A dynamic analysis method for ammonia combustion refrigeration system in large-scale park cooling

Abstract

The present application provides a kinetic analysis method for an ammonia combustion refrigeration system in a large-scale park cooling system, comprising: calculating the heat generated by ammonia combustion according to the geometric structure of an ammonia combustion chamber and the amount of fuel supplied, and analyzing the heat transfer mode between the combustion chamber and a desorption tower, including direct contact and inter-wall heat transfer, and obtaining and drawing the heat flux density distribution on the tower wall by simulating the heat transfer process; analyzing the desorption process of the refrigerant in the solution according to the flow state and the relationship curve between the pressure drop and the flow rate, and calculating the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower based on the solution concentration, the boiling point of the refrigerant and the vapor pressure; calculating the solution temperature distribution according to the heat flux density distribution, and obtaining the local heat transfer coefficient and the mass transfer efficiency distribution on the filler surface by analyzing the heat transfer and mass transfer processes in the desorption tower.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a dynamic analysis method for an ammonia combustion refrigeration system in large-scale park cooling. Background Art

[0002] In the cooling system of large-scale parks, the application of ammonia combustion refrigeration technology faces a difficult technical problem. As the core component of the system, the performance of the desorption tower directly affects the overall refrigeration efficiency. However, when we tried to increase the specific surface area of ​​the desorption tower to improve the heat transfer efficiency and refrigerant desorption rate, we encountered a dilemma. Reducing the tower diameter can increase the heat transfer area per unit volume, but it will also increase the flow resistance of the gas-liquid two-phase in the tower, affecting the mass transfer effect. On the contrary, although increasing the tower height can prolong the gas-liquid contact time, it will increase the equipment cost and floor space. In addition, there is a contradiction in the choice of operating pressure: high pressure is conducive to improving the desorption efficiency, but it will increase the manufacturing difficulty and operation risk of the equipment; low pressure may lead to insufficient desorption and affect the refrigeration effect. More complicated is the problem of uneven temperature distribution inside the desorption tower. The higher temperature at the bottom of the tower is conducive to the desorption of the refrigerant; while the lower temperature at the top of the tower may cause some of the desorbed refrigerant to be reabsorbed. The existence of this temperature gradient makes it difficult for the desorption process to achieve ideal efficiency. How to find the best balance between these conflicting factors while meeting the stringent requirements of large-scale parks for cooling capacity and energy efficiency has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present invention provides a dynamic analysis method for an ammonia combustion refrigeration system in a large-scale park cooling system, which mainly includes:

[0004] According to the geometric structure of the ammonia combustion chamber and the fuel supply, the heat generated by the ammonia combustion is calculated, and the heat transfer mode between the combustion chamber and the desorption tower is analyzed, including direct contact and inter-wall heat transfer. By simulating the heat transfer process, the heat flux density distribution on the tower wall is obtained and plotted;

[0005] Based on the filler type, specific surface area, porosity and bulk density parameters, the flow state of the absorbent solution in the desorption tower is analyzed, the gas-liquid two-phase flow rate and gas-liquid ratio distribution at different positions in the tower are obtained, the influence of the filler size and tower diameter ratio on the pressure drop is analyzed, and the relationship curve between the pressure drop and the flow rate is established;

[0006] According to the flow state and the relationship curve between pressure drop and flow rate, the desorption process of refrigerant in the solution is analyzed, and based on the solution concentration, refrigerant boiling point and vapor pressure, the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower are calculated;

[0007] According to the heat flux density distribution, the solution temperature distribution is calculated, and by analyzing the heat transfer and mass transfer process in the desorption tower, the local heat transfer coefficient and mass transfer efficiency distribution on the packing surface are obtained;

[0008] According to the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower, the desorption tower is integrated to obtain the distribution curve of the cumulative desorption amount of the refrigerant along the tower height, and the relationship between the mass transfer efficiency and the specific surface area of ​​the filler and the tower diameter is analyzed to determine the main factors affecting the mass transfer efficiency.

[0009] According to the local heat transfer coefficient of the packing surface and the main factors affecting the mass transfer efficiency, the geometric parameters of the packing specific surface area, tower diameter, tower height and solution flow rate operating parameters are systematically analyzed to set the parameter variation range. Through parameter scanning and sensitivity analysis, the influence of each parameter on the system performance is quantified to obtain the parameter analysis results.

[0010] According to the parameter analysis results, the packing specific surface area, tower diameter and tower height parameters under different solution flow conditions are optimized to obtain the optimal geometric parameter combination under each flow rate. Combined with the performance of the evaluation system under different cooling loads and parameter combinations, the operating characteristics of the system under various load conditions are obtained, including changes in energy efficiency ratio and response capabilities. A performance analysis report on the ammonia combustion refrigeration system is obtained, including the energy-saving potential and environmental impact assessment under various operating conditions.

[0011] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0012] The present invention discloses a method for optimizing the performance of an ammonia combustion refrigeration system. The method establishes a relationship model between pressure drop and flow rate by analyzing the heat transfer between the ammonia combustion chamber and the desorption tower, as well as the flow state of the absorbent solution in the desorption tower. The main factors affecting the mass transfer efficiency are determined by combining the refrigerant desorption process and the heat and mass transfer analysis. The optimal parameter combination under different cold load conditions is obtained by systematically analyzing and optimizing the geometric parameters such as the specific surface area of ​​the filler, the tower diameter, the tower height, and the operating parameters such as the solution flow rate. The present invention realizes the performance optimization of the ammonia combustion refrigeration system under various operating conditions, improves the energy efficiency ratio, enhances the system responsiveness, and evaluates the energy-saving potential and environmental impact, providing theoretical guidance and technical support for the design and operation of the ammonia combustion refrigeration system. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 The present invention is a flow chart of a kinetic analysis method for an ammonia combustion refrigeration system in large-scale park cooling.

[0014] Figure 2 It is a schematic diagram of a kinetic analysis method of an ammonia combustion refrigeration system in large-scale park cooling according to the present invention. DETAILED DESCRIPTION

[0015] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0016] like Figure 1-2 In this embodiment, a dynamic analysis method of an ammonia combustion refrigeration system in a large-scale park cooling system may specifically include:

[0017] S101. According to the geometric structure of the ammonia combustion chamber and the fuel supply, the heat generated by the ammonia combustion is calculated, and the heat transfer mode between the combustion chamber and the desorption tower is analyzed, including direct contact and inter-wall heat transfer. By simulating the heat transfer process, the heat flux density distribution on the tower wall is obtained and plotted.

