A virtual cold quantity-based refrigeration system optimization control method

Abstract

The present application belongs to the technical field of energy-saving control of refrigeration system, and particularly relates to a refrigeration system optimization control method based on virtual cold energy. The method comprises: collecting field historical data of the refrigeration system and preprocessing; establishing component mechanism model based on the composition of the refrigeration system, wherein the component mechanism model comprises evaporator mechanism model, compressor mechanism model, condenser mechanism model and throttling valve model, all component mechanism models are connected in series based on the law of conservation of mass, and the to-be-determined coefficients are calculated through genetic algorithm; combining the refrigeration system model and particle swarm optimization algorithm, calculating the optimal operation control parameters under the given virtual cold energy and weather parameters, so as to minimize the energy consumption of the refrigeration system; and issuing the optimal operation control parameters to the equipment controller; re-collecting data to calculate the system model error, and repeating the above steps when the error is too large. The present application solves the problem that the refrigeration system optimization control method cannot be implemented under the scenario of missing cold energy data.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of energy-saving control of refrigeration systems, and particularly relates to a virtual cold quantity-based refrigeration system optimization control method. BACKGROUND

[0002] In recent years, the cold chain logistics industry has experienced rapid development, with the market size continuously expanding. The construction pace of national backbone cold chain logistics bases and cold chain facilities in production and sales areas is steady, and the technical level of cold chain equipment has also been significantly improved. However, with the clear proposal of the "double carbon" target, the cold chain logistics industry also faces the urgent task of energy saving and emission reduction. This requirement prompts the design of cold chain logistics refrigeration warehouses to keep pace with the times and adapt to the green and low-carbon development trend.

[0003] Under this background, digital twin technology has received widespread attention due to its ability to significantly improve energy utilization efficiency and reduce energy consumption and emissions. By introducing a digital twin model, accurate simulation and prediction of the operating state of the cold storage can be achieved, thereby achieving optimized control and achieving the goal of energy saving and emission reduction.

[0004] However, most current refrigeration systems do not have flow measurement devices in actual applications, which leads to a key problem: the lack of a key parameter for energy-saving control, refrigerant mass flow. During the operation of the optimization control system, the refrigerant mass flow needs to be accurately calculated to evaluate the working state of the compressor and determine the heat exchange of the evaporator and condenser. Therefore, how to accurately calculate the refrigeration capacity and then build an effective digital twin model for the cold storage to achieve optimized control and energy-saving goals has become a technical problem that needs to be solved urgently. SUMMARY

[0005] To solve the above problems in the prior art, the application provides a virtual cold quantity-based refrigeration system optimization control method, which solves the problem that the refrigeration system optimization control method cannot be implemented in the cold storage and other cold quantity data missing scenarios.

[0006] The technical solution of the application to solve the above technical problems is as follows:

[0007] The application provides a virtual cold quantity-based refrigeration system optimization control method, which includes the following steps:

[0008] Step 100: Collecting field historical data of the refrigeration system and pre-processing the collected field historical data, wherein the field historical data includes weather parameters and operating parameters;

[0009] Step 200: Establish component mechanism models based on the composition of the refrigeration system. The component mechanism models include evaporator mechanism model, compressor mechanism model, condenser mechanism model, and throttling valve model. Based on the law of conservation of mass, connect all component mechanism models in series to form a refrigeration system model. Through iterative calculation, obtain the correct refrigerant mass flow rate at the compressor inlet and calculate the compressor power. Compare the calculated compressor power with the theoretical power and optimize the undetermined coefficients.

[0010] Step 300: Combining the refrigeration system model and particle swarm optimization algorithm, calculate the optimal operating control parameters under the given virtual cooling capacity and weather parameters to minimize the energy consumption of the refrigeration system;

[0011] Step 400: Execute step 100 again, and use the newly generated training samples to calculate the model error in the refrigeration system mechanism model. When the refrigeration system model error is greater than the preset threshold, repeat steps 200 to 400 above.

[0012] Furthermore, the compressor mechanism model is a semi-empirical thermodynamic model, and its calculation process includes suction throttling calculation process, leakage adiabatic mixing calculation process, suction constant pressure heating calculation process, multi-directional compression process, exhaust constant pressure cooling calculation process, and exhaust throttling calculation process; through iterative calculation, the undetermined coefficients and mass flow rate inside the compressor are solved.

[0013] Furthermore, the condenser mechanism model includes: dividing the condenser heat exchange process into multiple micro-elements, the heat exchange of each micro-element being calculated using the heat transfer coefficient and refrigerant temperature; and calculating the refrigerant enthalpy at the condenser outlet using the condenser heat exchange.

[0014] Furthermore, the throttling valve mechanism model states that the enthalpy values ​​of the refrigerant at the inlet and outlet before and after the throttling valve are equal, the enthalpy value of the refrigerant at the inlet before the throttling valve is the enthalpy value of the refrigerant at the outlet of the condenser, and the enthalpy value of the refrigerant at the outlet after the throttling valve is the enthalpy value of the refrigerant at the inlet of the evaporator.

