A variable air volume air conditioning system sensor online calibration method based on a physical model

Abstract

The application discloses a kind of variable air volume air conditioning system sensor online calibration methods based on physical model, comprising the following steps: S1, define variable air volume air conditioning system sensor multiple groups steady-state measurement value;S2, for variable air volume air conditioning system, according to energy conservation and pressure conservation these physical models and the steady-state measurement value obtained in S1 are modeled respectively to system sensor;S3, using Bayes theorem and Monte Carlo random sampling is calibrated to temperature, humidity, flow and differential pressure sensor in different calibration domain of variable air volume air conditioning system.The application combines historical data and the physical model of system, on the basis that new sensor is not removed or installed, for the sensor offset of different working environment and aging, after calibration, the accuracy of sensor measurement value has greatly promoted, guarantee the accuracy and stability of sensor bottom layer data, so that variable air volume air conditioning system more reliably operates.

Description

Technical Field

[0001] This invention relates to an online calibration method for sensors in a variable air volume (VAV) air conditioning system based on a physical model. It is an online calibration method for VAV air conditioning systems with rich information and different types of sensor faults, and belongs to the field of underlying sensor data processing technology for building energy consumption monitoring systems. Background Technology

[0002] With continuous economic and social development and the accelerating pace of urbanization, a large influx of people into cities has placed immense pressure on urban areas where land is extremely scarce. High-rise buildings have become the primary choice for alleviating housing pressure and addressing land scarcity. As building structures become increasingly complex, variable air volume (VAV) air conditioning systems are also becoming more complex. Statistics show that VAV air conditioning systems are widely used in residential and commercial buildings, industrial plants, and data centers, consuming significant amounts of energy; for example, the energy consumption of VAV air conditioning systems in buildings typically accounts for 50-60% of the total building energy consumption. To address the high energy consumption of building systems, various control methods are commonly used for intelligent regulation, such as continuous fine-tuning, analysis and optimization, fault detection and diagnosis based on building automation systems. Extensive research in this area has been conducted both domestically and internationally. In practice, the detection and control of VAV air conditioning systems rely heavily on monitoring data from various sensors within the system. Therefore, the accuracy of sensor measurements is crucial for the intelligent control of building systems. Currently, various methods and measures are being used to diagnose sensor faults, thereby improving the accuracy of sensor measurements. However, these methods and measures face two practical problems in the calibration process: 1) Focusing only on a single fault: Building energy systems are complex, with interconnected sensors. During operation, due to chain reactions, a fault in one sensor can lead to the failure of several others. Research on the diagnosis of concurrent faults in multiple sensors is currently scarce. 2) Existing sensor fault diagnosis methods focus only on diagnosis and do not address repair. After a period of operation, variable air volume (VAV) air conditioning systems will inevitably experience varying degrees of sensor drift. How to further repair the system after successful diagnosis is a crucial issue that must be addressed. Summary of the Invention

[0003] Based on the above problems, this invention combines historical data and the physical model of the system (some basic theorems such as energy conservation, pressure and flow conservation) to provide an online calibration method for sensors in a variable air volume air conditioning system based on a physical model. This method calibrates the sensor offset caused by different working environments and aging without removing or installing new sensors.

[0004] The technical solution of the present invention is as follows:

[0005] A method for online calibration of sensors in a variable air volume (VAV) air conditioning system based on a physical model, comprising the following steps:

[0006] S1. Define multiple sets of steady-state measurement values ​​for the sensors in the variable air volume (VAV) air conditioning system. The specific steps are as follows:

[0007] S1.1 Determine the type and quantity of sensors to be installed in the variable air volume (VAV) air conditioning system;

[0008] S1.2. Based on the sensor type and quantity, obtain the original sensor measurement values ​​from the variable air volume air conditioning system monitoring platform and construct a measurement value matrix;

[0009] S1.3. Based on indoor and outdoor temperature and humidity, clustering is used to cluster the measurement data datasets under different working conditions, thereby ensuring that the measurement data datasets under the same working conditions are in the same class.