[0018] The geometric structure parameters and fuel supply data of the ammonia combustion chamber are obtained, the ammonia combustion reaction rate is calculated, and the temperature distribution in the combustion chamber is obtained by combining the energy conservation equation; two heat transfer models, direct contact type and partition wall type, are established according to the temperature distribution of the combustion chamber, and finite element analysis is performed based on the two heat transfer models to obtain the temperature field distribution under the two heat transfer modes; the heat conduction partial differential equation is solved for the temperature distribution of the combustion chamber and the two heat transfer models, and the temperature value of each point in the desorption tower is obtained through iterative calculation; the heat flux density value of each point on the tower wall is calculated according to the temperature gradient data of the desorption tower by applying Fourier's heat conduction law; a continuous heat flux density distribution function is generated by a bilinear interpolation algorithm, and if the continuous heat flux density distribution function is generated, a thermal diagram of the heat flux density distribution on the tower wall is drawn.

[0019] Specifically, according to the geometric structure parameters of the ammonia combustion chamber and the fuel supply data, the Arrhenius equation is used to calculate the ammonia combustion reaction rate, and the temperature distribution in the combustion chamber is obtained by combining the energy conservation equation. For the heat transfer between the combustion chamber and the desorption tower, the finite element analysis is performed using COMSOL Multiphysics software, and two heat transfer models, direct contact and partition wall, are established respectively. The convection heat transfer model is used for the direct contact model, and the heat conduction model is used for the partition wall model. The temperature field distribution under the two heat transfer modes is obtained by solving. Based on the temperature distribution of the combustion chamber and the two heat transfer models, the Runge-Kutta method is implemented by MATLAB programming to solve the heat conduction partial differential equation. The temperature distribution of the combustion chamber and the results of the heat transfer model are used as boundary conditions. The temperature values ​​of each point in the desorption tower are obtained through iterative calculation, and the temperature gradient cloud map of the desorption tower is drawn. According to the temperature gradient data of the desorption tower, the Fourier heat conduction law is applied to calculate the heat flux density values ​​of each point on the tower wall, and the bilinear interpolation algorithm is used to generate a continuous heat flux density distribution function. The Python Matplotlib library is used to draw the heat flux density distribution thermodynamic map of the tower wall. The geometric structure of the ammonia combustion chamber is set as a cylinder with a diameter of 0.5 meters and a height of 1.2 meters. The fuel supply is 10 kg / hour. The Arrhenius equation is used to calculate the ammonia combustion reaction rate, with an activation energy of 60 kJ / mol and a pre-exponential factor of 1.2×10^6s^-1. Combined with the energy conservation equation, the temperature distribution in the combustion chamber is obtained by discretization through the finite difference method, and the highest temperature in the center reaches 1200℃. COMSOL Multiphysics software is used to construct two heat transfer models: direct contact and partition wall. The direct contact model adopts a convection heat transfer model, and the heat transfer coefficient is set to 500W / (m^2·K); the partition wall adopts a heat conduction model, and the heat transfer wall material is stainless steel with a thermal conductivity of 16W / (m·K). The temperature field distribution under the two heat transfer modes is obtained by solving the temperature field distribution under the two heat transfer modes. The temperature at the inlet of the desorption tower is 800℃ for the direct contact type and 750℃ for the partition wall type. Based on the above results, the fourth-order Runge-Kutta method was implemented using MATLAB programming to solve the two-dimensional heat conduction partial differential equation. The grid size was set to 0.01 meters, the time step was 0.1 seconds, and the iterative calculation was performed for 5000 steps to obtain the temperature values ​​at each point in the desorption tower. The built-in function contourf was used to draw the temperature gradient cloud map of the desorption tower, with a temperature range from 400℃ to 800℃. According to the temperature gradient data, the Fourier heat conduction law was used to calculate the heat flux density values ​​at each point on the tower wall, and the thermal conductivity was taken as 5W / (m·K). The bilinear interpolation algorithm was used, and the number of interpolation points was 100×100 to generate a continuous heat flux density distribution function. Finally, the pcolormesh function in Python's Matplotlib library was used to draw the heat flux density distribution thermodynamic map of the tower wall. The heat flux density ranged from 5000W / m^2 to 25000W / m^2, and the "hot" method was used for color mapping.

[0020] S102. Based on the filler type, specific surface area, porosity and bulk density parameters, analyze the flow state of the absorbent solution in the desorption tower, obtain the gas-liquid two-phase flow rate and gas-liquid ratio distribution at different positions in the tower, analyze the influence of the filler size and tower diameter ratio on the pressure drop, and establish the relationship curve between the pressure drop and the flow rate.

[0021] The filler type, specific surface area, porosity and packing density parameters of the desorption tower are obtained, and the pressure gradient at different positions in the desorption tower is calculated using the Stichlmair two-phase flow pressure drop model according to the parameters; the pressure distribution in the desorption tower is obtained by the numerical integration method, and the gas-liquid two-phase flow velocity distribution in the desorption tower is determined in combination with the continuity equation; a two-dimensional axisymmetric model of the desorption tower is established, and the porous medium parameters of the desorption tower filler area are set, and the porous medium parameters include the porosity, inertial resistance coefficient and viscous resistance coefficient calculated from the filler characteristics; the flow field information in the desorption tower is obtained, and the gas-liquid two-phase flow velocity and gas-liquid ratio distribution are extracted from the flow field information; the gas-liquid ratio distribution is feature identified to determine the flow state of the solution in the desorption tower; if the flow state of the solution in the desorption tower meets the preset conditions, the desorption tower pressure drop data is recorded; the relationship curve between the pressure drop data and the flow velocity is fitted by the least squares method to obtain the power function relationship between the pressure drop and the flow velocity.

