Energy-saving intelligent control method and device for cooling tower, electronic device and medium

Abstract

The application provides a cooling tower energy-saving intelligent control method, device, electronic device and medium, the method comprises: obtaining the target cooling parameter and real-time operation parameter of the cooling tower; inputting the real-time operation parameter of the cooling tower into a target algorithm model to obtain the current cooling parameter of the cooling tower; adjusting the real-time operation parameter to the target operation parameter according to the current cooling parameter and the target cooling parameter, and the target operation parameter is the operation parameter that makes the current cooling parameter meet the target cooling parameter and is the most energy-saving operation parameter of the cooling tower. The application can predict the cooling tower outlet water temperature based on the real-time operation parameter of the cooling tower, so that the cooling tower can avoid excessively low cooling tower outlet water temperature while meeting the cooling demand, which is beneficial to the evaporation of the heat of the cooling tower, thereby reducing energy consumption, improving energy efficiency, being conducive to energy saving and environmental protection, and meeting the concept of carbon neutralization and carbon peak.

Description

Technical Field

[0001] This application relates to the field of cooling tower technology, and in particular to a cooling tower energy-saving intelligent control method, device, electronic device and medium. Background Technology

[0002] Cooling towers, as energy conversion devices, are widely used in industrial production and daily life to provide cooling water to equipment. Cooling towers typically utilize the contact between water and air to dissipate heat generated by the equipment through evaporation, thereby lowering the water temperature through latent heat exchange. When the air in contact with water vapor becomes saturated, water molecules cannot evaporate further, reaching a dynamic equilibrium. At this point, the number of water molecules that evaporate equals the number of water molecules returning from the air to the water, and the water temperature remains constant. Ideally, the lowest outlet water temperature of a cooling tower equals the wet-bulb temperature of the outdoor air. Therefore, wet-bulb temperature is a crucial factor in controlling the outlet water temperature of a cooling tower. However, currently, there is a lack of methods to control the outlet water temperature of cooling towers based on ambient wet-bulb temperature, which is detrimental to heat evaporation and can easily lead to energy waste. Summary of the Invention

[0003] In view of this, it is necessary to provide a cooling tower energy-saving intelligent control method, device, electronic device and medium to solve the above-mentioned technical problem that the lack of means to control the cooling tower outlet water temperature through ambient wet-bulb temperature is not conducive to the heat evaporation of the cooling tower and leads to energy waste.

[0004] This application provides a method for intelligent energy-saving control of cooling towers, including:

[0005] Obtain the target cooling parameters and real-time operating parameters of the cooling tower;

[0006] The real-time operating parameters of the cooling tower are input into the target algorithm model to obtain the current cooling parameters of the cooling tower;

[0007] The real-time operating parameters of the cooling tower are adjusted to the target operating parameters based on the current cooling parameters and the target cooling parameters. The target operating parameters are the operating parameters of the cooling tower that save the most electricity among all the operating parameters that make the current cooling parameters meet the target cooling parameters.

[0008] Optionally, the target cooling parameters include the target outlet water temperature of the cooling tower, which is the sum of the ambient wet-bulb temperature and the preset temperature; the current cooling parameters include the predicted outlet water temperature of the cooling tower; and both the real-time operating parameters and the target operating parameters of the cooling tower include the fan operating status parameters of the cooling tower.

[0009] Optionally, if the fan is a fixed-frequency fan, the fan operating status parameters include the number of fixed-frequency fans turned on, and adjusting the real-time operating parameters of the cooling tower according to the current cooling parameters and the target cooling parameters includes:

[0010] The absolute value of the temperature difference is determined based on the predicted outlet water temperature and the target outlet water temperature, and the absolute value of the temperature difference is compared with a preset difference.

[0011] When the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, reduce the number of fixed-frequency fans that are turned on; or

[0012] When the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, the number of fixed-frequency fans turned on is increased.

[0013] Optionally, if the fan is a combination of a variable frequency fan and a fixed frequency fan, the operating status parameters of the fan include the number of fixed frequency fans turned on and the operating frequency of the variable frequency fan. Adjusting the real-time operating parameters of the cooling tower according to the current cooling parameters and the target cooling parameters includes:

[0014] The absolute value of the temperature difference is determined based on the predicted outlet water temperature and the target outlet water temperature, and the absolute value of the temperature difference is compared with a preset difference.

[0015] When the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, reduce the number of fixed-frequency fans and variable-frequency fans that are turned on and / or the operating frequency of the variable-frequency fans; or

[0016] When the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, the number of fixed-frequency fans and variable-frequency fans turned on and / or the operating frequency of the variable-frequency fans are increased.

[0017] Optionally, adjusting the real-time operating parameters of the cooling tower based on its current cooling parameters and the target cooling parameters further includes:

[0018] If multiple adjustable fan operating state parameters are obtained based on the predicted outlet water temperature and the target outlet water temperature, the total power consumption of the fan under each fan operating state parameter is determined.

[0019] Determine the operating status parameters of the fan with the lowest total power consumption, and adjust the number of fixed-frequency fans and variable-frequency fans turned on and / or the operating frequency of the variable-frequency fans based on the operating status parameters of the fan with the lowest total power consumption.

[0020] Optionally, the method further includes:

[0021] Acquire historical data of the cooling tower, including historical target cooling parameters and various historical operating parameters of the cooling tower;

[0022] The historical data is preprocessed to obtain a preset number of feature data;

[0023] The target algorithm model is obtained by training the algorithm model using the feature data.