[0015] Furthermore, the evaporator mechanism model is as follows: the evaporator is divided into multiple micro-elements, and the heat exchange of each micro-element is calculated using the corresponding heat transfer coefficient and temperature; the enthalpy of the refrigerant at the evaporator outlet is calculated using the heat exchange of the evaporator.

[0016] Furthermore, the correct refrigerant mass flow rate at the compressor inlet is obtained through iterative calculation of the refrigerant enthalpy at the evaporator outlet and the refrigerant enthalpy at the compressor inlet; combined with the compressor model, the actual power of the compressor is calculated and compared with the theoretical power, and the undetermined coefficients are optimized through a genetic algorithm.

[0017] Furthermore, the intake throttling calculation process is simplified to an isentropic nozzle process:

[0018]

[0019] In the above formula, h1 represents the specific enthalpy of the refrigerant after suction throttling; h0 represents the specific enthalpy at the compressor inlet. The compressor's gas delivery mass flow rate is represented by d1; d1 represents the refrigerant density after suction throttling; A su Indicates the inhalation throttling area;

[0020] The leakage adiabatic mixing calculation process is a mixing process of leakage mass flow rate and suction mass, and the specific enthalpy of the mixed refrigerant is calculated as follows:

[0021]

[0022] In the above formula, Indicates the compressor leakage mass flow rate; h represents the actual refrigerant mass flow rate entering the rotor compression chamber. leak,in h1 represents the specific enthalpy of the refrigerant at the inlet of the compressor leakage channel; h2 represents the specific enthalpy of the refrigerant after adiabatic mixing.

[0023] The calculation process for intake constant pressure heating is as follows: the intake gas is heated by the casing, the refrigerant temperature rises, and the enthalpy increases. Based on assumptions, the calculation is finally obtained through iteration using the law of conservation of energy.

[0024]

[0025] In the above formula, Q 23 Indicates the heat exchange capacity during intake preheating; AU suc The heat transfer coefficient of the intake gas is represented by t. w t2 represents the average casing temperature; t2 represents the refrigerant temperature on the suction side; c p2 h3 represents the specific heat capacity of the refrigerant on the suction side at constant pressure; h3 represents the specific enthalpy of the refrigerant after suction preheating.

[0026] The compression process is divided into a multi-stage compression process and a forced exhaust process:

[0027]

[0028] p4 = p3·BVR α ;

[0029] In the above formula, BVR represents the internal volume ratio of the compressor; α represents the polynomial index; v4 represents the refrigerant specific volume after the internal compression ends; v3 represents the refrigerant specific volume before the internal compression begins; p4 represents the pressure at the end of the internal compression; p3 represents the refrigerant pressure after suction throttling.

[0030] The calculation process for exhaust constant pressure cooling is similar to that for exhaust throttling and the intake process. The leakage of the twin-screw compressor is assessed using the virtual flow path connecting the intake and exhaust.

[0031]

[0032] In the above formula, Indicates compressor leakage; d leak Indicates the refrigerant density at the inlet of the leakage path; A leak Indicates the compressor leakage area; h leak,in Indicates the specific enthalpy of the refrigerant at the inlet of the leakage path; h leak,out Indicates the specific enthalpy of the refrigerant at the outlet of the leakage path;

[0033] Actual gas delivery mass flow rate:

[0034]

[0035] This model calculates the input power, mass flow rate, and refrigerant outlet temperature of the compressor based on the compressor's suction and discharge side operating conditions.

[0036]

[0037] Where t7 represents the refrigerant outlet temperature; f(n,load,p) evp ,t evp ,p con ) represents the input parameter set of the compressor model; n represents the compressor speed; load represents the load position; p evp t represents the evaporation pressure on the intake side. evp p represents the evaporation temperature on the intake side. con This indicates the condensation pressure on the exhaust side.

[0038] Furthermore, the condenser mechanism model divides the heat exchange process into N segments, with each segment having a heat exchange capacity of Q. con for:

[0039]

[0040] Among them, K con,i t represents the average heat transfer coefficient of each infinitesimal element. con,i t represents the average temperature of the refrigerant in each micro-element. sr,i This indicates the average temperature of each micro-element wind.

[0041] The enthalpy of the refrigerant at the condenser outlet, h3, is calculated using the heat exchange, where h2 represents the enthalpy of the refrigerant at the compressor outlet, calculated using a compressor model.

[0042]

[0043] In the above formula, This indicates the refrigerant mass flow rate of the compressor.

[0044] Furthermore, the evaporator mechanism model divides the heat exchange process into N segments, with each segment having a heat exchange capacity of Q. evp for:

[0045]

[0046] Among them, K evp,i t represents the average heat transfer coefficient of each infinitesimal element. evp,i t represents the average temperature of the refrigerant in each micro-element. sr,i d represents the average temperature of the cold water in each micro-element; d represents the diameter of the heat exchange tube; l represents the length of the heat exchange tube.

[0047] The enthalpy at the evaporator outlet (h) is calculated based on the heat exchange rate of the evaporator. 1' :

[0048]

[0049] In the above formula, This indicates the mass flow rate of refrigerant delivered by the compressor.