[0010] S1.4 Randomly select data from the measurement data dataset under the same working conditions to form the steady-state measurement data dataset required for calibration;

[0011] S2. For the variable air volume (VAV) air conditioning system, the system sensors are modeled based on physical models such as energy conservation and pressure conservation, as well as the steady-state measurement values ​​obtained in S1. The specific steps are as follows:

[0012] S2.1 Energy Conservation Model 1: Under air conditioning operation, the fresh air ratio of the entire variable air volume (VAV) air conditioning system is controlled by adjusting the valve sizes of the fresh air valve and return air valve; at this time, the fresh air energy plus the return air energy equals the mixed air energy; based on the above analysis, energy equations for fresh air, return air, and mixed air are established; the fresh air energy equation is determined by the fresh air temperature. T 1 and fresh air humidity Φ 1. Composition; The return air energy equation is composed of return air temperature. T 4 and return air humidity Φ 4 components; the mixed air energy equation is composed of mixed air temperature. T 2 and mixed air humidity Φ 2. Composition; Based on the fresh air ratio, establish the energy conservation equations for fresh air, mixed air, and return air;

[0013] S2.2 Energy Conservation Model 2: Under summer operating conditions, air flowing through the surface cooler exchanges heat fully with the chilled water in the coil. Assuming accurate sensor measurements, the heat carried away by the air should equal the heat lost by the chilled water. Based on the above analysis, energy conservation equations are established for both the air and chilled water sides. The air-side energy equation is determined by the air mixing temperature. T 2. Mixed air humidity Φ 2. Supply air temperature T 3. Supply air humidity Φ 3 and fresh air volume M 1. Composition; The water-side energy equation is based on the chilled water supply temperature.T 5. Chilled water return temperature T 6 and chilled water flow rate M Composed of 2 components;

[0014] S2.3, Energy Conservation Model 3: Under summer operating conditions, the refrigerant in the chiller unit removes energy from the chilled water through heat exchange. Then, it is transported to the condenser by the compressor to exchange heat with the cooling water, which in turn removes energy from the refrigerant, thus maintaining the refrigerant's previous state. Therefore, the energy conservation equation at this time is that the energy of the chilled water plus the compressor power consumption equals the energy of the cooling water. Based on the above analysis, energy conservation equations are established for both the chilled water side and the cooling water side. The energy equation for the chilled water side is determined by the chilled water supply temperature. T 5. Chilled water return temperature T 6 and chilled water flow rate M 2. Composition; The energy equation for the cooling water side is based on the cooling water supply temperature. T 7. Cooling water return temperature T 8 and cooling water flow rate M 3. Based on the compressor power, establish the energy conservation equations for the chilled water side and the cooling water side;

[0015] S2.4 Pressure Difference and Flow Rate Conservation Model: According to fluid mechanics theory, there is a linear relationship between chilled water flow rate and the pressure difference it consumes; based on this relationship, a conservation model of chilled water flow rate and pressure difference is constructed; similarly, for cooling water in the calibration domain, a conservation model is also established based on the relationship between pressure difference and flow rate.

[0016] S3. The temperature, humidity, flow rate, and differential pressure sensors of the variable air volume (VAV) air conditioning system are calibrated in different calibration domains using Bayes' theorem and Monte Carlo random sampling. The specific steps are as follows:

[0017] S3.1 Establish the prior distribution of offsets for each calibration domain and various types of sensors (mean is 0, assuming no offset occurred before calibration; standard deviation is the sensor accuracy).

[0018] S3.2 Define the calibration function, reference function and target function required for sensor calibration; where the calibration function is derived from the sensor measurement value and offset, and the reference function is derived from the energy conservation or pressure difference and flow conservation model containing the calibration function, as shown in equations (1)-(3);

[0019] Calibration function: (1)

[0020] Benchmark function: (2)

[0021] Objective function: (3)

[0022] in,S c For the calibration measurement values ​​of the sensor, R These are the original measurements from the sensor. x This is the offset. S b This is the reference output value of the sensor. f This is a model based on energy conservation or pressure difference and flow rate conservation. v The number of related variables, J(x) To calibrate the objective function of the problem, T The number of steady-state datasets, N To calibrate the number of sensors;

[0023] S3.3 Substitute the established objective function into the Bayesian likelihood function to obtain the probability density function of the posterior distribution, as shown in equations (4) and (5);

[0024] Likelihood function: (4)

[0025] Bayes' theorem: (5)

[0026] in, Let be the likelihood function of the offset. σ Standard deviation, As a prior distribution, For the posterior distribution of the offset, This is the normalization constant;

[0027] S3.4 Determine the number of iterations and lag point for Monte Carlo sampling, and use the prior average distribution of the offset of different types of sensors as the initial sample for sampling; generate a candidate sample from the probability density function based on the selected initial sample, and calculate the acceptance rate of the candidate sample, as shown in Equation (6).

