Global energy efficiency optimization method for heating ventilation air conditioning system
Through the global energy efficiency optimization method of HVAC system, load prediction, model identification and global optimization are used to optimize equipment parameters, and the problem of system energy efficiency deviation is solved, achieving maximum energy utilization efficiency improvement and optimal energy efficiency operation state.
Patent Information
- Application Number
- CN202411089224.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-09
- Publication Date
- 2024-12-06
AI Technical Summary
During the operation of HVAC systems, energy efficiency deviations are often caused by inaccurate design parameters, changes in system component performance, adjustment of operating parameters and insufficient energy efficiency monitoring, resulting in the lack of effective energy-saving optimization methods for the system to operate under low energy efficiency conditions for a long time. .
A method for finding global energy efficiency optimization in HVAC systems is proposed. Through load prediction, model identification and global optimization, a network model and global optimization model of the air conditioning system are established, and the equipment parameters are optimized using particle swarm algorithm to find the best operation of the system as a whole. Parameter values under operating conditions to achieve energy efficiency optimization.
Through appropriate control strategies and system parameter optimization, energy utilization efficiency is maximized and the HVAC system can be ensured to meet the indoor comfort requirements while operating under the overall optimal energy efficiency.
Smart Images

Figure CN119089769A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of global energy efficiency optimization of a heating, ventilation and air conditioning system, and in particular to a method for optimizing the global energy efficiency of a heating, ventilation and air conditioning system. Background Art
[0002] During the operation of the HVAC system, the actual operation often deviates from the design. This phenomenon will affect the system operation efficiency, causing the entire HVAC system to operate in a low-efficiency condition for a long time. The possible reasons for this phenomenon are as follows:
[0003] 1. Inaccurate design parameters: When designing an HVAC system, there may be inaccurate estimates of the parameters of the building or the environment in which it is used. For example, the building heat load, occupant density, equipment heat load and other parameters used in the design may not match the actual situation, which will result in the system capacity being too large or too small and unable to meet actual needs.
[0004] 2. Changes in system component performance: The performance of HVAC system components (such as cooling towers, air conditioning units, fans, etc.) may change during long-term operation. For example, filter blockage, refrigerant leakage, sensor drift and other problems may cause system performance to deteriorate.
[0005] 3. Adjustment of operating parameters: In order to cope with changes in actual conditions, operators may adjust operating parameters based on experience or needs, such as adjusting temperature set points, wind speed control, etc. These adjustments may be inconsistent with design parameters, resulting in deviations in system performance from design.
[0006] 4. Inadequate energy efficiency monitoring and maintenance: Lack of real-time monitoring and maintenance of HVAC system energy efficiency may lead to performance degradation. For example, sensors in the system may fail or be inaccurately calibrated, and problems that are not discovered and repaired in time may cause energy efficiency to deviate from the design.
[0007] At present, most HVAC systems have a large space for energy-saving optimization. In order to solve the above problems, the present invention comprehensively considers the system model complexity, model error, timeliness and system stability, and proposes a HVAC system global energy efficiency optimization method for optimizing the energy efficiency of the HVAC system. Summary of the invention
[0008] The purpose of the present invention is to provide a method for optimizing the global energy efficiency of a HVAC system in response to the above-mentioned problems. The method optimizes the global energy efficiency of the HVAC system according to the outdoor meteorological conditions, so that the system operates under the conditions with the best overall energy efficiency, thereby ensuring that the system achieves the goal of energy saving to the maximum extent.
[0009] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is as follows:
[0010] According to one aspect of the present invention, a method for optimizing the global energy efficiency of a heating, ventilation and air conditioning system is provided, comprising the following steps:
[0011] S1. Load forecasting: forecast the load of the building's air conditioning system based on historical outdoor meteorological parameters and air conditioning system energy consumption data;
[0012] S2. Model identification: Establish the network model of each device of the air-conditioning system according to the actual operating parameters and physical mechanism model of each device of the air-conditioning system;
[0013] S3. Global optimization: Based on the network model of each device, a global optimization model of the air-conditioning system is established, and the air-conditioning system is optimized according to the global optimization model to find the parameter values of each device that can make the air-conditioning system as a whole operate in the optimal condition.