[0022] Specifically, according to the filler type, specific surface area, porosity and bulk density parameters, the Stichlmair two-phase flow pressure drop model is used to calculate the pressure gradient at different positions in the desorption tower. The pressure distribution in the tower is obtained by numerical integration, and the gas-liquid two-phase flow velocity distribution is solved in combination with the continuity equation. The computational fluid dynamics software FLUENT is used to establish a two-dimensional axisymmetric model of the desorption tower, and the porous medium parameters in the filler area are set, including the porosity, inertial resistance coefficient and viscous resistance coefficient calculated from the filler characteristics. The physical properties of the absorbent solution and the inlet boundary conditions are input, and the Navier-Stokes equation is solved to obtain the flow field information in the tower. The gas-liquid two-phase flow velocity and gas-liquid ratio distribution are extracted from the flow field data, and the gas-liquid ratio contour map in the tower is generated by the interpolation function of MATLAB. The gas-liquid ratio distribution is feature-recognized by the K-means clustering algorithm to determine the flow state of the solution. The Python script was used to automate the calculation process, and the parameterized research function of FLUENT was used to realize batch calculation of different packing sizes and tower diameter ratios. The pressure drop data was recorded, and the relationship curve between pressure drop and flow rate was fitted by the least squares method to obtain the power function relationship between pressure drop and flow rate. At the same time, multiple regression analysis was used to establish the relationship model between pressure drop and packing size, tower diameter ratio and flow rate. In the implementation process, ceramic Raschig ring packing with a specific surface area of ​​350m 2 / m 3 , porosity is 0.75, bulk density is 650kg / m 3. The pressure gradient in the desorption tower was calculated using the Stichlmair two-phase flow pressure drop model. The tower is 10 meters high and is divided into 100 calculation units. The pressure drop in each unit is about 20Pa. The pressure distribution in the tower was obtained by trapezoidal numerical integration, and the pressure at the bottom of the tower was 1.2 bar. Combined with the continuity equation, the superficial velocity of the gas phase was calculated to be 2 m / s and that of the liquid phase to be 0.01 m / s. A two-dimensional axisymmetric model of the desorption tower was established in FLUENT. The tower diameter was 1 meter. The porous medium parameters of the packing area were set. The inertial resistance coefficient was 3.5 and the viscous resistance coefficient was 1.2×10 5 m -2 . Input absorbent solution density 1200kg / m 3 , viscosity 0.002Pa·s, and the flow field distribution in the tower was obtained. The griddata function of MATLAB was used to interpolate the flow field data and generate a 100×100 grid gas-liquid ratio contour map. The K-means clustering algorithm was used to set 3 cluster centers to perform feature recognition on the gas-liquid ratio distribution, and it was determined that the flow states of the solution in the top, middle and bottom of the tower were film flow, transition flow and bubbling flow, respectively. The parameterStudy module of Python was used to implement the parametric study of packing size 10-50mm and tower diameter ratio 0.1-0.5. The pressure drop data at 10 flow rates were recorded for each set of parameters. The polyfit function of numpy was used to fit the power function relationship between pressure drop and flow rate, and the exponent was about 1.8. Finally, the OLS function of the statsmodels library was used for multiple regression analysis to establish a relationship model between pressure drop and packing size, tower diameter ratio and flow rate, and the determination coefficient R 2 Reached 0.95.

[0023] S103. Analyze the desorption process of the refrigerant in the solution according to the flow state and the relationship curve between the pressure drop and the flow rate, and calculate the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower based on the solution concentration, the boiling point of the refrigerant and the vapor pressure.

[0024] A model establishment request carrying desorption tower structural parameters is received, the model establishment request is issued by FLUENT software; a fluid flow and heat transfer coupling model in the desorption tower is established according to the model establishment request, the model is used to solve the temperature distribution and velocity field in the tower; the activity calculation result of the refrigerant in the liquid phase is obtained, the activity calculation result is obtained by the NRTL activity coefficient model; the vapor pressure of the refrigerant at different temperatures and pressures is solved according to the activity calculation result; the gas-liquid interface mass transfer coefficient is calculated, the mass transfer coefficient is used to solve the mass transfer equation of the desorption process by applying the finite difference method; the cumulative desorption amount of each cross section in the tower is calculated, the cumulative desorption amount is used to calculate the amount of refrigerant vapor generated by using the mass conservation principle; a differential equation dM / dz=f(z, M, C) is set, wherein M is the amount of vapor generated, z is the tower height, and C is the solution concentration; the differential equation is solved, and the solution of the differential equation is used to obtain the distribution of the amount of refrigerant vapor generated in the tower height direction.

[0025] Specifically, according to the flow state and the relationship curve between pressure drop and flow velocity, the computational fluid dynamics software FLUENT was used to establish the fluid flow and heat transfer coupling model in the desorption tower, set the inlet boundary conditions and wall heat transfer conditions, and use the structured grid division method with a grid number of 100×500 to solve the temperature distribution and velocity field in the tower. The NRTL activity coefficient model was used to calculate the activity of the refrigerant in the liquid phase, and the vapor pressure of the refrigerant at different temperatures and pressures was solved in combination with the state equation. The local driving force at each position in the tower was obtained by interpolation. The gas-liquid interface mass transfer coefficient was calculated using the Onda correlation, which took into account the characteristic parameters of the packing such as specific surface area and porosity. Combined with the local driving force and mass transfer coefficient, the finite difference method was used to solve the mass transfer equation of the desorption process to obtain the local desorption rate at different positions in the tower. The cumulative desorption amount of each section in the tower is calculated by numerical integration, and the amount of refrigerant vapor generated is calculated using the principle of mass conservation. The differential equation dM / dz=f(z,M,C) is set, where M is the amount of vapor generated, z is the tower height, and C is the solution concentration. The differential equation is solved using the ode45 function of MATLAB to obtain the distribution of the amount of refrigerant vapor generated in the tower height direction, and the plot function is used to draw the curve of the amount of refrigerant vapor generated with the tower height. In the implementation, the desorption tower model is first established in FLUENT based on the flow state and pressure drop-flow velocity relationship curve obtained in the early stage. The tower is 10 meters high and 1 meter in diameter. The inlet boundary conditions are set as a solution flow rate of 0.5m / s and a temperature of 80℃; a gas flow rate of 2m / s and a temperature of 70℃. Using a 100×500 structured grid, the temperature distribution range in the tower is 70-85℃, the liquid phase flow rate is 0.3-0.6m / s, and the gas phase flow rate is 1.8-2.5m / s. The NRTL activity coefficient model is used to calculate the activity of refrigerants in the liquid phase, such as R134a, with an activity coefficient range of 0.8-1.2. Combined with the Peng-Robinson state equation, the vapor pressure range is calculated to be 0.5-1.5MPa. The local driving force at 100 locations in the tower is obtained using cubic spline interpolation. The Onda correlation is used to calculate the gas-liquid interface mass transfer coefficient, and the filler specific surface area is 250m 2 / m 3 , porosity 0.95, the mass transfer coefficient range is 0.001-0.005m / s. The finite difference method with central difference format and 1000 grids is used to solve the mass transfer equation and the local desorption rate is calculated to be 0.01-0.05kg / (m 3·s). The cumulative desorption amount was calculated by the trapezoidal integration method, and the total desorption amount at the top of the tower was 180kg / h. The differential equation dM / dz=0.05*(1-M / 180)*(1-z / 10) was set, where M is the steam generation (kg / h) and z is the tower height (m). The ode45 function of MATLAB was used to solve the equation, and the initial value M(0)=0 was set to obtain the steam generation at 10 points in the tower height direction. Finally, a curve graph was drawn, with the horizontal axis representing the tower height of 0-10m and the vertical axis representing the steam generation of 0-180kg / h. The curve was S-shaped, with the fastest growth in the middle of the tower.