[0024] Optionally, obtaining the historical data of the cooling tower includes:

[0025] The first server collects the underlying device data of the cooling tower by reading the gateway, and archives the underlying device data to the second server;

[0026] The third server is used to collect data from the second server on the underlying device data in an incremental manner, and the collected data is saved to the original data layer.

[0027] The data in the original data layer is preprocessed, and the preprocessed data is saved to the data warehouse;

[0028] The data in the data warehouse is row-to-column transformed to obtain a wide data table, which is then saved to a data mart as historical data to be preprocessed.

[0029] Optionally, the preprocessing of the historical data to obtain a preset number of feature data includes:

[0030] The historical data is copied to the local database corresponding to the device to which the target algorithm model belongs using data warehouse technology;

[0031] The historical data in the local database is integrated using Structured Query Language;

[0032] The integrated historical data is preprocessed to obtain a feature database, which includes multiple feature data.

[0033] Select the preset number of feature data from multiple feature data in the feature database.

[0034] Optionally, the preprocessing of the historical data to obtain a preset number of feature data further includes:

[0035] The feature database contains multiple feature data points, which are then input into an automatic machine learning algorithm to obtain a weighted ranking of the multiple feature data points.

[0036] The preset number of feature data are selected from the multiple feature data according to the weight sorting.

[0037] Optionally, the preprocessing of the historical data of the cooling tower to obtain a preset number of feature data further includes:

[0038] The feature database contains multiple feature data entries, which are input into the chi-square verification algorithm to output the preset number of feature data entries.

[0039] Optionally, the preprocessing of the historical data of the cooling tower to obtain a preset number of feature data further includes:

[0040] Data derivation is performed on a portion of the feature data from the preset number of feature data to obtain derived feature data that replaces the portion of feature data. The derived feature data and another portion of the feature data are used to train the algorithm model to obtain the target algorithm model.

[0041] Optionally, the cooling tower includes at least one fixed-frequency fan, and the portion of the characteristic data includes at least one characteristic data point: fixed-frequency fan operating status parameters obtained by preprocessing the fan operating status parameters of the at least one fixed-frequency fan.

[0042] The at least one feature data of the fixed-frequency fan operating status parameter is derived into a feature data of the number of fixed-frequency fans turned on, which is the derived feature data.

[0043] Optionally, the algorithm model is a regression algorithm model, which is any one of the support vector machine algorithm, random forest algorithm, gradient boosting decision tree algorithm and ridge regression algorithm.

[0044] Optionally, training the algorithm model using the feature data to obtain the target algorithm model includes:

[0045] The support vector machine algorithm, random forest algorithm, gradient boosting decision tree algorithm and ridge regression algorithm are trained using the feature data to obtain multiple candidate models;

[0046] Calculate the coefficient of determination, mean squared error, and mean absolute error for each candidate model based on the predicted and actual values ​​of each candidate model.

[0047] The target algorithm model is determined based on the determination coefficients, mean squared errors, and mean absolute errors of the multiple candidate models.

[0048] This application also provides a cooling tower energy-saving intelligent control device for intelligent control of the cooling tower. The cooling tower includes a water tower, a thermometer, a data collector, and multiple fans. The thermometer is located in the external environment where the water tower is situated and is used to sense the ambient wet-bulb temperature. The data collector is electrically connected to at least the multiple fans and is used to collect the operating status parameters of the multiple fans. The device includes:

[0049] A first processor, communicatively connected to the thermometer and the data acquisition unit, is used to determine the target cooling parameters of the cooling tower from the thermometer and the real-time operating parameters of the cooling tower from the data acquisition unit. The target cooling parameters include the sum of the ambient wet-bulb temperature and the preset temperature, and the real-time operating parameters include at least the operating status parameters of multiple fans of the cooling tower.

[0050] The first processor is further configured to input the real-time operating parameters of the cooling tower into the target algorithm model to obtain the current cooling parameters of the cooling tower; and to adjust the real-time operating parameters of the cooling tower to the target operating parameters based on the current cooling parameters and the target cooling parameters, wherein the target operating parameters are the most energy-efficient operating parameters of the cooling tower among all operating parameters that make the current cooling parameters meet the target cooling parameters.

[0051] This application also provides an electronic device, including:

[0052] Second processor; and

[0053] The second memory stores instructions that are loaded and executed by the second processor using the aforementioned cooling tower energy-saving intelligent control method.

[0054] This application also provides a computer-readable storage medium having at least one computer instruction stored thereon, the instruction being loaded and executed by a second processor as described above for the cooling tower energy-saving intelligent control method.

[0055] The energy-saving intelligent control method, device, electronic device and medium provided in this application can predict the cooling tower outlet water temperature based on the real-time operating parameters of the cooling tower, so that the cooling tower can meet the cooling demand while avoiding excessively low cooling tower outlet water temperature, which is conducive to the evaporation of heat from the cooling tower, thereby reducing energy consumption, improving energy efficiency, and promoting energy conservation and environmental protection, which is in line with the concept of carbon neutrality and carbon peaking. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0057] Figure 1 This is a schematic diagram of the application environment of the cooling tower provided in the preferred embodiment of this application.

[0058] Figure 2This is a schematic diagram of the structure of a cooling tower provided in one embodiment of this application.

[0059] Figure 3 This is a flowchart of the cooling tower energy-saving intelligent control method provided in a preferred embodiment of this application.

[0060] Figure 4 This is a processing framework diagram of a cooling tower energy-saving control method provided in an embodiment of this application.

[0061] Figure 5 This is a flowchart provided in one embodiment of the present application for adjusting the real-time operating parameters of the cooling tower to the target operating parameters based on the current cooling parameters and the target cooling parameters of the cooling tower.