[0028] Acceptance rate (6)

[0029] in, Let be the probability density function. Candidate samples generated from the probability density function, The previous candidate sample before the jump;

[0030] S3.5. Set a threshold, which can be any number between 0 and 1. When the acceptance rate is greater than the threshold, accept the transfer and proceed to the next iteration until the preset number of iterations is reached; otherwise, keep the original value; perform statistics on all samples that meet the condition of having an acceptance rate greater than the threshold and obtain the mean of these samples.

[0031] S3.6. Through S3.1-S3.5, obtain the average value of the offset of various types of sensors in each calibration domain, and then use the average value of these samples as the offset of the faulty sensor, add it to the original measurement value of the sensor, and realize the calibration of the fault of the sensor in different calibration domains.

[0032] The beneficial effects of this invention are as follows: This invention provides an online calibration method for different types of sensor faults that can effectively improve the data quality of variable air volume (VAV) air conditioning systems. It can calibrate all sensors in the VAV air conditioning system, and the accuracy of sensor measurements is greatly improved after calibration, ensuring the accuracy and stability of the sensor's underlying data, and making the VAV air conditioning system operate more reliably. Attached Figure Description

[0033] Figure 1 Flowchart for online calibration of sensors in a variable air volume (VAV) air conditioning system;

[0034] Figure 2 This is a diagram showing the sensor locations for a variable air volume (VAV) air conditioning system. Detailed Implementation

[0035] The specific embodiments of the present invention will be described in detail below with reference to the invention description and accompanying drawings.

[0036] Sensor locations for variable air volume (VAV) air conditioning systems, such as Figure 2 As shown, refer to Figure 1 The present invention provides an online calibration method for sensors in a variable air volume (VAV) air conditioning system based on a physical model, comprising the following steps:

[0037] S1. Define multiple sets of steady-state measurement values ​​for the sensors in the variable air volume (VAV) air conditioning system. The specific steps are as follows:

[0038] S1.1 Determine the type and quantity of sensors to be installed in the variable air volume (VAV) air conditioning system;

[0039] S1.2. Based on the sensor type and quantity, obtain the original sensor measurement values ​​from the variable air volume air conditioning system monitoring platform and construct a measurement value matrix;

[0040] S1.3. Based on indoor and outdoor temperature and humidity, clustering is used to cluster the measurement data datasets under different working conditions, thereby ensuring that the measurement data datasets under the same working conditions are in the same class.

[0041] S1.4 Randomly select data from the measurement data dataset under the same working conditions to form the steady-state measurement data dataset required for calibration;

[0042] S2. For the variable air volume (VAV) air conditioning system, the system sensors are modeled based on physical models such as energy conservation and pressure conservation, as well as the steady-state measurement values ​​obtained in S1. The specific steps are as follows:

[0043] S2.1 Energy Conservation Model 1: Under air conditioning operation, the fresh air ratio of the entire variable air volume (VAV) air conditioning system is controlled by adjusting the valve sizes of the fresh air valve and return air valve; at this time, the fresh air energy plus the return air energy equals the mixed air energy; based on the above analysis, energy equations for fresh air, return air, and mixed air are established; the fresh air energy equation is determined by the fresh air temperature. T 1. Fresh air humidity Φ The composition is as shown in the following formula; the return air energy equation is based on the return air temperature. T 4. Return air humidity Φ The mixture consists of four components, as shown in the following formula; the mixed air energy equation is based on the mixed air temperature. T 2. Mixed air humidity Φ The composition is as shown in the following formula; based on the fresh air ratio, the energy conservation equations for fresh air, mixed air, and return air are established as shown in the following formula;

[0044] Fresh air energy equation (7)

[0045] Mixed wind energy equation (8)

[0046] Return air energy equation (9)

[0047] Energy Conservation Equation 1 (10)

[0048] in, The enthalpy values ​​of fresh air, mixed air, and return air are expressed in kJ / kg. The specific heat at constant pressure of dry air is typically 1.005 kJ / (kg•℃); The temperatures of the fresh air, mixed air, and return air are in °C. The specific heat at constant pressure of water vapor is typically 1.84 kJ / (kg•℃); air humidity, % The pressure is the saturated water vapor pressure, in Pa; Atmospheric pressure, typically 101.325 kPa; For fresh air ratio.