[0026] S104. Calculate the solution temperature distribution based on the heat flux density distribution, and obtain the local heat transfer coefficient and mass transfer efficiency distribution on the packing surface by analyzing the heat transfer and mass transfer process in the desorption tower.

[0027] A three-dimensional temperature field numerical model in the desorption tower is established, and the numerical model includes the inlet temperature and the wall heat flow boundary conditions. Using the numerical model, the solution temperature distribution is solved. The physical properties of the solution and the refrigerant are obtained, and the heat transfer and mass transfer coupling model in the desorption tower is constructed using FLUENT according to the physical properties. The coupling model is used to simulate the gas-liquid two-phase flow. During the simulation process, the mass transfer term between the gas and liquid phases adopts the Lee model, and the turbulence model uses the RNGk-ε model. The velocity field and concentration field on the packing surface are obtained according to the simulation results. The temperature gradient and heat flux density on the packing surface are obtained, and the local heat transfer coefficient is calculated according to the temperature gradient and heat flux density. The relationship between the local heat transfer coefficient and the Reynolds number and the Prandtl number is obtained by fitting using the least squares method. The actual mass transfer flux and the ideal mass transfer flux on the packing surface are obtained, and the local mass transfer efficiency is calculated according to the actual mass transfer flux and the ideal mass transfer flux.

[0028] Specifically, according to the heat flux density distribution, the finite volume method is used to discretize the energy equation, and a three-dimensional temperature field numerical model in the desorption tower is established. The mesh division adopts a hexahedral structured grid with a grid number of 100×100×200. The inlet temperature and wall heat flux boundary conditions are set, and the solution temperature distribution is obtained by iterative solution using the SIMPLE algorithm. The computational fluid dynamics software FLUENT is used to construct a heat transfer and mass transfer coupling model in the desorption tower, and the physical properties of the solution and the refrigerant are set. The Euler-Euler two-fluid model is used to simulate the gas-liquid two-phase flow. The mass transfer term between the gas and liquid phases adopts the Lee model, and the turbulence model uses the RNGk-ε model to solve the velocity field and concentration field on the packing surface. Based on the temperature gradient and heat flux density on the packing surface, the modified Newton's cooling law q=h(Tw-Tb) is used to calculate the local heat transfer coefficient h, where q is the heat flux density, Tw is the wall temperature, and Tb is the main body temperature. The least squares method is used to fit the relationship between the local heat transfer coefficient h and the Reynolds number Re and Prandtl number Pr.

[0029] h=a·Re b ·Pr c

[0030] , a is the proportional constant in the formula, b is the exponent of the Reynolds number ReRe, which indicates the influence of the Reynolds number on the heat transfer coefficient. By comparing the actual mass transfer flux on the packing surface with the ideal mass transfer flux, the local mass transfer efficiency is calculated, and a three-layer BP neural network algorithm is used to establish a relationship model between the mass transfer efficiency and the packing geometric parameters, fluid properties and operating conditions. The input layer contains 10 nodes, the hidden layer has 20 nodes, and the output layer has 1 node. The network is trained using the Levenberg-Marquardt algorithm, and the contourf function of MATLAB is used to draw the distribution cloud map of the heat transfer coefficient and mass transfer efficiency. In the implementation process, firstly, according to the heat flux density distribution data obtained in the early stage, the finite volume method is used to construct a three-dimensional temperature field model of the desorption tower. The tower is 10 meters high and 1 meter in diameter, and is divided into a 100×100×200 hexahedral structured grid. The inlet temperature is set to 80°C, and the wall heat flux density range is 5000-25000W / m 2 The SIMPLE algorithm was used for 500 iterations, with a convergence criterion of 10 -6 The solution temperature distribution range is 70-90°C. The Euler-Euler two-fluid model is established in FLUENT, and the solution density is set to 1200 kg / m 3 , viscosity 0.002Pa·s, refrigerant density 4kg / m 3 , viscosity 1.2×10 -5 Pa·s. The Lee model coefficient is set to 0.1, and the RNGk-ε model constant Cμ is taken as 0.0845. The solution obtains the filler surface velocity field range of 0.1-1.0m / s, and the concentration field range of 0.1-0.5kg / m 3 Based on the surface temperature gradient and heat flux density of the filler, the local heat transfer coefficient is calculated in the range of 500-2000W / (m 2 ·K). The least squares method is used to fit the relationship h = 0.023Re 0 · 8 Pr 0 · 4 . By comparing the actual mass transfer flux with the ideal mass transfer flux, the local mass transfer efficiency range is 0.6-0.9. A three-layer BP neural network is constructed, with 10 nodes in the input layer including parameters such as filler specific surface area, porosity, liquid-gas ratio, 20 nodes in the hidden layer, and 1 node in the output layer for mass transfer efficiency. The network was trained using 100 sets of experimental data, with a learning rate of 0.01, 1000 training times, and the mean square error dropped to 0.001. Finally, the contourf function of MATLAB was used to draw the distribution cloud map of heat transfer coefficient and mass transfer efficiency. The horizontal and vertical coordinates are the tower diameter and tower height, and the color range is set to blue to red.