[0050] Furthermore, the correct refrigerant mass flow rate at the compressor inlet is obtained by calculating the refrigerant enthalpy at the evaporator outlet and the refrigerant enthalpy at the compressor inlet. Calculate compressor power using a compressor model:

[0051]

[0052] P loss,1 =a tl,1 ·P i ;

[0053] P loss,2 =a tl,2 ·μ·V sw ·ω 2 ;

[0054] P loss,3 =a tl,3 ·m loss ·Δh loss ;

[0055] P com =P i +P loss,1 +P loss,2 +P loss,3 ;

[0056] Among them, P i P represents the internal compression power of the compressor. loss,1 P represents the torque loss during compressor operation. loss,2 P represents the viscous loss during the compressor's operation. loss,3P represents the additional power loss during partial load operation; com This indicates the total input power of the compressor; w i This represents the internal compression work, calculated from the multi-compaction compression process; a tl,1 Indicates the torque friction loss coefficient; a tl,2 The coefficient of viscous friction loss is represented by μ, where μ represents the viscosity of the lubricating oil and ω represents the compressor speed; a tl,3 This represents the partial load loss factor, m. loss Δh represents the refrigerant mass flow rate bypassed during partial load operation. loss This represents the theoretical enthalpy loss of the bypassed refrigerant during partial load operation; the difference between the calculated and actual power is used to determine the coefficients to be measured, including the leakage area A, through a genetic algorithm and field data. leak Inlet and outlet throttling area, inlet and outlet heat transfer coefficient, loss coefficient a tl,1 a tl,2 a tl,3 M is the data sample size:

[0057]

[0058] Where err represents the difference between the compressor power and the actual power, P com,i P represents the calculated power of the compressor. real,i This indicates the actual power.

[0059] Furthermore, in step 300, a particle swarm optimization algorithm is used, given an evaporation pressure p. evp Compressor inlet temperature t0, refrigerant inlet temperature t sp By adjusting the compressor speed n and load, the value at a certain condensing temperature t can be obtained. con Under these conditions, the minimum power consumption P of the system is:

[0060] [P,n,load,t con ]=f(p evp ,t0,t sp ).

[0061] Compared with the prior art, the present invention has the following technical effects:

[0062] The control method of this invention is applied to refrigeration systems that are not equipped with flow measurement devices. Based on the semi-empirical model coefficient correction method, the mass flow rate through the compressor is calculated, thereby modeling the refrigeration system and solving the problem that the optimization control method of the refrigeration system cannot be implemented in the case of missing cooling capacity data. Attached Figure Description

[0063] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0064] Figure 1 This is a diagram of the refrigeration system of the water-cooled unit implemented in this invention;

[0065] Figure 2 This is a schematic diagram of the working process of the screw refrigeration compressor of the present invention;

[0066] Figure 3 A flowchart illustrating the construction of a refrigeration system model for implementing this invention;

[0067] Figure 4 This is a flowchart illustrating the implementation of the present invention. Detailed Implementation

[0068] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the technical solutions proposed according to the present invention are described in detail below with reference to the accompanying drawings and preferred embodiments. Specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form. Unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0069] The internal structure of a refrigeration system includes a compressor, condenser, expansion valve, and evaporator. The compressor draws in refrigerant vapor from the evaporator and compresses it into a high-pressure gas. During compression, the compressor consumes mechanical energy, which is converted into heat energy, causing the refrigerant vapor to become superheated. The high-temperature refrigerant vapor discharged from the compressor releases heat in the condenser, transferring it to the surrounding water or air, causing the refrigerant vapor to gradually condense into a liquid. The refrigerant liquid exiting the condenser is depressurized to the evaporation pressure by the expansion valve. After passing through the expansion valve, the liquid refrigerant begins to boil and vaporize due to the reduced pressure, absorbing heat and thus achieving the purpose of refrigeration.

[0070] like Figure 1 As shown, a specific example of the present invention is a water-cooled unit, wherein P e Q is the input power of the compressor. e For the evaporator cooling load, Q c For the condenser heat load, T water,in T represents the inlet temperature of the cold water. water,out p is the outlet temperature of the cold water.ref,in t is the compressor suction pressure. ref,in p is the compressor suction temperature. ref,out This refers to the compressor discharge pressure. However, this refrigeration system is not equipped with a flow measurement device, making it impossible to collect or calculate the mass flow rate of the working fluid flowing through various parts of the system or the required refrigeration load at the evaporator end over a certain period of time through on-site data acquisition or calculation.

[0071] In one embodiment of the present invention, reference is made to... Figures 1-4 This paper provides an optimized control method for a refrigeration system based on virtual cooling capacity, comprising the following steps:

[0072] Step 100: Collect on-site historical data of the refrigeration system and preprocess the collected on-site historical data, wherein the on-site historical data includes weather parameters and operating parameters;

[0073] Step 200: Establish component mechanism models based on the composition of the refrigeration system. The component mechanism models include evaporator mechanism model, compressor mechanism model, condenser mechanism model, and throttling valve model. Based on the law of conservation of mass, connect all component mechanism models in series to form a refrigeration system model. Through iterative calculation, obtain the correct refrigerant mass flow rate at the compressor inlet and calculate the compressor power. Compare the calculated compressor power with the theoretical power and optimize the undetermined coefficients.