[0062] Figure 6 This is a schematic diagram of the structure of a cooling tower provided in another embodiment of this application.

[0063] Figure 7 This is a flowchart provided in another embodiment of the present application for adjusting the real-time operating parameters of the cooling tower to the target operating parameters based on the current cooling parameters and the target cooling parameters of the cooling tower.

[0064] Figure 8 This is a schematic diagram of the structure of a cooling tower provided in another embodiment of this application.

[0065] Figure 9 This is a flowchart provided in one embodiment of the present application for adjusting the real-time operating parameters of the cooling tower to the target operating parameters based on the current cooling parameters and the target cooling parameters of the cooling tower.

[0066] Figure 10 This is a fitted curve diagram showing the relationship between the operating power of a wind turbine and the frequency of a wind turbine, provided in an embodiment of this application.

[0067] Figure 11 This is a flowchart of the method for establishing the target algorithm model provided in a preferred embodiment of this application.

[0068] Figure 12 This is a flowchart of obtaining historical data of a cooling tower provided in one embodiment of this application.

[0069] Figure 13 This is an architecture diagram for acquiring historical data of a cooling tower provided in one embodiment of this application.

[0070] Figure 14 This is a flowchart illustrating the process of preprocessing historical data to obtain a preset number of feature data, as provided in an embodiment of this application.

[0071] Figure 15A This is a graph showing the prediction results of a random forest model provided in an embodiment of this application.

[0072] Figure 15BThis is a graph showing the prediction results of a ridge regression model provided in an embodiment of this application.

[0073] Figure 15C This is a prediction result diagram of the gradient boosting decision tree algorithm model provided in an embodiment of this application.

[0074] Figure 15D This is a prediction result diagram of a support vector machine model provided in an embodiment of this application.

[0075] Figure 16 This is a schematic diagram of the evaluation metrics of an alternative model provided in an embodiment of this application.

[0076] Figure 17 This is a schematic diagram of the structure of an intelligent control device provided in an embodiment of this application.

[0077] Figure 18 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0078] To better understand the above-mentioned objectives, features, and advantages of this application, the application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0079] Numerous specific details are set forth in the following description to provide a thorough understanding of this application. The described embodiments are merely some, not all, of the embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0080] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0081] Please see Figure 1 The diagram shown is an application environment schematic of a cooling tower provided in a preferred embodiment of this application. The cooling tower 10 includes, but is not limited to, at least two water towers 11. Figure 1Only one water tower 11 is shown (the diagram of at least two water towers 11 will be described in detail below), along with a water supply pipe 12, a return water pipe 13, a water pump 141, a water pump 142, a thermometer 15, and a collector 16. Each water tower 11 is equipped with at least one fan 110, which will be described in detail below with reference to the accompanying drawings. The water pump 141 is located on the water supply pipe 12, the water pump 142 is located on the return water pipe 13, the thermometer 15 is located in the external environment where the water tower 11 is located, and the collector 16 is located on the water tower 11. The water supply pipe 12 and the return water pipe 13 are connected to the cooling device 2. The water pump 141 drives the cooling water output from the water tower 11 to be supplied to the cooling device 2 via the water supply pipe 12. Since the temperature of the cooling water is lower than that of the chilled water in the cooling device 2, the cooling water can absorb the heat from the chilled water in the cooling device 2, and then the water is returned to the water tower 11 by the water pump 142 for cooling, and then sent back to the cooling device 2 for recycling. Optionally, the cooling device 2 can be an ice machine, etc.

[0082] Please see Figure 2 The diagram shows a schematic of a cooling tower according to an embodiment of this application. At least two water towers 11 of the cooling tower 10 are arranged sequentially with a shared water supply pipe 12 and return pipe 13. The cooling tower 10 is connected to a cooling device 2. Fans 110 cool the water in the water towers 11. A water pump 141 transports the cooling water to the cooling device 2. After the cooling water absorbs heat from the chilled water in the cooling device 2, the water pump 142 returns the water to the water towers 11 for further cooling, forming a cycle. A thermometer 15 senses the ambient wet-bulb temperature, which is the lowest temperature achievable in the current environment solely through water evaporation. A data acquisition device 16 collects operating status parameters from multiple fans and can be a commercially available data acquisition device.

[0083] Please see Figure 3 The diagram shown is a flowchart of a cooling tower energy-saving control method provided in a preferred embodiment of this application. The order of steps in the flowchart can be changed, and some steps can be omitted, depending on different requirements.

[0084] Before explaining these steps, let's first introduce the background of cooling tower energy-saving control methods.

[0085] To save energy, the target outlet water temperature of a cooling tower is typically controlled as the sum of the ambient wet-bulb temperature and a preset temperature. The number and frequency of the cooling tower's fans operating at this temperature are examples of the "real-time operating parameters" explained below. However, the ambient wet-bulb temperature can change, and when it does, the target outlet water temperature of the cooling tower also changes. In this case, the target outlet water temperature before the change is referred to as an example of the "current cooling parameters" explained below, and the target outlet water temperature after the change is referred to as an example of the "target cooling parameters" explained below. To save energy, the "real-time operating parameters" corresponding to the "current cooling parameters" (e.g., the current number and frequency of the cooling tower's fans) need to be adjusted to meet the "target operating parameters" corresponding to the "target cooling parameters" (e.g., the target number and frequency of the cooling tower's fans, which both meet the requirements of the target cooling parameters and minimize the overall power consumption of the cooling tower). Therefore, how to adjust these parameters stably and quickly is a problem that urgently needs to be solved.