[0014] Preferably, in step S1, predicting the load of the air conditioning system of the building includes the following steps:
[0015] Choose appropriate input parameters: Choose appropriate input variables based on the building type;
[0016] Data cleaning: Clean the data of input variables and remove abnormal data;
[0017] Input parameter selection: According to the correlation between the input variables and the true value of energy consumption, the input parameters with greater correlation with the true value of energy consumption are selected as the input of the prediction model;
[0018] Prediction model training and parameter preservation: The training set and test set are randomly divided in a ratio of 8:2, the prediction model is trained and the model results are saved for subsequent calls.
[0019] Preferably, the input variables include outdoor temperature, outdoor humidity, a time period corresponding to the current energy consumption, and a day corresponding to the current energy consumption.
[0020] Preferably, the correlation between the input variable and the true value of energy consumption is represented by a Pearson correlation coefficient, and the Pearson correlation coefficient is ≥ 0.4.
[0021] Preferably, in step S2, the physical mechanism model includes a refrigerator model, a chilled water pump model and a cooling water pump model.
[0022] Preferably, according to the refrigerator model, the relationship expression of the load rate, the chilled water outlet temperature and the cooling water inlet temperature of the chiller is obtained, which is specifically:
[0023] COP=(a0*PLR 2 +b0*PLR+c0)*(a1*ELFT 2+b1*ELFT+c1)*(a2-b2
[0024] *CEFT-c2*PLR+d2*CEFT 2 +e2*CEFT*PLR-f2*PLR 2 )
[0025] Among them, ELFT represents the chilled water outlet temperature, in °C; CEFT represents the cooling water inlet temperature, in °C; PLR represents the partial load rate of the chiller; a0~f2 represent the fitting coefficients of different chillers.
[0026] Preferably, according to the chilled water pump model and the cooling water pump model, a relational expression of the water pump power on the frequency and a relational expression of the water pump flow on the frequency are obtained, which are specifically:
[0027] P=a0*Fre 3 +b0*Fre 2 +c0*Fre+d0
[0028] Q=a1*Fre+b1
[0029] Among them, P represents the power of the water pump, the unit is kW; Q represents the flow rate of the water pump, the unit is m 3 / h; Fre is the frequency of the water pump, the unit is Hz; a0~b1 are the fitting coefficients of different water pumps.
[0030] Preferably, in step S3, the optimization method of the global optimization model comprises the following steps:
[0031] Establishing a global optimization expression based on the refrigerator model, the chilled water pump model and the cooling water pump model;
[0032] The optimization constraints are clarified, and the constraints include load, outlet water temperature of chilled water, inlet water temperature of chilled water, chilled water temperature difference, refrigeration pump frequency and cooling pump frequency.
[0033] The particle swarm algorithm is used to optimize the global optimization expression and output the parameters of the chiller, chilled water pump and cooling water pump.
[0034] Preferably, the global optimization expression is:
[0035] P=P_chiller+P_cold_pump+P_cooling_pump+P_cooling_tower
[0036] =f(ELFT,CEFT,PLR)
[0037]
[0038] Where P_chiller represents the power of the refrigeration unit in kW; Q cooling It represents the predicted cooling load demand in kW; COP represents the relationship between the load rate, chilled water outlet temperature and cooling water inlet temperature of the chiller; Q cold_water Indicates the chilled water flow rate in m 3 / h;
[0039] c represents the specific heat capacity of water; Δt0 represents the temperature difference of chilled water; P_cold_pump represents the power of the chilled water pump, in kW; Q cooling_water Indicates cooling water flow rate, in m 3 / h; Δt1 represents the temperature difference of cooling water; P_cooling_pump represents the power of cooling water pump, in kW; P_cooling_tower represents the power of cooling tower, in kW.
[0040] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:
[0041] The present invention predicts the load of the building's air conditioning system based on historical outdoor meteorological parameters and air conditioning system energy consumption data, and establishes a HVAC equipment model expression based on the actual operating parameters of the HVAC system equipment and the physical mechanism model, and finally establishes a global optimization expression to optimize the system and find the parameter values of each device that can make the system as a whole in the optimal operating condition. Through appropriate control strategies and system parameter optimization, the system can maximize energy utilization efficiency while meeting indoor comfort requirements, ensuring that the HVAC system operates under the global optimal energy efficiency condition. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] Figure 1 It is a schematic diagram of the overall process of the present invention;
[0043] Figure 2 is a schematic diagram of the load forecasting process of the present invention;
[0044] Figure 3 It is a schematic diagram of the model identification process of the present invention;
[0045] Figure 4 It is a schematic diagram of the global optimization process of the present invention. DETAILED DESCRIPTION
[0046] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and preferred embodiments. However, it should be noted that many details listed in the specification are only for the purpose of enabling the reader to have a thorough understanding of one or more aspects of the invention, and these aspects of the present invention can be realized even without these specific details.