[0074] Step 300: Combining the refrigeration system model and particle swarm optimization algorithm, calculate the optimal operating control parameters under the given virtual cooling capacity and weather parameters to minimize the energy consumption of the refrigeration system; and send the optimal operating control parameters to the equipment controller.

[0075] Step 400: Execute step 100 again, and use the newly generated training samples to calculate the model error in the refrigeration system mechanism model. When the refrigeration system model error is greater than the preset threshold, repeat steps 200 to 400 above.

[0076] The following is a detailed explanation of each of the above steps:

[0077] Step 100: Collect on-site historical data of the refrigeration system and preprocess the collected on-site historical data, wherein the on-site historical data includes weather parameters and operating parameters.

[0078] As an example, this step may include the following steps:

[0079] Step 110: Collect on-site historical data of the refrigeration system, including weather parameters and operating parameters.

[0080] The operating parameters include pressure, temperature, compressor power, load, and frequency data. Only pressure and temperature sensors are installed in the refrigeration system, and compressor power, load, and frequency data are collected on the compressor side; the mass flow rate of the working fluid or the flow rate of the refrigerant on the other side are not collected.

[0081] Pressure sensors are used to monitor the pressure on the high-pressure and low-pressure sides of the refrigeration system. Temperature sensors are used to measure the temperature of key components of the refrigeration system, such as the evaporator outlet and condenser inlet. Temperature data is crucial for ensuring the system can provide the required cooling capacity and for monitoring the system's energy efficiency ratio.

[0082] Compressor power, load, and frequency reflect its operating status and efficiency. Power data reveals the compressor's energy consumption, while load and frequency data help understand the compressor's workload and regulation capabilities.

[0083] Step 120: Preprocess the collected on-site historical data as training samples for the subsequent refrigeration system model; wherein, the preprocessing includes data noise reduction, data filtering, data repair, data reduction and steady-state processing.

[0084] Data denoising aims to reduce random errors and noise in collected historical field data to improve data accuracy and reliability. Data denoising can employ wavelet thresholding and empirical mode decomposition (EMD). Wavelet transform decomposes a signal into components of different frequencies; by setting a threshold, wavelet coefficients below that threshold can be filtered out, thus removing noise. Empirical mode decomposition (EMD) is an adaptive signal decomposition method that decomposes complex signals into a series of intrinsic mode functions (IMFs); by selectively removing certain IMFs, noise suppression can be achieved.

[0085] Data filtering aims to remove outliers and unexpected data points. Data filtering can employ Mahalanobis distance and k-nearest neighbors (kNN) methods. Mahalanobis distance considers the correlation between data points and can more accurately identify outliers that are inconsistent with the overall data distribution; by setting a threshold, points exceeding that threshold can be filtered out. kNN filters based on the similarity between data points; for each data point, its distance to its k nearest neighbors is calculated; if a point's distance is significantly greater than other points, it may be considered an outlier and filtered out.

[0086] Data repair aims to fill in missing values ​​or repair damaged data. Data repair methods include linear interpolation and the Expectation-Maximization (EM) algorithm. Linear interpolation estimates missing values ​​by connecting known data points with straight lines. The Expectation-Maximization (EM) algorithm is used to estimate model parameters in the presence of missing data; by iteratively updating the parameters and missing values, data repair can be achieved.

[0087] Data reduction aims to reduce the dimensionality and complexity of data while retaining as much information as possible. Methods for data reduction include Latin hypercube sampling and Z-score normalization. Latin hypercube sampling is a method for generating multidimensional uniformly distributed samples; this method reduces the amount of data while preserving important features. Z-score normalization standardizes data by calculating the standard deviation of each data point from the mean, giving the data a uniform scale, which facilitates subsequent processing and analysis.

[0088] Steady-state processing aims to ensure that data is stable over time, meaning that the statistical properties of the data remain unchanged over a certain period of time. This is particularly important for time series analysis and forecasting.

[0089] Step 200: Establish component mechanism models based on the composition of the refrigeration system. The component mechanism models include evaporator mechanism model, compressor mechanism model, condenser mechanism model, and throttling valve model. Based on the law of conservation of mass, connect all component mechanism models in series to form a refrigeration system model. Through iterative calculation, obtain the correct refrigerant mass flow rate at the compressor inlet and calculate the compressor power. Compare the calculated compressor power with the theoretical power and optimize the undetermined coefficients.

[0090] As an example, this step may include the following steps:

[0091] Step 210: Establish a compressor mechanism model. The compressor mechanism model is a semi-empirical model of a twin-screw compressor. Its calculation process can be divided into suction throttling calculation process, leakage adiabatic mixing calculation process, suction constant pressure heating calculation process, multi-directional compression process, exhaust constant pressure cooling calculation process, and exhaust throttling calculation process. Through iterative calculation, solve for the undetermined coefficients and mass flow rate inside the compressor.

[0092] Reference Figure 2 0-1 represents the intake throttling calculation process, 1-2 represents the leakage adiabatic mixing calculation process, 2-3 represents the intake constant pressure heating calculation process, 3-5 represents the multi-directional compression process, 5-6 represents the exhaust constant pressure cooling calculation process, and 7-8 represents the exhaust throttling calculation process.