[0031] S105. According to the local desorption rate at different positions in the desorption tower and the amount of refrigerant vapor generated, the desorption tower is integrated to obtain the distribution curve of the cumulative desorption amount of the refrigerant along the tower height, and the relationship between the mass transfer efficiency and the specific surface area of ​​the filler and the tower diameter is analyzed to determine the main factors affecting the mass transfer efficiency.

[0032] The local desorption rate and refrigerant vapor generation data at different positions in the desorption tower are obtained; the desorption tower is integrated by a trapezoidal numerical integration method according to the data to obtain the cumulative desorption amount distribution of the refrigerant; the cumulative desorption amount data are interpolated by cubic spline to generate a distribution curve of the cumulative desorption amount along the tower height; the mean value and standard deviation of the local mass transfer efficiency are calculated, and if the mean value plus or minus two times the standard deviation exceeds a preset threshold, it is determined to be the optimal mass transfer area with the highest desorption efficiency; a relationship model between the mass transfer efficiency and the specific surface area of ​​the filler and the tower diameter is established by a multivariate regression analysis method, with the mass transfer efficiency as the dependent variable and the specific surface area of ​​the filler and the tower diameter as the independent variables; the importance of each factor to the mass transfer efficiency is evaluated, and the importance ranking of each factor is determined by calculating the average impurity reduction.

[0033] Specifically, according to the local desorption rate and refrigerant vapor generation data at different positions in the desorption tower, the trapezoidal numerical integration method is used to integrate the desorption tower, and the local desorption amount is accumulated along the tower height to obtain the cumulative desorption amount distribution of the refrigerant. The cumulative desorption amount data is interpolated by cubic spline using the spline function of MATLAB to generate a high-precision distribution curve of the cumulative desorption amount along the tower height, and the plot function is used to draw the curve, with the horizontal axis being the tower height and the vertical axis being the cumulative desorption amount. Combined with the mass transfer efficiency distribution data, the moving window method is used to calculate the mean and standard deviation of the local mass transfer efficiency. The window size is set to 10% of the tower height, and the sliding step is 20% of the window size. The optimal mass transfer area with the highest desorption efficiency is determined by setting a threshold (mean value plus or minus two times the standard deviation), and the area is marked on the cumulative desorption amount distribution curve. The relationship model between mass transfer efficiency and specific surface area of ​​filler and tower diameter was established by multiple regression analysis. The stepwise regression algorithm was used to screen significant variables. The critical value of the F test was used as the standard for entry and elimination, α=0.05. Multicollinearity diagnosis was performed by calculating the variance inflation factor (VIF). Variables with VIF greater than 10 were eliminated. The main factors affecting mass transfer efficiency were determined by comparing standardized regression coefficients. The importance of each factor to mass transfer efficiency was evaluated using the random forest algorithm. The number of trees was set to 100 and the minimum number of leaf node samples was set to 5. The importance of each factor was ranked by calculating the average impurity reduction to verify the results of the multiple regression analysis. In the implementation process, the desorption tower was first integrated. The tower was 10 meters high and divided into 100 calculation units. The local desorption rate of each unit ranged from 0.01 to 0.05 kg / (m3 ·s). The trapezoidal rule was used for numerical integration to obtain the cumulative desorption distribution, and the cumulative desorption at the top of the tower was 180kg / h. The spline function of MATLAB was used to perform cubic spline interpolation on these 100 data points to generate 1000 interpolation points to form a smooth cumulative desorption distribution curve. The moving window method was applied to the mass transfer efficiency data, with a window size of 1 meter and a sliding step of 0.2 meters. The local mass transfer efficiency was calculated to have an average value of 0.75 and a standard deviation of 0.08. The threshold was set to the mean value plus or minus twice the standard deviation of 0.59-0.91, and it was determined that the optimal mass transfer area was located at a tower height of 3.5-6.5 meters. A multivariate regression analysis was performed, and the independent variables included the filler specific surface area of ​​200-400m 2 / m 3 and tower diameter 0.5-2.0m, and the dependent variable is mass transfer efficiency. In the stepwise regression, the critical value of the F test was set to 4.0, and the two variables of packing specific surface area and tower diameter were retained after screening. The calculated VIF values ​​were all less than 5, and there was no serious multicollinearity. The standardized regression coefficient showed that the effect of packing specific surface area 0.72 on mass transfer efficiency was greater than that of tower diameter -0.45. The random forest algorithm was used to evaluate the importance of factors, with 100 trees and a minimum leaf node sample number of 5. The results showed that the importance scores of packing specific surface area were 0.68 and tower diameter were 0.32, which verified the regression analysis results.

[0034] S106. Based on the local heat transfer coefficient of the packing surface and the main factors affecting the mass transfer efficiency, a systematic analysis is conducted on the geometric parameters of the packing specific surface area, tower diameter, tower height, and solution flow rate operating parameters to set the parameter variation range. Through parameter scanning and sensitivity analysis, the influence of each parameter on the system performance is quantified to obtain parameter analysis results.

[0035] Obtain uniformly distributed sample points generated by the Latin hypercube sampling method; construct a simplified model of the desorption tower based on the sample points, wherein the simplified model uses a second-order polynomial to fit the relationship between system performance and parameters; determine the importance ranking of each parameter in the Pareto diagram, and if the parameter importance exceeds a preset threshold, it is determined to be a key parameter; calculate the sensitivity coefficient for the key parameter, and the sensitivity coefficient is obtained by the central difference method; use the NSGA-II algorithm for multi-objective optimization, and the multi-objective optimization combines the three performance indicators of total heat transfer, desorption efficiency and pressure drop at the same time to obtain a Pareto optimal solution set, and select the best parameter combination from the Pareto optimal solution set.

