Cascade refrigeration system-based centralized cooling optimization control method and related device

Abstract

The invention discloses a centralized cooling optimization control method based on a cascade refrigeration system and a related device, and belongs to the technical field of refrigeration systems.The optimization control method comprises the steps that on-site historical data of a compressor, an interstage heat exchanger and an evaporative condenser are collected, and the collected data are preprocessed; training a pre-established compressor semi-empirical model, an inter-stage heat exchanger discrete model and an evaporative condenser distribution parameter model by using the preprocessed field historical data; respectively establishing neural network models for the compressor, the interstage heat exchanger and the evaporative condenser; carrying out cold load distribution on the compressors with different performances by adopting a plurality of numerical value methods, selecting a starting sequence of the compressors with different performances, and carrying out variable-load variable-frequency control on the compressors with different performances; and an optimization algorithm is adopted to enable the energy efficiency ratio of the whole concentrated cooling system based on the cascade refrigeration system to be the highest, and optimization control is completed. The method can accurately predict and control the highly integrated centralized refrigeration system, and realizes energy conservation and emission reduction of the refrigeration house.

Description

Technical Field

[0001] The present invention belongs to the technical field of refrigeration systems, and particularly relates to an optimized control method and related device for centralized cooling based on a cascade refrigeration system. Background Art

[0002] In recent years, more and more attention has been paid to the problem of improving the actual operating energy efficiency of refrigeration systems. With the continuous maturity of cold chain logistics and intelligent technologies, the cold storage industry also urgently needs to strengthen its work in energy conservation and upgrading. By introducing advanced energy-saving equipment and technologies, the energy utilization efficiency of cold storages can be improved, energy consumption and emissions can be reduced, thereby achieving the goal of sustainable development.

[0003] The excellent performance of the cascade refrigeration system at low evaporation temperatures makes it the first choice in the cold storage industry. However, in industrial refrigeration, the lack of some sensors makes it difficult to directly use data models to achieve prediction and control.

[0004] Moreover, in current industrial refrigeration technologies, refrigeration equipment is highly integrated. The parallel evaporative condenser is responsible for multiple integrated skids, one integrated skid is responsible for multiple cold ends with different evaporation pressures, and multiple compressors in parallel are required at the cold ends to ensure a certain cooling load. Therefore, a control method is needed to control the optimal intermediate heat exchange temperature, optimal condensation temperature, and the number of parallel evaporative condensers opened, as well as the load position, frequency, and number of parallel compressors opened through the evaporation pressure at each cold end, so as to optimize the efficiency of the entire centralized cooling system. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimized control method and related device for centralized cooling based on a cascade refrigeration system in view of the above problems in the prior art, which can accurately predict and control a highly integrated centralized refrigeration system and achieve energy conservation and emission reduction in cold storages.

[0006] In order to achieve the above prior art, the present invention has the following technical solutions:

[0007] In a first aspect, an optimized control method for centralized cooling based on a cascade refrigeration system is provided, including:

[0008] Collecting on-site historical data of compressors, intermediate heat exchangers, and evaporative condensers, and preprocessing the collected data;

[0009] Training a pre-established semi-empirical model of the compressor using the preprocessed on-site historical data of the compressor;

[0010] Training a pre-established discrete model of the intermediate heat exchanger and a distributed parameter model of the evaporative condenser using the on-site historical data of the intermediate heat exchanger and the evaporative condenser respectively;

[0011] Based on the data generated by the semi-empirical model of the compressor, the discrete model of the inter-stage heat exchanger, and the distributed parameter model of the evaporative condenser, neural network models are established for the compressor, the inter-stage heat exchanger, and the evaporative condenser respectively;

[0012] Based on the established neural network models, various numerical methods are used to allocate the cooling load for compressors with different performances, select the starting sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances;

[0013] An optimization algorithm is used to maximize the energy efficiency ratio of the centralized cooling system based on the cascade refrigeration system, and the optimal control is completed.

[0014] As a preferred solution, the on-site historical data of the compressor includes compressor power, load position, frequency, suction pressure, and discharge pressure;

[0015] The on-site historical data of the evaporative condenser includes the power of the evaporative condenser and the ambient temperature and humidity;

[0016] The methods for preprocessing the collected data include wavelet threshold denoising, data steady-state processing, outlier removal, and duplicate value removal.

[0017] As a preferred solution, the compressor semi-empirical model establishes the suction throttling process, the leakage adiabatic mixing process, the suction constant pressure heating process, the polytropic compression process, the discharge constant pressure cooling process, and the discharge throttling process;

[0018] The expression of the suction throttling process is as follows:

[0019]

[0020] In the formula, h 1 is the enthalpy of the refrigerant after throttling, h 0 is the enthalpy of the refrigerant before throttling, is the refrigerant flow rate inhaled into the compressor, A su is the suction throttling area, d 1 is the density of the refrigerant after throttling;

[0021] The leakage adiabatic mixing process is the mixing process of the leakage mass flow and the inhaled mass, is the mass after mixing, h 2 is the enthalpy of the refrigerant after mixing, and the expression is as follows:

[0022]

[0023] The suction constant pressure heating process is obtained by iteration through the law of conservation of energy:

[0024]

[0025] In the formula, Q 23 is the heat obtained after the refrigerant is heated by the housing, h 3 is the enthalpy of the refrigerant after being heated, c p2 is the specific heat capacity at constant pressure of the refrigerant before being heated, -AU is the suction heat transfer coefficient, t w is the housing temperature, t 2 is the temperature of the refrigerant before being heated;

[0026] suc

[0027] The expression for the polytropic compression process is as follows:

[0028]

[0029] p 4 = p 3 ·BVR α

[0030] In the formula, BVR is the internal volume ratio of the compressor, α is the polytropic exponent, v 3 and p 3 are the specific volume and pressure of the refrigerant before being compressed respectively, v 4 and p 4 are the specific volume and pressure of the refrigerant after being compressed respectively;

[0031] The exhaust constant-pressure cooling process and the exhaust throttling process are similar to the suction process. For the calculation of the leakage of a twin-screw compressor, it is divided into the leakage at the first tooth groove and the leakage at the rotor contact. The calculation expressions are as follows:

[0032]

[0033] The actual inlet mass flow rate is:

[0034]

[0035] In the formula, is the leakage mass, A leak is the leakage area of the compressor, h leak,in is the enthalpy value of the refrigerant before leakage, d leak is the density of the refrigerant before leakage, h leak,out is the enthalpy value of the refrigerant after leakage;

[0036] The mass flow rate through the compressor and the outlet temperature of the compressor are calculated by the inlet pressure and outlet pressure of the compressor:

[0037]

[0038] In the formula, t 7 is the outlet temperature of the compressor, n is the rotational speed of the compressor, load is the load position of the compressor, pevp is the evaporation pressure, p con is the condensation pressure.