[0093] The intake throttling calculation process can be simplified to an isentropic nozzle process:

[0094]

[0095] In the above formula, h1 represents the specific enthalpy of the refrigerant after suction throttling; h0 represents the specific enthalpy at the compressor inlet. The compressor's gas delivery mass flow rate is represented by d1; d1 represents the refrigerant density after suction throttling; A su This indicates the inhalation throttling area.

[0096] The leakage adiabatic mixing calculation process is a mixing process of leakage mass flow rate and suction mass, and the specific enthalpy of the mixed refrigerant is calculated as follows:

[0097]

[0098] In the above formula, Indicates the compressor leakage mass flow rate; h represents the actual refrigerant mass flow rate entering the rotor compression chamber. leak,in h1 represents the specific enthalpy of the refrigerant at the inlet of the compressor leakage channel; h2 represents the specific enthalpy of the refrigerant after adiabatic mixing.

[0099] The calculation process for intake constant pressure heating is as follows: the intake gas is heated by the casing, the refrigerant temperature rises, and the enthalpy increases. Based on assumptions, the calculation is finally obtained through iteration using the law of conservation of energy.

[0100]

[0101] In the above formula, Q 23 Indicates the heat exchange capacity during intake preheating; AU suc The heat transfer coefficient of the intake gas is represented by t. w t2 represents the average casing temperature; t2 represents the refrigerant temperature on the suction side; c p2 h3 represents the specific heat capacity of the refrigerant on the suction side at constant pressure; h3 represents the specific enthalpy of the refrigerant after suction preheating.

[0102] The multi-phase compression process can be divided into a multi-phase compression process and a forced exhaust process:

[0103]

[0104] p4 = p3·BVR α ;

[0105] In the above formula, BVR represents the internal volume ratio of the compressor; α represents the polynomial index; v4 represents the refrigerant specific volume after the internal compression ends; v3 represents the refrigerant specific volume before the internal compression begins; p4 represents the pressure at the end of the internal compression; and p3 represents the refrigerant pressure after suction throttling.

[0106] The exhaust constant pressure cooling calculation process is similar to the exhaust throttling calculation process and the intake process. The leakage of the twin-screw compressor is evaluated by the virtual flow path connecting the intake and exhaust:

[0107]

[0108] In the above formula, Indicates compressor leakage; d leak Indicates the refrigerant density at the inlet of the leakage path; A leak Indicates the compressor leakage area; hleak,in Indicates the specific enthalpy of the refrigerant at the inlet of the leakage path; h leak,out This indicates the specific enthalpy of the refrigerant at the outlet of the leakage channel.

[0109] Actual gas delivery mass flow rate:

[0110]

[0111] This model calculates the input power, mass flow rate, and refrigerant outlet temperature of the compressor based on the compressor's suction and discharge side operating conditions.

[0112]

[0113] Where t7 represents the refrigerant outlet temperature; f(n,load,p) evp ,t evp ,p con ) represents the input parameter set of the compressor model; n represents the compressor speed; load represents the load position; p evp t represents the evaporation pressure on the intake side. evp p represents the evaporation temperature on the intake side. con This indicates the condensation pressure on the exhaust side.

[0114] Step 220: Establish a condenser mechanism model, which is a discrete model of an air-cooled condenser; divide the condenser heat exchange process into multiple micro-elements, and calculate the heat exchange of each micro-element using the heat transfer coefficient and refrigerant temperature; calculate the refrigerant enthalpy at the condenser outlet using the condenser heat exchange.

[0115] Divide the heat exchange process into N segments, with each segment having a heat exchange capacity of Q. con for:

[0116]

[0117] Among them, K con,i t represents the average heat transfer coefficient of each infinitesimal element. con,i t represents the average temperature of the refrigerant in each micro-element. sr,i d represents the average temperature of each micro-element airflow; l represents the diameter of the heat exchange tube; and h represents the length of the heat exchange tube. The enthalpy of the refrigerant at the condenser outlet, h3, can be calculated from the heat exchange volume, where h2 is the enthalpy at the compressor outlet, calculated using a compressor model.

[0118]

[0119] Step 230: Establish a throttling valve mechanism model: The enthalpy values ​​of the refrigerant at the inlet and outlet before and after the throttling valve are equal.

[0120] The throttle valve mechanism is modeled as an isenthalpic model, where the enthalpy at the inlet and outlet of the throttle valve is equal.

[0121] h3 = h4;

[0122] In the above formula, h4 represents the enthalpy of the refrigerant after the throttling valve.

[0123] Step 240: Establish an evaporator mechanism model, which is a discrete model of a shell-and-tube evaporator; divide the evaporator into multiple micro-elements, calculate the heat transfer of each micro-element using the corresponding heat transfer coefficient and temperature; and calculate the enthalpy of the refrigerant at the evaporator outlet using the heat transfer of the evaporator.

