A refrigeration station energy efficiency model fitting method and device

Abstract

The application discloses a refrigeration station energy efficiency model fitting method and device, obtains a chilled water unit operation power model, a refrigeration pump operation power model, a cooling pump operation power model, a cooling tower fan operation power model and refrigerating capacity of the chilled water unit; constructs a total operation power model of the refrigeration station by using the chilled water unit operation power model, the refrigeration pump operation power model, the cooling pump operation power model and the cooling tower fan operation power model; and establishes the refrigeration station energy efficiency model according to the refrigerating capacity and the total operation power model; the application combines a data driving method and a physical modeling method, fully gives play to respective advantages of machine learning and the physical modeling method, improves fitting precision and accuracy of the refrigeration station energy efficiency model, and lays a foundation for subsequent energy saving and optimal control of the refrigeration station.

Description

Technical Field

[0001] This invention belongs to the field of energy efficiency analysis technology for refrigeration plants, and particularly relates to a method and device for fitting an energy efficiency model for refrigeration plants. Background Technology

[0002] Air conditioning systems are increasingly widely used in various types of buildings. While improving people's living standards, air conditioning systems also cause a large amount of energy consumption. The energy consumption of refrigeration plant systems accounts for 60% to 80% of the total energy consumption of air conditioning systems. The energy efficiency model fitting of refrigeration plants is related to the energy consumption calculation of the entire system and subsequent energy-saving optimization control. Therefore, the study of energy efficiency model fitting of refrigeration plant systems is of great significance to air conditioning systems.

[0003] Currently, the main methods for fitting energy efficiency models of refrigeration plants include: 1) Statistical regression method: By collecting and analyzing a large amount of experimental data, statistical regression methods are used to fit the model, such as linear regression, multinomial regression, or nonlinear regression; 2) Machine learning method: Machine learning algorithms, such as neural networks, support vector machines, decision trees, etc., are used to train the model to fit the energy efficiency model of the refrigeration plant; 3) Physical modeling method: Based on the physical principles and energy balance equations of the refrigeration plant, a mathematical model is established, and experimental data is used for parameter fitting and correction.

[0004] However, due to the large number of devices and complex structure of the refrigeration station system, as well as its complex characteristics such as nonlinearity, large time lag, time variation and strong coupling, it has brought difficulties to mechanism modeling and energy efficiency model fitting. For such a complex nonlinear model of the refrigeration station, the effect of a single fitting method is limited and cannot accurately capture the dynamic characteristics of the system. Summary of the Invention

[0005] The purpose of this invention is to provide a method and apparatus for fitting an energy efficiency model of a refrigeration plant, so as to improve the accuracy of the fitting model of the refrigeration plant.

[0006] This invention adopts the following technical solution: a method for fitting an energy efficiency model of a refrigeration station, comprising the following steps:

[0007] The operating power models of chiller units, chilled water pumps, cooling pumps, cooling tower fans, and chiller unit cooling capacity were obtained. The chiller unit operating power model was fitted using a support vector machine regression algorithm with multi-core learning, and the fitted dataset included cooling capacity, outdoor wet-bulb temperature, water supply temperature, and chiller unit energy efficiency.

[0008] A total operating power model for the refrigeration plant is constructed by using the operating power models of the chiller unit, chilled water pump, cooling pump, and cooling tower fan.

[0009] An energy efficiency model for the refrigeration station is established based on the refrigeration capacity and total operating power model.

[0010] Furthermore, the chiller unit operating power model was fitted using a support vector machine regression algorithm with multi-kernel learning, resulting in the following:

[0011] Construct an energy efficiency model for the chiller unit;

[0012] Based on the fitted dataset, the parameters of the chiller unit energy efficiency model are solved using the support vector machine regression algorithm; among them, the sigmoid kernel function and the Gaussian radial basis kernel function are selected during the solution process.

[0013] Based on the solved energy efficiency model of the chiller unit, a power model for the chiller unit's operation is constructed.

[0014] Furthermore, the operating power model of the chiller unit is as follows:

[0015] P set =Q set / COP set ,

[0016] Among them, P set Q represents the operating power of the chiller unit. set For cooling capacity, COP set Energy efficiency of chiller units;

[0017] The energy efficiency of the chiller unit is obtained through the chiller unit energy efficiency model, which is as follows:

[0018] COP set (x i )=ω·φ(x i )+b,

[0019] Among them, COP set (x i ) indicates that the input quantity is x i The corresponding chiller unit energy efficiency, x i Includes cooling capacity, outdoor wet-bulb temperature, and supply water temperature; ω is the weight vector; b is the bias constant; φ(x) i ) represents the kernel function, which includes the sigmoid kernel function φ1 and the Gaussian radial basis kernel function φ2.

