A cooling tower outlet water temperature prediction method, system and optimization control method

Abstract

This invention provides a method, system, and optimized control method for predicting the outlet water temperature of a cooling tower. The prediction method includes the following steps: obtaining the outdoor wet-bulb temperature t of the cooling tower. s Inlet water temperature t in The cooling water flow rate Q and the actual operating power N of the cooling tower fan; the outdoor wet-bulb temperature t s Inlet water temperature t in The cooling water flow rate Q and the actual operating power N of the cooling tower fan are input into a pre-established regression prediction model to obtain the predicted cooling tower outlet water temperature t. out The regression prediction model is expressed as: where a, b, c, d, and e are regression coefficients. This method eliminates the need for additional airflow monitoring equipment, enabling accurate prediction of cooling tower outlet water temperature.

Description

Technical Field

[0001] This invention belongs to the field of air conditioning and refrigeration technology, and specifically relates to a method and system for predicting the outlet water temperature of a cooling tower, as well as an optimized control method for an air conditioning cold source system. Background Technology

[0002] In air conditioning systems, cooling towers are key heat dissipation devices, and their outlet water temperature directly affects the condensing pressure of chillers and the energy efficiency of system operation. Accurate prediction of outlet water temperature is of great significance for the optimized control of air conditioning cold source systems.

[0003] In existing technologies, empirical models based on multiple regression analysis have been widely used and researched due to their advantages such as simple structure, low computational cost, and ease of engineering deployment. Related research has established a multiple regression model between cooling tower outlet water temperature and inlet water temperature, cooling water flow rate, cooling tower air volume, and outdoor wet-bulb temperature, with the following expression:

[0004]

[0005] In the formula, t out The outlet water temperature of the cooling tower is ℃; t s Outdoor wet-bulb temperature, °C; t in Q is the inlet water temperature of the cooling tower, in °C; Q is the actual operating flow rate of the cooling water, in m³. 3 / h; G is the actual operating air volume of the cooling tower, kg / s; a, b, c, d, and e are regression coefficients.

[0006] However, the engineering application of this model is significantly limited: its input parameters depend on the actual operating air volume G of the cooling tower. In most actual engineering sites, due to the high cost of air volume monitoring, the difficulty of sensor installation, and the limited conditions for on-site modification, air volume is not monitored in real time. As a result, although the model is theoretically valid, it is difficult to put into practical application due to the lack of key parameters. Summary of the Invention

[0007] This invention provides a method, system, and optimized control method for predicting the outlet water temperature of a cooling tower, which eliminates the need for additional airflow monitoring equipment and enables accurate prediction of the cooling tower outlet water temperature.

[0008] The technical solution of the present invention is as follows:

[0009] A method for predicting the outlet water temperature of a cooling tower includes the following steps:

[0010] S1: Obtain the outdoor wet-bulb temperature t of the cooling tower. s Inlet water temperature t in Cooling water flow rate Q and actual operating power N of cooling tower fan;

[0011] S2: The outdoor wet-bulb temperature t s Inlet water temperature t in The cooling water flow rate Q and the actual operating power N of the cooling tower fan are input into a pre-established regression prediction model to obtain the predicted cooling tower outlet water temperature t. out ;

[0012] The expression for the regression prediction model is:

[0013] ;

[0014] In the formula, a, b, c, d, and e are regression coefficients.

[0015] Furthermore, in the method for predicting the outlet water temperature of the cooling tower, the regression coefficients a, b, c, d, and e are obtained by fitting historical operating data using a multiple linear regression method.

[0016] Furthermore, in the method for predicting the outlet water temperature of the cooling tower, before fitting the regression coefficients using the multiple linear regression method, a step of cleaning the historical operating data is also included.

[0017] Furthermore, in the method for predicting the outlet water temperature of the cooling tower, the data cleaning includes removing abnormal data caused by manual intervention and / or overlapping periods of equipment switching.

[0018] Furthermore, in the method for predicting the cooling tower outlet water temperature, the actual operating power N of the cooling tower fan is obtained through one of the following methods:

[0019] Directly measure the input power of the cooling tower fan;

[0020] Read the power value from the frequency converter of the cooling tower fan;

[0021] It is calculated based on the current and voltage of the cooling tower fan.