[0039] As a preferred solution, the discrete model of the inter-stage heat exchanger is established by dividing the heat exchange process into N segments, and the heat exchange amount of each segment is calculated according to the following formula:

[0040]

[0041] In the formula, K i is the average heat transfer coefficient of each micro-element, d is the diameter of the heat exchanger tube, t ht,i is the average temperature of the refrigerant at the high temperature stage of each segment, t lt,i is the average temperature of the low temperature stage of each micro-element, and Δh is the enthalpy difference between the inlet and outlet of the heat exchanger; is the refrigerant flow rate sucked into the compressor;

[0042] By establishing the energy conservation and heat conduction equations for each discrete unit, the micro-element length l N is gradually calculated; starting from the inlet, it is compared with the tube length l of the heat exchanger, and by iterating the evaporation temperature of the high temperature stage until the total length of the micro-element length l N is equal to the tube length l of the heat exchanger, and the heat exchange temperature difference of the corresponding heat exchanger is obtained;

[0043] t ht,evp = f(t lt,con , t lt,dis , m lt,ref )

[0044] In the formula, t ht,evp is the evaporation temperature of the high temperature stage, t lt,con is the condensation temperature of the low temperature stage, t lt,dis is the outlet temperature of the low temperature stage compressor, and m lt,ref is the mass flow rate of the low temperature stage.

[0045] As a preferred solution, the distributed parameter model of the evaporative condenser is established according to the heat exchange process, including:

[0046] For the refrigerant in the tube to dissipate heat to the water film, the following expression is established:

[0047] dQ con = k·(t ref - t w )·dF w,wa

[0048] In the formula, Q con is the total heat exchange amount of the evaporative condenser, k is the total heat transfer coefficient, F w,wa is the wetted surface area of the wall, t ref is the refrigerant temperature, t wis the wall temperature; driven by the temperature difference, the water film exchanges heat with the flowing air outside the tube, and the following expression is established:

[0049] dQ a,w = α a,w ·(t w - t a )·dF a,w

[0050] In the formula, Q a,w is the heat exchanged between the water film and the flowing air outside the tube, α a,w is the heat transfer coefficient of air-water, F a,w is the air-water surface area, t a is the air temperature;

[0051] Water evaporates into the air and causes the air humidity to increase, and the air outlet side is saturated wet air. The following expression is established:

[0052] dm wv = β·(x”(t w ) - x)·dF a,w

[0053] Among them, m wv is the mass of water vapor, β is the mass transfer coefficient, x”(t w ) is the moisture content of saturated water vapor at the wall temperature, and x is the moisture content of the air;

[0054] There is the following energy conservation equation:

[0055] -dm wv = dm w = m da dx

[0056] m da dh a = -h wv dm w,v -dQ a,w

[0057] m da dh a +dQ con = m w dh w +h w dm w

[0058] -dQ con = m ref dh ref

[0059] Among them, m da is the dry air mass, m wis the mass of water, h a is the enthalpy of moist air, h wv is the enthalpy of water vapor, h w is the enthalpy of water, m ref is the refrigerant flow rate, h ref is the refrigerant enthalpy;

[0060] Differential equations are established as follows for the moisture content of air, air temperature, water film temperature, and refrigerant enthalpy along the tube length:

[0061]

[0062] In the formula, f is the ratio of the wall area to the outside area of the tube, B is the effective length of a single tube, c a is the specific heat capacity of moist air, c wv is the specific heat capacity of water vapor, r 0 is the latent heat of vaporization;

[0063] And the boundary conditions are:

[0064] t a (l = L) = t a,in

[0065] x(l = L) = x in

[0066] h ref (l = 0) = h ref,in = f coolprop (t ref = t ref,in ; p ref = p ref,con )

[0067] t w (l = 0) = t w (l = L)

[0068] By solving the differential equations, the expression of the distributed parameter model of the evaporative condenser is obtained:

[0069]

[0070] In the formula, Q is the high-temperature stage heat load, t ht,dis is the high-temperature stage refrigerant inlet temperature, which is also equal to the compressor outlet temperature, t con is the condensation temperature, t air,in air inlet temperature, is the air inlet humidity, q v,air is the air inlet flow rate, m ht,ref is the high-temperature stage mass flow rate.

[0071] As a preferred solution, in the step of establishing neural network models for the compressor, the inter-stage heat exchanger, and the evaporative condenser respectively based on the data generated by the semi-empirical model of the compressor, the discrete model of the inter-stage heat exchanger, and the distributed parameter model of the evaporative condenser, a fully connected neural network is adopted to build a two-layer BP neural network model with a hidden layer containing the poslin positive linear function and a linear output layer;

[0072] For the compressor model, the input and output of the neural network model are as follows:

[0073]

[0074] For the evaporative condenser, when other conditions are known, the required inlet flow rate of the air is obtained, and thus the fan power of the required evaporative condenser is obtained. The input and output of the neural network model are as follows:

[0075]

[0076] For the inter-stage heat exchanger, under the known low-temperature stage operating conditions, the optimal inter-stage heat transfer temperature difference δ t is obtained, and the expression is as follows:

[0077] δ t = f(t lt,con , t lt,dis , m lt,ref )

[0078] where δ t is the inter-stage heat exchanger temperature difference, t lt,con is the low-temperature stage condensation temperature, t lt,dis is the low-temperature stage compressor outlet temperature, and m lt,ref is the low-temperature stage refrigerant flow rate.

[0079] As a preferred solution, in the step of using various numerical methods to distribute the cooling load for compressors with different performances, select the startup sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances based on the established neural network model:

[0080] For two different compressors, calculate based on the Brent method, and adaptively select the bisection method or the parabolic interpolation method according to the current interval and function value during each iteration, and allocate the same cooling load to the same compressor;

[0081] For different compressors, solve based on the sequential quadratic programming SQP method. The initial value gives the same cooling load to each compressor, and gradually approaches the optimal solution by transforming the non-linear programming problem into a series of quadratic programming problems.

[0082] As a preferred solution, in the step of using an optimization algorithm to maximize the energy efficiency ratio of the centralized cooling system based on the cascade refrigeration system and complete the optimization control, based on the particle swarm optimization algorithm, the inter-stage heat exchange temperature and the high-stage condensation temperature of multiple skids, the corresponding load positions and speeds of each compressor, and the number of evaporative condensers turned on are optimized and controlled through the evaporation pressure, cooling load, and ambient temperature; the cooling load is an unobservable quantity and is calculated through the average cooling load over a period of history.