[0086] To address this, this application uses an artificial intelligence (AI) model (as an example of the "target algorithm model" described below) to receive real-time operating parameter predictions to derive the current cooling parameters. Then, based on the required target cooling parameters and the energy consumption fitting curve described below, the target operating parameters corresponding to the target cooling parameters are derived. Specifically, this will be explained in detail through the following steps S101-S103.

[0087] S101, obtain the target cooling parameters and real-time operating parameters of the cooling tower.

[0088] Please see Figure 4 The diagram shown is a processing framework diagram of a cooling tower energy-saving control method provided in an embodiment of this application. In the "Real-time Data Prediction" box, real-time operating parameters are obtained as input data for real-time data prediction. These real-time operating parameters include... Figure 4 The "Real-time Data Prediction" box lists the outside air temperature, wet-bulb temperature, fan frequency, etc. These data affect the current cooling parameters of the cooling tower. That is, if these data change, the current cooling parameters of the cooling tower will also change. Conversely, if these data do not change, the current cooling parameters of the cooling tower will not change. The target cooling parameters are obtained as the goal for energy-saving control of the cooling tower. That is, the current cooling parameters of the cooling tower are adjusted to meet the requirements of the target cooling parameters by adjusting the real-time operating parameters.

[0089] In one embodiment, the target cooling parameter is the target outlet water temperature of the cooling tower. Obtaining the target cooling parameter includes: acquiring the ambient wet-bulb temperature sensed by thermometer 15, and determining the target outlet water temperature as the sum of the ambient wet-bulb temperature and a preset temperature. Optionally, the preset temperature is 3°C. The preset temperature is selected from a preset temperature range, which is a performance indicator of the cooling tower 11. Typically, the preset temperature range of a cooling tower 11 is determined at the factory, and the conventional preset temperature range is ±5°C. At this point, the purpose of energy-saving control of the cooling tower is to ensure that the actual outlet water temperature equals the target outlet water temperature or that the temperature fluctuation range of the actual outlet water temperature is within the upper and lower fluctuation range of the target outlet water temperature (this upper and lower fluctuation range is set according to actual needs, for example, target outlet water temperature ±0.5°C; further, if the target outlet water temperature is 25°C, then an actual outlet water temperature equal to 25°C or within a temperature fluctuation range of 25±0.5°C is considered to meet the requirements), while simultaneously ensuring the cooling tower is most energy-efficient.

[0090] It should be noted that the target outlet water temperature is within the design outlet water temperature range of cooling tower 10. Optionally, the design outlet water temperature range is 16-32℃, that is, greater than or equal to 16℃ and less than or equal to 32℃. If the target outlet water temperature is lower than the lower limit of the design outlet water temperature range, the target outlet water temperature is the lower limit of the design outlet water temperature range, for example, 16℃. If the target outlet water temperature is higher than the upper limit of the design outlet water temperature range, the target outlet water temperature is the upper limit of the design outlet water temperature range, for example, 32℃.

[0091] In one embodiment, the target outlet water temperature of each water tower 11 can be set separately. Specifically, according to a first order, the target outlet water temperature of the later water tower 11 is set to be greater than the target outlet water temperature of the earlier water tower 11. For example, according to the distance from the cooling device 2 among all the water towers 11 from closest to farthest, all the water towers are defined in chronological order as the first water tower (i.e., the first water tower), the second water tower, ..., the last water tower. In this case, the first order is... Figure 2 The sequence shown from right to left indicates that the target outlet water temperature of the first water tower 11 is the ambient wet-bulb temperature + 3°C, and the target outlet water temperature of the second water tower 11 is the ambient wet-bulb temperature + 3.5°C. In this article, "first sequence" refers to the order from the first water tower to the last water tower. For example, the first sequence is... Figure 2 The order shown is from right to left; in other words, the first order is the order from closest to furthest from the cooling device 2, with the earlier water tower being closer to the cooling device 2 than the later water tower.

[0092] It should be noted that since the outlet water temperature requires a certain delay to stabilize after the first water tower is turned on, the target outlet water temperature of the second water tower is set to the ambient wet-bulb temperature + 3.5℃, slightly higher than the target outlet water temperature of the first water tower. This avoids setting the target temperatures of both water towers to the same level, which would cause frequent starting and stopping of the water towers and thus cause oscillation within the cooling tower system. Avoiding oscillation within the cooling tower system achieves energy savings, as the energy consumption during oscillation is greater than during normal operation or standby. Furthermore, because the target temperature of the later water tower is set higher, the requirements for the number or frequency of fans are lower, further achieving energy savings.

[0093] In other embodiments, the target outlet water temperature for each water tower may also be set to be the same.

[0094] In one embodiment, real-time operating parameters include, but are not limited to, ambient wet-bulb temperature, ambient temperature, cooling tower outlet water temperature, cooling tower return water temperature, and fan operating status parameters. Fan operating status parameters include the operating frequency of the variable frequency fan and / or the number of fixed frequency fans in operation.

[0095] S102, input the real-time operating parameters of the cooling tower into the target algorithm model to obtain the current cooling parameters of the cooling tower.

[0096] In one embodiment, the target cooling parameters of the cooling tower are determined from a thermometer, and the real-time operating parameters of the cooling tower are determined from a data acquisition device. The target cooling parameters include the sum of the ambient wet-bulb temperature and the preset temperature, and the real-time operating parameters include at least the operating status parameters of multiple fans of the cooling tower.

[0097] like Figure 4 As shown, in one embodiment, real-time operating parameters are input into the target algorithm model, and the target algorithm model outputs the current cooling parameters. The current cooling parameters are the predicted outlet water temperature of the cooling tower.

