat exchange tubes N t , heat exchange tube spacing P t , the specific process is:

[0162] (1) The tube flow area calculation formula is:

[0163] M t = ρ t · A t · w t

[0164]

[0165] In the formula: M t , M s Indicated as tube side, shell side mass flow (M t , M s Known quantity), kg / s; ρ t Indicated as tube side fluid density (known quantity), kg / m 3 ; A t Indicated as total heat exchange tube flow cross-sectional area (unknown quantity, At belongs to intermediate variable, variable is wt and di and Nt, wherein (di = do-2 tube wall thickness), ), m 2 ; N indicates the number of heat exchange tubes (unknown quantity; w t Indicated as tube side fluid flow velocity, m / s (unknown quantity;

[0166] (4) The tube number calculation formula is:

[0167]

[0168] In the formula, L represents the length of the heat exchange tube, m; I represents the ratio of the length of the heat exchange tube to the diameter of the shell, I is generally an integer, and the provisions in GB1512014 are adopted for selection. Generally, the ratio of the length of the heat exchange tube to the diameter of the shell of a horizontal tube-shell heat exchanger is between 6 and 10, and that of a vertical tube-shell heat exchanger is between 4 and 6.

[0169] (5) The total number of heat exchange tubes is calculated according to the formula:

[0170] N t = Z t · N

[0171] In the formula, N t represents the total number of heat exchange tubes; Z t represents the number of tube passes, generally Z t is between 1 and 4; and N represents the number of heat exchange tubes.

[0172] (6) The tube spacing is calculated according to the formula:

[0173] P t ≥ d o

[0174] In the formula, P t represents the tube spacing; and do represents the outer diameter of the heat exchange tube.

[0175] S7.2.2. Calculate the shell side structure parameters, including the shell side inner diameter D s , the baffle diameter D b , the baffle spacing I s , the number of baffles N b , and the specific process is as follows:

[0176] (1) The shell side inner diameter is calculated according to the formula:

[0177] D s = (b-1) d o +2b′

[0178] In the formula, b′ represents the distance from the center of the outermost tube on the center line of the tube bundle to the inner wall of the shell, generally b′ = (1-1.5) d o ; b represents the number of tubes along the diagonal of the hexagon, and the tubes are arranged in a triangular shape. The tubes are arranged in a square shape. N t represents the total number of heat exchange tubes; and D s should be an integer.

[0179] (2) The baffle diameter is calculated according to the formula:

[0180] The baffle diameter is generally larger than the shell inner diameter Ds Small gap to ensure that the baffle can be smoothly installed in the shell.

[0181] D b = D s - 2 x δ

[0182] In the formula, D b is expressed as the baffle diameter, and δ gap is generally 5-8 mm;

[0183] (3) The baffle spacing calculation formula is:

[0184]

[0185] In the formula, l s is expressed as the baffle spacing, m; M s is expressed as the mass flow rate of the shell-side fluid, kg / s; p s is expressed as the density of the shell-side fluid, kg / m 3 ; w s is expressed as the flow rate of the shell-side fluid, m / s; D s is expressed as the shell inside diameter, m; N t is expressed as the total number of heat exchange tubes; d o is expressed as the heat exchange tube outer diameter, m;

[0186] (4) The baffle number calculation formula is:

[0187]

[0188] In the formula, N b is expressed as the number of baffles; L is expressed as the heat exchange tube length, m; l s is expressed as the baffle spacing, m;

[0189] The final output optimization parameters include:

[0190] Optimal geometric parameters: heat exchange tube outer diameter Heat exchange tube length L * , heat exchanger tube number Heat exchange tube wall thickness Heat exchange tube spacing Shell diameter Baffle diameter Baffle spacing Baffle number , number of heat exchange tubes

[0191] Flow parameters: heat exchange tube internal flow rate

[0192] Performance indicators: overall heat transfer coefficient K * , tube side resistance shell side resistance is expressed by a table as follows:

[0193]

[0194]

[0195] Other steps and parameters are the same as one of the first nine embodiments.