[0083] In a second aspect, a centralized cooling optimization control system based on a cascade refrigeration system is provided, including:

[0084] A field historical data acquisition and processing module, configured to acquire the field historical data of compressors, inter-stage heat exchangers, and evaporative condensers, and preprocess the acquired data;

[0085] A compressor model training module, configured to train a pre-established semi-empirical model of the compressor using the preprocessed field historical data of the compressor;

[0086] An inter-stage heat exchanger model and an evaporative condenser model training module, configured to train a pre-established discrete model of the inter-stage heat exchanger and a distributed parameter model of the evaporative condenser respectively using the field historical data of the inter-stage heat exchanger and the evaporative condenser;

[0087] A neural network model establishment module, configured to establish neural network models for the compressor, the inter-stage heat exchanger, and the evaporative condenser respectively based on the data generated by the semi-empirical model of the compressor, the discrete model of the inter-stage heat exchanger, and the distributed parameter model of the evaporative condenser;

[0088] A compressor cooling load distribution module, configured to distribute the cooling load for compressors with different performances using various numerical methods based on the established neural network models, select the starting sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances;

[0089] A centralized cooling optimization module, configured to use an optimization algorithm to maximize the energy efficiency ratio of the centralized cooling system based on the cascade refrigeration system and complete the optimization control.

[0090] In a third aspect, a computer-readable storage medium is provided, in which at least one instruction is stored, and the at least one instruction is executed by a processor in an electronic device to implement the centralized cooling optimization control method based on the cascade refrigeration system.

[0091] Compared with the prior art, the present invention has at least the following beneficial effects:

[0092] The present invention optimizes the control of a centralized cooling system based on a cascade refrigeration system. In the system, multiple compressors with different performances, multiple inter-stage heat exchangers, and multiple evaporative condensers are connected in parallel to provide cooling for cold ends with different evaporation temperatures. The present invention collects the on-site historical data of the compressors, inter-stage heat exchangers, and evaporative condensers, and after preprocessing the collected data, trains the pre-established semi-empirical model of the compressor, the discrete model of the inter-stage heat exchanger, and the distributed parameter model of the evaporative condenser. Based on the data generated by the semi-empirical model of the compressor, the discrete model of the inter-stage heat exchanger, and the distributed parameter model of the evaporative condenser, neural network models are established for the compressor, the inter-stage heat exchanger, and the evaporative condenser respectively. Then, a variety of numerical methods are used to distribute the cooling load among compressors with different performances, select the starting sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances. In addition, an optimization algorithm is used to maximize the energy efficiency ratio of the entire centralized cooling system based on the cascade refrigeration system, thus completing the optimization control. The present invention is based on a semi-empirical model and a data-driven model to improve the prediction accuracy of the refrigeration system. When the cooling load, ambient temperature and humidity, and evaporation pressure are given, the number of the best parallel compressors to be started, the load positions and motor frequencies of each compressor, the number of evaporative condensers to be started, and the optimal condensation temperature and inter-stage heat exchange temperature can be calculated, solving the problem that it is difficult to optimize the control of the centralized cooling system based on the cascade refrigeration system and achieving energy conservation and emission reduction in cold storages. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and those of ordinary skill in the art can obtain other relevant drawings without creative efforts based on these drawings.

[0094] Figure 1 Schematic diagram of the low-temperature stage structure of the centralized cooling system based on the cascade refrigeration system according to the embodiment of the present invention;

[0095] Figure 2 Schematic diagram of the high-temperature stage structure of the centralized cooling system based on the cascade refrigeration system according to the embodiment of the present invention;

[0096] Figure 3 Schematic diagram of the cooling load distribution for 2 identical compressors and 1 different compressor according to the embodiment of the present invention;

[0097] Figure 4 Schematic diagram of the cooling load distribution for 3 different compressors according to the embodiment of the present invention;

[0098] Figure 5 Comparison chart of the energy efficiency ratio between the optimized control of the method according to the embodiment of the present invention and the actual system. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0099] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, those of ordinary skill in the art can also obtain other embodiments without creative efforts.

[0100] The embodiment of the present invention proposes a centralized cooling optimization control method based on a cascade refrigeration system, which controls the optimal heat exchange temperature, the optimal condensation temperature of the skid-mounted intermediate heat exchanger, the number of parallel evaporative condensers opened, as well as the load position, frequency, and the number of parallel compressors opened through the evaporation pressure at each cold end, so as to optimize the efficiency of the entire centralized cooling system.

[0101] The centralized cooling optimization control method based on the cascade refrigeration system in the embodiment of the present invention mainly includes the following steps:

[0102] S1. Collect the on-site historical data of the compressor, the intermediate heat exchanger, and the evaporative condenser, and preprocess the collected data;

[0103] S2. Use the preprocessed on-site historical data of the compressor to train the pre-established semi-empirical model of the compressor;

[0104] S3. Use the on-site historical data of the intermediate heat exchanger and the evaporative condenser to train the pre-established discrete model of the intermediate heat exchanger and the distributed parameter model of the evaporative condenser respectively;

[0105] S4. Based on the data generated by the semi-empirical model of the compressor, the discrete model of the intermediate heat exchanger, and the distributed parameter model of the evaporative condenser, establish neural network models for the compressor, the intermediate heat exchanger, and the evaporative condenser respectively;

[0106] S5. Based on the established neural network models, use various numerical methods to distribute the cooling load among compressors with different performances, select the starting sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances;

[0107] S6. Use an optimization algorithm to maximize the energy efficiency ratio of the entire centralized cooling system based on the cascade refrigeration system, and complete the optimization control.

[0108] The centralized cooling system based on the cascade refrigeration system in the embodiment of the present invention has multiple compressors with different performances connected in parallel, multiple intermediate heat exchangers connected in parallel, and multiple evaporative condensers connected in parallel, so as to supply cooling to multiple cold ends with different evaporation temperatures.

[0109] In a possible implementation, the on-site historical data of the compressor includes compressor power, load position, frequency, suction pressure, and discharge pressure; the on-site historical data of the evaporative condenser includes evaporative condenser power and ambient temperature and humidity;

[0110] The method for preprocessing the collected data in step S1 includes: wavelet threshold denoising, data steady-state processing, outlier removal, and duplicate value removal. By preprocessing the collected data, the uncertainty of the real data and the repeatability of the data are minimized to the greatest extent, and some data under different working conditions are selected as the training set of the model.