[0098] In one embodiment, the target algorithm model is a regression algorithm model, which can be any one of the support vector machine algorithm, random forest algorithm, gradient boosting decision tree algorithm and ridge regression algorithm.

[0099] In one embodiment, the real-time operating parameters also include ambient wet-bulb temperature, ambient temperature, cooling tower outlet water temperature, and cooling tower return water temperature. Inputting the real-time operating parameters of the cooling tower into the target algorithm model to obtain the current cooling parameters of the cooling tower includes: inputting the ambient wet-bulb temperature, ambient temperature, cooling tower outlet water temperature, cooling tower return water temperature, and fan operating status parameters into the regression algorithm model, and outputting the predicted outlet water temperature of the cooling tower through the regression algorithm model.

[0100] S103, adjust the real-time operating parameters of the cooling tower to the target operating parameters based on the current cooling parameters and the target cooling parameters of the cooling tower.

[0101] In one embodiment, the target operating parameter is the operating parameter that saves the cooling tower's power consumption among all operating parameters that make the current cooling parameters meet the target cooling parameters.

[0102] like Figure 4 As shown, in one embodiment, the real-time operating parameters of the cooling tower are adjusted to the target operating parameters based on the predicted and target outlet water temperatures. The adjusted real-time operating parameters can be the current fan operating status parameters of the cooling tower, and the target operating parameters can be the target fan operating status parameters of the cooling tower. That is, the target operating parameters are the fan operating status parameters that are most energy-efficient among all fan operating status parameters that make the current fan operating status parameters meet the target fan operating status parameters.

[0103] like Figure 2 As shown, in the first embodiment of the cooling tower, at least two water towers have identical structures, and the fans installed on the water towers are all fixed-frequency fans. The fan operating status parameters include the number of fixed-frequency fans that are turned on. Please refer to [link / reference]. Figure 5 As shown, adjusting the real-time operating parameters of the cooling tower to the target operating parameters based on the current and target cooling parameters includes:

[0104] S201, determine the absolute value of the temperature difference based on the predicted outlet water temperature and the target outlet water temperature, and compare the absolute value of the temperature difference with the preset difference.

[0105] In one embodiment, determining the absolute value of the temperature difference based on the predicted outlet water temperature and the target outlet water temperature includes: calculating the temperature difference obtained by subtracting the target outlet water temperature from the predicted outlet water temperature, and determining the absolute value of the temperature difference.

[0106] S202, when the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, reduce the number of fixed frequency fans that are turned on.

[0107] In one embodiment, if the temperature difference is less than 0 and the absolute value of the temperature difference is greater than a preset difference, at least two fans 110 of the water tower 11 are shut down as needed in a second order opposite to the first order, until the absolute value of the temperature difference is less than or equal to the preset difference.

[0108] Specifically, first, the fans 110 of the left water tower 11 are turned off one by one. After each fan 110 is turned off, it is determined whether the absolute value of the temperature difference is less than or equal to a preset difference. If the absolute value of the temperature difference is less than or equal to the preset difference, the fan 110 is turned off. If all the fans 110 of the left water tower 11 have been turned off and the absolute value of the temperature difference is still greater than the preset difference, then the fans 110 of the right water tower 11 are turned off one by one until the absolute value of the temperature difference is less than or equal to the preset difference.

[0109] It should be noted that if the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, it means that the predicted outlet water temperature is less than the target outlet water temperature and the difference between the predicted outlet water temperature and the target outlet water temperature is large, indicating that the predicted outlet water temperature is too low. Therefore, the cooling effect on the water in the water tower is reduced by reducing the number of fixed frequency fans turned on, so as to save energy.

[0110] S203, when the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, increase the number of fixed frequency fans that are turned on.

[0111] In one embodiment, if the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than a preset difference, at least two fans 110 of the water tower 11 are turned on as needed in a first order until the absolute value of the temperature difference is less than or equal to the preset difference.

[0112] Specifically, first, the fans 110 of the right water tower 11 are turned on one by one. After each fan 110 is turned on, it is determined whether the absolute value of the temperature difference is less than or equal to a preset difference. If the absolute value of the temperature difference is less than or equal to the preset difference, the fan 110 is turned off. If all the fans 110 of the right water tower 11 are turned on and the absolute value of the temperature difference is still greater than the preset difference, then the fans 110 of the left water tower 11 are turned on one by one until the absolute value of the temperature difference is less than or equal to the preset difference.

[0113] It should be noted that if the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, it means that the predicted outlet water temperature is greater than the target outlet water temperature, and the difference between the predicted outlet water temperature and the target outlet water temperature is large, which cannot meet the cooling requirements of cooling device 2. Therefore, the water in the water tower is cooled by increasing the number of fixed frequency fans to reduce the outlet water temperature and meet the cooling requirements of cooling device 2.

[0114] Please see Figure 6 As shown, in the second embodiment of the cooling tower, at least two water towers have the same structure, and the fans installed on at least two water towers are a combination of variable frequency fans and fixed frequency fans. At least two water towers 11 are arranged in sequence in a manner that shares a water supply pipe 12 and a return water pipe 13.

[0115] In the second embodiment of the cooling tower, the fan operating status parameters include the number of fixed-frequency fans in operation and the operating frequency of the variable-frequency fans. Please refer to [link / reference]. Figure 7 As shown, adjusting the real-time operating parameters of the cooling tower to the target operating parameters based on the current and target cooling parameters includes:

[0116] S301, determine the absolute value of the temperature difference based on the predicted outlet water temperature and the target outlet water temperature, and compare the absolute value of the temperature difference with the preset difference.

[0117] S302, when the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, reduce the number of fixed frequency fans and variable frequency fans turned on and / or the operating frequency of variable frequency fans.

