Energy demand distribution method for air conditioning system and air conditioning system
By establishing a model of the air-conditioning system and using particle swarm optimization algorithm, the optimal intermediate temperature value is determined and the energy consumption of multi-stage refrigeration air-conditioning system is optimized, and the energy consumption optimization of multi-stage refrigeration system is solved, and the energy efficiency of the air-conditioning system is optimized is achieved. and improve operational efficiency.
Patent Information
- Application Number
- CN202411075355.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-06
- Publication Date
- 2024-12-03
AI Technical Summary
In a multi-stage refrigeration and air conditioning system, how to optimize the sum of energy consumption of each stage of refrigeration system to achieve global energy consumption optimal.
By establishing an enthalpy pressure graph model, refrigeration capacity demand model, compression mechanism cooling load model and energy consumption model of the air conditioning system, the particle swarm optimization algorithm is used to calculate the optimal total energy consumption and determine the optimal intermediate temperature value, and then for at least two levels Refrigeration energy needs to be distributed.
The overall energy efficiency of the air conditioning system is achieved, the total energy consumption is reduced, and the system's operating efficiency and cost-effectiveness are improved.
Smart Images

Figure CN119063163A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to an energy demand distribution method for an air-conditioning system and an air-conditioning system, and in particular to a method for distributing energy demands at various levels in an air-conditioning system and an air-conditioning system using the method.
[0002] Background
[0003] For independent air conditioning systems, such as refrigeration systems, the relationship between energy efficiency and compressor frequency is that it first increases and then decreases, and different air conditioning systems use different compressors, so the energy efficiency and frequency curves are also different. If single-stage refrigeration is used, the required compressor capacity is relatively large and the cost is high. Therefore, in some cases, the air conditioning system has multi-stage refrigeration, such as two-stage refrigeration, where two compressors can share the total cooling capacity. When combining multi-stage refrigeration together, it is necessary to consider the optimal energy efficiency of the combination. Summary of the invention
[0004] It is necessary to explore a method to minimize the sum of energy consumption of each stage of refrigeration system in an air-conditioning system with multi-stage refrigeration, so as to achieve optimal global energy consumption.
[0005] Therefore, according to a first aspect of the present application, there is provided a method for allocating energy demand for an air conditioning system, wherein the air conditioning system has at least two stages of refrigeration, and is characterized by comprising the following steps:
[0006] S01, prepare data;
[0007] S02, selecting an enthalpy-pressure diagram model and performing training on the enthalpy-pressure diagram model;
[0008] S03, establishing a cooling capacity demand relationship model between the air side cooling capacity and the refrigerant side cooling capacity of the air conditioning system;
[0009] S04, establishing a target optimization model and selecting an optimization algorithm;
[0010] S05, calculating the optimal total energy consumption and determining the optimal intermediate temperature value; and
[0011] S06: Allocate energy requirements for the at least two stages of refrigeration according to the optimal intermediate temperature value.
[0012] According to the first aspect of the present application, it is characterized in that step S03 comprises:
[0013] S03-1, establish a cooling capacity demand model;
[0014] S03-2, establish a compressor refrigeration load model and a compressor energy consumption model;
[0015] S03-3, training and fitting a refrigeration capacity demand model, a compressor refrigeration load model, and a compressor energy consumption model; and
[0016] S03-4, perform heat balance calculation on the indoor air side and the refrigerant side.
[0017] According to the first aspect of the present application, it is characterized in that: the refrigeration capacity demand model is related to the intermediate temperature value; the compressor refrigeration load model is related to the compressor speed, evaporating temperature and condensing temperature; and the compressor energy consumption model is related to the compressor speed, evaporating temperature and condensing temperature.
[0018] According to a first aspect of the present application, it is characterized in that the evaporation temperature and the condensation temperature are related to the refrigerant type, cooling method, design of the condenser or evaporator, ambient temperature and ambient humidity.
[0019] According to the first aspect of the present application, it is characterized in that: in step S03-3, the least squares method is used for model training and fitting.
[0020] According to a first aspect of the present application, it is characterized in that: in step S01, the data includes refrigeration demand parameters and refrigeration source cycle parameters.
[0021] According to the first aspect of the present application, it is characterized in that: in step S02, the enthalpy-pressure diagram model includes a pressure-enthalpy line model, an isothermal line model and an isentropic line model; and the enthalpy-pressure diagram model is trained using the least squares method.
[0022] According to the first aspect of the present application, it is characterized in that: in step S04, the target optimization model is set as a fitness function; and the optimization algorithm is a particle swarm optimization algorithm.
[0023] According to the first aspect of the present application, it is characterized in that: in step S05, multiple total energy consumptions are calculated based on the multiple optimization solutions obtained by the particle swarm optimization algorithm in step S04, wherein the minimum value of the multiple total energy consumptions corresponds to the optimal total energy consumption, and the intermediate temperature value corresponding to the optimal total energy consumption is the optimal intermediate temperature value.
[0024] According to the first aspect of the present application, it is characterized in that: the total energy consumption includes compressor energy consumption, and the compressor energy consumption is related to the compressor speed, evaporation temperature and condensation temperature.
[0025] According to the first aspect of the present application, it is characterized in that: the total energy consumption includes the power consumption of the outdoor condensing fan, and the power consumption of the outdoor condensing fan can be determined according to the enthalpy-pressure diagram model, compressor speed, evaporation temperature, condensation temperature and fan performance curve parameters.