[0124] By dividing the heat exchange process into N segments, each segment can exchange a heat of Q. evp for:

[0125]

[0126] Among them, K evp,i t represents the average heat transfer coefficient of each infinitesimal element. evp,i t represents the average temperature of the refrigerant in each micro-element. sr,i The average temperature of the cold water in each micro-element is given by d, where d represents the diameter of the heat exchange tube and l represents the length of the heat exchange tube.

[0127] The enthalpy h of the refrigerant at the evaporator outlet can be calculated from the heat exchange in the evaporator. 1' :

[0128]

[0129] In the above formula, This indicates the mass flow rate of refrigerant delivered by the compressor.

[0130] Step 250: Obtain the correct refrigerant mass flow rate at the compressor inlet by iteratively calculating the refrigerant enthalpy at the evaporator outlet and the refrigerant enthalpy at the compressor inlet; combine the compressor model to calculate the actual power of the compressor and compare it with the theoretical power; optimize the undetermined coefficients using a genetic algorithm.

[0131] By calculating the enthalpy of the refrigerant at the evaporator outlet (h) 1' Iterate the correct mass flow rate using the compressor inlet refrigerant enthalpy h1. Calculate compressor power using a compressor model:

[0132]

[0133] P loss,1 =a tl,1 ·P i ;

[0134] P loss,2 =a tl,2 ·μ·V sw ·ω 2 ;

[0135] P loss,3 =a tl,3 ·m loss ·Δh loss ;

[0136] P com =P i +P loss,1 +P loss,2 +P loss,3 ;

[0137] Among them, P i P represents the internal compression power of the compressor. loss,1 P represents the torque loss during compressor operation. loss,2 P represents the viscous loss during the compressor's operation. loss,3 P represents the additional power loss during partial load operation; com This indicates the total input power of the compressor; w i This represents the internal compression work, calculated from the multi-compaction compression process; a tl,1 Indicates the torque friction loss coefficient; a tl,2 The coefficient of viscous friction loss is represented by μ, where μ represents the viscosity of the lubricating oil and ω represents the compressor speed; a tl,3 This represents the partial load loss factor, m. loss Δh represents the refrigerant mass flow rate bypassed during partial load operation. loss This represents the theoretical enthalpy loss of the refrigerant that is bypassed during partial load operation.

[0138] The difference between the actual and predicted power is calculated, and the coefficients to be measured are determined using a genetic algorithm and field data. Specifically, firstly, the possible value ranges of each coefficient are limited based on engineering experience. Then, an initial parameter set is randomly generated within the feasible region using a genetic algorithm. Based on the initial parameter set, the compressor power is predicted. The difference between the predicted and actual values ​​is evaluated: if the accuracy requirements are met or the number of iterations exceeds the upper limit, the currently obtained coefficient value is output; if the accuracy requirements are not met, the genetic algorithm is used to adjust the parameter values, and the above process is repeated until the coefficients to be measured are finally determined.

[0139] Including the leakage area A leak Inlet and outlet throttling area, inlet and outlet heat transfer coefficient, loss coefficient a tl,1 a tl,2 a tl,3 And so on, where M is the data sample size:

[0140]

[0141] Where err represents the difference between the compressor power and the actual power, P com,i P represents the calculated power of the compressor.real,i This indicates the actual power.

[0142] Step 300: Combining the refrigeration system model and particle swarm optimization algorithm, calculate the optimal operating control parameters under the given virtual cooling capacity and weather parameters to minimize the energy consumption of the refrigeration system; and send the optimal operating control parameters to the equipment controller.

[0143] Particle swarm optimization (PSO) is a swarm intelligence-based optimization technique that simulates the foraging behavior of birds, finding the optimal solution to a problem through information sharing and cooperation among individuals within the group. PSO can be used to adjust compressor speed and load to minimize system power consumption at a given condensing temperature, specifically including:

[0144] (1) Define the optimization problem:

[0145] The objective is to minimize the system's power consumption P; the variables are compressor speed n and load; the constraint is the evaporation pressure p. evp Compressor inlet temperature t0, refrigerant inlet temperature t sp And the average cooling load over a period of time:

[0146] [P,n,load,t con ]=f(p evp ,t0,t sp Q evp ).

[0147] (2) Initialization of Particle Swarm Optimization Algorithm:

[0148] Particle swarm: Initializes a group of particles, each particle representing a set of compressor speeds and load positions;

[0149] Velocity: Each particle is assigned a random velocity, which determines the direction and step size of the particle's movement in the search space;

[0150] Position: The position of a particle corresponds to a point in the search space, i.e., a specific set of compressor speed and load values;

[0151] Individual optimal position: Record the best position found by each particle so far, that is, the compressor speed and load position corresponding to the minimum power consumption;

[0152] Global best position: Records the best position found so far in the entire particle swarm.

[0153] (3) Iterative process:

[0154] Evaluation: Calculate the objective function value, i.e. power consumption, for each particle, and update its individual best position and global best position.

[0155] Update speed: The velocity of each particle is updated based on its current velocity, individual best position, and global best position. This involves three parameters: inertia weight, cognitive learning factor, and social learning factor.

[0156] Update position: Update the position of each particle according to the new velocity.