[0196] Simulation analysis combined with the first ten embodiments

[0197] The application discloses a heat exchanger optimization design system based on a genetic algorithm, comprising a design index input module and a heat exchanger parameter optimization module. The system takes seven key design parameters, including the number of tube passes, the outer diameter of the tube, the length of the heat exchange tube, the flow rate in the tube, the flow rate on the shell side, the thickness of the tube wall and the number of heat exchange tubes, as optimization variables, and establishes a fitness function with the heat transfer coefficient and the pressure drop ratio as the target. By introducing a mixed coding strategy, the continuous parameters and integer parameters are differentiated, and an improved genetic algorithm is used to realize the iterative evolution of the population through coding, selection, crossover and mutation and other intelligent optimization operations, and finally converges to the global optimal parameter combination that meets the engineering constraint conditions.

[0198] The system integrates a high-precision heat and mass transfer model, fully considers the complex flow characteristics of the shell side fluid and the engineering constraint conditions, and effectively solves the precision loss problem caused by model simplification in traditional methods. The application significantly improves the heat exchanger performance optimization efficiency and global search ability, and is suitable for efficient thermal design in complex industrial scenes.

[0199] The above is only the preferred embodiment of the application, and it should be understood that the application is not limited to the above specific embodiments. Although the application has been disclosed as above with the preferred embodiment, it is not intended to limit the application. Any person skilled in the art can make some changes or modifications to the disclosed technical content without departing from the scope of the technical solution of the application, and make equivalent embodiments with equivalent changes, as long as they do not depart from the technical solution of the application, and the technical essence of the application is within the spirit and principles of the application. Any simple modification, equivalent replacement and improvement of the above embodiments are still within the protection scope of the technical solution of the application.

Claims

1. A heat exchanger optimization design system based on genetic algorithm, characterized in that: include: Design index input module, heat exchanger parameter optimization module; The design index input module is used to input necessary parameters for heat exchanger design; The necessary parameters for heat exchanger design include: necessary thermal design parameter values, optimization parameters and optimization parameter ranges; The design index input module is also used to generate an initial population according to the optimization parameters; The heat exchanger thermal calculation module is used to optimize the optimal heat exchanger parameters using a genetic algorithm based on the initial population and the necessary parameter values ​​for thermal design.

2. The heat exchanger optimization design system based on genetic algorithm according to claim 1, characterized in that: The necessary parameters for thermal design include: shell side working fluid flow Ms, shell side fluid inlet temperature Ts,i, shell side fluid outlet temperature Ts,o, shell side fluid inlet pressure Ps,i, tube side fluid inlet temperature Tt,i, tube side fluid outlet temperature Tw,o, tube side fluid inlet pressure Pw,i, heat transfer tube thermal conductivity λ w ; The optimization parameters include: the number of pipe passes Z t , tube outer diameter d o , heat exchange tube length L, flow velocity in the tube w t , shell side velocity w s , tube wall thickness δ w , the number of heat exchange tubes N; The optimization parameter range includes: the range of tube pass number, Z t ∈[Z t，min , Z t，max ]; Z t，min Indicates the minimum value of the tube, Z t，max Indicates the maximum value of the pipe; Tube outer diameter range, d o ∈[d o，min , d o，max ];d o，min Indicates the minimum outer diameter of the tube, d o，max Indicates the maximum outer diameter of the tube Heat exchange tube length range, L∈[L min , L max ]; L min Indicates the minimum length of the heat exchange tube, L max Indicates the maximum length of the heat exchange tube Flow rate range in the tube, w t ∈[w t，min , w t，max ];w t，min Indicates the minimum flow velocity in the pipe, w t，max Indicates the maximum flow velocity in the pipe Shell side velocity range, w s ∈[w s，min , w s，max ];w s，min Indicates the minimum shell side velocity, w s，max Indicates the maximum shell side velocity Pipe wall thickness range, δ w ∈[δ w，min , δ w，max ];δ w，min Indicates the minimum wall thickness, δ w，max Indicates the maximum wall thickness The range of the number of heat exchange tubes, N∈[N min , N max ]; N min Indicates the minimum number of heat exchange tubes, N max Indicates the maximum number of heat exchange tubes.