[0049] S2.2 Energy Conservation Model 2: Under summer operating conditions, air flowing through the surface cooler exchanges heat fully with the chilled water in the coil. Assuming accurate sensor measurements, the heat carried away by the air should equal the heat lost by the chilled water. Based on the above analysis, energy conservation equations are established for both the air and chilled water sides. The air-side energy equation is determined by the air mixing temperature. T 2. Mixed air humidity Φ 2. Supply air temperature T 3. Supply air humidity Φ 3 and fresh air volume M1. Composition; The water-side energy equation is based on the chilled water supply temperature. T 5. Chilled water return temperature T 6 and chilled water flow rate M Composed of 2 components;

[0050] Energy Conservation Equation 2 (11)

[0051] in, The specific heat at constant pressure of chilled water is typically 4.186 kJ / (kg•℃).

[0052] S2.3, Energy Conservation Model 3: Under summer operating conditions, the refrigerant in the chiller unit removes energy from the chilled water through heat exchange. Then, it is transported to the condenser by the compressor to exchange heat with the cooling water, which in turn removes energy from the refrigerant, thus maintaining the refrigerant's previous state. Therefore, the energy conservation equation here is that the energy of the chilled water plus the compressor power consumption equals the energy of the cooling water. Based on the above analysis, energy conservation equations are established for both the chilled water side and the cooling water side. The chilled water side energy equation is determined by the chilled water supply temperature. T 5. Chilled water return temperature T 6 and chilled water flow rate M 2. Composition; The energy equation for the cooling water side is based on the cooling water supply temperature. T 7. Cooling water return temperature T 8 and cooling water flow rate M 3. Composition; Based on the compressor power, establish the energy conservation equations for the chilled water side and the cooling water side, as shown in the following equation;

[0053] Energy Conservation Equation 3 (12)

[0054] in, The value is the compressor power, expressed in kJ.

[0055] S2.4 Pressure Difference and Flow Rate Conservation Model: According to fluid mechanics theory, there is a linear relationship between chilled water flow rate and the pressure difference it consumes; based on this relationship, a conservation model of chilled water flow rate and pressure difference is constructed, as shown in the following equation; Similarly, for cooling water in the calibration domain, a conservation model is also established based on the relationship between pressure difference and flow rate, as shown in the following equation;

[0056] Chilled water pressure difference and flow rate conservation (13)

[0057] Cooling water pressure difference and flow rate conservation (14)

[0058] in, The pressure difference across the chilled water pump and the cooling water pump is expressed in Pa. For the impedance of chilled water and cooling water pipes and valves, Pa / (m) 3 / h) 2 ; For chilled water and cooling water flow rates, m 3 / h.

[0059] S3. The temperature, humidity, flow rate, and differential pressure sensors of the variable air volume (VAV) air conditioning system are calibrated in different calibration domains using Bayes' theorem and Monte Carlo random sampling. The specific steps are as follows:

[0060] S3.1 Establish the prior distribution of offsets for each calibration domain and various types of sensors (mean is 0, assuming no offset occurred before calibration; standard deviation is the sensor accuracy).

[0061] S3.2 Define the calibration function, reference function and target function required for sensor calibration; where the calibration function is derived from the sensor measurement value and offset, and the reference function is derived from the energy conservation or pressure difference and flow conservation model containing the calibration function, as shown in equations (1)-(3);

[0062] Calibration function: (1)

[0063] Benchmark function: (2)

[0064] Objective function: (3)

[0065] in, S c For the calibration measurement values ​​of the sensor, R These are the original measurements from the sensor. x This is the offset. S b This is the reference output value of the sensor. f This is a model based on energy conservation or pressure difference and flow rate conservation. v The number of related variables, J(x) To calibrate the objective function of the problem, T The number of steady-state datasets, N To calibrate the number of sensors;

[0066] S3.3 Substitute the established objective function into the Bayesian likelihood function to obtain the probability density function of the posterior distribution, as shown in equations (4) and (5);

[0067] Likelihood function: (4)

[0068] Bayes' theorem: (5)

[0069] Let be the likelihood function of the offset. σ Standard deviation, As a prior distribution, For the posterior distribution of the offset, This is the normalization constant;

[0070] S3.4 Determine the number of iterations and lag point for Monte Carlo sampling, and use the prior average distribution of the offset of different types of sensors as the initial sample for sampling; generate a candidate sample from the probability density function based on the selected initial sample, and calculate the acceptance rate of the candidate sample, as shown in Equation (6).