[0026] According to the second aspect of the present application, an air-conditioning system is provided, characterized in that it includes at least two stages of refrigeration and a controller, wherein each stage of the at least two stages of refrigeration includes: a compressor, a condenser, a throttle valve and an evaporator, and the controller is configured to allocate energy demands for the at least two stages of refrigeration according to the energy demand allocation method for an air-conditioning system as described in the first aspect of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0027] These and other features and advantages of the present application may be better understood by reading the following detailed description with reference to the accompanying drawings, in which like reference numerals refer to like parts throughout, wherein:
[0028] Figure 1 is a schematic diagram of an air conditioning system with two-stage refrigeration;
[0029] Figure 2 is a general logic block diagram of the energy demand allocation method for the air conditioning system;
[0030] Figure 3A A specific flow chart for enthalpy-pressure diagram model training;
[0031] Figure 3B This is a schematic diagram of a common physical property parameter table;
[0032] Figure 3C for Figure 3B A simplified schematic diagram of the physical property parameter table shown;
[0033] Figure 4 A specific flow chart for heat balance calculation on the air side and refrigerant side of the air conditioning system;
[0034] Figure 5 A specific flow chart for finding the optimal solution with minimum energy consumption;
[0035] Figure 6 A block diagram showing the specific details of the controller. DETAILED DESCRIPTION
[0036] The purpose of the present invention is to provide an energy demand allocation method and an air conditioning system for an air conditioning system, which can establish a refrigeration system energy consumption model and an algorithm to find the optimal point of system energy efficiency, so that the entire system can always operate in the current optimal state. The energy demand allocation method of the present application can be used for a multi-stage heating or cooling air conditioning system. The following takes a two-stage cooling air conditioning system as an example to illustrate the specific implementation of the present application.
[0037] Figure 1 It is a schematic diagram of an air conditioning system with two refrigeration units.
[0038] like Figure 1As shown, the air conditioning system 100 has two refrigeration sections, namely, a primary refrigeration section 102 and a secondary refrigeration section 104. The primary refrigeration section 102 has a primary air inlet side 132, a primary air outlet side 134 and a primary refrigeration side 136; the secondary refrigeration section 104 has a secondary air inlet side 142, a secondary air outlet side 144 and a secondary refrigeration side 146. Air enters the primary refrigeration section 102 from the primary air inlet side 132. In the primary refrigeration section 102, the primary refrigeration side 136 cools the air, thereby changing the temperature and humidity of the air. The air that has undergone primary cooling is discharged from the primary refrigeration section 102 from the primary air outlet side 134, and then enters the secondary refrigeration section 104 from the secondary air inlet side 142. In the secondary refrigeration section 104, the secondary refrigeration side 146 continues to cool the air, thereby further changing the temperature and humidity of the air. The air that has undergone secondary cooling is discharged from the secondary refrigeration section 104 from the secondary air outlet side 144. For example, the original temperature of the air is 35°C and the original humidity is 50%. The air needs to be cooled to 10°C and 95% humidity. An intermediate cooling temperature needs to be set, such as 20°C and 70% humidity, so that the primary cooling section 102 and the secondary cooling section 104 operate according to their respective cooling targets to achieve the optimal overall energy efficiency of the air-conditioning system 100.
[0039] Continue as Figure 1 As shown, the primary refrigeration side 136 includes a primary evaporator 111, a primary compressor 112, a primary condenser 113 and a primary throttle valve 114. The primary evaporator 111, the primary compressor 112, the primary condenser 113 and the primary throttle valve 114 are fluidly connected in sequence to form a first refrigerant circuit. Similarly, the secondary refrigeration side 146 includes a secondary evaporator 121, a secondary compressor 122, a secondary condenser 123 and a secondary throttle valve 124. The secondary evaporator 121, the secondary compressor 122, the secondary condenser 123 and the secondary throttle valve 124 are fluidly connected in sequence to form a second refrigerant circuit. A primary condensing fan 131 is arranged near the primary condenser 113, and a secondary condensing fan 132 is arranged near the secondary condenser 123 to promote heat exchange between the primary condenser 113 and the secondary condenser 123 and the external environment.
[0040] Continue as Figure 1 As shown, the air-conditioning system 100 also has a controller 160, which is communicatively connected to the first-stage evaporator 111, the first-stage compressor 112, the first-stage condenser 113, the first-stage throttle valve 114, the second-stage evaporator 121, the second-stage compressor 122, the second-stage condenser 123 and the second-stage throttle valve 124 to collect relevant parameter data; the controller 160 can also collect parameters from various sensors on the first refrigerant circuit and the second refrigerant circuit; through the collected data, after modeling calculations, the appropriate operating frequency of the two-stage compressor and the appropriate operating frequency of the two-stage condensing fan are obtained, and then the operating frequency of the two-stage compressor and the two-stage condensing fan are adjusted.
[0041] In the first refrigerant circuit, the refrigerant is compressed into a high-pressure, high-temperature state in the first-stage compressor 112 and then discharged into the first-stage condenser 113. Then, the refrigerant exchanges heat with cooling water in the first-stage condenser 113 to release heat and is condensed into a high-pressure, liquid state, and then discharged into the first-stage throttle valve 114. In the first-stage throttle valve 114, the refrigerant is expanded and throttled into a medium-pressure two-phase state, and then flows into the first-stage evaporator 111 to exchange heat with the air in the first-stage refrigeration section 102, thereby achieving the effect of first-stage refrigeration. The refrigerant cycle in the second refrigerant circuit is the same as that in the first refrigerant circuit, and will not be repeated here.
[0042] Figure 2 The figure is a general logic block diagram of the energy demand allocation method for air conditioning system.
[0043] like Figure 2 As shown, in step 202, data is prepared first. A large amount of data comes from data provided by compressor and fan manufacturers and simulation data. These data are explored, cleaned, extracted, reduced, and organized into the form required for model and mathematical calculation, and the density of data is expanded to make the training model more accurate. The data that needs to be obtained or prepared include refrigeration demand data and refrigeration source cycle parameters.