[0020] Furthermore, the operating power model of the chilled pump is as follows:

[0021]

[0022] Among them, P pumpf For the operating power of the chilled pump, γ f V is the specific weight of the fluid in the refrigeration pump. pumpf H represents the flow rate of the chilled pump. fFor the head of the refrigeration pump, η pumpf For the efficiency of the chilled pump, η pumpf,d For the refrigeration pump drive efficiency, η pumpf,g For the efficiency of the chilled water pump motor, η pumpf,f For inverter efficiency.

[0023] Furthermore, the cooling pump operating power model is as follows:

[0024]

[0025] Among them, P pumpc For the operating power of the cooling pump, γ c V is the density of the fluid in the cooling pump. pumpc H is the flow rate of the cooling pump. c For the cooling pump head, η pumpc For cooling pump efficiency, η pumpc,d For the cooling pump drive efficiency, η pumpc,g For the efficiency of the cooling pump motor, η pumpc,f For the efficiency of the cooling pump frequency converter.

[0026] Furthermore, the operating power model for the cooling tower fan is as follows:

[0027] P tower =u2·k tower ,

[0028] Among them, P tower Let u2 be the operating power of the cooling tower fan, u2 be the fitting parameter, and k be the operating power of the cooling tower fan. tower This refers to the fan speed ratio of the cooling tower.

[0029] Furthermore, the calculation method for the cooling tower fan speed ratio is as follows:

[0030] k tower =V a / u1,

[0031] Among them, V a U is the airflow rate, and u1 is the fitting parameter.

[0032] Furthermore, the airflow rate is obtained using a simplified cooling tower model, which is as follows:

[0033]

[0034] Among them, Q ch For the heat dissipation of the cooling tower, d1, d2, and d3 are fitting parameters, V w T represents the cooling water flow rate. cin T represents the inlet temperature of the cooling water. w This refers to the outdoor wet-bulb temperature.

[0035] Another technical solution of the present invention: a refrigeration station energy efficiency model fitting device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-mentioned method.

[0036] The beneficial effects of this invention are as follows: This invention combines data-driven and physical modeling methods, giving full play to the advantages of each of the machine learning and physical modeling methods. Moreover, the multi-kernel learning support vector machine regression algorithm achieves better regression results by combining the advantages of different kernel functions, improving the accuracy and precision of the energy efficiency model fitting of the refrigeration station, reducing the consumption of computing resources, making the fitting method more practical and feasible, and laying the foundation for the subsequent energy saving and optimized control of the refrigeration station. Attached Figure Description

[0037] Figure 1 This is a flowchart of the model fitting process for the chiller unit in an embodiment of the present invention;

[0038] Figure 2 This is a schematic diagram of the comprehensive energy efficiency calculation method for a refrigeration station in an embodiment of the present invention. Detailed Implementation

[0039] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0040] This invention discovers that combining data-driven and physical modeling methods can improve the fitting accuracy and stability of the energy efficiency model of a refrigeration plant, fully tapping the energy-saving potential of the refrigeration plant. The energy-consuming equipment in a refrigeration plant system mainly includes chillers, chilled water pumps, cooling pumps, and cooling tower fans. Chillers account for approximately 55% of the total system energy consumption, chilled water pumps and cooling pumps account for about 38%, and cooling tower fans account for 7%. Furthermore, the energy efficiency of each piece of equipment is interrelated.

[0041] The refrigeration station system mainly consists of a chiller unit, a chilled water circuit, and a cooling water circuit. The chiller unit is the core component, and it mainly consists of four parts: an evaporator, a compressor, a condenser, and an expansion valve. The chiller unit is mainly responsible for generating cooling capacity, and its working principle is a cycle in which the refrigerant continuously absorbs and releases heat.

[0042] First, the liquid refrigerant absorbs heat from the chilled water return in the evaporator and turns into a gaseous state. Then, it is compressed into a high-temperature, high-pressure gas by the compressor. Next, the high-temperature, high-pressure gas transfers heat to the cooling water through heat exchange in the condenser. The cooling water carries the heat to the cooling tower and discharges it. The gaseous refrigerant cools down and turns into a liquid state. Then, the low-temperature, high-pressure liquid refrigerant flows through the expansion valve under the system's drive, and becomes a low-temperature, low-pressure liquid refrigerant that is sent back to the evaporator to enter the next cycle.