3. The heat exchanger optimization design system based on genetic algorithm according to claim 2, characterized in that: The initial population is generated according to the necessary design parameters of the heat exchanger. The specific process is as follows: Step A1: Generate a coding matrix for each optimized parameter. The specific process of generating a coding matrix for one of the optimized parameters is as follows: Randomly generating T parameter values ​​within an optimization parameter range of the optimization parameter, and performing binary encoding on the T parameter values ​​to obtain T binary sequences of the optimization parameter; combining the T binary sequences row by row to form a coding matrix, where T is a positive integer; Step A2: Use the Latin hypercube sampling method to process the encoding matrix of each optimization parameter to generate M initial individuals. The genetic structure of each individual is composed of the binary encoding of each optimization parameter in series, where M is a positive integer. Based on the M initial individuals, an initial population is generated, the population size is set to M, the maximum number of iterations is Ngmax, and the individual gene structure is expressed by the formula: [X = [x1, x2, x3, x4, x5, x6, x7]], x1 represents the number of tube passes Z t The binary code of x2 represents the outer diameter of the tube d o x3 represents the binary code of the heat exchange tube length L; x4 represents the flow velocity w in the tube t Binary code; x5 represents the shell side flow velocity w s The binary code of x6 represents the wall thickness δ w x7 represents the binary code of the number of heat exchange tubes N.

4. The heat exchanger optimization design system based on genetic algorithm according to claim 3, characterized in that: The heat exchanger thermal calculation module uses a genetic algorithm to optimize the optimal heat exchanger parameters based on the initial population and the necessary thermal design parameter values. The specific process is as follows: S1: Decode each individual in the initial population into a real parameter value, and calculate the fitness value of each individual in the initial population based on the real parameter value and the necessary parameter value of thermal design; S2: Arrange each individual in the initial population in ascending order according to fitness value; S3: According to the arrangement order of each individual in the initial population, the initial population is subjected to mixed selection processing to select M1 elite individuals to form an elite population; S4: Perform crossover genetic treatment and mutation genetic treatment in sequence according to the elite population to obtain M2 offspring individuals; form the offspring population; S5: Calculate the fitness value of each offspring individual in the offspring population; take the offspring individual with the highest fitness value as the optimal offspring individual; S6: Set the stopping condition and judge whether the stopping condition is met according to the fitness value of the optimal offspring individual. When the stopping condition is met, the optimal offspring individual is used as the output individual; Enter S7; When the stopping condition is not met, the offspring population is used as the initial population for the next iteration and returns to S2; S7: Decode the output individual into the optimal optimization parameters, and calculate the optimal heat exchanger parameters according to the optimal optimization parameters.

5. The heat exchanger optimization design system based on genetic algorithm according to claim 4, characterized in that: In S1, the individual x i The variables are decoded into the real parameter value of the i-th optimization parameter, which is expressed as: For continuous parameters: For integer parameters: Among them, x i Represents the real value after decoding the i-th optimization parameter; a i The minimum value of the i-th optimization parameter; b i Indicates the maximum value of the i-th optimization parameter; n i Indicates the number of bits of code for the i-th optimization parameter; v i represents the decimal value of the binary gene segment corresponding to the i-th optimization parameter; represents the floor function; Among them, the continuous parameters include: tube outer diameter d o , heat exchange tube length L, flow velocity in the tube w t , shell side velocity w s , tube wall thickness δ w ; Integer parameters include: pipe number Z t , the number of heat exchange tubes N; The fitness value of an individual in the initial population is calculated in S1 and is expressed as follows: Where K represents the overall heat transfer coefficient of the heat exchanger; ΔP t Indicates the tube side pressure drop, ΔP s Indicates the shell side pressure drop.