[0047] See also Figures 1 to 4 The present invention provides a method for optimizing the global energy efficiency of a HVAC system, and the technical solution is as follows:
[0048] A method for optimizing the global energy efficiency of a heating, ventilation and air conditioning system comprises the following steps:
[0049] S1. Load forecasting: Based on historical outdoor meteorological parameters and air conditioning system energy consumption data, the load of the building's air conditioning system is predicted.
[0050] like Figure 2 As shown, specifically, first, select appropriate input variables of the prediction model according to the building type. The input variables include meteorological data and time parameters. The meteorological data includes outdoor temperature and outdoor humidity. The time parameters include the hours corresponding to the current energy consumption and the day of the week corresponding to the current energy consumption.
[0051] Then, the data of the input variables is cleaned. A variety of methods can be used to clean abnormal data, including but not limited to empty value cleaning and over-limit cleaning.
[0052] Then, the input parameters are selected, and the input parameters of the prediction model are selected according to the size of the Pearson correlation coefficient between the input parameters and the true value of energy consumption. In this embodiment, the Pearson correlation coefficient must be ≥0.4. For example, the Pearson correlation coefficient between the outdoor temperature and the true value of energy consumption is 0.6, which means that the outdoor temperature has a strong positive correlation with energy consumption. When the outdoor temperature rises, the air conditioning system needs to consume more energy for cooling to maintain a comfortable indoor temperature. The Pearson correlation coefficient between the outdoor humidity and the true value of energy consumption is 0.5. Higher outdoor humidity may increase the dehumidification burden of the air conditioning system, thereby increasing energy consumption. The Pearson correlation coefficient between rainfall and the true value of energy consumption is 0.2. Generally speaking, rainfall has a relatively small impact on the energy consumption of the HVAC system, so the correlation is weak. The Pearson correlation coefficient between the nearby noise level and the true value of energy consumption is 0.1, and the noise level is usually not directly significantly related to the energy consumption of the air conditioning system. In this embodiment, the input parameters with a Pearson correlation coefficient of ≥0.4 are selected as the input variables of the prediction model.
[0053] Finally, the prediction model is trained using the selected input variables, and the model output parameter results are saved. The input parameters include outdoor temperature, outdoor humidity, the number of hours corresponding to the current energy consumption, and the day of the week corresponding to the current energy consumption. The model training uses historical outdoor meteorological parameters and system energy consumption data. Regarding the selection of the prediction model, the gradient boosting decision tree (XGBoost) algorithm model, the random forest (Random Forest) model, the support vector machine (Support Vector Machine, SVM), and the artificial neural network (Artificial Neural Network, ANN) can be selected. In this embodiment, the gradient boosting decision tree (XGBoost) algorithm model is used as the prediction model. The training set and the test set are randomly divided in a ratio of 8:2, the gradient boosting decision tree algorithm model is trained, and the model output results are saved for subsequent calls.
[0054] S2. Model identification: Based on the actual operating parameters and physical mechanism model of the HVAC system equipment, the HVAC equipment model expression is established by comprehensively considering the model complexity and errors caused by simplification.
[0055] like Figure 3 As shown, specifically, the physical mechanism model includes a refrigerator model, a chilled water pump model, and a cooling water pump model. According to the chiller COP at different load rates corresponding to different cooling water inlet temperatures under different chilled water inlet and outlet temperature conditions provided by the manufacturer, the chiller model is identified. Finally, the relationship expression of the chiller COP with respect to load rate, chilled water outlet temperature, and cooling water inlet temperature is obtained:
[0056] COP=(a0*PLR 2 +b0*PLR+c0)*(a1*ELFT 2 +b1*ELFT+c1)*(a2-b2
[0057] *CEFT-c2*PLR+d2*CEFT 2 +e2*CEFT*PLR-f2*PLR 2 )
[0058] Among them, ELFT stands for EvaporatorLeavingFluidTemp, which is the chilled water outlet temperature in °C;
[0059] CEFT stands for Condenser Entering Fluid Temp, which is the cooling water inlet temperature in °C;
[0060] PLR stands for Partial load rate, which is the partial load rate of the chiller;
[0061] a0~f2 represent the fitting coefficients, and different chillers need to be determined based on specific data fitting.