[0043] The chilled water pump and chilled water piping together form the chilled water circuit. Matching the refrigerant circulation process described above, the chilled water first exchanges heat with the refrigerant at the evaporator, is cooled, and then pressurized by the chilled water pump and sent into the chilled water piping. It then flows to the air handling unit at the air conditioning terminal, where it exchanges heat with the air entering the room, removing heat from the air. The cooled air then enters the room, achieving the purpose of lowering the indoor temperature. Simultaneously, the chilled water absorbs heat from the air, its temperature rises, and the heated chilled water then re-enters the evaporator to exchange heat with the refrigerant, repeating the above process.

[0044] The cooling pump, cooling water pipes, and cooling tower together form the cooling water circuit. During system operation, the cooling water absorbs heat released by the refrigerant at the condenser, thus increasing its temperature. Then, the cooling pump pumps the heated cooling water into the cooling tower, allowing it to exchange heat with the external environment. Finally, the cooled cooling water returns to the chiller unit to begin the next cycle.

[0045] This invention discloses a method for fitting an energy efficiency model of a refrigeration station, such as... Figure 1 As shown, the process includes the following steps: obtaining the operating power model of the chiller unit, the operating power model of the chilled water pump, the operating power model of the cooling pump, the operating power model of the cooling tower fan, and the cooling capacity of the chiller unit; wherein, the operating power model of the chiller unit is obtained by fitting a support vector machine regression algorithm with multi-kernel learning, and the fitted dataset includes cooling capacity, outdoor wet-bulb temperature, supply water temperature, and chiller unit energy efficiency; constructing a total operating power model of the refrigeration station based on the operating power model of the chiller unit, the operating power model of the chilled water pump, the operating power model of the cooling pump, and the operating power model of the cooling tower fan; and establishing an energy efficiency model of the refrigeration station based on the cooling capacity and the total operating power model.

[0046] This invention combines data-driven and physical modeling methods, giving full play to the advantages of both machine learning and physical modeling methods. Furthermore, the multi-kernel learning support vector machine regression algorithm achieves better regression results by combining the advantages of different kernel functions, thereby improving the accuracy and precision of the energy efficiency model fitting for refrigeration plants, reducing the consumption of computing resources, and making the fitting method more practical and feasible. This lays the foundation for subsequent energy saving and optimized control of refrigeration plants.

[0047] In this embodiment of the invention, the energy efficiency model for the refrigeration station is specifically as follows:

[0048]

[0049] Among them, COP system Q is the system energy efficiency ratio (i.e., the overall energy efficiency of the refrigeration station). set The cooling capacity of the chiller unit is expressed in kW (P). total This represents the total operating power of all equipment in the refrigeration station, expressed in kW.

[0050] Based on the above analysis, we can conclude that:

[0051] P total =P set +P pumpf +P pumpc +P tower (2)

[0052] Among them, P set P represents the operating power of the chiller unit. pumpf P is the operating power of the chilled water pump. pumpc P is the operating power of the cooling pump. tower The values ​​represent the operating power of the cooling tower fan, all in kW.

[0053] In one embodiment, such as Figure 2 As shown, the chiller unit operating power model is obtained by fitting a support vector machine regression algorithm with multi-kernel learning, including: constructing a chiller unit energy efficiency model; solving for the parameters of the chiller unit energy efficiency model using a support vector machine regression algorithm based on the fitted dataset; wherein, the sigmoid kernel function and the Gaussian radial basis kernel function are selected during the solution process; and constructing a chiller unit operating power model based on the solved chiller unit energy efficiency model.

[0054] For the model fitting of the chiller unit, this embodiment uses a multi-core learning support vector machine regression algorithm. Support vector regression (SVR) is a regression method based on support vector machines. Unlike traditional regression methods, SVR focuses on finding a boundary such that most training samples are located inside the boundary, and the distance from the boundary to the training samples is minimized.