[0118] In one embodiment, if the temperature difference is less than 0 and the absolute value of the temperature difference is greater than a preset difference, the fixed-frequency fan that is turned on is turned off, and the frequency of the variable-frequency fan is adjusted until the absolute value of the temperature difference is less than or equal to the preset difference.

[0119] Specifically, if the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, the fixed-frequency fan that is turned on will be turned off first, and the frequency of the variable-frequency fan will be gradually increased or decreased. At the same time, it will be determined whether the absolute value of the temperature difference is less than or equal to the preset difference. If the absolute value of the temperature difference is less than or equal to the preset difference, the frequency of the variable-frequency fan will be stopped from being increased or decreased, or the fixed-frequency fan that is turned on will be stopped from being turned off.

[0120] S303, when the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, increase the number of fixed frequency fans and variable frequency fans turned on and / or the operating frequency of the variable frequency fans.

[0121] In one embodiment, if the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than a preset difference, the variable frequency fan is turned on, and the frequency of the variable frequency fan changes according to the difference between the outlet water temperature and the target temperature.

[0122] Specifically, if the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, the variable frequency fan is turned on and the frequency of the variable frequency fan is gradually increased. At the same time, it is determined whether the absolute value of the temperature difference is less than or equal to the preset difference. If the absolute value of the temperature difference is less than or equal to the preset difference, the frequency of the variable frequency fan is stopped from being increased. If the absolute value of the temperature difference is still greater than the preset difference, the frequency of the variable frequency fan continues to be increased until the frequency upper limit is reached.

[0123] If the variable frequency fan reaches its frequency limit, the fixed frequency fan will be activated. The frequency of the variable frequency fan will change in coordination with that of the fixed frequency fan according to the temperature difference, until the absolute value of the temperature difference is less than or equal to a preset value. One or more fixed frequency fans can be activated as needed.

[0124] Specifically, if the variable frequency fan reaches its frequency limit and the absolute value of the temperature difference is still greater than the preset difference, the fixed frequency fan is turned on and the frequency of the variable frequency fan is reduced until the absolute value of the temperature difference is less than or equal to the preset difference.

[0125] It should be noted that if the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, it means that the predicted outlet water temperature is greater than the target outlet water temperature, and the difference between the predicted outlet water temperature and the target outlet water temperature is large, which cannot meet the cooling requirements of cooling device 2. Therefore, the frequency of the variable frequency fan is increased first. If the frequency of the variable frequency fan reaches the upper limit and the predicted outlet water temperature is still too high, the fixed frequency fan is turned on, and the frequency of the variable frequency fan is reduced based on the temperature difference to avoid energy waste caused by excessive fan frequency.

[0126] Please see Figure 8 As shown, in the third embodiment of the cooling tower, at least two water towers 11 arranged sequentially in one cooling tower 10 and at least two water towers 11 arranged sequentially in another cooling tower 10 are connected in parallel. One cooling tower 10 and the other cooling tower 10 share a water supply main pipe 112 and are respectively connected to two water supply pipes 12 via the water supply main pipe 112. The two water supply pipes 12 are used to supply water to the two cooling devices 2 respectively. One end of each of the two return water pipes 13 is connected to a return water main pipe 113, and the other end of each of the two return water pipes 13 is respectively connected to the two cooling devices 2 so as to supply the return water of the two cooling devices 2 to one cooling tower 10 and the other cooling tower 10 respectively via the return water main pipe 113.

[0127] In the third embodiment of the cooling tower, at least two of the two cooling towers 10 have identical structures, and the fans 110 of the at least two cooling towers 11 are a combination of fixed-frequency fans and variable-frequency fans. The two cooling towers 10 control the operating status of their respective fans 110 based on the difference between the outlet water temperature of the water supply pipe 12 and the target temperature, with the control logic prioritizing the activation of the fans. Figure 8 The fan 110 in the upper middle cooling tower 10 prioritizes shutting down the fan 110 that is turned on. Other control logic is the same as described above and will not be repeated here.

[0128] like Figure 4 As shown, in the "Control" box, after determining the fan operating parameters based on the predicted and target outlet water temperatures, energy consumption is fitted to the fan. Please refer to [link / reference]. Figure 9 As shown, adjusting the real-time operating parameters of the cooling tower based on its current and target cooling parameters also includes:

[0129] S401, if multiple adjustable fan operating status parameters are obtained based on the predicted outlet water temperature and the target outlet water temperature, determine the total power consumption of the fan under each fan operating status parameter.

[0130] S402, determine the operating status parameters of the fan with the lowest total power consumption, and adjust the number of fixed-frequency fans and variable-frequency fans and / or the operating frequency of the variable-frequency fans based on the operating status parameters of the fan with the lowest total power consumption.

[0131] In one embodiment, the power consumption of a fixed-frequency fan is the product of its rated power and time. The power consumption of a variable-frequency fan is (X / f) max ) 3 The product of rated power and time. Where X is the current frequency of the variable frequency fan, and f... max This is the maximum frequency of the variable frequency fan. The total power consumption of cooling tower 10 is the sum of the power consumption of all fixed frequency fans and variable frequency fans.

[0132] In another embodiment, the power consumption of the fixed-frequency fan is the product of the fan's aging coefficient, rated power, and time. The power consumption of the variable-frequency fan is (X / f) max ) 3 The product of the fan aging coefficient, rated power, and time. Where X is the current frequency of the variable frequency fan, and f... max This represents the maximum frequency of the variable frequency fan. The fan aging factor is a pre-set percentage based on the fan's condition, which can include the fan's usage time and health status.