[0036] Specifically, according to the local heat transfer coefficient of the packing surface and the main factors affecting the mass transfer efficiency, the parameter space is established, the geometric parameters of the packing specific surface area, tower diameter, tower height and the range of change of the solution flow rate operating parameters are set, and the Latin hypercube sampling method is implemented by the pyDOE library to generate 100 uniformly distributed sample points. The simplified model of the desorption tower is constructed by the response surface method, and the relationship between the system performance and the parameters is fitted by a second-order polynomial. The parameter scanning is implemented by Python script to obtain the system performance indicators under different parameter combinations, such as total heat transfer, desorption efficiency and pressure drop. The parameter scanning results are processed, and the variance analysis is performed by the statsmodels library to calculate the F value and p value of each parameter. The importance ranking of each parameter is intuitively displayed by drawing a Pareto chart to determine the key parameters affecting the system performance. For the key parameters, a local sensitivity analysis is performed, and the sensitivity coefficient is calculated by the central difference method. The step size is 1% of the parameter range. The contourf function of MATLAB is used to draw a two-dimensional sensitivity contour map to quantify the degree of influence of parameter changes on system performance. At the same time, the Morris screening method is implemented by the SALib library to perform global sensitivity analysis and verify the results of local sensitivity analysis. Finally, the NSGA-II algorithm was used for multi-objective optimization, taking into account the three performance indicators of total heat transfer, desorption efficiency and pressure drop, and the Pareto optimal solution set was obtained, from which the best parameter combination was selected. In the implementation process, the parameter variation range was first set to 200-400m 2 / m 3 , tower diameter 0.5-2.0m, tower height 5-15m, solution flow rate 1000-5000kg / h. Use the lhs function of the pyDOE library to generate 100 Latin hypercube sample points. Construct a second-order polynomial response surface model to fit system performance indicators, such as total heat transfer Q = a0 + a1x1 + a2x2 + a3x3 + a4x4 + a12x1x2 + ... + a11x1 2 +..., where x1-x4 are four parameters. By scanning the parameters, the performance index range is 50-200kW for total heat transfer, 60-90% for desorption efficiency, and 0.5-5kPa / m for pressure drop. The ols function of statsmodels.api is used for variance analysis to calculate the statistical F value and p value. The results show that the probability p value of the filler specific surface area and the solution flow rate is <0.05, which is a significant parameter. A local sensitivity analysis is performed on these two key parameters using the central difference method with a step size of 1% of the parameter range. The sensitivity coefficient is calculated to be 0.5kW / (m 2 / m 3), with a solution flow rate of 0.03kW / (kg / h). The Morris function of the SALib library was used for global sensitivity analysis, and the results verified the accuracy of the local analysis. Finally, the NSGA-II algorithm was used for multi-objective optimization, with a population size of 100 and 500 iterations. Twenty Pareto optimal solutions were obtained, from which the optimal parameter combination of total heat transfer of 180kW, desorption efficiency of 85%, and pressure drop of 2kPa / m was selected: filler specific surface area of ​​350m 2 / m 3 , tower diameter 1.2m, tower height 10m, solution flow rate 3500kg / h.

[0037] S107. Based on the parameter analysis results, the packing specific surface area, tower diameter and tower height parameters under different solution flow conditions are optimized to obtain the optimal geometric parameter combination under each flow rate. Combined with the performance of the evaluation system under different cooling loads and parameter combinations, the operating characteristics of the system under various load conditions are obtained, including changes in energy efficiency ratio and response capabilities. A performance analysis report on the ammonia combustion refrigeration system is obtained, including energy-saving potential and environmental impact assessment under various operating conditions.

[0038] Obtain the parameters of packing specific surface area, tower diameter and tower height, and perform multi-objective optimization on the parameters; simplify the dynamic model through the ammonia combustion refrigeration system according to the optimized geometric parameter combination; input the optimized geometric parameter combination, set different cooling load conditions, solve the differential equation group, and record the energy efficiency ratio, cooling capacity and response time performance indicators; use the polyfit function of the numpy library to perform polynomial fitting to obtain the energy efficiency ratio change curve; if the energy efficiency ratio change curve meets the preset conditions, determine the optimal operating load range of the system; calculate environmental impact indicators such as carbon emissions and energy consumption; construct a fuzzy comprehensive evaluation model and set the triangle membership function; quantitatively evaluate the system performance under different operating conditions through the fuzzy weighted average method; judge the degree of influence of different parameters on the system performance; if the degree of influence exceeds the preset threshold, verify the robustness of the optimization result.

[0039] Specifically, according to the parameter analysis results, the NSGA-II algorithm is implemented by using the DEAP library to perform multi-objective optimization on the parameters of the packing specific surface area, tower diameter and tower height under different solution flow conditions. The population size is set to 100, the crossover probability is 0.9, the mutation probability is 0.1, and 500 generations are iterated to obtain the Pareto optimal solution set under each flow rate, from which the geometric parameter combination with the best comprehensive performance is selected. The simplified dynamic model of the ammonia combustion refrigeration system is constructed using the scipy.integrate library of Python, the optimized geometric parameter combination is input, different cooling load conditions are set, the odeint function is used to solve the differential equation group, the operation process of the system under partial load conditions is simulated, and the performance indicators such as energy efficiency ratio, cooling capacity and response time are recorded. Based on the simulation results, the polynomial fitting is performed using the polyfit function of numpy, and the relationship model between system performance and cooling load is established to obtain the energy efficiency ratio change curve. The optimal operating load range of the system is determined by calculating the first-order derivative of the energy efficiency ratio, and the dynamic response capability of the system is evaluated using the step response method. Combined with system performance data, OpenLCA software is used to conduct life cycle assessment, calculate environmental impact indicators such as carbon emissions and energy consumption, build a fuzzy comprehensive evaluation model, set the triangle membership function, and use the fuzzy weighted average method to quantitatively evaluate system performance under different operating conditions, and generate an ammonia combustion refrigeration system performance analysis report. The SALib library is used to implement Sobol sensitivity analysis, evaluate the impact of different parameters on system performance, and verify the robustness of the optimization results.

[0040] In the implementation process, the NSGA-II algorithm was first implemented using the DEAP library to optimize the five flow points within the solution flow range of 1000-5000 kg / h. Taking the flow rate of 3000 kg / h as an example, after 500 generations of iterations, the Pareto optimal solution set was obtained, from which the filler with a specific surface area of ​​350 m 2 / m 3 , tower diameter 1.2m, tower height 10m. Then, the simplified dynamic model was constructed using the scipy.integrate library, the cooling load range was set to 20-100kW, the step size was 10kW, and the odeint function was used to solve the differential equations to obtain the data points of energy efficiency ratio changing with load. The polyfit function of numpy was used to fit the 4th order polynomial, and the fitting degree R 2 Reach 0.99, and get the energy efficiency ratio curve COP = -2e-7Q 4 +5e-5Q 3 -0.004Q 2+0.15Q+1.2, where Q is the cooling load. The optimal operating load range is determined to be 60-80kW by derivation. The OpenLCA software was used for life cycle assessment, and it was calculated that under rated conditions, the system's carbon emissions for every 1kWh of cooling is 0.5kgCO2 equivalent, and the energy consumption is 1.8kWh. A fuzzy comprehensive evaluation model was constructed, and three evaluation indicators, energy efficiency ratio, response time and environmental impact, were set. The triangular membership function was used with weights of 0.4, 0.3 and 0.3 respectively. The fuzzy weighted average method was used to obtain a comprehensive system performance score of 0.85. Finally, the SALib library was used to implement Sobol sensitivity analysis, and the first-order sensitivity indexes of the packing specific surface area, tower diameter and tower height to the system performance were calculated to be 0.45, 0.30 and 0.25 respectively, which verified the rationality of the optimization results.