[0022] A system for predicting the outlet water temperature of a cooling tower, comprising:

[0023] The data acquisition module is used to obtain the outdoor wet-bulb temperature, inlet water temperature, cooling water flow rate, and actual operating power of the cooling tower fan in real time.

[0024] The regression calculation module is equipped with a regression prediction model, which is used to calculate the predicted cooling tower outlet water temperature based on the data acquired by the data acquisition module.

[0025] The output module is used to output the prediction results calculated by the regression calculation module.

[0026] A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method for predicting the outlet water temperature of a cooling tower.

[0027] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the program to implement the method for predicting the outlet water temperature of a cooling tower.

[0028] An optimized control method for an air conditioning cooling source system includes:

[0029] The cooling tower outlet water temperature is predicted using the aforementioned method.

[0030] Based on the predicted outlet water temperature, adjust one or more of the following: the operating frequency of the cooling water pump, the operating frequency of the cooling tower fan, or the operating parameters of the chiller unit.

[0031] Furthermore, in the optimized control method for the air conditioning cold source system, the operating frequency of the cooling water pump, the operating frequency of the cooling tower fan, or the operating parameters of the chiller unit are adjusted to optimize the total operating energy consumption of the air conditioning cold source system.

[0032] The beneficial effects of this invention are as follows:

[0033] This invention provides a method for predicting the outlet water temperature of a cooling tower. It uses the actual operating power N of the cooling tower fan to replace the air volume G, which is difficult to measure directly, and derives G ∝ N using the fan similarity law. 1 / 3 The relationship was established and incorporated into the regression prediction model, thus avoiding the cost and maintenance difficulties of installing air volume sensors.

[0034] The method for predicting the outlet water temperature of the cooling tower adopts a multiple linear regression model. The model has a simple structure and low computational cost, and can be easily deployed in building automation systems, programmable logic controllers or embedded systems without the need for high-performance hardware support.

[0035] The method for predicting the outlet water temperature of this cooling tower is achieved by introducing N. 1 / 3 This feature enables the model to accurately reflect the impact of changes in fan power on the outlet water temperature. At the same time, data cleaning removes outlier data, further improving the model's prediction accuracy.

[0036] This optimized control method for air conditioning cooling source systems, based on the predicted cooling tower outlet water temperature, minimizes the overall operating energy consumption of the cooling source system by coordinating the adjustment of cooling water pump frequency, cooling tower fan frequency, and chiller unit operating parameters, while dynamically balancing heat dissipation capacity and energy consumption demand. This significantly improves the energy efficiency of the air conditioning system, reduces carbon dioxide emissions, and contributes to the green and low-carbon development of the HVAC industry. Attached Figure Description

[0037] Figure 1 This is a flowchart of a method for predicting the outlet water temperature of a cooling tower according to the present invention;

[0038] Figure 2 This is a flowchart of the model training process for a method for predicting the outlet water temperature of a cooling tower according to the present invention.

[0039] Figure 3 This is a curve comparing the calculated and measured values ​​of the cooling tower outlet water temperature in Embodiment 1 of the present invention;

[0040] Figure 4 This is a curve comparing the calculated and measured values ​​of the cooling tower outlet water temperature in Embodiment 2 of the present invention;

[0041] Figure 5 This is a curve comparing the calculated and measured values ​​of the cooling tower outlet water temperature in Embodiment 3 of the present invention. Detailed Implementation

[0042] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The advantages and features of the present invention will become clearer from the following description and claims. It should be noted that the drawings are all in a very simplified form and use non-precise proportions, and are only used to facilitate and clarify the illustration of the embodiments of the present invention.

[0043] like Figure 1 As shown, this embodiment provides a method for predicting the outlet water temperature of a cooling tower, including the following steps: S1-S2.

[0044] S1: Data Acquisition. Acquire the outdoor wet-bulb temperature t of the cooling tower. s Inlet water temperature t in Cooling water flow rate Q and actual operating power N of cooling tower fan.

[0045] Among them, outdoor wet-bulb temperature t s It can be obtained through a wet-bulb temperature sensor placed near the cooling tower, or calculated by combining meteorological data with dry-bulb temperature and relative humidity. Inlet water temperature t in Temperature can be obtained through a temperature sensor installed on the cooling tower's inlet pipe. Cooling water flow rate Q can be obtained through a flow meter installed on the cooling water pipeline, or estimated through the pump frequency and performance curve.