[0071] Acceptance rate (6)

[0072] in Let be the probability density function. Candidate samples generated from the probability density function, The previous candidate sample before the jump;

[0073] S3.5. Set a threshold, which can be any number between 0 and 1. When the acceptance rate is greater than the threshold, accept the transfer and proceed to the next iteration until the preset number of iterations is reached; otherwise, keep the original value; perform statistics on all samples that meet the condition of having an acceptance rate greater than the threshold and obtain the mean of these samples.

[0074] S3.6. Through S3.1-S3.5, the average displacement values ​​of various types of sensors within each calibration domain are obtained. These sample averages are then used as the offset of the faulty sensor and added to the original sensor measurement value to achieve sensor fault calibration within different calibration domains. The results of this embodiment are shown in Table 1. During the calibration process, different physical models included all sensors of the variable air volume air conditioning system, and all sensors were calibrated. After calibration, the sensor accuracy improved by more than 80%. Additionally, the supply air humidity, which was not initially measured, was also improved. Φ 3 and air volume M 1. Even after calibration, relatively accurate calibration values ​​can be obtained, which greatly ensures the accuracy of the underlying data.

[0075] Table 1 Sensor calibration results

[0076]

Claims

1. A method for online calibration of sensors in a variable air volume (VAV) air conditioning system based on a physical model, characterized in that, The steps are as follows: S1. Define multiple sets of steady-state measurement values ​​for the sensors in the variable air volume (VAV) air conditioning system. The specific steps are as follows: S1.1 Determine the type and quantity of sensors to be installed in the variable air volume (VAV) air conditioning system; S1.

2. Based on the sensor type and quantity, obtain the original sensor measurement values ​​from the variable air volume air conditioning system monitoring platform and construct a measurement value matrix; S1.

3. Based on indoor and outdoor temperature and humidity, clustering is used to cluster the measurement data datasets under different working conditions, thereby ensuring that the measurement data datasets under the same working conditions are in the same class. S1.4 Randomly select data from the measurement data dataset under the same working conditions to form the steady-state measurement data dataset required for calibration; S2. For the variable air volume (VAV) air conditioning system, the system sensors are modeled based on physical models such as energy conservation and pressure conservation, as well as the steady-state measurement values ​​obtained in S1. The specific steps are as follows: S2.1 Energy Conservation Model 1: Under air conditioning operation, the fresh air ratio of the entire variable air volume (VAV) air conditioning system is controlled by adjusting the valve sizes of the fresh air valve and return air valve; at this time, the fresh air energy plus the return air energy equals the mixed air energy; based on the above analysis, energy equations for fresh air, return air, and mixed air are established; the fresh air energy equation is determined by the fresh air temperature. T 1 and fresh air humidity Φ 1. Composition; The return air energy equation is composed of return air temperature. T 4 and return air humidity Φ 4 components; the mixed air energy equation consists of the mixed air temperature. T 2 and mixed air humidity Φ 2. Composition; Based on the fresh air ratio, establish the energy conservation equations for fresh air, mixed air, and return air; S2.2 Energy Conservation Model 2: Under summer operating conditions, air flows through the surface cooler and exchanges heat with the chilled water in the coil. If the sensor measurement is accurate, the heat carried away by the air should be equal to the heat lost by the chilled water. Based on the above analysis, energy conservation equations are established for the air side and the chilled water side. The air-side energy equation is derived from the air-mixing temperature. T 2. Mixed air humidity Φ 2. Supply air temperature T 3. Supply air humidity Φ 3 and fresh air volume M 1. Composition; The water-side energy equation is based on the chilled water supply temperature. T 5. Chilled water return temperature T 6 and chilled water flow rate M Composed of 2 components; S2.3, Energy Conservation Model 3: Under summer operating conditions, the refrigerant in the chiller unit removes energy from the chilled water through heat exchange. Then, it is transported to the condenser by the compressor to exchange heat with the cooling water, which in turn removes energy from the refrigerant, thus maintaining the refrigerant's previous state. Therefore, the energy conservation equation at this time is that the energy of the chilled water plus the compressor power consumption equals the energy of the cooling water. Based on the above analysis, energy conservation equations are established for both the chilled water side and the cooling water side. The energy equation for the chilled water side is determined by the chilled water supply temperature. T 5. Chilled water return temperature T 6 and chilled water flow rate M 2. Composition; The energy equation for the cooling water side is based on the cooling water supply temperature. T 7. Cooling water return temperature T 8 and cooling water flow rate M 3. Based on the compressor power, establish the energy conservation equations for the chilled water side and the cooling water side; S2.4 Pressure Difference and Flow Rate Conservation Model: According to fluid mechanics theory, there is a linear relationship between chilled water flow rate and the pressure difference it consumes; based on this relationship, a conservation model of chilled water flow rate and pressure difference is constructed; similarly, for cooling water in the calibration domain, a conservation model is also established based on the relationship between pressure difference and flow rate. S3. The temperature, humidity, flow rate, and differential pressure sensors of the variable air volume (VAV) air conditioning system are calibrated in different calibration domains using Bayes' theorem and Monte Carlo random sampling. The specific steps are as follows: S3.1 Establish the prior distribution of offsets for each calibration domain and various types of sensors; the mean is 0, indicating that no offset occurred before calibration; the standard deviation is the sensor accuracy. S3.2 Define the calibration function, reference function and target function required for sensor calibration; where the calibration function is derived from the sensor measurement value and offset, and the reference function is derived from the energy conservation or pressure difference and flow conservation model containing the calibration function, as shown in equations (1)-(3); Calibration function: (1) Benchmark function: (2) Objective function: (3) in, S c For the calibration measurement values ​​of the sensor, R These are the original measurements from the sensor. x This is the offset. S b This is the reference output value of the sensor. f This is a model based on energy conservation or pressure difference and flow rate conservation. v The number of relevant variables, J(x) To calibrate the objective function of the problem, T The number of steady-state datasets, N To calibrate the number of sensors; S3.3 Substitute the established objective function into the Bayesian likelihood function to obtain the probability density function of the posterior distribution, as shown in equations (4) and (5); Likelihood function: (4) Bayes' theorem: (5) in, Let be the likelihood function of the offset. σ Standard deviation, For the prior distribution, For the posterior distribution of the offset, This is a normalization constant; S3.4 Determine the number of iterations and lag point for Monte Carlo sampling, and use the prior average distribution of the offset of different types of sensors as the initial sample for sampling; generate a candidate sample from the probability density function based on the selected initial sample, and calculate the acceptance rate of the candidate sample, as shown in Equation (6). Acceptance rate (6) in, Let be the probability density function. Candidate samples generated from the probability density function, The previous candidate sample before the jump; S3.