[0044] Among them, the refrigeration demand data include: atmospheric pressure, fresh air dry bulb temperature, fresh air wet bulb temperature, saturated water vapor pressure at wet bulb temperature, saturated humidity, fresh air humidity, fresh air enthalpy, fresh air density, intermediate dry bulb temperature, intermediate wet bulb temperature, saturated water vapor pressure at intermediate wet bulb temperature, intermediate saturated humidity, intermediate humidity, intermediate enthalpy, intermediate density, outlet air dry bulb temperature, outlet air wet bulb temperature, saturated water vapor pressure at outlet air wet bulb temperature, outlet air saturated humidity, outlet air humidity, outlet air enthalpy, outlet air density, total cooling capacity, primary cooling capacity and secondary cooling capacity; the refrigeration source cycle parameters include: speed, cooling capacity coefficient, power coefficient, evaporation temperature, cold energy temperature, cooling capacity, power and flow rate of different compressors. The air volume (Q), air pressure (H), power (N) and efficiency (η) of different axial flow fans and the refrigerant pressure-enthalpy diagram data. The above refrigeration demand data or refrigeration source cycle parameters can be acquired through corresponding sensors (not shown in the figure) or corresponding components in the air-conditioning system 100, and then the collected data is transmitted to the controller 160.
[0045] Among them, the dry bulb temperature refers to the value read on the dry bulb thermometer exposed to the air but not directly exposed to the sun. The temperature mentioned in the weather forecast refers to the dry bulb temperature of the day. The wet bulb temperature takes into account the effect of water evaporation on the temperature. It is the temperature measured by wrapping the mercury bulb of the thermometer with wet gauze. Since water evaporation absorbs heat, the wet bulb temperature is usually lower than the dry bulb temperature. When the water vapor in the air reaches saturation, the dry bulb temperature is equal to the wet bulb temperature. Therefore, the wet bulb temperature can reflect the humidity of the air to a certain extent.
[0046] After completing the operation of step 202 , the operation proceeds to step 204 .
[0047] In step 204, an enthalpy-pressure diagram model is selected and the enthalpy-pressure diagram model is trained, wherein the enthalpy-pressure diagram model includes a pressure-enthalpy line model, an isotherm line model, and an isentropic line model. For specific model training steps, please refer to Figures 3A-3C Description.
[0048] After completing the operation of step 204 , the operation proceeds to step 206 .
[0049] In step 206, a model of the relationship between the cooling capacity of the air side and the cooling capacity demand of the refrigerant side of the air conditioning system is established to perform heat balance calculation. For specific steps, please refer to Figure 4 Description.
[0050] After completing the operation of step 206 , the operation proceeds to step 208 .
[0051] In step 208, a target optimization model is established and an optimization algorithm is selected. For specific steps, see Figure 5 Description.
[0052] After completing the operation of step 208 , the operation proceeds to step 210 .
[0053] In step 210, the optimal total energy consumption (P d ) and determine the optimal intermediate temperature value (t 中间最优 ), where the total refrigeration power consumption is the sum of the primary refrigeration section compressor power consumption, the secondary refrigeration section compressor power consumption, the indoor fan power consumption, the outdoor fan power consumption, the expansion valve power consumption and the heat exchange power consumption.
[0054] After completing the operation of step 210 , the operation proceeds to step 212 .
[0055] In step 212, the optimal intermediate temperature value (t 中间最优 ) Energy demand distribution is performed on at least two-stage refrigeration units. In the embodiment of the present application, two-stage refrigeration is adopted, so energy demand distribution needs to be performed on the two-stage refrigeration units.
[0056] Figure 3AThe specific flow chart for enthalpy-pressure diagram model training is as follows: Figure 3B This is a schematic diagram of a common physical property parameter table. Figure 3C for Figure 3B The simplified schematic diagram of the physical property parameter table shown is
[0057] The enthalpy-pressure data of the corresponding types of refrigerants are known and selected. We can draw corresponding curves based on the physical parameter data, including pressure-enthalpy curves, isotherms, and isentropic curves. These curves can be fitted with data through function models to establish pressure-enthalpy curve models, isotherm models, and isentropic curve models. According to the characteristics and requirements of the data, select a suitable polynomial function to fit the data, or customize the fitting function according to specific actual needs.
[0058] like Figure 3A As shown, taking the pressure-enthalpy curve model fitting training process as an example, in step 302, pressure-enthalpy training data is obtained, and after completing the operation of step 302, the operation proceeds to step 304.
[0059] In step 304, by observing the distribution of the data, the appropriate fitting times are selected, and the polynomial degree is selected. The higher the polynomial degree, the more accurate the fitting result, but the greater the amount of calculation. The polynomial degree can be selected according to the specific situation, and usually an n-order polynomial can be selected as follows:
[0060]
[0061] After completing the operation of step 304 , the operation proceeds to step 306 .
[0062] In step 306, the enthalpy-pressure diagram model is trained using the least squares method, the difference between the actual value and the predicted value of each observation point is squared, and all square values are added to obtain the residual square sum. The sum of the squares of the output values of each function and the true result is fitted and established as follows:
[0063]
[0064] After completing the operation of step 306 , the operation proceeds to step 308 .
[0065] In step 308, the polynomial fitting process is to obtain the coefficient a n , which minimizes the respective loss values. By iteratively optimizing the residual sum of squares, the model parameters closest to the actual observed data can be found to minimize the sum of squared loss deviations.
[0066] The loss deviation function takes partial derivative of the coefficients:
[0067]
[0068] After completing the operation of step 308 , the operation proceeds to step 310 .
[0069] In step 310, after partial derivatives are taken for the coefficients, the polynomial derivative is simplified to a matrix:
[0070]
[0071] It can be obtained that A=(P T *P) -1 *P T *H, the corresponding items in matrix A are the coefficients of the fitted pressure-enthalpy curve.
[0072] After completing the operation of step 310 , the operation proceeds to step 312 .
[0073] In step 312, a pressure-enthalpy curve model is obtained according to the coefficients of the fitted pressure-enthalpy curve calculated in step 310. The model is evaluated by calculating the error term of the model. If the prediction effect of the model is not good, it is necessary to collect and optimize data again and perform an iterative optimization process.
[0074] Use the same method to find out the coefficients of the isotherm model and the isentropic model.