[0046] The actual operating power N of the cooling tower fan can be obtained through one of the following methods:

[0047] Direct power measurement: The input power of the cooling tower fan is directly measured using a power transmitter;

[0048] Read from the frequency converter: If the fan is driven by a frequency converter, the current output power can be read from the frequency converter;

[0049] Calculation based on current and voltage: This is obtained by measuring the current and voltage of the cooling tower fan motor and combining this with the power factor calculation.

[0050] Step S1 obtains key operating parameters of the cooling tower through various flexible methods. In particular, it uses easily acquired fan power instead of difficult-to-measure air volume, providing a reliable data foundation for subsequent high-precision prediction models without the need to add dedicated air volume sensors. This significantly reduces the hardware cost and maintenance difficulty of the system.

[0051] S2: Model prediction. The outdoor wet-bulb temperature t... s Inlet water temperature t in The cooling water flow rate Q and the actual operating power N of the cooling tower fan are input into a pre-established regression prediction model to obtain the predicted cooling tower outlet water temperature t. out .

[0052] The expression for the regression prediction model is:

[0053] ;

[0054] In the formula, a, b, c, d, and e are regression coefficients obtained by fitting historical operating data.

[0055] The specific principle derivation is as follows:

[0056] Cooling tower ventilation systems typically have relatively simple and fixed airflow paths, and their resistance characteristics remain basically stable during operation. The system's resistance characteristic curve does not change, satisfying the application conditions of the similarity law. According to the fan similarity law, the cooling tower fan power N and the cooling tower air volume G satisfy the following relationship:

[0057]

[0058] In the formula, N is the actual operating power of the cooling tower fan (kW); N' is the rated power of the cooling tower fan (kW); G is the actual operating air volume of the cooling tower (kg / s); and G' is the rated air volume of the cooling tower (kg / s).

[0059] From the above formula, the expression for the air volume during the operation of the cooling tower can be derived:

[0060]

[0061] When the cooling tower model is determined, G' and N' are both known quantities, and their values ​​can be included in the parameter d to be identified through regression. Therefore, the air volume G and N 1 / 3 They are directly proportional, that is, G ∝ N 1 / 3 .

[0062] Based on the principle of heat and mass exchange, the outlet water temperature t of the cooling tower out Mainly affected by outdoor wet-bulb temperature t s Inlet water temperature t in The influence of cooling water flow rate Q and cooling tower air volume G. Substituting the above proportional relationships into N... 1 / 3 Instead of directly measuring the air volume G, a multiple linear regression model is used for modeling, resulting in the regression prediction model of this application:

[0063]

[0064] In the formula, a, b, c, d, and e are regression coefficients obtained by fitting historical operating data.

[0065] The above method, in this invention, introduces the cube root term N of the cooling tower fan power. 1 / 3 Instead of directly measuring air volume, a cooling tower outlet water temperature prediction model based on multiple linear regression was constructed. This model achieves high-precision prediction without the need to install air volume sensors, significantly reducing system costs and maintenance difficulty, and providing a reliable basis for energy-saving optimization control of air conditioning cold source systems.

[0066] like Figure 2 As shown, in a preferred embodiment, the method for predicting the outlet water temperature of the cooling tower further includes S3.

[0067] S3: Model Building and Updates. During the initial model building, it is necessary to collect historical operational data over a period of time, including the outdoor wet-bulb temperature t of the cooling tower. s Inlet water temperature t in The data collected should include the cooling water flow rate Q, the actual operating power N of the cooling tower fan, and the corresponding actual outlet water temperature (as training labels). The collected data should cover various operating conditions of the cooling tower, such as different loads, different ambient temperatures, and different fan operating frequencies.

[0068] Before fitting regression coefficients using the multiple linear regression method, it is necessary to clean the historical data and remove outliers. Outliers mainly include:

[0069] Data resulting from manual intervention: such as manually starting and stopping fans, manually adjusting valves, etc.