5. Set a threshold, which can be any number between 0 and 1. When the acceptance rate is greater than the threshold, accept the transfer and proceed to the next iteration until the preset number of iterations is reached; otherwise, keep the original value; perform statistics on all samples that meet the condition of having an acceptance rate greater than the threshold and obtain the mean of these samples. S3.

6. Through S3.1-S3.5, obtain the average value of the offset of various types of sensors in each calibration domain, and then use the average value of these samples as the offset of the faulty sensor, add it to the original measurement value of the sensor, and realize the calibration of the fault of the sensor in different calibration domains.

2. The online calibration method for sensors in a variable air volume (VAV) air conditioning system based on a physical model according to claim 1, characterized in that, In step S2, The specific equations involved in the energy conservation model 1 are as follows: the fresh air energy equation is shown in equation (7); the return air energy equation is shown in equation (9); the mixed air energy equation is shown in equation (8); based on the fresh air ratio, the energy conservation equations for fresh air, mixed air and return air are established as shown in equation (10); Fresh air energy equation (7) Mixed wind energy equation (8) Return air energy equation (9) Energy Conservation Equation 1 (10) in, The enthalpy values ​​of fresh air, mixed air, and return air are expressed in kJ / kg. The specific heat at constant pressure of dry air; The temperatures of the fresh air, mixed air, and return air are in °C. The specific heat at constant pressure of water vapor; air humidity, % The pressure is the saturated water vapor pressure, in Pa; Atmospheric pressure; For fresh air ratio; The equations involved in energy conservation model 2 are as follows: Energy Conservation Equation 2 (11) in, The specific heat at constant pressure of chilled water; The equations involved in energy conservation model 3 are as follows: Energy Conservation Equation 3 (12) in, The compressor power is expressed in kJ. The equations involved in the pressure difference and flow rate conservation model are as follows: Chilled water pressure difference and flow rate conservation (13) Cooling water pressure difference and flow rate conservation (14) in, The pressure difference across the chilled water pump and the cooling water pump is expressed in Pa. For the impedance of chilled water and cooling water pipes and valves, Pa / (m) 3 / h) 2 ; For chilled water and cooling water flow rates, m 3 / h.

Images