[0075] like Figure 3B-3C As shown in the figure, a common physical property parameter table is shown, which maps the liquid saturation line and the gas saturation line to the pressure coordinate axis, with the horizontal axis representing the enthalpy value and the vertical axis representing the pressure. In the physical property parameter table, polynomial fitting is used, which is divided into fitting into a pressure liquid saturation model and a pressure gas saturation model. On this physical property parameter table, the compressor refrigeration cycle process can be depicted, which are the compression process ( Figure 3C State 1-State 2), condensation process ( Figure 3C State 2-State 3), expansion process ( Figure 3C State 3-State 4), evaporation process ( Figure 3C The evaporation process corresponds to a constant evaporation temperature (T 蒸发 ), the condensation process corresponds to a constant condensation temperature (T 冷凝 ).
[0076] After training the pressure-enthalpy curve model, isotherm model, and isentropic curve model, the initial evaporation temperature t 蒸发 and condensation temperature t 冷凝 The corresponding two isotherms are t 蒸发 and t 冷凝 The liquid saturation and vapor saturation pressures corresponding to the two isotherms are P(T 蒸发 ), P(T 冷凝 ), the enthalpy values corresponding to the two saturation pressures are H(P(T 蒸发)), H(P(T 冷凝 )), the entropy slope corresponding to the condensation temperature is S(T 蒸发 ), the enthalpy value corresponding to the evaporation temperature at state 1 is h1. The corresponding isentropic line in the pressure-enthalpy diagram is P(T)-P(T 蒸发 )=S(T 蒸发 )(h-h1).
[0077] Specifically, since the pressure corresponding to the pressure-enthalpy curve and the enthalpy value is sometimes a one-to-many mapping, and the enthalpy value corresponding to the pressure is sometimes also a one-to-many mapping, a piecewise polynomial function is used to fit the piecewise curve, with the saturated liquid pressure-enthalpy curve as one curve and the saturated steam pressure-enthalpy curve as another curve, where the pressure is the independent variable and the enthalpy value is the dependent variable. The piecewise polynomial of the pressure-enthalpy curve model is established as follows:
[0078]
[0079] Isotherms and isentropic lines have a one-to-one mapping relationship within their value range, and are also fitted using polynomial functions.
[0080] There is a linear relationship between isotherm temperature and pressure, which can be fitted into a temperature-pressure model P(T 蒸发 ), P(T 冷凝 ), the isotherm model polynomial is established as follows:
[0081]
[0082] The isentropic line can be fitted into a linear model between pressure and enthalpy (corresponding to the compression process from state 1 to state 2). The slope model of the isentropic line curve selects the polynomial as follows:
[0083]
[0084] Then the isentropic line model polynomial is established as follows:
[0085] P(T)-P(T 蒸发 )=S(T 蒸发 )(h-h1),
[0086] Where h1 is the enthalpy value corresponding to the evaporation temperature at state 1, that is:
[0087] h1=H(P(T 蒸发 )).
[0088] The isentropic line intersects the condensation line at state 2, which means that the condensation temperature (T 冷凝 ) is brought into the isentropic line model, that is:
[0089] P(T 冷凝 )-P(T 蒸发 )=S(T蒸发 )(h-h1),
[0090] The functional relationship of the isentropic line can be obtained:
[0091]
[0092] Here h is the enthalpy value h2 at state 2, that is:
[0093]
[0094] Then we can get the enthalpy value h3 at state 3, that is:
[0095] h3=H(P(T 冷凝 )),
[0096] The enthalpy value at state 4 is equal to the enthalpy value at state 3, that is:
[0097] h4=H(P(T 冷凝 )).
[0098] Thus, the evaporation temperature t 蒸发 and condensation temperature t 冷凝 The corresponding enthalpy values at state 1-state 4 are:
[0099] h1=H(P(T 蒸发 )),
[0100]
[0101] h3=H(P(T 冷凝 )),
[0102] h4=H(P(T 冷凝 )).
[0103] Figure 4 It is a specific flow chart for heat balance calculation on the air side and refrigerant side of the air conditioning system.
[0104] like Figure 4 As shown, in step 402, a refrigeration demand model is established.
[0105] First, the cooling capacity is provided by two independent refrigeration cycles. The primary evaporator 111 and the secondary evaporator 121 exchange heat with the wind side respectively, so the total cooling capacity is the sum of the primary cooling capacity and the secondary cooling capacity. Among them, the cooling capacity is based on a large number of different required air volumes, fresh air data, air supply data, and intermediate state data made by air side experiments, and these experimental data are cleaned and fitted.
[0106] Based on the large amount of data measured by the enthalpy difference method, the prepared data was used to train the cooling capacity demand model corresponding to different air temperature differences, and the following relationship was obtained:
[0107] Q i =F(Vol,P,t 干球 ,t 湿球 ),
[0108] Since this application uses a two-stage refrigeration cycle to refrigerate and control the fresh air temperature, the first-stage refrigeration capacity model and the second-stage refrigeration model are respectively:
[0109] QQ1=F(Vol,P,t 新风干球 ,t 新风湿球 ,t 中间干球 ,t 中间湿球 ),
[0110] Q2=G(Vol,P,t 中间干球 ,,t 中间湿球 ,t 出风干球 , t 出风湿球 ),
[0111] Among them, the variable Vol is the air volume, the variable P is the atmospheric pressure, and t 新风湿球 is the fresh air wet bulb temperature, t 新风干球 is the fresh air dry bulb temperature, variable t 中间干球 is the output dry bulb temperature of the first stage of two-stage refrigeration, variable t 中间湿球 is the output wet bulb temperature of the first stage of two-stage refrigeration, t 出风干球 is the output dry bulb temperature of the second stage of two-stage refrigeration, t 出风湿球 It is the output wet bulb temperature of the second stage of two-stage refrigeration. The fresh air dry bulb temperature and fresh air wet bulb temperature can be detected by the temperature sensor. The outlet dry bulb temperature and outlet wet bulb temperature are the target values set by the system and are known values for the entire dehumidifier.