[0070] Data resulting from equipment switching overlap: such as transitional data during frequency changes of fan inverters and non-steady-state data during the start-up and shutdown of cooling towers;

[0071] The sensor malfunction resulted in clearly erroneous data.

[0072] After data cleaning, multiple linear regression was used to fit the regression coefficients a, b, c, d, and e. Specifically, the actual outlet water temperature was used as the dependent variable, and t was used as the variable. s t in Q, N 1 / 3 Using as the independent variable, a multiple linear regression equation is established, and the regression coefficients are solved using the least squares method.

[0073] After the training model is established, it can be periodically updated online using newly added operational data to maintain the model's prediction accuracy. During updates, a sliding window method or recursive least squares method can be used, refitting only with data from the most recent period to adapt to the slow changes in cooling tower performance.

[0074] In step S3, by cleaning historical operating data and fitting model coefficients using a multiple linear regression method, a cooling tower outlet water temperature prediction model based on the cube root of fan power is established. Through an online update mechanism, the model can adapt to the slow changes in cooling tower performance, thereby maintaining high-precision prediction capability over the long term.

[0075] In a preferred embodiment, the actual operating power N of the cooling tower fan is obtained through one of the following methods:

[0076] Method 1: Directly measure the input power of the cooling tower fan.

[0077] A power transmitter or smart meter can be installed at the power input of the cooling tower fan to directly measure the fan's input power. The power transmitter can output analog or digital signals, which can be connected to a data acquisition module. This method offers high measurement accuracy but requires additional hardware.

[0078] Method 2: Read the power value from the frequency converter of the cooling tower fan.

[0079] If the cooling tower fan is driven by a frequency converter, the current output power can be read from the frequency converter via a communication interface. Common communication protocols include Modbus, BACnet, and Profibus. This method requires no additional hardware, is low-cost, and easy to implement.

[0080] Method 3: Calculated based on the current and voltage of the cooling tower fan.

[0081] The operating current and voltage of the cooling tower fan motor are measured, and the input power of the fan is calculated by combining the motor's power factor. The calculation formula is:

[0082]

[0083] In the formula, U is the line voltage, kV; I is the line current, A; The power factor.

[0084] This method only requires the addition of a current transformer and a voltage sampling circuit, resulting in lower costs and making it suitable for fixed-frequency fans without frequency converters.

[0085] By acquiring wind turbine power through three flexible methods—direct measurement, inverter reading, or current and voltage calculation—this method provides a low-cost, highly reliable, and easily implemented source of power data for prediction models without the need for air volume sensors, significantly improving the method's engineering adaptability and application value.

[0086] This embodiment also provides a system for predicting the outlet water temperature of a cooling tower, including a data acquisition module, a regression calculation module, and an output module.

[0087] The data acquisition module is used to acquire the outdoor wet-bulb temperature, inlet water temperature, cooling water flow rate, and actual operating power of the cooling tower fan in real time. The data acquisition module may include a wet-bulb temperature sensor, an inlet water temperature sensor, a flow meter, and a power acquisition unit. The wet-bulb temperature sensor is located near the air inlet of the cooling tower and is used to measure the outdoor wet-bulb temperature. The inlet water temperature sensor is located on the cooling tower inlet pipe and is used to measure the inlet water temperature. The flow meter is located on the cooling water pipeline and is used to measure the cooling water flow rate. The power acquisition unit is used to acquire the actual operating power of the cooling tower fan and may be a power transmitter, a communication interface with a frequency converter, or a current and voltage acquisition module.

[0088] The regression calculation module is equipped with a regression prediction model, used to calculate the predicted cooling tower outlet water temperature based on the data acquired by the data acquisition module. The regression calculation module receives real-time data sent by the data acquisition module and calculates the predicted cooling tower outlet water temperature t based on the regression prediction model. out The regression calculation module can be a controller in a building automation system, a programmable logic controller, an embedded system, or an industrial computer.

[0089] The output module is used to output the prediction results calculated by the regression calculation module. The output module may include a display screen, a communication interface, or an analog output interface. The prediction results can be displayed in real time on a local display screen, sent to a higher-level monitoring system via the communication interface, or sent as control signals to cooling water pumps, cooling tower fans, or chiller units to achieve optimized control.