[0133] In other embodiments, the total power consumption of the cooling tower 10 can also be obtained through multiple power lines connected to the cooling tower 10. For example, the total power consumption of the cooling tower 10 can be obtained by connecting multiple power lines connected to the cooling tower 10 through an electricity meter (not shown).

[0134] Furthermore, if the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, the number of fixed-frequency fans and variable-frequency fans that are turned on is controlled, and the frequency of the variable-frequency fans is controlled, until the absolute value of the temperature difference is less than or equal to the preset difference, and the total power consumption of the cooling tower is minimized.

[0135] Specifically, if the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference, the analysis can consider multiple combinations of the number of fixed-frequency fans, the number of variable-frequency fans, and the frequency of variable-frequency fans that make the absolute value of the temperature difference less than or equal to the preset difference. The total power consumption of each combination is calculated, the combination with the minimum total power consumption is determined, and the number of fixed-frequency fans and the frequency of variable-frequency fans are controlled based on the data of the combination with the minimum total power consumption.

[0136] If the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, control the number of fixed-frequency fans and variable-frequency fans that are turned on, and control the frequency of the variable-frequency fans, until the absolute value of the temperature difference is less than or equal to the preset difference, and minimize the total power consumption of the cooling tower.

[0137] Specifically, if the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference, the analysis can consider multiple combinations of the number of fixed-frequency fans, the number of variable-frequency fans, and the frequency of variable-frequency fans that make the absolute value of the temperature difference less than or equal to the preset difference. The total power consumption of each combination is calculated, the combination with the minimum total power consumption is determined, and the number of fixed-frequency fans and the frequency of variable-frequency fans are controlled based on the data of the combination with the minimum total power consumption.

[0138] Please see Figure 10 As shown, the fit between fan operating power (unit: watts, W) and fan frequency (unit: hertz, Hz) is high. Furthermore, the power consumption of the cooling tower increases with increasing fan operating power. The frequency of fixed-frequency and variable-frequency fans can be represented by the number of fixed-frequency and variable-frequency fans that are turned on. For example, if the frequency of a fixed-frequency fan is 50Hz, it can be represented by 1 fixed-frequency fan being turned on. If the maximum operating frequency of a variable-frequency fan is 50Hz, it can be represented by 2 variable-frequency fans being turned on, with each variable-frequency fan operating at an actual frequency of 30Hz. Thus, determining the number of fixed-frequency fans turned on, the number of variable-frequency fans turned on, and the frequency of the variable-frequency fans that minimizes total power consumption corresponds to determining the sum of the frequencies of the minimum fixed-frequency and variable-frequency fans that ensure the absolute value of the temperature difference is less than or equal to a preset difference. Conversely, the number of fixed-frequency fans turned on, the number of variable-frequency fans turned on, and the frequency of the variable-frequency fans can be determined based on the sum of the frequencies of the minimum fixed-frequency and variable-frequency fans.

[0139] Through the above fitting process, the fan operating state parameters (number of fixed-frequency fans and variable-frequency fans turned on, frequency of variable-frequency fans) that minimize the total power consumption of the cooling tower, i.e., the fan doing the least work, are obtained. Then, the actual operating state of the fans is controlled back based on the fan operating state parameters that minimize the work done by the cooling tower.

[0140] Please see Figure 11 The diagram shown is a flowchart of a method for establishing a target algorithm model according to a preferred embodiment of this application. The order of steps in the flowchart can be changed, and some steps can be omitted, depending on different requirements.

[0141] S501, retrieve historical data of the cooling tower.

[0142] like Figure 4 As shown, in the "Historical Data Modeling" box, the first step is to collect data and obtain all historical data of the cooling tower.

[0143] In one embodiment, the historical data includes the cooling tower's historical target cooling parameters and various historical operating parameters.

[0144] In one embodiment, please refer to Figure 12 As shown, the historical data obtained from the cooling tower includes:

[0145] S5011, the first server is used to collect the underlying device data of the cooling tower 10 by reading the first gateway, and the underlying device data is archived to the second server.

[0146] Please see Figure 13 The diagram shown illustrates an architecture for acquiring historical data from a cooling tower according to an embodiment of this application. Optionally, the first server 30 is a SCADA (Supervisory Control And Data Acquisition) server, the second server 40 is an archiving server, the first gateway 31 is a core switch, and the underlying devices 32 include an aggregation switch 321, a monitoring system 322, a second gateway 323, an access switch 324, the cooling tower 10 itself, and instruments 325 of the cooling tower 10. The first server 30 collects underlying device data of the cooling tower 10 from the underlying devices 32 through the first gateway 31. The underlying device data of the cooling tower 10 includes historical target cooling parameters and historical operating parameters of the cooling tower 10.

[0147] In one embodiment, the monitoring system 322 runs configuration software (IFIX) and communicates with the access switch 324. Through the access switch 324, it obtains historical target cooling parameters and historical operating parameters of the cooling tower 10, such as historical target outlet water temperature and historical fan operating status parameters. The second gateway 323 obtains historical operating parameters from the instruments 325 of the cooling tower 10 via an RS485 interface, such as ambient wet-bulb temperature, ambient temperature, cooling tower outlet water temperature, and cooling tower return water temperature. The monitoring system 322 transmits the obtained historical operating parameters of the cooling tower to the aggregation switch 321 via the OPC (OLE for Process Control) protocol. The second gateway 323 also transmits the obtained historical operating parameters to the aggregation switch 321 via the Modbus TCP communication protocol. The first server 30 reads the historical operating parameters of the cooling tower obtained by the monitoring system 322 and the second gateway 323 through the first gateway 31 and archives the historical operating parameters of the cooling tower to the archive server.