[0157] Check constraints: Ensure the new location meets all given constraints, such as evaporation pressure, compressor inlet temperature, refrigerant inlet temperature, etc. If not, make necessary adjustments or reinitialize the particle.

[0158] Iteration: Repeat the above steps until a predetermined number of iterations is reached or a certain stopping criterion is met, such as the improvement in power consumption being less than a certain threshold.

[0159] (4) Output of results:

[0160] Output the globally optimal position, i.e., the compressor speed and load value corresponding to minimizing power consumption; output the system power consumption at this time.

[0161] Step 400: Execute step 100 again, and use the newly generated training samples to calculate the model error in the refrigeration system mechanism model. When the refrigeration system model error is greater than the preset threshold, repeat steps 200 to 400 above.

[0162] Each optimized data is collected and compared with the program's calculation results. When the error is too large, the test coefficients are retrained to ensure the accuracy of the program.

[0163] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for optimizing and controlling a refrigeration system based on virtual cooling capacity, characterized in that, Includes the following steps: Step 100: Collect on-site historical data of the refrigeration system and preprocess the collected on-site historical data, wherein the on-site historical data includes weather parameters and operating parameters; Step 200: Establish component mechanism models based on the composition of the refrigeration system. The component mechanism models include evaporator mechanism model, compressor mechanism model, condenser mechanism model, and throttling valve model. Based on the law of conservation of mass, connect all component mechanism models in series to form a refrigeration system model. Through iterative calculation, obtain the correct refrigerant mass flow rate at the compressor inlet and calculate the compressor power. Compare the calculated compressor power with the theoretical power and optimize the undetermined coefficients. Step 300: Combining the refrigeration system model and particle swarm optimization algorithm, calculate the optimal operating control parameters under the given virtual cooling capacity and weather parameters to minimize the energy consumption of the refrigeration system; Step 400: Execute step 100 again, and use the newly generated training samples to calculate the model error in the refrigeration system mechanism model. When the refrigeration system model error is greater than the preset threshold, repeat steps 200 to 400 above.

2. The refrigeration system optimization control method based on virtual cooling capacity according to claim 1, characterized in that, The compressor mechanism model includes the following calculation processes: intake throttling calculation process, leakage adiabatic mixing calculation process, intake constant pressure heating calculation process, multi-directional compression process, exhaust constant pressure cooling calculation process, and exhaust throttling calculation process; through iterative calculation, the undetermined coefficients and mass flow rate inside the compressor are solved.

3. The method for optimizing and controlling a refrigeration system based on virtual cooling capacity according to claim 2, characterized in that, The condenser mechanism model includes: dividing the condenser heat exchange process into multiple micro-elements, and calculating the heat exchange of each micro-element using the heat transfer coefficient and refrigerant temperature; and calculating the refrigerant enthalpy at the condenser outlet using the heat exchange of the condenser.

4. The method for optimizing and controlling a refrigeration system based on virtual cooling capacity according to claim 3, characterized in that, The throttling valve model has the same enthalpy values ​​for the refrigerant before and after the throttling valve. The enthalpy value of the refrigerant at the inlet before the throttling valve is the enthalpy value of the refrigerant at the outlet of the condenser, and the enthalpy value of the refrigerant at the outlet after the throttling valve is the enthalpy value of the refrigerant at the inlet of the evaporator.

5. The refrigeration system optimization control method based on virtual cooling capacity according to claim 4, characterized in that, The evaporator mechanism model is as follows: the evaporator is divided into multiple micro-elements, and the heat exchange of each micro-element is calculated using the corresponding heat transfer coefficient and temperature; the enthalpy of the refrigerant at the evaporator outlet is calculated using the heat exchange of the evaporator.