[0041] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the concept of the present application. For example, the above features are replaced with the technical features with similar functions disclosed in the present application (but not limited to) to form a technical solution.

Claims

1. A dynamic analysis method for ammonia combustion refrigeration system in large-scale park cooling, characterized in that: The method comprises: calculating the heat generated by ammonia combustion according to the geometric structure of the ammonia combustion chamber and the amount of fuel supplied, and analyzing the heat transfer mode between the combustion chamber and the desorption tower, including direct contact heat transfer and inter-wall heat transfer, and obtaining and drawing the heat flux density distribution on the tower wall by simulating the heat transfer process; analyzing the flow state of the absorbent solution in the desorption tower based on the filler type, specific surface area, porosity and bulk density parameters, obtaining the gas-liquid two-phase flow rate and gas-liquid ratio distribution at different positions in the tower, analyzing the influence of the filler size and the tower diameter ratio on the pressure drop, and establishing a relationship curve between the pressure drop and the flow rate; analyzing the desorption process of the refrigerant in the solution according to the flow state and the relationship curve between the pressure drop and the flow rate, and calculating the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower based on the solution concentration, the boiling point of the refrigerant and the vapor pressure; calculating the solution temperature distribution according to the heat flux density distribution, and obtaining the local heat transfer coefficient and the mass transfer efficiency distribution on the filler surface by analyzing the heat transfer and mass transfer process in the desorption tower; and calculating the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower according to the local desorption rate at different positions in the desorption tower. The absorption rate and the amount of refrigerant vapor generated are integrated for the desorption tower to obtain the distribution curve of the cumulative desorption amount of the refrigerant along the tower height, and the relationship between the mass transfer efficiency and the specific surface area of ​​the packing and the tower diameter is analyzed to determine the main factors affecting the mass transfer efficiency; according to the local heat transfer coefficient of the packing surface and the main factors affecting the mass transfer efficiency, a systematic analysis is conducted on the geometric parameters of the specific surface area of ​​the packing, the tower diameter, and the tower height, as well as the solution flow rate operating parameters, in order to set the parameter variation range, and through parameter scanning and sensitivity analysis, the influence of each parameter on the system performance is quantified to obtain the parameter analysis results; according to the parameter analysis results, the parameters of the specific surface area of ​​the packing, the tower diameter and the tower height under different solution flow conditions are optimized to obtain the optimal geometric parameter combination under each flow rate, and the performance of the system under different cold loads and parameter combinations is combined with the evaluation system. The operating characteristics of the system under various load conditions, including changes in energy efficiency ratio and response capabilities, are obtained, and a performance analysis report of the ammonia combustion refrigeration system is obtained, including energy saving potential and environmental impact assessment under various operating conditions; The method comprises the following steps: analyzing the flow state of the absorbent solution in the desorption tower based on the filler type, specific surface area, porosity and bulk density parameters, obtaining the gas-liquid two-phase flow rate and gas-liquid ratio distribution at different positions in the tower, analyzing the influence of the filler size and the tower diameter ratio on the pressure drop, and establishing a relationship curve between the pressure drop and the flow rate, including: obtaining the filler type, specific surface area, porosity and bulk density parameters of the desorption tower, and calculating the pressure gradient at different positions in the desorption tower using the Stichlmair two-phase flow pressure drop model according to the parameters; obtaining the pressure distribution in the desorption tower by a numerical integration method, and determining the gas-liquid two-phase flow in the desorption tower in combination with the continuity equation; and obtaining the pressure distribution in the desorption tower by a numerical integration method. velocity distribution; establishing a two-dimensional axisymmetric model of the desorption tower, setting the porous medium parameters of the desorption tower packing area, wherein the porous medium parameters include the porosity and the inertial resistance coefficient and the viscous resistance coefficient calculated from the packing characteristics; obtaining the flow field information in the desorption tower, and extracting the gas-liquid two-phase flow velocity and the gas-liquid ratio distribution from the flow field information; performing feature recognition on the gas-liquid ratio distribution to determine the flow state of the solution in the desorption tower; if the flow state of the solution in the desorption tower meets the preset conditions, recording the pressure drop data of the desorption tower; fitting the relationship curve between the pressure drop data and the flow velocity by the least squares method, and obtaining the power function relationship between the pressure drop and the flow velocity; The method includes performing integral calculation on the desorption tower according to the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower, obtaining a distribution curve of the cumulative desorption amount of the refrigerant along the tower height, and combining the relationship between the mass transfer efficiency, the specific surface area of ​​the filler, and the tower diameter to determine the main factors affecting the mass transfer efficiency, including: obtaining the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower; performing integral calculation on the desorption tower according to the data using a trapezoidal numerical integration method to obtain the cumulative desorption amount distribution of the refrigerant; performing cubic spline interpolation on the cumulative desorption amount data to generate a distribution curve of the cumulative desorption amount along the tower height; calculating the average value and standard deviation of the local mass transfer efficiency, and if the average value plus or minus two times the standard deviation exceeds a preset threshold, it is determined to be the optimal mass transfer area with the highest desorption efficiency; using a multiple regression analysis method to establish a relationship model between the mass transfer efficiency, the specific surface area of ​​the filler, and the tower diameter, with the mass transfer efficiency as the dependent variable, and the specific surface area of ​​the filler and the tower diameter as the independent variables; evaluating the importance of each factor to the mass transfer efficiency, and determining the importance ranking of each factor by calculating the average impurity reduction; According to the local heat transfer coefficient of the filler surface and the main factors affecting the mass transfer efficiency, a systematic analysis is performed on the geometric parameters of the filler specific surface area, tower diameter, tower height and solution flow rate operating parameters to set the parameter variation range, and the influence of each parameter on the system performance is quantified through parameter scanning and sensitivity analysis to obtain parameter analysis results, including: obtaining uniformly distributed sample points generated by the Latin hypercube sampling method; constructing a simplified model of the desorption tower according to the sample points, and the simplified model uses a second-order polynomial to fit the relationship between the system performance and the parameters; judging the importance ranking of each parameter in the Pareto diagram, if the parameter importance exceeds a preset threshold, it is determined to be a key parameter; calculating the sensitivity coefficient for the key parameter, and the sensitivity coefficient is obtained by the central difference method; using the NSGA-II algorithm for multi-objective optimization, the multi-objective optimization combines the three performance indicators of total heat transfer, desorption efficiency and pressure drop at the same time to obtain a Pareto optimal solution set, and selecting the best parameter combination from the Pareto optimal solution set.