[0112] After completing the operation of step 402 , the operation proceeds to step 404 .
[0113] In step 404, a compressor refrigeration load model and a compressor energy consumption model are established.
[0114] The compressors used in the refrigeration cycle of the dehumidifier are provided by various manufacturers, with numerous types and parameters. The performance parameters of different compressors are also provided. These large amounts of data include cooling capacity coefficient, power coefficient, flow coefficient, and speed. The model selected in this application uses a ternary polynomial, and uses the data of these compressors to train their respective cooling capacity models and power models.
[0115] The compressor refrigeration load model uses the compressor speed (N P ), evaporation temperature (T蒸发 ), condensation temperature (T 冷凝 ) as the input parameter, and the cooling capacity as the output parameter. Since the mapping of the model is a many-to-one relationship, the model selects the multivariate linear fitting as follows:
[0116] Q 压缩机i =Q Pi (N P ,T 蒸发 ,T 冷凝 ),
[0117] The compressor power model uses the compressor speed (N P ), evaporation temperature (T 蒸发 ), condensation temperature (T 冷凝 ) as the input parameter, cooling capacity as the output parameter, and power as the output parameter. Since the mapping of the model is a many-to-one relationship, the model selects multivariate linear fitting as follows:
[0118] P 压缩机i =P Pi (N P ,T 蒸发 ,T 冷凝 ).
[0119] After completing the operation of step 404 , the operation proceeds to step 406 .
[0120] In step 406, training and fitting of the refrigeration capacity demand model, the compressor refrigeration load model, and the compressor energy consumption model are performed.
[0121] Q above i =F(Vol,P,t 干球 ,t 湿球 ), Q 压缩机i =Q Pi (N P ,T 蒸发 ,T 冷凝 ) and P 压缩机i =P Pi (N P ,T 蒸发 ,T 冷凝 ) Each of the three models has its own data that can be trained into its own model to fit the relationship between multiple independent variables and dependent variables. This application uses multivariate linear regression and the least squares method to supervise the learning and fitting data. The training steps are as follows:
[0122] Step 1, initialization: select an initial multiple linear regression formula f(x1,x2,…,x n ), where f(x1,x2,…,x n ) in this application are specifically F(Vol,P,t 干球 ,t湿球 ,),Q Pi (N P ,T 蒸发 ,T 冷凝 ), P Pi (N P ,T 蒸发 ,T 冷凝 )Three multiple linear regression formulas.
[0123] Step 2: Calculate the residual: For each data point (x 1i ,x 2i ,…,x ni ,y i ), calculate the residual e i =y i -f((x 1i ,x 2i ,…,x ni ).
[0124] Step 3: Update the formula: According to the goal of minimizing the sum of squares of the residuals, update the formula f(x1,x2,…,x n ).
[0125] Step 4: Repeat steps 2 and 3 until the formula converges or the preset number of iterations is reached.
[0126] Step 5: After training, verify the accuracy and finally determine the model Q i =F(Vol,P,t 干球 ,t 湿球 ), Q 压缩机i =Q Pi (N P ,T 蒸发 ,T 冷凝 ) and P 压缩机i =P Pi (N P ,T 蒸发 ,T 冷凝 ).
[0127] After completing the operation of step 406 , the operation proceeds to step 408 .
[0128] In step 408, heat balance calculations are performed on the indoor air side and the refrigerant side.
[0129] The two refrigeration cycles are designed with their own evaporator heat exchanger plates. When the heat exchanger plates are working, they are only related to the specific temperatures and temperature differences on both sides. Therefore, the model we established only needs to consider the temperatures on both sides. The heat transfer coefficient is related to the structure of the evaporator heat exchanger plate. After the evaporator heat exchanger plate structure is designed, the heat transfer coefficients on the air side and the refrigerant side can be calculated based on the temperatures on the air side and the refrigerant side, that is, the air side heat transfer coefficient model: , heat transfer coefficient model on the refrigerant side: β i =ψ(T 蒸发 ).
[0130] The cooling demand is the heat transfer from the cooling capacity provided by the evaporator through the heat exchange equipment. Therefore, the heat provided by the cooling capacity of any single compressor after passing through the heat exchange plate is exactly equal to the cooling demand on the air side, that is:
[0131] Q 压缩机i (N P ,T 蒸发 ,T 冷凝 )=α i β i Q i .
[0132] Here, let the cooling capacity difference Φ i (N P ,T 蒸发 ,T 冷凝 )=Q 压缩机i (N P ,T 蒸发 ,T 冷凝 )-α i β i Q i ,
[0133] Among them, Φ i (N P ,T 蒸发 ,T 冷凝 ) is the difference between the cooling capacity supplied by the compressor and the cooling capacity required on the demand side, and the difference model of the cooling capacity supply and demand balance is obtained. There are three variables in the formula: compressor speed (N P ), evaporation temperature (T 蒸发 ), condensation temperature (T 冷凝 ), these three parameters determine whether our cooling supply and demand are balanced.
[0134] Evaporation temperature (T 蒸发 ), condensation temperature (T 冷凝 ) calculation requires comprehensive consideration of multiple factors, including refrigerant type, cooling method, condenser and evaporator design, and operating environment. Here, the refrigerant type, cooling method, condenser and evaporator design are all known and can be calculated. The operating environment mainly refers to the operating ambient temperature and ambient humidity.
[0135] In order to allow the evaporator to transfer heat effectively with the fresh air, the evaporation temperature has a quantitative difference relationship with the supply air temperature, and the condensation temperature has a quantitative difference relationship with the fresh air temperature. Generally, the evaporation temperature is ΔT lower than the supply air temperature, and the condensation temperature is ΔT′ higher than the fresh air temperature, that is:
[0136] T 蒸发1 =t 新风干球 -ΔT1,
[0137] T 冷凝1 =t 新风干球 +ΔT′1,
[0138] T 蒸发2 =t 中间干球 -ΔT2,
[0139] T 冷凝2 =t 中间干球 +ΔT′2,
[0140] Among them, ΔT1∈(t1,t2), ΔT2∈(t3,t4), ΔT′1∈(t5,t6), ΔT′2∈(t7,t8).