[0062] Based on the field test data, the model identification of the chilled water pump and the cooling water pump is carried out. The relationship between the pump power and the frequency and the flow rate and the frequency is obtained, that is,
[0063] P=a0*Fre 3 +b0*Fre 2 +c0*Fre+d0
[0064] Q=a1*Fre+b1
[0065] Where P represents the power of the water pump, in kW; Q represents the flow rate of the water pump, in m 3 / h; Fre represents the frequency of the water pump, in Hz; a0~b1 represent the fitting coefficients, which need to be determined based on specific data fitting for different water pumps.
[0066] S3. Global optimization: Combine the above expressions to form a global optimization expression for the HVAC system, and use the particle swarm optimization algorithm to optimize the system and find the parameter values of each device that can make the system as a whole operate in the optimal condition.
[0067] like Figure 4 As shown, specifically, a global optimization expression is established, namely:
[0068] P=P_chiller+P_cold_pump+P_cooling_pump+P_cooling_tower
[0069] =f(ELFT,CEFT,PLR)
[0070]
[0071] Where P_chiller represents the power of the refrigeration unit in kW; Q cooling It represents the predicted cooling load demand in kW; COP represents the relationship between the load rate, chilled water outlet temperature and cooling water inlet temperature of the chiller; Q cold_water Indicates the chilled water flow rate in m 3 / h;
[0072] c represents the specific heat capacity of water; Δt0 represents the temperature difference of chilled water; P_cold_pump represents the power of the chilled water pump, in kW; Q cooling_water Indicates cooling water flow rate, in m 3 / h; Δt1 represents the temperature difference of cooling water; P_cooling_pump represents the power of cooling water pump, in kW; P_cooling_tower represents the power of cooling tower, in kW.
[0073] By adding up the power of the main equipment on the cold source side mentioned above, we can get the global total power P, in kW.
[0074] The constraints of the optimization algorithm are clarified through the following parameter characteristics.
[0075] Load, based on weather forecasts, can be used to obtain the load demand in the future. The cooling capacity of the chiller should not be less than the load demand in the load forecast;
[0076] The outlet temperature of chilled water should not be too low (wasting electricity) or too high (ineffective cooling).
[0077] The inlet water temperature of the chilled water can avoid unexpected situations at the end;
[0078] Reasonable chilled water temperature difference to avoid too small chilled water flow;
[0079] Reasonable refrigeration pump frequency to avoid surge;
[0080] Reasonable cooling pump frequency to avoid surge.
[0081] After determining the constraints, the particle swarm is used to optimize the above global optimization expression. Specifically, the Python language instantiates the particle swarm optimization algorithm, sets the optimization goal, optimization constraints, optimization independent variables and model parameters of the function, where:
[0082] The optimization goal is min(P), which is to minimize the overall system energy consumption;
[0083] The optimization constraints are the constraints for optimizing the parameters listed above, namely, the constraints of load, chilled water outlet water temperature, chilled water inlet water temperature, chilled water temperature difference, refrigeration pump frequency, and cooling pump frequency;
[0084] The optimization independent variables are the optimization load rate PLR, the chilled water outlet temperature ELFT and the cooling water inlet temperature CEFT, and the output of the particle swarm optimization algorithm is the optimal result of the above independent variables;
[0085] On this basis, the chilled water flow and cooling water flow can be calculated. According to the relationship between the water pump flow and the frequency, i.e., Q = f (Fre), the frequency of the chilled water pump and the cooling water pump can be calculated.
[0086] Based on the optimization results, the following can be set: the load rate PLR of the refrigeration host, the chilled water outlet temperature ELFT, the cooling water inlet temperature CEFT, and the frequencies of the chilled water pump and the cooling water pump.
[0087] According to the above steps, the optimal independent variables under the corresponding objectives are obtained, namely, the load rate PLR, the chilled water outlet temperature ELFT and the cooling water inlet temperature CEFT.
[0088] The present invention achieves the goal of meeting indoor comfort requirements while maximizing energy efficiency through appropriate control strategies and system parameter optimization, ensuring that the HVAC system operates under conditions with optimal global energy efficiency.
[0089] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.
Claims
1. A method for optimizing the global energy efficiency of a HVAC system, characterized in that: The following steps are involved: S1. Load forecasting: forecast the load of the building's air conditioning system based on historical outdoor meteorological parameters and air conditioning system energy consumption data; S2. Model identification: Establish the network model of each device of the air-conditioning system according to the actual operating parameters and physical mechanism model of each device of the air-conditioning system; S3. Global optimization: Based on the network model of each device, a global optimization model of the air-conditioning system is established, and the air-conditioning system is optimized according to the global optimization model to find the parameter values of each device that can make the air-conditioning system as a whole operate in the optimal condition.