[0055] In this embodiment, the influencing factors are first analyzed, and the energy efficiency COP of the chiller unit is evaluated. set After analyzing the influencing factors, it was found that the main influencing factor is the cooling load Q. set (That is, the cooling capacity of the chiller unit), outdoor wet-bulb temperature T w and water supply temperature T eo Then, data acquisition and processing are carried out. The data collected from the operation of the refrigeration station is saved and processed. Abnormal data is removed and missing values ​​are interpolated. For the given training dataset, preprocessing operations such as feature scaling and standardization are performed to ensure the scale consistency of each feature.

[0056] Then, a model is defined by fitting the data through a hyperplane constructed in the feature space, which is determined by the support vectors. The hyperplane can be represented as a linear or nonlinear function.

[0057] Next, we construct the objective function, which consists of two parts: one part is the constraint condition to ensure that the training samples are inside the hyperplane, and the other part is to minimize the distance from the hyperplane to the training samples (x, y). This gives us an optimization problem, and by solving this problem, we can find the optimal hyperplane.

[0058] Finally, the optimization problem is solved using the quadratic programming method in convex optimization theory. The optimal solution is obtained by solving the dual problem, and the fitting parameters are obtained.

[0059] The idea behind the Support Vector Machine (SVM) regression algorithm is to map the dataset to a high-dimensional feature space H using a kernel function φ, and then perform linear regression in this space. This cleverly transforms the original nonlinear problem into a linear problem in a high-dimensional space, achieving the effect of linear regression in the original space. Specifically, the functional form (i.e., the chiller unit energy efficiency model) can be expressed as:

[0060] COP set (x i )=ω·φ(x i )+b (3)

[0061] Among them, COP set (x i ) indicates that the input quantity is x i The corresponding chiller unit energy efficiency (in terms of output), x i Includes cooling capacity, outdoor wet-bulb temperature, and supply water temperature; ω is the weight vector; b is the bias constant; φ(x) i ) represents the kernel function, which includes the sigmoid kernel function φ1 and the Gaussian radial basis kernel function φ2.

[0062] Thus, linear regression in the high-dimensional feature space corresponds to nonlinear regression in the low-dimensional space, and avoids the calculation of the dot product of ω and φ in the high-dimensional space. ω and b in equation (3) can be estimated by minimizing equation (4).

[0063]

[0064] The first term (ω·ω) represents the function COP. set (x i The complexity of ); the second term represents empirical risk, where |y i -COP set (x i )| ε =max{0,|y i -f(x i)|-ε} is the insensitive loss function, which is used to represent the decision function with sparse points. ε is the maximum error allowed by the regression. ε controls the number of support vectors and the generalization ability. The larger the value, the fewer the support vectors. C is a positive constant, which reflects the trade-off between the complexity of the function class and the average loss on the training set. Using the duality principle, the Lagrangian function and the kernel function are introduced at the same time to transform the minimization of equation (4) into an optimization problem, and the bias b is solved according to the Kuhn-Tucker theorem.

[0065] In this embodiment of the invention, the support vector machine regression algorithm is improved by incorporating multi-kernel learning and combining the advantages of different kernel functions to achieve better regression results. Therefore, two kernel functions are selected: the sigmoid kernel function φ1 and the Gaussian radial basis kernel function φ2.

[0066] φ1(x1,x2)=tanh(g1*x1*x2+coef) (5)

[0067] φ2(x1,x2)=exp(-g2*||x1-x2|| 2 (6)

[0068] Where x1 and x2 are different input samples x i g1, g2, and coef are the parameters of the kernel function, respectively.

[0069] After establishing the above multi-core learning support vector machine regression algorithm model, then using cold load Q... set Outdoor wet-bulb temperature T w and water supply temperature T eo As input, the energy efficiency COP of the chiller unit. set To output the model, we train it to obtain the trained parameters, which involves using a multi-kernel learning support vector machine regression algorithm to calculate the energy efficiency COP of the chiller unit. set Perform fitting calculations.

[0070] Therefore, the operating power model of the chiller unit can be obtained as follows:

[0071] P set =Q set / COP set (7)

[0072] Q ch =P set *(1+COP set (8)

[0073] Among them, P set Q represents the operating power of the chiller unit. set For cooling capacity, COP set Energy efficiency of chiller units.

[0074] In one embodiment, the operating power model of the chilled pump is as follows:

[0075]

[0076] Among them, P pumpf For the operating power of the chilled pump, γ f V is the specific weight of the fluid in the refrigeration pump. pumpf H represents the flow rate of the chilled pump. f For the head of the refrigeration pump, η pumpf For the efficiency of the chilled pump, η pumpf,d For the refrigeration pump drive efficiency, η pumpf,g For the efficiency of the chilled water pump motor, η pumpf,f For inverter efficiency.