6. The heat exchanger optimization design system based on genetic algorithm according to claim 5, characterized in that: The calculation formula of the overall heat transfer coefficient K of the heat exchanger is: Where, α t Expressed as the heat transfer coefficient of flow in the tube; α s Expressed as the heat transfer coefficient inside the tube shell; r s，i Expressed as the thermal resistance of dirt in the pipe, r s，o Expressed as the heat resistance of dirt outside the tube; λ w Expressed as the thermal conductivity of the pipe wall material; d o Expressed as the outer diameter of the tube, δ w Indicates the thickness of the pipe wall; d i Expressed as the inner diameter of the tube; The heat transfer coefficient of the flow in the tube ɑ t The calculation formula is: Now t =0.023Re t 0.8 Per t n Where, Nu t Expressed as the tube side Nusselt number; Re t Expressed as the tube side Reynolds number; Pr t Expressed as the tube-side Planck number; λ t Expressed as the thermal conductivity of the tube side fluid; n represents the intermediate variable, The heat transfer coefficient ɑ in the tube shell s The calculation formula is: Where, j H is the baffle window correction factor, j c is the side flow correction factor, j l is the leakage correction factor, j b is the bundle stagger correction factor; Pr s is the shell-side Planck number; λ s is the thermal conductivity of the shell side fluid; μ s Expressed as the dynamic viscosity of the shell side fluid, μ w Expressed as the dynamic viscosity of the fluid at the shell side wall; The tube side pressure drop ΔP t The calculation formula is: Where Δp i is the resistance along the way; Δp r is the bending resistance; is the viscosity correction factor, f i represents the friction coefficient inside the tube, ρ t represents the fluid density on the tube side; w t Indicates the fluid flow rate on the tube side; L indicates the length of the heat exchange tube, Z t Indicates the tube temperature, When Re≥2100, Re represents the Reynolds number of the fluid in the tube; When Re<2100, μ f Indicates the dynamic viscosity of the fluid on the tube side, μ w Indicates the dynamic viscosity of the fluid at the tube side wall The shell side pressure drop ΔP s The calculation formula is: When no baffle is used and the smooth heat exchange tubes are arranged in a bundle, the shell side resistance is calculated using the following formula: Where, N represents the number of heat exchange tubes; w s It is expressed as the shell side flow velocity; N is the number of heat exchange tubes; μ is the dynamic viscosity of the shell side fluid; μ w It is expressed as the dynamic viscosity of the fluid at the shell side wall; ρ is the fluid density; When no baffle is used and the smooth heat exchange tubes are arranged in a bundle, the shell side resistance is calculated using the following formula: Where w s It is expressed as the shell side flow velocity, m / s; N is the number of heat exchange tubes; μ is the dynamic viscosity of the shell side fluid; ρ is the fluid density;.

7. The heat exchanger optimization design system based on genetic algorithm according to claim 6, characterized in that: In S3, the initial population is subjected to mixed selection according to the arrangement order of each individual in the initial population, and M1 elite individuals are selected to form an elite population. The specific process is as follows: S3.

1. According to the ranking order of each individual in the initial population, select the top m1 individuals as elite individuals; S3.

2. Perform roulette wheel selection on the individuals not selected in S3.1, and select m2 individuals as elite individuals, ultimately obtaining M1 elite individuals; forming an elite population; M1 = m1 + m2.

8. The heat exchanger optimization design system based on genetic algorithm according to claim 7, characterized in that: In S4, crossover genetic processing and variation genetic processing are sequentially performed on the elite population to obtain M2 offspring individuals, which form an offspring population. The specific process is as follows: S4.1: Set the crossover probability p c , according to the crossover probability p c Perform cross-genetic operations on the elite population to obtain individuals after cross-genetic operation; In S4.1, two elite individuals in the elite population undergo a crossover genetic operation to obtain two individuals after crossover genetics. The specific process is as follows: Perform random length code swapping at a random intersection of the structural parameter codes of the two elite individuals to generate two individuals with new corresponding heat exchanger structural parameters as two new heat exchanger individuals; The two elite individuals undergoing cross-inheritance treatment are separated by one position; S4.2: Set the mutation probability p m , according to the mutation probability p m Perform mutation genetic operations on the individuals after crossover inheritance to obtain M2 offspring individuals; The specific process of the mutation genetic operation is: mutating the random positions of the structural parameter encodings of the individual heat exchangers.

9. The heat exchanger optimization design system based on genetic algorithm according to claim 8, characterized in that: The stop condition in S6 is expressed as follows: |F gen -F gen-100 / F gen |<1% Where, F gen Indicates the fitness value of the optimal offspring individual; F gen-100 represents the optimal fitness within 100 iterations.