[0141] Therefore, the evaporation temperature and condensation temperature can be taken into account in the cooling capacity difference Φ i (N P ,T 蒸发 ,T 冷凝 ) transforms the cooling capacity model, and obtains the first-stage cooling capacity difference and the second-stage cooling capacity difference as follows:
[0142] Φ1(N P ,T 蒸发 ,T 冷凝 )=Φ1(N P ,t 中间干球 ,ΔT1,ΔT2)=Φ1(N P ,t 新风干球 -ΔT1,t 新风干球 +ΔT′1),
[0143] Φ2(N P ′,T 蒸发 ′,T 冷凝 ′)=Φ2(N P ,t 中间干球 ,ΔT′1,ΔT′2)=Φ2(N P ′,t 中间干球 -ΔT2,t 中间干球 +ΔT′2).
[0144] Figure 5 A specific flow chart for finding the optimal solution with minimum energy consumption.
[0145] like Figure 5 As shown in the figure, the difference between the supply and demand of the cooling capacity consumed by the two-stage refrigeration is taken as the optimization target, and the minimum value of the optimization target is calculated:
[0146] Min[Φ1(N P ,t 中间干球 ,ΔT1,ΔT2)]1,
[0147] Min[Φ2(N P ′,t 中间干球 ,ΔT′1,ΔT′2)]2,
[0148] in, t 中间干球 ∈(t 出风干球 ,t 新风干球 ).
[0149] The fitness function set in this application is the objective function to be optimized:
[0150] Fit(Φ1)=Φ1(N P ,t 中间干球 ,ΔT1,ΔT2),
[0151] Among them, t 中间干球 ∈(t 出风干球 ,t 新风干球 ), ΔT1∈(t1,t2),ΔT2∈(t3,t4),
[0152] Fit(Φ2)=Φ2(N P ,t 中间干球 ,ΔT′1,ΔT′2),
[0153] Among them, t 中间干球 ∈(t 出风干球 ,t 新风干球 ), ΔT′1∈(t5,t6), ΔT′2∈(t7,t8).
[0154] The particle swarm optimization algorithm is used to solve the optimal solution for the minimum value of the two-stage refrigeration energy consumption function, and the particle swarm optimization algorithm position is set as:
[0155] t={N P ,t 中间 , ΔT1, ΔT2},
[0156] The speed update formula is:
[0157] v id (t+1)=wv id (t)+c1r1(pBest id -x id (t))+c2r2(gBest d -x id (t)),
[0158] The position update formula is:
[0159] x id (t+1)=x id (t)+v id(t+1),
[0160] Among them, v id (t+1) is the velocity of particle i in generation t+1, x id (t+1) is the position of particle i in the t+1 generation; w is the inertia weight, the number of iterations decreases linearly, the learning factors c1 and c2 are 2, the uniformly distributed random numbers r1 and r2 are 0.5, and pBest id is the historical optimal position of the particle, gBest d is the global optimal position of the group.
[0161] The particle swarm algorithm is used to find the optimal solution with minimum energy consumption. The specific optimization steps are as follows:
[0162] At step 502 , particle swarm calculation is entered, and the operation proceeds to step 504 .
[0163] In step 504, all particles are initialized (initial possible solutions), including the size of the group N, and their speed v id and position assignment x id After completing the operation of step 504 , the operation proceeds to step 506 .
[0164] In step 506 , the individual's historical best pBest is set as the current position, and the best individual in the group is used as the current gBest. After completing the operation of step 506 , the operation proceeds to step 508 .
[0165] In step 508, in each generation of evolution, the fitness function value of each particle is calculated as a criterion for judging the quality. After completing the operation of step 508, the operation proceeds to step 510.
[0166] In step 510, for each particle, use its fitness value and individual extreme value pBest id Compare, if the current fitness function value is better than the historical optimal value, update pBest id After completing the operation of step 510 , the operation proceeds to step 512 .
[0167] In step 512, for each particle, its fitness value and the global extreme value gBest are used to calculate the d Compare, if the current fitness function value is better than the global historical optimal value, update gBest d After completing the operation of step 512 , the operation proceeds to step 514 .
[0168] In step 514, the d-th dimension velocity and position of each particle i are updated according to the velocity update formula and the position update formula, respectively. After completing the operation of step 514, the operation proceeds to step 516.
[0169] In step 516, it is determined whether the particle has reached the convergence condition, that is, the absolute value of the current objective function | Φ i |Whether it is within the tolerance value ΔQ. If the particle reaches the convergence condition, the operation goes to step 518 to add the particle to the solution set; if the particle does not reach the convergence condition, the particle is discarded, and the operation goes to step 520 to determine whether the maximum number of cycles K is met. max .
[0170] In step 518, after adding the particle to the solution set, the operation goes to step 520 to determine whether the maximum number of loops K is met. max .
[0171] In step 520, it is determined whether the maximum number of cycles K is satisfied. max If the maximum number of cycles K is not met max , the operation goes to step 522, after setting k=k+1, and then goes to step 508, recalculates the fitness function value of each particle, and continues the iterative calculation; if the maximum number of cycles K is met max , the operation goes to step 524, outputs the particle solution set, and obtains the optimal solution under the condition of the minimum value of the two-stage refrigeration energy consumption function.
[0172] The difference between the supply and demand of the refrigeration capacity consumed by the two-stage refrigeration in this application is taken as the optimization target. Through the above particle swarm algorithm, multiple optimization solutions {N P ,t 中间 , ΔT1, ΔT2}, and each optimal solution can also be used to calculate each group {N P ,T 蒸发 ,T 冷凝}. Based on multiple optimization solutions, multiple total energy consumption (P d ).