[0090] The aforementioned prediction system acquires key operating parameters through a data acquisition module, which is powered by a built-in N... 1 / 3 The regression calculation module of the regression model predicts the outlet water temperature in real time, and the output module applies the results to display or optimization control, realizing the integrated approach from data acquisition to prediction application, and providing a plug-and-play, high-precision, low-cost intelligent prediction solution for air conditioning cold source systems.

[0091] This embodiment also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for predicting the outlet water temperature of the cooling tower. The computer-readable storage medium can be any medium with program storage capabilities, such as ROM, RAM, magnetic disk, optical disk, USB flash drive, or portable hard drive.

[0092] This embodiment also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor, when executing the program, implements the method for predicting the outlet water temperature of the cooling tower. The processor can be any chip or circuit with program execution capabilities, such as a central processing unit, digital signal processor, microcontroller, or ARM processor.

[0093] This embodiment also provides an optimized control method for an air conditioning cold source system, including the following steps: S10-S20.

[0094] S10: Using the aforementioned method for predicting the cooling tower outlet water temperature, the cooling tower outlet water temperature t is predicted. out Get the current outdoor wet-bulb temperature t of the cooling tower. s Inlet water temperature t in The cooling water flow rate Q and the actual operating power N of the cooling tower fan are input into the regression prediction model to calculate the predicted outlet water temperature t of the cooling tower. out .

[0095] S20: Based on the predicted outlet water temperature, adjust one or more of the following: the operating frequency of the cooling water pump, the operating frequency of the cooling tower fan, or the operating parameters of the chiller unit.

[0096] Specifically, based on the predicted outlet water temperature t out The optimization and control of various devices in the air conditioning cooling source system includes one or more of the following control strategies:

[0097] Strategy 1: Optimized control of cooling water pumps

[0098] Adjust the operating frequency of the cooling water pump based on the predicted outlet water temperature. For example:

[0099] When the predicted outlet water temperature is low, it indicates that the cooling tower has sufficient heat dissipation capacity. The frequency of the cooling water pump can be appropriately reduced and the cooling water flow rate reduced to save pump energy consumption.

[0100] When the predicted outlet water temperature is high, it indicates that the cooling tower's heat dissipation capacity is insufficient. The frequency of the cooling water pump can be appropriately increased to enhance the cooling water flow rate and improve the heat dissipation effect.

[0101] Strategy 2: Optimized Control of Cooling Tower Fans

[0102] Adjust the operating frequency of the cooling tower fan based on the predicted outlet water temperature. For example:

[0103] When the predicted outlet water temperature is low, it indicates that the cooling tower has excessive heat dissipation capacity. The fan frequency and air volume can be appropriately reduced to save fan energy consumption.

[0104] When the predicted outlet water temperature is high, it indicates that the cooling tower's heat dissipation capacity is insufficient. The fan frequency and air volume can be appropriately increased to enhance the heat dissipation effect.

[0105] Strategy 3: Optimized Control of Chiller Units

[0106] Adjust the operating parameters of the chiller unit based on the predicted outlet water temperature. For example:

[0107] When the predicted outlet water temperature is low, it indicates that the cooling conditions are good. The set outlet water temperature of the chiller unit can be appropriately increased to improve the unit's energy efficiency.

[0108] When the predicted outlet water temperature is high, it indicates that the cooling conditions are poor. The set outlet water temperature of the chiller unit can be appropriately reduced to ensure the cooling effect of the air conditioning terminal.

[0109] In a system where multiple chiller units operate in parallel, the number of units to be put into operation can be determined based on the predicted outlet water temperature.

[0110] Collaborative optimization control

[0111] To achieve optimal overall system energy consumption, the three strategies mentioned above can be synergistically optimized. Specifically, an energy consumption model for the air conditioning cooling source system is established, using the cooling water pump frequency, cooling tower fan frequency, and chiller unit operating parameters as optimization variables, and minimizing the total system energy consumption as the objective function. An optimization algorithm is then used to solve for the optimal control parameters. The goal of the optimization is to minimize the total operating energy consumption of the air conditioning cooling source system.