2. A method for optimizing the global energy efficiency of a HVAC system according to claim 1, characterized in that: In step S1, the load prediction of the air conditioning system of the building includes the following steps: Choose appropriate input parameters: Choose appropriate input variables based on the building type; Data cleaning: Clean the data of input variables and remove abnormal data; Input parameter selection: According to the correlation between the input variables and the true value of energy consumption, the input parameters with greater correlation with the true value of energy consumption are selected as the input of the prediction model; Prediction model training and parameter preservation: The training set and test set are randomly divided in a ratio of 8:2, the prediction model is trained and the model results are saved for subsequent calls.
3. A method for optimizing the global energy efficiency of a HVAC system according to claim 2, characterized in that: The input variables include outdoor temperature, outdoor humidity, a time period corresponding to the current energy consumption, and a day corresponding to the current energy consumption.
4. A method for optimizing the global energy efficiency of a HVAC system according to claim 2, characterized in that: The correlation between the input variable and the true value of energy consumption is represented by the Pearson correlation coefficient, and the Pearson correlation coefficient is ≥ 0.
4.
5. A method for optimizing the global energy efficiency of a HVAC system according to claim 1, characterized in that: In step S2, the physical mechanism model includes a refrigerator model, a chilled water pump model and a cooling water pump model.
6. A method for optimizing the global energy efficiency of a HVAC system according to claim 5, characterized in that: According to the refrigerator model, the relationship expression of the load rate, chilled water outlet temperature and cooling water inlet temperature of the chiller is obtained, which is specifically: COP=(a0*PLR 2 +b0*PLR+c0)*(a1*ELFT 2 +b1*ELFT+c1)*(a2-b2*CEFT-c2*PLR+d2*CEFT 2 +e2*CEFT*PLR-f2*PLR 2 ) Among them, ELFT represents the chilled water outlet temperature, in °C; CEFT represents the cooling water inlet temperature, in °C; PLR represents the partial load rate of the chiller; a0~f2 represent the fitting coefficients of different chillers.
7. A method for optimizing the global energy efficiency of a HVAC system according to claim 6, characterized in that: According to the freezing water pump model and the cooling water pump model, the relationship between the water pump power and the frequency and the relationship between the water pump flow and the frequency are obtained, which are specifically: P=a0*Fre 3 +b0*Fre 2 +c0*Fre+d0 Q=a1*Fre+b1 Among them, P represents the power of the water pump, the unit is kW; Q represents the flow rate of the water pump, the unit is m 3 / h; Fre is the frequency of the water pump, in Hz; a0~b1 are the fitting coefficients of different water pumps.
8. A method for optimizing the global energy efficiency of a HVAC system according to claim 7, characterized in that: In step S3, the optimization method of the global optimization model comprises the following steps: Establishing a global optimization expression based on the refrigerator model, the chilled water pump model and the cooling water pump model; The optimization constraints are clarified, and the constraints include load, outlet water temperature of chilled water, inlet water temperature of chilled water, chilled water temperature difference, refrigeration pump frequency and cooling pump frequency. The particle swarm algorithm is used to optimize the global optimization expression and output the parameters of the chiller, chilled water pump and cooling water pump.
9. A method for optimizing the global energy efficiency of a HVAC system according to claim 8, characterized in that: The global optimization expression is: P=P_chiller+P_cold_pump+P_cooling_ump+P_cooling_tower =f(ELFT,CEFT,PLR) Where P_chiller represents the power of the refrigeration unit in kW; Q cooling It represents the predicted cooling load demand in kW; COP represents the relationship between the load rate, chilled water outlet temperature and cooling water inlet temperature of the chiller; Q cold_water Indicates the chilled water flow rate in m 3 / h; c represents the specific heat capacity of water; Δt0 represents the temperature difference of chilled water; P_cold_pump represents the power of the chilled water pump, in kW; Q cooling_water Indicates cooling water flow rate, in m 3 / h; Δt1 represents the temperature difference of cooling water; P_cooling_pump represents the power of cooling water pump, in kW; P_cooling_tower represents the cooling tower power in kW.
Citation Information
Cited By
Energy-saving optimization control method and system for water chilling unit of central air conditioner
CN119412784A
Smart energy control system
CN120403034A