[0111] In a possible implementation, the compressor semi-empirical model in step S2 establishes an intake throttling process, a leakage adiabatic mixing process, an intake constant-pressure heating process, a polytropic compression process, an exhaust constant-pressure cooling process, and an exhaust throttling process;

[0112] The expression of the intake throttling process is as follows:

[0113]

[0114] In the formula, h 1 is the enthalpy of the refrigerant after throttling, h 0 is the enthalpy of the refrigerant before throttling, is the refrigerant flow rate inhaled into the compressor, A su is the intake throttling area, d 1 is the density of the refrigerant after throttling;

[0115] The leakage adiabatic mixing process is the mixing process of the leakage mass flow rate and the inhaled mass, and the inhaled mass is the refrigerant flow rate inhaled into the compressor is the mass after mixing, h 2 is the enthalpy of the refrigerant after mixing, and the expression is as follows:

[0116]

[0117] The intake constant-pressure heating process is obtained by iteration through the law of conservation of energy:

[0118]

[0119] In the formula, Q 23 is the heat obtained by the refrigerant after being heated by the casing, h 3 is the enthalpy of the refrigerant after being heated, c p2 is the specific heat capacity at constant pressure of the refrigerant before being heated, -AU is the intake heat transfer coefficient, t w is the casing temperature, t 2 is the temperature of the refrigerant before being heated;

[0120] suc

[0121] The expression for the polytropic compression process is as follows:

[0122]

[0123] p 4 = p 3 ·BVR α

[0124] Wherein, BVR is the internal volume ratio of the compressor, α is the polytropic exponent, v 3 and p 3 are respectively the specific volume and pressure of the refrigerant before compression, v 4 and p 4 are respectively the specific volume and pressure of the refrigerant after compression;

[0125] The constant-pressure exhaust cooling process and the exhaust throttling process are similar to the suction process. For the calculation of the leakage of a twin-screw compressor, it is divided into the leakage of the first tooth groove and the leakage at the rotor contact. The calculation expressions are as follows:

[0126]

[0127] The actual inlet mass flow rate is:

[0128]

[0129] Wherein, is the leakage mass, A leak is the leakage area of the compressor, h leak,in is the enthalpy value of the refrigerant before leakage, d leak is the density of the refrigerant before leakage, h leak,out is the enthalpy value of the refrigerant after leakage;

[0130] The mass flow rate through the compressor and the outlet temperature of the compressor are calculated by the inlet pressure and outlet pressure of the compressor:

[0131]

[0132] Wherein, t 7 is the outlet temperature of the compressor, n is the rotational speed of the compressor, load is the load position of the compressor, p evp is the evaporation pressure, p con is the condensation pressure.

[0133] In a possible implementation manner, the discrete model of the inter-stage heat exchanger in step S3 is obtained by dividing the heat exchange process into N segments, and the heat transfer amount of each segment is calculated according to the following formula:

[0134]

[0135] Wherein, Ki is the average heat transfer coefficient of each microelement, d is the diameter of the heat exchanger tube, and t ht,i is the average temperature of the high-temperature stage refrigerant in each section, and t lt,i is the average temperature of the low-temperature stage in each microelement, and Δh is the enthalpy difference between the inlet and outlet of the heat exchanger;

[0136] By establishing the energy conservation and heat conduction equations for each discrete unit, the microelement length l is gradually calculated N ; starting from the inlet, it is compared with the heat exchanger tube length l. By iterating the high-temperature stage evaporation temperature until the total length of the microelement length l N is equal to the heat exchanger tube length l, the heat transfer temperature difference of the corresponding heat exchanger is obtained;

[0137] t ht,evp = f(t lt,con , t lt,dis , m lt,ref )

[0138] In the formula, t ht,evp is the high-temperature stage evaporation temperature, t lt,con is the low-temperature stage condensation temperature, t lt,dis is the low-temperature stage compressor outlet temperature, and m lt,ref is the low-temperature stage mass flow rate.

[0139] In a possible implementation manner, the distributed parameter model of the evaporative condenser described in step S3 is established according to the heat transfer process, including:

[0140] First, the refrigerant in the tube dissipates heat to the water film, and the following expression is established:

[0141] dQ con = k·(t ref - t w )·dF w,wa

[0142] In the formula, Q con is the total heat transfer of the evaporative condenser, k is the total heat transfer coefficient, F w,wa is the wetted surface area of the wall, t ref is the refrigerant temperature, and t w is the wall temperature; driven by the temperature difference, the water film exchanges heat with the flowing air outside the tube, and the following expression is established:

[0143] dQ a,w = α a,w ·(t w - t a )·dF a,w

[0144] In the formula, Q a,w is the heat exchanged between the water film and the flowing air outside the tube, and αa,w is the heat transfer coefficient of air - water, F a,w is the air - water surface area, t a is the air temperature;

[0145] Water evaporates into the air, causing the air humidity to increase, and the air outlet side is saturated humid air. The following expression is established:

[0146] dm wv = β·(x”(t w ) - x)·dF a,w

[0147] where m wv is the mass of water vapor, β is the mass transfer coefficient, x”(t w ) is the moisture content of saturated water vapor at the wall temperature, and x is the moisture content of the air;

[0148] The following energy conservation equation exists:

[0149] -dm wv = dm w = m da dx

[0150] m da dh a = -h wv dm w,v -dQ a,w

[0151] m da dh a +dQ con = m w dh w +h w dm w

[0152] -dQ con = m ref dh ref

[0153] where m da is the dry air mass, m w is the mass of water, h a is the enthalpy of humid air, h wv is the enthalpy of water vapor, h w is the enthalpy of water, m ref is the refrigerant flow rate, h ref is the enthalpy of the refrigerant.

[0154] Differential equations are established as follows for the moisture content of air, air temperature, water film temperature, and refrigerant enthalpy along the tube length:

[0155]

[0156] In the formula, f is the ratio of the wall area to the outside area of the tube, B is the effective length of a single tube, and c a is the specific heat capacity of humid air, and c wv is the specific heat capacity of water vapor, and r 0 is the latent heat of vaporization;

[0157] and the boundary conditions:

[0158] t a (l = L) = t a,in

[0159] x(l = L) = x in

[0160] h ref (l = 0) = h ref,in = f coolprop (t ref = t ref,in ; p ref = p ref,con )

[0161] t w (l = 0) = t w (l = L)

[0162] By solving the differential equation, the expression of the distributed parameter model of the evaporative condenser is obtained:

[0163]

[0164] In the formula, Q is the high-temperature stage heat load, and t ht,dis is the high-temperature stage refrigerant inlet temperature, which is also equal to the compressor outlet temperature, and t con is the condensation temperature, and t air,in is the air inlet temperature, is the air inlet humidity, q v,air is the air inlet flow rate, m ht,ref is the high-temperature stage mass flow rate;

[0165] In a possible implementation manner, step S4 adopts a fully connected neural network to build a two-layer BP neural network model with a hidden layer containing the poslin positive linear function and a linear output layer;

[0166] For the compressor model, the input and output of the neural network model are:

[0167]

[0168] For an evaporative condenser, when other conditions are known, the required inlet flow rate of the air is obtained, and thus the fan power of the required evaporative condenser can be obtained. The input and output of the neural network model are as follows:

[0169]

[0170] For an inter-stage heat exchanger, when the operating conditions of the low-temperature stage are known, the optimal inter-stage heat transfer temperature difference δ t is obtained, and the expression is as follows:

[0171] δ t = f(t lt,con , t lt,dis , m lt,ref )

[0172] where δ t is the temperature difference of the inter-stage heat exchanger, t lt,con is the condensation temperature of the low-temperature stage, t lt,dis is the outlet temperature of the low-temperature stage compressor, and m lt,ref is the refrigerant flow rate of the low-temperature stage.