[0112] The aforementioned optimized control method for the air conditioning cooling source system, based on the predicted cooling tower outlet water temperature, achieves the minimization of the overall operating energy consumption of the cooling source system by coordinating the adjustment of the cooling water pump frequency, cooling tower fan frequency, and chiller unit operating parameters, thereby dynamically balancing heat dissipation capacity and energy consumption demand, and significantly improving the energy efficiency level of the air conditioning system.

[0113] The method for predicting the outlet water temperature of the cooling tower according to the present invention will be further described in detail below with reference to specific implementation examples.

[0114] Example 1: Application verification in a five-star hotel

[0115] This embodiment verifies the application of the method of the present invention in the air conditioning cooling system of a five-star hotel. The hotel is located in a hot-summer, warm-winter region, with a building area of ​​approximately 156,000 square meters, and the air conditioning cooling system is installed in the hotel's refrigeration room.

[0116] The air conditioning cooling system is equipped with three 4220kW centrifugal chiller units and two 2638kW centrifugal chiller units, and the cooling tower fans are controlled by frequency converters.

[0117] Step 1: Data Collection. The hotel's operational data from August was selected as the validation dataset. The collected parameters included: the outdoor wet-bulb temperature (t) of the cooling tower. s Cooling tower inlet water temperature t in The cooling water flow rate Q, the actual operating power N of the cooling tower fan, and the actual outlet water temperature of the cooling tower.

[0118] Data was collected every 5 minutes, resulting in approximately 8,900 data points. Data from August 1st to 16th was used as the training set to fit regression coefficients; data from August 17th to 31st was used as the test set to validate the model's predictive performance.

[0119] Step 2: Model Fitting. After cleaning the training set data and removing outliers caused by human intervention or overlapping equipment switching periods, regression coefficients are obtained using multiple linear regression. The established prediction model is as follows:

[0120]

[0121] Step 3: Verify the prediction effect. Substitute the test set data into the above model to calculate and predict the cooling tower outlet water temperature, and compare it with the actual operating cooling tower outlet water temperature. The coefficient of determination R between the two sets of data is calculated. 2 The value of 0.914 indicates that the model has a good fit.

[0122] Step 4: Typical day comparison analysis. Figure 3 The graph shows a comparison between the model-predicted and actual measured values ​​of the cooling tower outlet water temperature on August 28th. As can be seen from the graph, the calculated and measured values ​​of the cooling tower outlet water temperature generally show a consistent trend and good agreement. The average relative error is 0.59%, and the maximum relative error is 1.51%.

[0123] The predicted values ​​accurately follow changes in the measured values, maintaining high prediction accuracy during both high-load periods during the day and low-load periods at night.

[0124] The above results verify the accuracy and effectiveness of the prediction model of this invention in the air conditioning cold source system of a five-star hotel.

[0125] Example 2: Application Verification in an Office Building

[0126] This embodiment verifies the application of the method of the present invention in the air conditioning cooling system of a Grade A office building. The office building is located in a hot-summer, cold-winter region, with a building area of ​​approximately 38,000 square meters. The air conditioning cooling system is equipped with two screw chillers, two cooling water pumps, and two crossflow cooling towers. The cooling tower fans are controlled by frequency converters.

[0127] Step 1: Data Collection. The operational data for this office building in August was selected as the validation dataset. The collected parameters included: the outdoor wet-bulb temperature (t) of the cooling tower. s Cooling tower inlet water temperature t in The cooling water flow rate Q, the actual operating power N of the cooling tower fan, and the actual outlet water temperature of the cooling tower.

[0128] Data was collected every 10 minutes, resulting in approximately 4400 data points. Data from August 1st to 11th was used as the training set to fit regression coefficients; data from August 12th to 31st was used as the test set to validate the model's predictive performance.

[0129] Step 2: Model Fitting. The training set data is cleaned, and regression coefficients are obtained using multiple linear regression. The established prediction model is as follows:

[0130]

[0131] Step 3: Verification of Prediction Results. Substitute the test set data into the above model to calculate and predict the cooling tower outlet water temperature, and compare this predicted temperature with the actual operating cooling tower outlet water temperature. The coefficient of determination R between the two sets of data is calculated. 2 The value of 0.906 indicates that the model has a good fit.