[0173] As described above, in step 210, the optimal total energy consumption (P d最优 ) and determine the optimal intermediate temperature value (t 中间最优 ), where the total refrigeration power consumption is the sum of the primary refrigeration section compressor power consumption, the secondary refrigeration section compressor power consumption, the indoor fan power consumption, the outdoor fan power consumption, the expansion valve power consumption and the heat exchange power consumption. The calculation formula is as follows:
[0174] Total power consumption of two-stage refrigeration cycle = power consumption of compressor in first-stage refrigeration section +
[0175] Secondary refrigeration compressor power consumption +
[0176] Indoor fan power consumption+
[0177] Outdoor fan power consumption+
[0178] Expansion valve power consumption+
[0179] Heat exchange power consumption.
[0180] Among them, indoor fan power consumption: Generally, the air supply volume of the indoor unit remains unchanged. This application considers that the power of the indoor unit is a fixed value under the set working conditions, that is, P 室内风机 .
[0181] The expansion valve is divided into thermal expansion valve and electronic expansion valve. The thermal expansion valve consumes almost no energy, while the energy consumed by the electronic expansion valve is not large and relatively fixed. Therefore, this application sets the power consumption of the expansion valve to be fixed, that is, P 膨胀阀 .
[0182] Heat exchange energy consumption: The operation itself does not generate power consumption, but when transferring heat between different media, there will be energy loss. This energy consumption comes from the refrigerant and depends on many factors, such as the heat transfer capacity of the heat transfer medium, flow rate, working conditions, materials, etc. The specific value needs to be determined according to the specific specifications and working conditions of the heat exchanger. This patent does not consider the power consumption generated by the heat exchanger itself, but only considers the impact of the heat transfer model on the heat transfer coefficient.
[0183] Compressor power consumption calculation model: The refrigerant of the compressor has been selected. First, the power provided by the compressor must meet the needs of the refrigeration cycle. Second, the refrigeration cycle has some losses. We have compiled a large amount of compressor speed, evaporation temperature and condensation temperature data based on the experimental data and empirical formulas provided by the manufacturer. Based on this data, we can use the polynomial training model to fit the data and find the power model of the compressor under different speeds, different evaporation temperatures and condensation temperature data:
[0184] P 压缩机i (N P ,T 蒸发 ,T 冷凝 ).
[0185] Outdoor condensing fan power consumption: Use the outdoor fan to blow the heat of the condenser, so that the condenser can exchange heat with the outside environment. According to the refrigeration principle, the mass flow rate of the refrigerant in the compressor is:
[0186]
[0187] Where Q 压缩机i (N P ,T 蒸发 ,T 冷凝 ) represents the compressor refrigeration load.
[0188] Because the mass flow rate of the compressor is equal to the mass flow rate of the condenser, the refrigerant releases heat Q to the outside in the condenser 冷凝 =q m ×(h2-h3).
[0189] According to the mathematical formula for heat exchange design of air-cooled condensers, the air volume V of the fan on the condenser can be calculated. 冷凝风机 , the ambient temperature is measured by the actual ambient temperature as t a1 , the condenser air outlet temperature can be measured t a2 , after heat exchange calculation between outdoor air side and refrigerant device, the required air volume can be obtained ρ a and C pa These are the thermophysical parameters of air. a and C pa It can be obtained through query, so it is considered as known data. The fan performance curve parameters include the main performance parameters of the fan, such as the curve of the relationship between air volume Q, air pressure H, power N and efficiency η. According to the fan performance curve parameters, the air pressure H corresponding to the fan air volume can be obtained, and thus the power consumption of the condensing fan can be obtained according to the following formula:
[0190] P 冷凝风机 =ρVH / 3600η,
[0191] Among them, P 冷凝风机 represents the power of the axial flow fan, ρ represents the air density, V represents the air volume, H represents the wind pressure, and η represents the efficiency of the fan.
[0192] The power calculation model P consumed by the single-stage refrigeration section si for:
[0193] P si =P 压缩机i (N P ,T 蒸发 ,P 冷凝 )+P 冷凝风机i +P 室内风机 +P 膨胀阀 ,
[0194] The power calculation model P consumed by the two-stage refrigeration section d for:
[0195] P d =P s1 +P s2 ,
[0196] according to Figure 5 The particle swarm algorithm shown can obtain multiple optimization solutions and the power consumed by the above two-stage refrigeration section. The number of loop calculations is the number of optimization solutions, and finally multiple total energy consumption (P d ) in the minimum value Min[P d ], at this time, the energy consumption of the entire dehumidifier is the lowest, and the corresponding intermediate temperature t 中间 This is the optimal setting of the intermediate temperature t中间最优 .
[0197] Set the optimal intermediate temperature t 中间最优 Input into the air conditioning system, so that the first-stage cooling can set the intermediate temperature t 中间最优 Using the target temperature for cooling operation can optimize the energy demand distribution of the refrigeration unit and minimize the total energy consumption.
[0198] Figure 6 FIG. 1 is a block diagram of the controller in more detail, showing the main components of the controller 160. The controller 160 is capable of storing and executing Figure 2 , 3A , 4 and 5, the procedures shown in the flow chart, storage and calling as Figure 2 , 3A , 4 and 5.
[0199] like Figure 6 As shown, the controller 160 includes a bus 602, a processor 604, a memory 606, an input interface 608, and an output interface 610. The processor 604, the memory 606, the input interface 608, and the output interface 610 are connected to the bus 602. The processor 604 can read a program (or instruction) from the memory 606 and execute the program (or instruction) to process data; the processor 604 can also write data or a program (or instruction) into the memory 606. The memory 606 can store programs (instructions) or data. By executing the instructions in the memory 606, the processor 604 can control the memory 606, the input interface 608, and the output interface 610.
[0200] In the present application, the memory 606 can store the Figure 2 , 3A , 4 and 5 and the operating parameters required to execute the procedures.
[0201] The input interface 608 is configured to receive parameters from various sensor feedback signals or user input, and convert the data of these parameters into signals recognizable by the processor 604 and store them in the memory 606 .