[0077] For the refrigeration pump, this embodiment adopts a physical mechanism modeling method. First, a mathematical model is established, then data is collected and processed, and parameters are fitted based on this data.

[0078] Refrigeration pump flow rate V pumpf calculate:

[0079]

[0080] Where c is the specific heat capacity of water, which is taken as 4.2 in this embodiment; dT is the temperature difference of chilled water, which is taken as 7 in this embodiment.

[0081] The formulas for chilled water pipeline curves, chilled water pump HV curves, and chilled water pump operating power are as follows:

[0082]

[0083]

[0084]

[0085]

[0086] Among them, H f For refrigeration pump head; S f The impedance of the refrigeration system piping network; a pumpf b pumpf and c pumpf All are fitting parameters; k pumpf η is the speed ratio of the refrigeration pump. pumpf For the efficiency of the chilled pump, η pumpf,d For transmission efficiency, η pumpf,g For the efficiency of the electric motor, η pumpf,f For inverter efficiency, g f1 and g f2All are fitting parameters; f0, f1, f2, and f3 are all fitting parameters, γ f The specific weight of the fluid in the refrigeration pump, in kN / m³. 3 When water is a fluid, take 9.8.

[0087] For cooling pumps and cooling towers, a physical mechanism modeling method is also used. First, a mathematical model is established, then data is collected and processed, and parameters are fitted based on this data.

[0088] The model in this embodiment of the invention needs to treat the cooling pump and cooling tower as a whole for modeling, and then calculate the heat exchange capacity of the chiller unit as Q. ch The minimum total operating power of the cooling tower and cooling pump at that time is taken as the operating power of the entire cooling system.

[0089] Specifically, the cooling pump operating power model is as follows:

[0090]

[0091] Among them, P pumpc For the operating power of the cooling pump, γ c V is the density of the fluid in the cooling pump. pumpc H is the flow rate of the cooling pump. c For the cooling pump head, η pumpc For cooling pump efficiency, η pumpc,d For the cooling pump drive efficiency, η pumpc,g For the efficiency of the cooling pump motor, η pumpc,f For the efficiency of the cooling pump frequency converter.

[0092] The formulas for cooling water pipeline curves, cooling pump HV curves, and cooling pump operating power are as follows:

[0093]

[0094]

[0095]

[0096]

[0097] Among them, S c Given the resistance of the cooling system piping network, a pumpc b pumpc and c pumpc k are the fitting parameters. pumpc η is the speed ratio of the cooling pump. pumpc For the efficiency of the cooling pump, g c1 and g c2 Here, c0, c1, c2, and c3 are the fitting parameters; γ cThe specific weight of the fluid, in kN / m³ 3 When water is a fluid, take 9.8.

[0098] For the cooling pump speed ratio k pumpc Generally speaking, it is related to the adjustable range of the frequency converter. In this embodiment, it is taken as 0.5 to 1. Therefore, k is respectively... pumpc =0.5 and k pumpc Substituting 1 into formulas (16) and (17), the corresponding cooling pump flow rate V is calculated. pumpc1 and V pumpc2 This refers to the range of flow rates for the cooling pump.

[0099] If there are n cooling pumps, then the flow rate range of each of the n cooling pumps needs to be calculated. The sum of the flow rates of all cooling pumps equals the cooling water flow rate.

[0100] The airflow rate is then obtained using a simplified cooling tower model, which is as follows:

[0101]

[0102] Among them, Q ch For the heat dissipation of the cooling tower, d1, d2, and d3 are fitting parameters, V w T represents the cooling water flow rate. cin T represents the inlet temperature of the cooling water. w The outdoor wet-bulb temperature is used. In this embodiment of the invention, the cooling water inlet temperature is 1 to 5°C higher than the outdoor wet-bulb temperature.

[0103] If there are m cooling towers, then it is necessary to calculate the operating power of each cooling tower when 1 to m cooling towers are turned on.

[0104] The calculation method for the cooling tower fan speed ratio is as follows:

[0105] k tower =V a / u1 (21)

[0106] Among them, V a U is the airflow rate, and u1 is the fitting parameter.