[0132] Step 4: Typical day comparison analysis. Figure 4 The graph shows a comparison between the model's predicted and actual measured values ​​of the cooling tower outlet water temperature on August 13th. As can be seen from the graph, the calculated and measured values ​​of the cooling tower outlet water temperature generally match well, and the model accurately captures the dynamic changes in the outlet water temperature. The average relative error is 0.54%, and the maximum relative error is 1.75%. The predicted values ​​maintain high accuracy during both the morning load increase period and the afternoon peak load period.

[0133] The above results verify the applicability and accuracy of the prediction model of this invention in office building air conditioning cold source systems.

[0134] Example 3: Application verification in a factory

[0135] This embodiment verifies the application of the method of the present invention in the air conditioning system of an electronics factory. The factory is located in a hot-summer, cold-winter region, and the air conditioning system is located in a refrigeration room. The air conditioning system is equipped with two 2813kW centrifugal chillers and one 1044kW screw chiller, and the cooling tower fans are controlled by frequency converters.

[0136] Step 1: Data Acquisition. The factory's June operating data was selected as the validation dataset. The collected parameters included: the outdoor wet-bulb temperature (t) of the cooling tower. s Cooling tower inlet water temperature t in The cooling water flow rate Q, the actual operating power N of the cooling tower fan, and the actual outlet water temperature of the cooling tower.

[0137] Data was collected every 15 minutes, resulting in approximately 2800 data points. Data from June 1st to 16th was used as the training set to fit regression coefficients; data from June 17th to 30th was used as the test set to validate the model's predictive performance.

[0138] Step 2: Model Fitting. The training set data is cleaned, and regression coefficients are obtained using multiple linear regression. The established prediction model is as follows:

[0139]

[0140] Step 3: Verification of Prediction Results. Substitute the test set data into the above model to calculate and predict the cooling tower outlet water temperature, and compare this predicted temperature with the actual operating cooling tower outlet water temperature. The coefficient of determination R between the two sets of data is calculated. 2 The value of 0.888 indicates that the model has a good fit.

[0141] Step 4: Typical day comparison analysis. Figure 5 The graph shows a comparison between the model's predicted and actual measured values ​​of the cooling tower outlet water temperature on June 20th. The graph indicates that the calculated and measured values ​​of the cooling tower outlet water temperature generally follow the same trend, demonstrating that the model accurately captures the variation pattern of the outlet water temperature. The average relative error is 0.58%, and the maximum relative error is 0.93%, validating the effectiveness of introducing N. 1 / 3 This item can accurately reflect the impact of changes in fan power on the outlet water temperature.

[0142] The above results verify the robustness and accuracy of the prediction model of this invention in industrial air conditioning systems.

[0143] Based on the above three application cases, the validation results of the prediction model of this invention in different types of buildings are summarized as follows:

[0144]

[0145] The verification results above show that:

[0146] High prediction accuracy: the coefficient of determination R for three cases 2 All values ​​reached above 0.88, the average relative error was within 0.6%, and the maximum relative error did not exceed 1.75%, indicating that the prediction model of this invention has high prediction accuracy.

[0147] Wide applicability: The method of this invention has achieved good prediction results in air conditioning cold source systems of different types of buildings such as hotels, office buildings, and factories, verifying its wide applicability.

[0148] Operating condition adaptability: Under different wind turbine operating modes such as variable frequency regulation and dual-speed switching, the model maintains high prediction accuracy, verifying the effectiveness of introducing N. 1 / 3 This item can accurately reflect the impact of changes in fan power on the outlet water temperature.

[0149] Engineering practicality: The model only needs to collect conventional operating parameters such as outdoor wet-bulb temperature, inlet water temperature, cooling water flow rate and fan power, without the need to install air volume sensors, and has significant engineering practical value.

[0150] The prediction method of this invention is also applicable to other types of cooling towers, including but not limited to: by ventilation method: natural ventilation cooling towers, mechanical ventilation cooling towers; by the contact method between hot water and air: wet cooling towers, dry cooling towers, dry-wet cooling towers; by the flow direction of hot water and air: counter-flow cooling towers, cross-flow cooling towers, mixed-flow cooling towers; by application: air conditioning cooling towers, industrial cooling towers.

[0151] Example 4: Application Effect of Optimized Control Method

[0152] The embodiment applies the optimized control method of the present invention to a real-world air conditioning cooling source system and evaluates the application effect.