[0173] In a possible implementation manner, in step S5, multiple numerical methods are used to allocate the cooling load for compressors with different performances. For two different compressors, the calculation is based on the Brent method. At each iteration, the bisection method or the parabolic interpolation method is adaptively selected according to the current interval and function value, and the same cooling load is allocated to the same compressor;

[0174] For different compressors, the solution is based on the sequential quadratic programming SQP method. The initial value gives the same cooling load to each compressor, and the optimal solution is gradually approximated by transforming the nonlinear programming problem into a series of quadratic programming problems.

[0175] In a possible implementation manner, in step S6, based on the particle swarm optimization algorithm, the inter-stage heat transfer temperature of multiple skids, the high-temperature stage condensation temperature, the load positions and speeds of the corresponding compressors, and the number of evaporative condensers turned on are optimized and controlled through the evaporation pressure, cooling load, and ambient temperature; the cooling load is an unobservable quantity and is calculated based on the average cooling load over a certain period of history.

[0176] Please refer to Figure 1 and Figure 2 . The centralized cooling optimization control method based on the cascade refrigeration system in the embodiment of the present invention is described by taking a food freezing and processing factory as an example.

[0177] As Figure 1As shown in the figure, in the low-temperature stage structure of the system, the evaporation temperature of the quick-freezing unit is -40°C. The end of the horizontal evaporation unit is the brine line, and the evaporation temperature is -34°C to -32°C. Heat exchange occurs in the 1,2# intermediate heat exchanger. The evaporation temperatures of the freezing unit and the refrigeration unit are -33°C, and heat exchange occurs in the #3 intermediate heat exchanger. Among them, the 2,3# compressors are parallel compressors, the 4-6# compressors are parallel compressors, and the 7,8# compressors are parallel compressors.

[0178] As Figure 2 shown in the figure, in the high-temperature stage structure of the system, the end of the ice water and air-conditioning plate heat exchanger is water, and the evaporation temperature is 0°C. The parallel compressors 12,14# absorb heat in the #3 intermediate heat exchanger, and the 9-11,13# parallel compressors absorb heat in the 1,2# intermediate heat exchanger.

[0179] Among them, the ammonia compressor is optimized for variable load, and the carbon dioxide compressor is optimized for variable frequency and variable load.

[0180] According to the centralized cooling optimization control method based on the cascade refrigeration system in the embodiment of the present invention, first, collect the data of each compressor, the inlet and outlet pressures of the compressor, the power of the compressor, and the rotational speed and load position of the compressor, and establish a model for each compressor:

[0181] The suction throttling process can be simplified to an isentropic nozzle process, where h 1 is the refrigerant enthalpy after throttling, h 0 is the refrigerant enthalpy before throttling, is the refrigerant flow rate inhaled by the compressor, A su is the suction throttling area, d 1 is the refrigerant density expression after throttling as follows:

[0182]

[0183] The leakage mass flow and the mixing process of the inhaled mass are the leakage adiabatic mixing calculation processes. Among them, the inhaled mass is the refrigerant flow rate inhaled by the compressor is the mass after mixing, h 2 is the refrigerant enthalpy after mixing. The initial value of the leakage mass flow can be assumed for iterative calculation:

[0184]

[0185] The inhaled gas will be heated by the casing of the machine, the refrigerant temperature will rise, and the enthalpy value will increase. Among them, Q 23 is the heat obtained by the refrigerant after being heated by the casing, h 3 is the enthalpy of the refrigerant after being heated, c p2 is the specific heat capacity at constant pressure of the refrigerant before being heated, -AU suc is the suction heat transfer coefficient, tw is the temperature of the housing, t 2 is the temperature of the refrigerant before heating, and finally obtained by iteration through the law of conservation of energy:

[0186]

[0187] The compression process of a twin-screw compressor can be divided into a polytropic compression process and a forced exhaust process, where BVR is the internal volume ratio of the compressor, α is the polytropic exponent, v 3 and p 3 are the specific volume and pressure of the refrigerant before compression, respectively, and v 4 and p 4 are the specific volume and pressure of the refrigerant after compression, respectively:

[0188]

[0189] p 4 = p 3 ·BVR α

[0190] The calculation process of constant-pressure cooling during exhaust and the calculation process of exhaust throttling are similar to the suction process. The calculation of leakage in a twin-screw compressor is divided into the leakage of the first tooth groove and the leakage at the rotor contact. Among them is the leakage mass, A leak is the leakage area of the compressor, h leak,in is the enthalpy value of the refrigerant before leakage, d leak is the density of the refrigerant before leakage, h leak,out is the enthalpy value of the refrigerant after leakage:

[0191]

[0192] Actual inlet mass flow rate:

[0193]

[0194] The model calculates the mass flow rate through the carbon dioxide compressor and the outlet temperature of the compressor based on the inlet pressure and outlet pressure of the compressor:

[0195]

[0196] The ammonia compressor is a constant-frequency compressor, and its motor speed is fixed at 2960 r / min. This model can calculate the mass flow rate through the ammonia compressor and the outlet temperature of the compressor based on the inlet pressure and outlet pressure of the compressor:

[0197]

[0198] Provide a high-quality data volume for the data model database through a semi-empirical model. Establish a data model for the compressor using these newly generated data and the original data, and use the data model to improve the optimization speed of the system:

[0199]

[0200] By collecting the heat exchange temperature and heat exchange temperature difference of the inter-stage heat exchanger, calculating the required cooling load on the carbon dioxide side and ammonia side through the compressor model, and calculating the heat leakage coefficient of the inter-stage heat exchanger, an inter-stage heat exchanger model can be established:

[0201] The discrete model of the inter-stage heat exchanger divides the heat exchange process into N segments, and the heat exchange amount of each segment is:

[0202]

[0203] Among them, K i is the average heat transfer coefficient of each micro-element segment, t ht,i is the average refrigerant temperature of each micro-element segment on the ammonia side, t lt,i is the average temperature of each micro-element segment on the carbon dioxide side.

[0204] By establishing the energy conservation and heat conduction equations for each discrete unit, gradually calculate the micro-element length l N . Starting from the inlet, compare it with the heat exchanger tube length l, and through iterative high-temperature stage evaporation temperature until the total length of the micro-element length l N is equal to the heat exchanger tube length l, the heat exchange temperature difference of the heat exchanger can be obtained. Among them, is the low-temperature stage condensation temperature, is the low-temperature stage compressor outlet temperature:

[0205]

[0206] Provide a high-quality data volume for the data model database through a semi-empirical model. Establish a data model for the inter-stage heat exchanger using these newly generated data and the original data, and use the data model to improve the optimization speed of the system:

[0207]

[0208] For the distributed parameter model of the evaporative condenser, first, the refrigerant in the tube dissipates heat to the water film. Among them, k is the total heat transfer coefficient, F w,wa is the wetted surface area of the wall, t ref is the refrigerant temperature, t w is the wall temperature:

[0209] Q con =k·(t ref -t w )·dFw,wa

[0210] Driven by the temperature difference, the water film exchanges heat with the flowing air outside the tube. Among them, α a,w is the heat transfer coefficient of air-water, F a,w is the air-water surface area, and t a is the air temperature.