6. The method for optimizing and controlling a refrigeration system based on virtual cooling capacity according to claim 2, characterized in that, The intake throttling calculation process is simplified to an isentropic nozzle process: In the above formula, h1 represents the specific enthalpy of the refrigerant after suction throttling; h0 represents the inlet enthalpy of the compressor. The compressor's gas delivery mass flow rate is represented by d1; d1 represents the refrigerant density after suction throttling; A su Indicates the inhalation throttling area; The leakage adiabatic mixing calculation process is a mixing process of leakage mass flow rate and suction mass, and the specific enthalpy of the mixed refrigerant is calculated as follows: In the above formula, Indicates the compressor leakage mass flow rate; h represents the actual refrigerant mass flow rate entering the rotor compression chamber. leak,in h1 represents the specific enthalpy of the refrigerant at the compressor leakage channel inlet; h2 represents the specific enthalpy of the refrigerant after adiabatic mixing. The calculation process for intake constant pressure heating is as follows: the intake gas is heated by the casing, the refrigerant temperature rises, and the enthalpy increases. Based on assumptions, the calculation is finally obtained through iteration using the law of conservation of energy. In the above formula, Q 23 This indicates that the refrigerant is preheated by the casing, resulting in heat exchange; AU suc The heat transfer coefficient of the intake gas is represented by t. w t2 represents the casing temperature; t2 represents the suction-side refrigerant temperature; c p2 This indicates the specific heat capacity of the refrigerant on the suction side at constant pressure; m tot h3 represents the actual refrigerant mass flow rate entering the rotor compression chamber; h3 represents the specific enthalpy of the refrigerant after suction preheating. The multi-phase compression process is divided into a multi-phase compression process and a forced exhaust process: p4=p3·BVR α ; In the above formula, BVR represents the internal volume ratio of the compressor; α represents the polynomial index; v4 represents the refrigerant specific volume after the internal compression process is completed; v3 represents the refrigerant specific volume before the internal compression process begins; p4 represents the internal compression final pressure; p3 represents the refrigerant pressure after suction throttling. The calculation process for exhaust constant pressure cooling is similar to that for exhaust throttling and the intake process; the leakage of the twin-screw compressor is assessed by the virtual flow path connecting the intake and exhaust: In the above formula, Indicates the compressor leakage mass flow rate; d leak Indicates the refrigerant density at the inlet of the leakage path; A leak h represents the area of ​​the compressor leakage channel. leak,in This indicates the specific enthalpy of the refrigerant at the inlet of the compressor leakage channel; h leak,out This indicates the specific enthalpy of the refrigerant at the outlet of the compressor leakage channel; Actual gas delivery mass flow rate: This model calculates the input power, mass flow rate, and refrigerant outlet temperature of the compressor based on the compressor's suction and discharge side operating conditions. Where t7 represents the refrigerant outlet temperature; f(n,load,p) evp ,p con ) represents the input parameter set of the compressor model; n represents the compressor speed; load represents the load position; p evp t represents the evaporation pressure on the intake side. evp p represents the evaporation temperature on the intake side. con This indicates the condensation pressure on the exhaust side.

7. The method for optimizing and controlling a refrigeration system based on virtual cooling capacity according to claim 3, characterized in that, The condenser mechanism model divides the heat exchange process into N segments, with each segment having a heat exchange capacity of Q. con for: Among them, K con,i t represents the average heat transfer coefficient of each infinitesimal element. con,i t represents the average temperature of the refrigerant in each micro-element. sr,i This indicates the average temperature of the refrigerant in each micro-element. The enthalpy of the refrigerant at the condenser outlet, h3, is calculated using the heat exchange, where h2 represents the enthalpy of the refrigerant at the compressor outlet, calculated using a compressor model. In the above formula, This indicates the refrigerant mass flow rate of the compressor.

8. The method for optimizing and controlling a refrigeration system based on virtual cooling capacity according to claim 4, characterized in that, The evaporator mechanism model divides the heat exchange process into N segments, with each segment having a heat exchange capacity of Q. evp for: Among them, K evp,i t represents the average heat transfer coefficient of each infinitesimal element. evp,i t represents the average temperature of the refrigerant in each micro-element. sr,i d represents the average temperature of each micro-element; d represents the diameter of the heat exchange tube; l represents the length of the heat exchange tube. The enthalpy at the evaporator outlet (h) is calculated based on the heat exchange rate of the evaporator. 1' : In the above formula, This indicates the mass flow rate of refrigerant delivered by the compressor.

9. The refrigeration system optimization control method based on virtual cooling capacity according to claim 5, characterized in that, The correct refrigerant mass flow rate at the compressor inlet is determined by calculating the refrigerant enthalpy at the evaporator outlet and the refrigerant enthalpy at the compressor inlet. Calculate compressor power: P loss,1 =a tl,1 ·P i ; P loss,2 =a tl,2 ·m·V sw ·oh 2 ; P loss,3 =a tl,3 ·m loss ·Δh loss ; P com =P i +P loss,1 +P loss,2 +P loss,3 ; Among them, P i P represents the internal compression power of the compressor. loss,1 P represents the torque loss during compressor operation. loss,2 P represents the viscous loss during the compressor's operation. loss,3 P represents the additional power loss during partial load operation; com This indicates the total input power of the compressor; w i This represents the internal compression work, calculated from the multi-compaction compression process; a tl,1 Indicates the torque friction loss coefficient; a tl,2 The coefficient of viscous friction loss is represented by μ, where μ represents the viscosity of the lubricating oil and ω represents the compressor speed; a tl,3 This represents the partial load loss factor, m. loss Δh represents the refrigerant mass flow rate bypassed during partial load operation. loss This represents the theoretical enthalpy loss caused by the bypassed refrigerant during partial load operation; The difference between the compressor power and the actual power is calculated, and the coefficients to be measured, including the leakage area A, are determined using a genetic algorithm and field data. leak Inlet and outlet throttling area, inlet and outlet heat transfer coefficient, loss coefficient a tl,1 a tl,2 a tl,3 M is the data sample size: Where err represents the difference between the compressor power and the actual power, P com,i P represents the calculated power of the compressor. real,i This indicates the actual power.

10. The refrigeration system optimization control method based on virtual cooling capacity according to claim 1, characterized in that, In step 300, the particle swarm optimization algorithm is used to determine the evaporation pressure p. evp Compressor inlet temperature t0, refrigerant inlet temperature t sp By adjusting the compressor speed n and load, the value at a certain condensing temperature t can be obtained. con Under these conditions, the minimum power consumption P of the system is: [P,n,load,t con ]=f(p evp ,t0,t sp )。

Images