2. The method according to claim 1, wherein: The method calculates the heat generated by ammonia combustion according to the geometric structure and fuel supply of the ammonia combustion chamber, analyzes the heat transfer mode between the combustion chamber and the desorption tower, including direct contact heat transfer and partition heat transfer, and obtains and draws the heat flux density distribution on the tower wall by simulating the heat transfer process, including: obtaining the geometric structure parameters and fuel supply data of the ammonia combustion chamber, calculating the ammonia combustion reaction rate, and solving the energy conservation equation to obtain the temperature distribution in the combustion chamber; establishing two heat transfer models, direct contact and partition heat transfer models, according to the temperature distribution of the combustion chamber, and performing finite element analysis based on the two heat transfer models to obtain the temperature field distribution under the two heat transfer modes; solving the heat conduction partial differential equation for the combustion chamber temperature distribution and the two heat transfer models, and obtaining the temperature value of each point on the desorption tower through iterative calculation; calculating the heat flux density value of each point on the tower wall according to the temperature gradient data of the desorption tower by applying Fourier's heat conduction law; generating a continuous heat flux density distribution function by using a bilinear interpolation algorithm, and if the continuous heat flux density distribution function is generated, drawing a thermodynamic diagram of the heat flux density distribution on the tower wall.

3. The method according to claim 1, wherein: The method analyzes the desorption process of the refrigerant in the solution according to the flow state and the relationship curve between the pressure drop and the flow rate, and calculates the local desorption rate and the amount of refrigerant vapor generated at different positions in the desorption tower based on the solution concentration, the boiling point of the refrigerant and the vapor pressure, including: receiving a model establishment request carrying the structural parameters of the desorption tower, the model establishment request is issued by the FLUENT software; establishing a fluid flow and heat transfer coupling model in the desorption tower according to the model establishment request, the model is used to solve the temperature distribution and flow velocity field in the tower; obtaining the activity calculation result of the refrigerant in the liquid phase, the activity calculation result is obtained by the NRTL activity coefficient model; Solving the vapor pressure of the refrigerant at different temperatures and pressures according to the activity calculation results; calculating the gas-liquid interface mass transfer coefficient, which is used to solve the mass transfer equation of the desorption process using a finite difference method; Calculate the cumulative desorption amount of each section in the tower, and the cumulative desorption amount is used to calculate the amount of refrigerant vapor generated using the principle of mass conservation; set the differential equation dM / dz=f(z, M, C), where M is the vapor generation, z is the tower height, and C is the solution concentration; solve the differential equation, and the solution of the differential equation is used to obtain the distribution of the refrigerant vapor generation in the tower height direction.

4. The method according to claim 1, wherein: The method calculates the solution temperature distribution according to the heat flux density distribution, and obtains the local heat transfer coefficient and mass transfer efficiency distribution on the packing surface by analyzing the heat transfer and mass transfer process in the desorption tower, including: establishing a three-dimensional temperature field numerical model in the desorption tower, the numerical model including inlet temperature and wall heat flow boundary conditions; using the numerical model to solve and obtain the solution temperature distribution; obtaining the physical properties of the solution and the refrigerant, and using FLUENT to construct a heat transfer and mass transfer coupling model in the desorption tower according to the physical properties; using the coupling model to simulate the gas-liquid two-phase flow, and the mass transfer term between the gas and liquid phases in the simulation process adopts the Lee model, and the turbulence model selects the RNGk-ε model; obtaining the velocity field and concentration field on the packing surface according to the simulation results; obtaining the temperature gradient and heat flux density on the packing surface, and calculating the local heat transfer coefficient according to the temperature gradient and heat flux density; using the least squares method to fit the relationship between the local heat transfer coefficient and the Reynolds number and the Prandtl number; obtaining the actual mass transfer flux and the ideal mass transfer flux on the packing surface, and calculating the local mass transfer efficiency according to the actual mass transfer flux and the ideal mass transfer flux.

5. The method according to claim 1, wherein: According to the parameter analysis results, the packing specific surface area, tower diameter and tower height parameters under different solution flow conditions are optimized to obtain the optimal geometric parameter combination under each flow rate, and the performance of the evaluation system under different cooling loads and parameter combinations is combined to obtain the operating characteristics of the system under various load conditions, including energy efficiency ratio changes and responsiveness, and obtain an ammonia combustion refrigeration system performance analysis report, including energy saving potential and environmental impact assessment under various operating conditions, including: obtaining the packing specific surface area, tower diameter and tower height parameters, and performing multi-objective optimization on the parameters; according to the optimized geometric parameter combination, by simplifying the dynamics of the ammonia combustion refrigeration system dynamic model; input the optimized geometric parameter combination, set different cooling load conditions, solve the differential equations, and record the energy efficiency ratio, cooling capacity and response time performance indicators; use the polyfit function of the numpy library to perform polynomial fitting to obtain the energy efficiency ratio change curve; if the energy efficiency ratio change curve meets the preset conditions, determine the optimal operating load range of the system; calculate environmental impact indicators such as carbon emissions and energy consumption; construct a fuzzy comprehensive evaluation model and set the triangle membership function; quantitatively evaluate the system performance under different operating conditions through the fuzzy weighted average method; determine the degree of influence of different parameters on system performance; If the degree of influence exceeds a preset threshold, the robustness of the optimization result is verified.

Images