[0211] dQ a,w = α a,w ·(t w - t a )·dF a,w

[0212] Finally, the water evaporates into the air and causes the air humidity to increase, and the air outlet side is assumed to be saturated humid air. Among them, m w,v is the mass of water vapor, and β is the mass transfer coefficient:

[0213] dm w,v = β(x”(t w ) - x)·dF a,w

[0214] According to the law of conservation of energy, among them, m da is the mass of dry air, m w is the mass of water, h a is the enthalpy value of humid air, h wv is the enthalpy value of water vapor, h w is the enthalpy value of water, m ref is the refrigerant flow rate, and h ref is the enthalpy value of the refrigerant:

[0215] - dm wv = dm w = m da dx

[0216] m da dh a = - h wv dm w,v - dQ a,w

[0217] m da dh a + dQ con = m w dh w + h w dm w

[0218] - dQ con = m ref dh ref

[0219] Combined with the basic equations of heat and mass transfer, the following differential equations can be established for the enthalpy of air humidity, air temperature, water film temperature, and refrigerant along the tube length. Here, f is the ratio of the wall area to the outer tube area, B is the effective length of a single tube, m da is the dry air mass, c a is the specific heat capacity of air, c wv is the specific heat capacity of water vapor, r 0 is the latent heat of vaporization:

[0220]

[0221] And through the boundary conditions:

[0222] t a (l = L) = t a,in

[0223] x(l = L) = x in

[0224] h ref (l = 0) = h ref,in = f coolprop (t ref = t ref,in ; p ref = p ref,con )

[0225] t w (l = 0) = t w (l = L)

[0226] By solving the differential equations, the equations of the evaporative condenser can be obtained. Here, t ht,dis is the inlet temperature of the high-temperature stage refrigerant, i.e., the outlet temperature of the compressor, t con is the condensation temperature, t air,in is the air inlet temperature, q v,air is the air inlet flow rate, m ht,ref is the high-temperature stage mass flow rate:

[0227]

[0228] Provide high-quality data volume for the data model database through the distributed parameter model. Establish a data model for the evaporative condenser using these newly generated data and the original data, and use the data model to improve the optimization speed of the system:

[0229]

[0230] Based on the data model, conduct energy-saving optimization on the system. The optimization objective is the total energy consumption of the system. First, the evaporation pressure of each unit is the set value. The required cooling capacity of each unit is calculated based on the average cooling capacity over a past period of time. The required cooling capacity of the cold storage can be used to obtain the required while is the value to be optimized by the particle swarm optimization algorithm. At this time, the rotational speed, load position, and outlet temperature of the carbon dioxide compressor can be obtained.

[0231]

[0232] Through the Brent method of the compressor parallel optimization algorithm, the parallel optimization schemes of the 2, 3# compressors and the 7, 8# compressors can be obtained, that is, the mass flow rate allocated to each compressor minimizes the power consumption of the compressor unit. As Figure 3 shown.

[0233] Through the SQP algorithm of the compressor parallel optimization algorithm, the parallel optimization scheme of the 4 - 6# compressors can be obtained, that is, the mass flow rate allocated to each compressor minimizes the power consumption of the compressor unit. As Figure 4 shown.

[0234] Through the discrete model of the inter - stage heat exchanger, the required heat transfer temperature difference δ can be obtained under the conditions of the carbon dioxide side condensation temperature, carbon dioxide outlet temperature, and carbon dioxide mass flow rate t , and the evaporation temperature on the ammonia side can be obtained The evaporation pressure on the ammonia side is calculated by calculating the saturation pressure

[0235]

[0236] The required cooling load on the ammonia side, that is, the required mass flow rate on the ammonia side, can be calculated through the heat leakage coefficient of the heat exchanger while is the value to be optimized by the particle swarm optimization algorithm. At this time, the load position and outlet temperature of the ammonia compressor can be obtained:

[0237]

[0238] Similar to the carbon dioxide compressor, through the compressor parallel optimization algorithm, the mass flow rate distribution scheme that minimizes the power consumption of the compressor unit can be obtained. Among them, 9 - 11# are the same compressors. In the parallel optimization of the same compressors, the power consumption is always the lowest when the load positions are equal. Therefore, the Brent method can also be used to solve it.

[0239] Through the outlet temperature of the ammonia compressor, the heat load Q on the ammonia side and the ambient air temperature t air,in , and the air humidity are the observed values. At this time, through the data model of the evaporative condenser, the required air flow rate of the evaporative condenser can be obtained, and the fan power of the evaporative condenser can be obtained.

[0240]

[0241] Through the particle swarm optimization algorithm, the value to be optimized is the carbon dioxide condensation temperature of the 1st and 2nd inter-stage heat exchangers The carbon dioxide condensation temperature of the 3rd inter-stage heat exchanger The ammonia-side condensation temperature t con , and the target value is the sum P of the compressor power and the evaporative condenser power tot , and then the optimal carbon dioxide compressor load positions and speeds, the load position of the optimal ammonia compressor, as well as the speed and the number of fans turned on of the evaporative condenser under this working condition of the system can be obtained.

[0242] Figure 5 This is a comparison chart of the optimized control of the method of the embodiment of the present invention and the actual system energy efficiency ratio. It can be seen from the chart that the present invention has obvious energy-saving effects, with an average energy-saving energy consumption reduction of 14.6% and a maximum energy consumption reduction of 30.4%

[0243] Another embodiment of the present invention also proposes a centralized cooling optimization control system based on a cascade refrigeration system, including:

[0244] A field historical data acquisition and processing module, which is used to collect the field historical data of compressors, inter-stage heat exchangers and evaporative condensers, and preprocess the collected data;

[0245] A compressor model training module, which is used to train a pre-established compressor semi-empirical model using the preprocessed field historical data of the compressor;

[0246] An inter-stage heat exchanger model and an evaporative condenser model training module, which are used to train a pre-established discrete model of the inter-stage heat exchanger and a distributed parameter model of the evaporative condenser respectively using the field historical data of the inter-stage heat exchanger and the evaporative condenser;

[0247] A neural network model establishment module, which is used to establish neural network models for compressors, inter-stage heat exchangers and evaporative condensers respectively based on the data generated by the compressor semi-empirical model, the discrete model of the inter-stage heat exchanger and the distributed parameter model of the evaporative condenser;

[0248] A compressor cooling load distribution module, which is used to distribute the cooling load of compressors with different performances by using various numerical methods based on the established neural network model, select the startup sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances;

[0249] A centralized cooling optimization module, which is used to use an optimization algorithm to make the energy efficiency ratio of the entire centralized cooling system based on the cascade refrigeration system the highest and complete the optimization control.