[0202] The processor 604 is configured to input the signal parameters stored in the memory 606 into the program, thereby calculating the optimal set intermediate temperature (optimal intermediate dry-bulb temperature), and then obtaining the optimal operating frequency of the two-stage compressor and the optimal operating frequency of the two-stage fan.
[0203] The output interface 610 is configured to receive the required valve parameters from the processor 604, and output them to each level of compressors and each level of condensing fans through the control line 612, thereby adjusting the operating frequency of each level of compressors and each level of condensing fans so that the entire air-conditioning system is always operating in the current optimal state.
[0204] The energy demand allocation method for an air conditioning system and the air conditioning system using the method of the present application can at least achieve the following beneficial technical effects:
[0205] First, consider the optimal energy efficiency of the system as a whole: by establishing an energy consumption model for the refrigeration system, the algorithm is used to find the optimal energy efficiency point of the system, so that the entire system is always running in the current optimal state.
[0206] Second, the advantages of algorithm optimization: Compared with the existing fixed dew point or fixed mechanical control technology, the use of machine learning and optimization algorithm technology reduces the complexity of optimizing the intermediate temperature, and can quickly find the most suitable intermediate temperature, so that the energy consumption of two-stage refrigeration reaches the optimal state, with fast response speed, saving manpower and material resources, and lower costs.
[0207] Although the present application has been described in conjunction with the examples of the embodiments summarized above, it is obvious to those of ordinary skill in the art that various alternatives, modifications, changes, improvements and / or substantially equivalent solutions, whether known or currently or foreseeable in the near future, may be apparent. In addition, the technical effects and / or technical problems described in this specification are exemplary rather than restrictive; so the disclosure in this specification may be used to solve other technical problems and have other technical effects and / or may solve other technical problems. Therefore, the examples of the embodiments of the present application as stated above are intended to be illustrative rather than restrictive. Various changes may be made without departing from the spirit or scope of the present application. Therefore, the present application is intended to include all known or earlier developed alternatives, modifications, changes, improvements and / or substantially equivalent solutions.
Claims
1. A method for allocating energy demand for an air conditioning system having at least two stages of refrigeration, characterized in that The following steps are involved: S01, prepare data; S02, selecting an enthalpy-pressure diagram model and performing training on the enthalpy-pressure diagram model; S03, establishing a cooling capacity demand relationship model between the air side cooling capacity and the refrigerant side cooling capacity of the air conditioning system; S04, establishing a target optimization model and selecting an optimization algorithm; S05, calculating the optimal total energy consumption and determining the optimal intermediate temperature value; as well as S06: Allocate energy requirements for the at least two stages of refrigeration according to the optimal intermediate temperature value.
2. The energy demand allocation method for an air conditioning system according to claim 1, characterized in that Step S03 includes: S03-1, establish a cooling capacity demand model; S03-2, establish a compressor refrigeration load model and a compressor energy consumption model; S03-3, training and fitting a refrigeration capacity demand model, a compressor refrigeration load model, and a compressor energy consumption model; and S03-4, perform heat balance calculation on the indoor air side and the refrigerant side.
3. The energy demand allocation method for an air conditioning system according to claim 2, characterized in that: The cooling capacity demand model is related to the intermediate temperature value; The compressor refrigeration load model is related to the compressor speed, evaporation temperature and condensation temperature; and The compressor energy consumption model is related to the compressor speed, evaporation temperature and condensation temperature.
4. The energy demand allocation method for an air conditioning system according to claim 3, characterized in that: The evaporation temperature and the condensation temperature are related to the type of refrigerant, the cooling method, the design of the condenser or evaporator, the ambient temperature and the ambient humidity.
5. The energy demand allocation method for an air conditioning system according to claim 2, characterized in that: In step S03-3, the least squares method is used for model training and fitting.
6. The energy demand allocation method for an air conditioning system according to claim 1, characterized in that: In step S01, the data includes refrigeration demand parameters and refrigeration source cycle parameters.
7. The energy demand allocation method for an air conditioning system according to claim 1, characterized in that: In step S02, the enthalpy-pressure diagram model includes a pressure-enthalpy line model, an isotherm line model and an isentropic line model; and The enthalpy-pressure diagram model is trained using the least squares method.
8. The energy demand allocation method for an air conditioning system according to claim 1, characterized in that: In step S04, the target optimization model is set as a fitness function; and The optimization algorithm is a particle swarm optimization algorithm.
9. The energy demand allocation method for an air conditioning system according to claim 8, characterized in that: In step S05, multiple total energy consumptions are calculated based on the multiple optimization solutions obtained by the particle swarm optimization algorithm in step S04. The minimum value among the multiple total energy consumptions corresponds to the optimal total energy consumption, and the intermediate temperature value corresponding to the optimal total energy consumption is the optimal intermediate temperature value.
10. The energy demand allocation method for an air conditioning system according to claim 9, characterized in that: The total energy consumption includes compressor energy consumption, and the compressor energy consumption is related to compressor speed, evaporation temperature and condensation temperature.
11. The energy demand allocation method for an air conditioning system according to claim 9, characterized in that: The total energy consumption includes the power consumption of an outdoor condensing fan, and the power consumption of the outdoor condensing fan can be determined according to the enthalpy-pressure diagram model, the compressor speed, the evaporation temperature, the condensation temperature and the fan performance curve parameters.
12. An air conditioning system, characterized in that The method comprises at least two stages of refrigeration and a controller, wherein each stage of refrigeration in the at least two stages of refrigeration comprises: A compressor, a condenser, a throttle valve and an evaporator, and the controller is configured to distribute energy demand for the at least two stages of refrigeration according to the energy demand distribution method for an air-conditioning system as described in any one of claims 1 to 11.
Citation Information
Cited By
Refrigerating system optimization control method based on virtual cooling capacity
CN119879452A
A virtual cold quantity-based refrigeration system optimization control method
CN119879452B