[0153] The cooling system is equipped with three centrifugal chillers, three cooling water pumps, and three cooling towers. Before applying the method of this invention, the cooling water pumps and cooling tower fans operated at fixed frequencies, with the number of units in operation determined based on experience. After applying the method of this invention, the cooling water pumps and cooling tower fans are frequency-controlled based on the predicted cooling tower outlet water temperature, while the operating parameters of the chillers are optimized.

[0154] After a month of comparative operation, the optimized control method of this invention resulted in: a reduction of approximately 15% in cooling water pump energy consumption; a reduction of approximately 22% in cooling tower fan energy consumption; an improvement of approximately 6% in chiller unit energy efficiency; and an improvement of approximately 7% in the overall system energy efficiency ratio. This verifies the effectiveness and energy-saving effect of the optimized control method of this invention.

[0155] The optimized control method of the present invention is also applicable to other types of air conditioning cold source systems, including but not limited to: by chiller unit type: centrifugal chiller unit, screw chiller unit, scroll chiller unit, absorption chiller unit; by system form: primary pump system, secondary pump system, variable flow system, constant flow system.

[0156] The above description is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the present invention in any way. Any changes or modifications made by those skilled in the art based on the above disclosure shall fall within the protection scope of the claims.

Claims

1. A method for predicting the outlet water temperature of a cooling tower, characterized in that, Includes the following steps: S1: Obtain the outdoor wet-bulb temperature t of the cooling tower. s Inlet water temperature t in Cooling water flow rate Q and actual operating power N of cooling tower fan; S2: The outdoor wet-bulb temperature t s Inlet water temperature t in The cooling water flow rate Q and the actual operating power N of the cooling tower fan are input into a pre-established regression prediction model to obtain the predicted cooling tower outlet water temperature t. out ; The expression for the regression prediction model is: ； In the formula, a, b, c, d, and e are regression coefficients.

2. The method for predicting the outlet water temperature of a cooling tower as described in claim 1, characterized in that, The regression coefficients a, b, c, d, and e are obtained by fitting historical operating data using a multiple linear regression method.

3. The method for predicting the outlet water temperature of a cooling tower as described in claim 2, characterized in that, Before fitting the regression coefficients using the multiple linear regression method, a data cleaning step is also included for the historical operating data.

4. The method for predicting the outlet water temperature of a cooling tower as described in claim 3, characterized in that, The data cleaning includes removing abnormal data caused by human intervention and / or overlapping periods of equipment switching.

5. The method for predicting the outlet water temperature of a cooling tower as described in claim 1, characterized in that, The actual operating power N of the cooling tower fan is obtained through one of the following methods: Directly measure the input power of the cooling tower fan; Read the power value from the frequency converter of the cooling tower fan; It is calculated based on the current and voltage of the cooling tower fan.

6. A system for predicting the outlet water temperature of a cooling tower, characterized in that, include: The data acquisition module is used to obtain the outdoor wet-bulb temperature, inlet water temperature, cooling water flow rate, and actual operating power of the cooling tower fan in real time. The regression calculation module is equipped with a regression prediction model of the cooling tower outlet water temperature prediction method as described in any one of claims 1-5, which is used to calculate the predicted cooling tower outlet water temperature based on the data acquired by the data acquisition module. The output module is used to output the prediction results calculated by the regression calculation module.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the method for predicting the outlet water temperature of the cooling tower as described in any one of claims 1-5.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for predicting the outlet water temperature of the cooling tower as described in any one of claims 1-5.

9. An optimized control method for an air conditioning cooling source system, characterized in that, include: The cooling tower outlet water temperature is predicted using the method for predicting the cooling tower outlet water temperature as described in any one of claims 1-5. Based on the predicted outlet water temperature, adjust one or more of the following: the operating frequency of the cooling water pump, the operating frequency of the cooling tower fan, or the operating parameters of the chiller unit.

10. The optimized control method for an air conditioning cold source system as described in claim 9, characterized in that, The operating frequency of the cooling water pump, the operating frequency of the cooling tower fan, or the operating parameters of the chiller unit are adjusted to optimize the total operating energy consumption of the air conditioning cold source system.

Images