[0250] Another embodiment of the present invention further provides a computer-readable storage medium, in which at least one instruction is stored, and the at least one instruction is executed by a processor in an electronic device to implement the centralized cooling optimization control method based on the cascade refrigeration system.

[0251] Exemplarily, the instructions stored in the memory can be divided into one or more modules / units, and the one or more modules / units are stored in the computer-readable storage medium and executed by the processor to complete the centralized cooling optimization control method based on the cascade refrigeration system of the present invention. The one or more modules / units can be a series of computer-readable instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the server.

[0252] The electronic device can be a computing device such as a smart phone, a notebook, a palm computer, and a cloud server. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the electronic device may further include more or fewer components, or combine certain components, or different components. For example, the electronic device may further include input / output devices, network access devices, a bus, etc.

[0253] The processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may also be any conventional processor, etc.

[0254] The memory may be an internal storage unit of the server, such as the hard disk or memory of the server. The memory may also be an external storage device of the server, such as a plug-in hard disk equipped on the server, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the memory may also include both the internal storage unit and the external storage device of the server. The memory is used to store the computer-readable instructions and other programs and data required by the server. The memory may also be used to temporarily store the data that has been output or will be output.

[0255] It should be noted that for the content such as information interaction and execution process between the above-mentioned module units, since it is based on the same concept as the method embodiment, for its specific functions and the technical effects brought, reference can be specifically made to the method embodiment part, and details will not be repeated here.

[0256] Those skilled in the art can clearly understand that for the convenience and conciseness of description, only the above division of each functional unit and module is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit. In addition, the specific names of each functional unit and module are only for the convenience of mutual distinction and do not limit the protection scope of this application. The specific working process of the units and modules in the above system can refer to the corresponding process in the foregoing method embodiment, and details will not be repeated here.

[0257] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above method embodiments of this application, a computer program can be used to instruct the relevant hardware to complete. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can at least include: any entity or device that can carry the computer program code to the photographing device / terminal device, recording medium, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium. For example, a USB flash drive, a mobile hard disk, a magnetic disk or an optical disc, etc.

[0258] In the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed or recorded in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0259] The above-described embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present application, and should all be included within the protection scope of the present application.

Claims

1. A centralized cooling optimization control method based on a cascade refrigeration system, characterized in that: include: Collect the field historical data of compressors, interstage heat exchangers and evaporative condensers, and pre-process the collected data; Using pre-processed compressor field historical data to train a pre-established compressor semi-empirical model; The pre-established discrete model of the interstage heat exchanger and the distributed parameter model of the evaporative condenser are trained using the field historical data of the interstage heat exchanger and the evaporative condenser respectively; Based on the data generated by the semi-empirical model of the compressor, the discrete model of the interstage heat exchanger and the distributed parameter model of the evaporative condenser, neural network models are established for the compressor, interstage heat exchanger and evaporative condenser respectively; Based on the established neural network model, a variety of numerical methods are used to distribute the cooling load to compressors with different performances, select the startup sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances; The optimization algorithm is used to maximize the energy efficiency ratio of the entire centralized cooling system based on the cascade refrigeration system and complete the optimization control.

2. The centralized cooling optimization control method based on the cascade refrigeration system according to claim 1 is characterized in that: The on-site historical data of the compressor includes compressor power, load, frequency, suction pressure and exhaust pressure; The on-site historical data of the evaporative condenser includes the evaporative condenser power and the ambient temperature and humidity; The methods for preprocessing the collected data include wavelet threshold denoising, data steady-state processing, outlier removal, and duplicate value removal.

3. The centralized cooling optimization control method based on the cascade refrigeration system according to claim 1 is characterized in that: The semi-empirical model of the compressor establishes an intake throttling process, a leakage adiabatic mixing process, an intake constant pressure heating process, a multi-party compression process, an exhaust constant pressure cooling process and an exhaust throttling process; The expression of suction throttling process is as follows: In the formula, h1 is the refrigerant enthalpy after throttling, h0 is the refrigerant enthalpy before throttling, is the refrigerant flow rate sucked into the compressor, A su is the suction throttling area, d1 is the refrigerant density after throttling; The leakage adiabatic mixing process is the leakage mass flow rate Mixing process with the suction mass, is the mass after mixing, h2 is the enthalpy of the refrigerant after mixing, and the expression is as follows: The constant pressure heating process of air intake is obtained by iterating the law of conservation of energy: In the formula, Q 23 is the heat obtained after the refrigerant is heated by the casing, h3 is the enthalpy of the refrigerant after heating, c p2 is the constant pressure specific heat capacity of the refrigerant before being heated, -AU is the suction heat transfer coefficient, t w is the casing temperature, t2 is the temperature of the refrigerant before being heated; suc The expression of the multi-party compression process is as follows: Wherein, BVR is the volume ratio of the compressor, α is the polynomial index, v3 and p3 are the specific volume and pressure of the refrigerant before compression, v4 and p4 are the specific volume and pressure of the refrigerant after compression; The exhaust constant pressure cooling process and the exhaust throttling process are similar to the suction process. The calculation of the twin-screw compressor leakage is divided into the leakage of the first tooth groove and the leakage calculation at the rotor contact point. The calculation expression is as follows: The actual inlet mass flow rate is: In the formula, is the leakage quality, A leak is the compressor leakage area, h leak,in is the refrigerant enthalpy before leakage, d leak is the density of refrigerant before leakage, h leak,out is the enthalpy of refrigerant after leakage; The mass flow rate through the compressor and the compressor outlet temperature are calculated by the compressor inlet pressure and outlet pressure: Where t7 is the compressor outlet temperature, n is the compressor speed, load is the compressor load, p evp is the evaporation pressure, p con is the condensation pressure.

4. The centralized cooling optimization control method based on the cascade refrigeration system according to claim 1 is characterized in that: The interstage heat exchanger discrete model divides the heat exchange process into N sections, and the heat exchange amount of each section is calculated according to the following formula: In the formula, K i is the average heat transfer coefficient of each microelement, d is the diameter of the heat exchanger tube, t ht,i is the average temperature of the refrigerant in each high-temperature stage, t lt,i is the average temperature of the low temperature stage of each microelement, Δh is the enthalpy difference between the inlet and outlet of the heat exchanger; is the refrigerant flow rate sucked into the compressor; By establishing the energy conservation and heat conduction equations for each discrete unit, the microelement length l is calculated step by step. N ; Calculate from the inlet, compare with the heat exchanger tube length l, and iterate the high-temperature stage evaporation temperature until the microelement length l N The total length of is equal to the heat exchanger tube length l, and the heat exchange temperature difference of the corresponding heat exchanger is obtained; t ht,evp =f(t lt,con ,t lt,dis ,m lt,ref ) Where, t ht,evp is the high temperature stage evaporation temperature, t lt,con is the low temperature stage condensation temperature, t lt,dis is the outlet temperature of the low temperature compressor, m lt,ref is the low temperature mass flow rate.