10. The heat exchanger optimization design system based on genetic algorithm according to claim 9, characterized in that: In S7, the output individual is decoded into the optimal optimization parameters, and the optimal heat exchanger parameters are calculated based on the optimal optimization parameters. The specific process is as follows: S7.1: Calculate the working fluid flow rate M on the tube side based on the optimal optimization parameters t , heat exchange area A of the heat exchanger, and overall heat transfer coefficient K of the heat exchanger, which can be expressed as follows: Where: ΔT s Expressed as the shell side inlet and outlet temperature difference, ΔT t Indicates the inlet and outlet temperature difference of the tube side, °C; It is expressed as the constant pressure specific heat of the shell side and tube side fluid, kJ / (kgK); K is the overall heat transfer coefficient of the heat exchanger, W / (m 2 ·K); A represents the heat exchange area of ​​the heat exchanger, m 2 ; ΔT m Expressed as heat transfer logarithmic temperature difference, ℃; M s Indicates the shell side working medium flow rate, (kg / s); M t Tube side working fluid flow rate; The heat exchange logarithmic temperature difference ΔT m The calculation formula is expressed as: Where: T s，ave Expressed as the average temperature of the shell side fluid, °C; T s，i Expressed as the shell side fluid inlet temperature; T s，o Expressed as the shell side fluid outlet temperature; T t，i Expressed as the tube side fluid inlet temperature; T t，o Expressed as the tube side fluid outlet temperature; T t，ave Expressed as the average temperature of the fluid on the tube side, °C; ΔT m Expressed as qualitative temperature, ℃ S7.2: According to the working medium flow rate M on the tube side t , calculate the heat exchanger structural parameters based on the heat exchange area A of the heat exchanger; The heat exchanger structural parameters include: tube side structural parameters and shell side structural parameters; S7.2.

1. Calculate the tube side structural parameters, including: tube side flow area A t 、Number of pipe passes Z t , total number of heat exchange tubes N t , heat exchange tube spacing P t The specific process is: The calculation formula of the tube flow area is expressed as: M t =ρ t ·A t ·w t Where: M t 、M s Expressed as tube side and shell side mass flow rate, kg / s; ρ t Expressed as the tube side fluid density, kg / m 3 ; A t Expressed as the total cross-sectional area of ​​the heat exchange tube, m 2 ; N represents the number of heat exchange tubes; w t Indicates the fluid flow rate on the tube side, m / s; The calculation formula for the number of tube passes is expressed as: Where: L represents the length of the heat exchange tube, m; l represents the ratio of the heat exchange tube length to the shell diameter, The calculation formula for the total number of heat exchange tubes is expressed as: N t =Z t ·N Where: N t Expressed as the total number of heat exchange tubes; Z t is the number of tube passes; N is the number of heat exchange tubes; The tube spacing calculation formula is expressed as: P t ≥d o Where P t It represents the distance between heat exchange tubes; do represents the outer diameter of heat exchange tubes; S7.2.

2. Calculate the shell side structural parameters, including: shell side inner diameter D s , baffle diameter D b , baffle spacing l s , number of baffles N b ; The shell side inner diameter calculation formula is expressed as: D s =(b-1)d o +2b’ Where b' is the distance from the center of the outermost tube on the center line of the tube bundle to the inner wall of the shell. The baffle diameter calculation formula is expressed as: D b =D s -2×δ Where D b It is expressed as the baffle diameter, and δ represents the gap; The baffle spacing calculation formula is expressed as: Where, l s Expressed as the baffle spacing; M s Expressed as the mass flow rate of the shell side fluid; ρ s Expressed as the density of the shell side fluid; w s Expressed as the flow rate of the shell side fluid; D s Expressed as the inner diameter of the shell; N t Expressed as the total number of heat exchange tubes; d o Expressed as the outer diameter of the heat exchange tube, The calculation formula for the number of baffles is expressed as: Where N b It is the number of baffles; L is the length of the heat exchange tube; l s Expressed as the baffle spacing.

Images