[0107] Therefore, the operating power model of the cooling tower fan is as follows:

[0108] P tower =u2·k tower (twenty two)

[0109] Among them, P tower Let u2 be the operating power of the cooling tower fan, u2 be the fitting parameter, and k be the operating power of the cooling tower fan. tower This refers to the fan speed ratio of the cooling tower.

[0110] First, calculate Q using formula (20).ch T w T w T-size 1 to 5 cin and cooling pump flow rate V pumpc1 To V pumpc2 The system calculates all airflow rates and then uses the formula to determine the operating power of the cooling tower for each airflow rate. Finally, it calculates and determines Q. ch T w T cin Find the minimum total operating power of the cooling pump and cooling tower, and record the corresponding cooling pump flow rate V at this time. minc The speed ratio of the cooling tower fan to k tower .

[0111] The present invention also discloses a refrigeration station energy efficiency model fitting device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method.

[0112] The present invention implements all or part of the processes in the methods of the above embodiments, which can be accomplished by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a storage device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0113] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

Claims

1. A method for fitting an energy efficiency model of a refrigeration station, characterized in that, Includes the following steps: The chiller unit operating power model, chilled water pump operating power model, cooling pump operating power model, cooling tower fan operating power model, and chiller unit cooling capacity are obtained; wherein, the chiller unit operating power model is obtained by fitting a support vector machine regression algorithm with multi-core learning, and the fitted dataset includes the cooling capacity, outdoor wet-bulb temperature, water supply temperature, and chiller unit energy efficiency. A total operating power model for the refrigeration plant is constructed by using the operating power models of the chiller unit, chilled water pump, cooling pump, and cooling tower fan. Establish an energy efficiency model for the refrigeration station based on the aforementioned cooling capacity and total operating power model; The chiller unit operating power model was fitted using a support vector machine regression algorithm with multi-kernel learning, resulting in the following: Construct an energy efficiency model for the chiller unit; Based on the fitted dataset, the parameters of the chiller unit energy efficiency model are solved using the support vector machine regression algorithm; wherein, the sigmoid kernel function and the Gaussian radial basis kernel function are selected during the solution process. Based on the solved chiller unit energy efficiency model, construct the chiller unit operating power model; The operating power model of the chiller unit is as follows: ， in, This refers to the operating power of the chiller unit. The cooling capacity, Energy efficiency of chiller units; The energy efficiency of the chiller unit is obtained through a chiller unit energy efficiency model, which is as follows: ， in, Indicates the input quantity is The corresponding chiller unit energy efficiency, This includes cooling capacity, outdoor wet-bulb temperature, and water supply temperature. For weight vectors, The bias constant is The kernel function includes the sigmoid kernel function. and Gaussian radial basis kernel function .

2. The energy efficiency model fitting method for a refrigeration station as described in claim 1, characterized in that, The operating power model of the chilled pump is as follows: ， in, For the operating power of the chilled pump, The specific gravity of the fluid in the refrigeration pump. This refers to the flow rate of the chilled water pump. For the head of the refrigeration pump, For chilled pump efficiency, For the refrigeration pump drive efficiency, For the efficiency of the chilled water pump motor, For inverter efficiency.

3. The energy efficiency model fitting method for a refrigeration station as described in claim 2, characterized in that, The operating power model of the cooling pump is as follows: ， in, For the operating power of the cooling pump, The specific gravity of the fluid in the cooling pump. For cooling pump flow rate, For cooling pump head, For cooling pump efficiency, For the cooling pump drive efficiency, For the efficiency of the cooling pump motor, For the efficiency of the cooling pump frequency converter.

4. The method for fitting an energy efficiency model of a refrigeration station as described in claim 3, characterized in that, The operating power model of the cooling tower fan is as follows: ， in, For the operating power of the cooling tower fan, For the fitting parameters, This refers to the fan speed ratio of the cooling tower.

5. The method for fitting an energy efficiency model of a refrigeration station as described in claim 4, characterized in that, The calculation method for the cooling tower fan speed ratio is as follows: ， in, For airflow, These are the fitting parameters.

6. The method for fitting an energy efficiency model for a refrigeration station as described in claim 5, characterized in that, The airflow rate is obtained using a simplified cooling tower model, which is as follows: ， in, For the heat dissipation of the cooling tower, , and All are fitted parameters. For cooling water flow rate, This refers to the inlet temperature of the cooling water. This refers to the outdoor wet-bulb temperature.

7. A device for fitting an energy efficiency model of a refrigeration station, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 6.

Images