5. The centralized cooling optimization control method based on the cascade refrigeration system according to claim 1 is characterized in that: The evaporative condenser distributed parameter model is established according to the heat exchange process, including: For the refrigerant in the tube to dissipate heat to the water film, the expression is as follows: dQ con =k·(t ref -t w )·dF w,wa In the formula, Q con is the total heat transfer of the evaporative condenser, k is the total heat transfer coefficient, F w,wa is the wetted surface area of ​​the wall, t ref is the refrigerant temperature, t w is the wall temperature; driven by the temperature difference, the water film exchanges heat with the flowing air outside the tube, and the expression is established as follows: dQ a,w =α a,w ·(t w -t a )·dF a,w In the formula, Q a,w is the heat exchanged between the water film and the flowing air outside the tube, α a,w is the air-water heat transfer coefficient, F a,w is the air-water surface area, t a is the air temperature; Water evaporates into the air and causes the air humidity to increase, and the air outlet side is saturated humid air. The expression is established as follows: dm wv =β·(x”(t w )-x)·dF a,w Among them, m wv is the mass of water vapor, β is the mass transfer coefficient, x”(t w ) is the moisture content of saturated water vapor at the wall temperature, and x is the moisture content of air; There is the following energy conservation equation: -dm wv =dm w =m da dx m da dh a =-h wv dm w,v -dQ a,w m da dh a +dQ con =m w dh w +h w dm w -dQ con =m ref dh ref Among them, m da is the dry air mass, m w is the mass of water, h a is the enthalpy of moist air, h wv is the enthalpy of water vapor, h w is the enthalpy of water, m ref is the refrigerant flow rate, h ref is the refrigerant enthalpy; The following differential equations are established for air humidity, air temperature, water film temperature and refrigerant enthalpy along the tube length: In the formula, f is the ratio of the wall area to the outer area of ​​the tube, B is the effective length of a single tube, and c is a is the specific heat capacity of moist air, c wv is the specific heat capacity of water vapor, r0 is the latent heat of vaporization; And the boundary conditions: t a (l=L)=t a,in x(l=L)=x in h ref (l=0)=h ref,in =f coolprop (t ref =t ref,in ;p ref =p ref,con ) t w (l=0)=t w (l=L) By solving the differential equation, the distributed parameter model expression of the evaporative condenser is obtained: Where Q is the high temperature heat load, t ht,dis is the inlet temperature of the high-temperature refrigerant, which is also equal to the compressor outlet temperature, t con is the condensation temperature, t air,in Air inlet temperature, is the air inlet humidity, q v,air is the air inlet flow rate, m ht,ref is the high temperature grade mass flow rate.

6. The centralized cooling optimization control method based on the cascade refrigeration system according to claim 1 is characterized in that: In the step of establishing neural network models for the compressor, interstage heat exchanger and evaporative condenser respectively based on the data generated by the semi-empirical model of the compressor, the discrete model of the interstage heat exchanger and the distributed parameter model of the evaporative condenser, a fully connected neural network is used to build a double-layer BP neural network model containing a hidden layer of a poslin positive linear function and a linear output layer; For the compressor model, the input and output of the neural network model are: For the evaporative condenser, the required wind inlet flow rate is obtained under other known conditions, thereby obtaining the required fan power of the evaporative condenser. The input and output of the neural network model are: For the interstage heat exchanger, under the known low temperature stage working conditions, the optimal interstage heat exchange temperature difference δ is obtained t , the expression is as follows: δ t =f(t lt,con ,t lt,dis ,m lt,ref ) Among them, δ t is the temperature difference between the heat exchangers, t lt,con is the low temperature stage condensation temperature, t lt,dis is the outlet temperature of the low temperature compressor, m lt,ref is the low temperature refrigerant flow rate.

7. The centralized cooling optimization control method based on the cascade refrigeration system according to claim 1 is characterized in that: In the steps of distributing cooling loads to compressors with different performances based on the established neural network model, using a variety of numerical methods, selecting the startup sequence of compressors with different performances, and performing variable load and variable frequency control on compressors with different performances: For two different compressors, the calculation is based on the Brent method. In each iteration, the binary method or the parabolic interpolation method is adaptively selected according to the current interval and function value to allocate the same cooling load to the same compressor. For different compressors, the solution is based on the sequential quadratic programming (SQP) method. The initial value is given to each compressor with the same cooling load. The optimal solution is gradually approached by converting the nonlinear programming problem into a series of quadratic programming problems.

8. The centralized cooling optimization control method based on the cascade refrigeration system according to claim 1 is characterized in that: In the step of using the optimization algorithm to maximize the energy efficiency ratio of the entire centralized cooling system based on the cascade refrigeration system and completing the optimization control, based on the particle swarm optimization algorithm, the inter-stage heat exchange temperature and the high-temperature stage condensing temperature of multiple skids, the corresponding load and speed of each compressor, and the number of evaporative condensers opened are optimized and controlled through evaporation pressure, cooling load and ambient temperature; the cooling load is an unobservable quantity and is calculated by the average cooling load over a period of history.

9. A centralized cooling optimization control system based on a cascade refrigeration system, characterized in that: include: On-site historical data collection and processing module, used to collect on-site historical data of compressors, interstage heat exchangers and evaporative condensers, and pre-process the collected data; A compressor model training module, used for training a pre-established compressor semi-empirical model using pre-processed compressor field historical data; Interstage heat exchanger model and evaporative condenser model training modules, used to use the field historical data of the interstage heat exchanger and the evaporative condenser to respectively train the pre-established interstage heat exchanger discrete model and the evaporative condenser distributed parameter model; A neural network model building module is used to build neural network models for the compressor, interstage heat exchanger and evaporative condenser respectively based on data generated by the compressor semi-empirical model, the interstage heat exchanger discrete model and the evaporative condenser distributed parameter model; The compressor cooling load distribution module is used to distribute cooling loads to compressors with different performances based on the established neural network model using a variety of numerical methods, select the startup sequence of compressors with different performances, and perform variable load and variable frequency control on compressors with different performances; The centralized cooling optimization module is used to use optimization algorithms to maximize the energy efficiency ratio of the entire centralized cooling system based on the cascade refrigeration system and complete optimization control.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores at least one instruction, and the at least one instruction is executed by a processor in an electronic device to implement the centralized cooling optimization control method based on a cascade refrigeration system as described in any one of claims 1 to 8.

Images