[0148] S5012, the third server 50 is used to collect data from the second server 40 in the form of incremental extraction of the underlying device data, and the collected data is saved to the original data layer 511.

[0149] In one embodiment, the third server 50 is an AI server, which includes an AI database 51. The AI ​​database 51 includes a raw data layer 511, a data warehouse 512, and a data mart 513.

[0150] In one embodiment, incremental extraction refers to the collection of changed underlying device data from the second server 40 by the third server 50 at a preset frequency. The changes may include additions and / or modifications.

[0151] S5013, preprocess the data in the original data layer 511, and save the preprocessed data to the data warehouse 512.

[0152] In one embodiment, the preprocessing of the data in the original data layer 511 includes data cleaning, which includes filling in missing values ​​and data, replacing invalid data, deduplicating duplicate data, and normalizing the data, such as retaining an appropriate number of decimal places and converting code values.

[0153] S5014: Perform row-to-column transformation on the data in data warehouse 512 to obtain a wide data table, and save the wide data table to data mart 513 as historical data to be preprocessed. The historical data to be preprocessed is used for subsequent algorithm model training and prediction.

[0154] S502, preprocess historical data to obtain a preset number of feature data.

[0155] Please see Figure 14 As shown, in one embodiment, preprocessing historical data to obtain a preset amount of feature data includes:

[0156] S5021 uses data warehouse technology to copy historical data to the local database corresponding to the device to which the target algorithm model belongs.

[0157] It should be noted that copying historical data from the original underlying database to the local database reduces the storage pressure on the underlying database.

[0158] S5022 integrates historical data from the local database using Structured Query Language (SQL).

[0159] It should be noted that since all historical data comes from different tables and is collected at different frequencies, structured query language is used to integrate all data into a single table column based on the minimum time dimension. Optionally, the minimum time dimension can be minutes, seconds, hours, etc.

[0160] S5023, preprocess the integrated historical data to obtain a feature database, which includes multiple feature data.

[0161] In one embodiment, the preprocessing of the integrated historical data includes cleaning such as null value filling and outlier removal to obtain a feature database related to the cooling tower system, thereby obtaining multiple feature data, each of which is a type of historical operating parameter.

[0162] S5024: Select a preset number of feature data from multiple feature data in the feature database.

[0163] like Figure 4 As shown, after data collection, feature processing is performed on the collected historical data to establish a training set for the target algorithm model. Thus, in one embodiment, preprocessing historical data to obtain a preset number of feature data further includes: inputting multiple feature data items included in the feature database into an automatic machine learning algorithm to obtain a weighted ranking of the multiple feature data items, and selecting a preset number of feature data items from the multiple feature data items based on the weighted ranking.

[0164] In one embodiment, the automatic machine learning algorithm is H2O automl. The input data of the automatic machine learning algorithm consists of multiple historical operating parameters, and the output data consists of the weight of each historical operating parameter. The multiple historical operating parameters are input into the automatic machine learning algorithm, which outputs the weight of each historical operating parameter. The output weights are sorted in descending order. Based on the weight sorting, a preset number of historical operating parameters of different types are determined from the multiple historical operating parameters as feature data. Optionally, the preset number is 5, that is, the feature data may include 5 types of operating parameters: ambient wet-bulb temperature, ambient temperature, cooling tower outlet water temperature, cooling tower return water temperature, and fan operating status parameters.

[0165] In another embodiment, preprocessing the historical data of the cooling tower to obtain a preset number of feature data further includes: inputting multiple feature data entries from the feature database into a chi-square verification algorithm, and outputting the preset number of feature data entries. The chi-square calculation formula is as follows:

[0166]

[0167] In the above calculation formula, x 2 Here is the chi-square value, f0 is the actual frequency, and f e This represents the expected frequency. It should be noted that the chi-square value is used to characterize the correlation between the operating parameter and the target cooling parameter. The actual frequency is the number of times each operating parameter is actually correlated with the target cooling parameter in the historical data, while the expected frequency is the number of times each operating parameter is expected to be correlated with the target cooling parameter in the historical data.

[0168] Specifically, the process involves inputting multiple feature data points from the feature database into the chi-square verification algorithm and outputting a preset number of feature data points. This includes inputting multiple feature data points from the feature database into the chi-square verification algorithm, outputting the chi-square value of each feature data point through the chi-square verification algorithm, and if the chi-square value is less than a preset threshold, then it is determined that the historical operating parameters of the type corresponding to the chi-square value are related to the target cooling parameters, and this feature data point is one of the preset number of feature data points.

[0169] In one embodiment, preprocessing historical data of the cooling tower to obtain a preset number of feature data further includes: performing data derivation on a portion of the feature data from the preset number of feature data to obtain derived feature data that replaces a portion of the feature data.

[0170] In one embodiment, the cooling tower includes at least one fixed-frequency fan, and a portion of the feature data includes at least one feature data of fixed-frequency fan operating status parameters obtained by preprocessing the fan operating status parameters of at least one fixed-frequency fan, and the at least one feature data of fixed-frequency fan operating status parameters is derived into a feature data of the number of fixed-frequency fans turned on as derived feature data.

[0171] For example, part of the feature data is the operating frequency of multiple fixed-frequency fans. Since the operating frequency of fixed-frequency fans is fixed, the operating frequency of multiple fixed-frequency fans can be used to derive the number of fixed-frequency fans that are turned on.

[0172] For example, by acquiring ambient wet-bulb temperature, ambient temperature, cooling tower outlet water temperature, or cooling tower return water temperature at different times, temperature data can be derived.

[0173] S503 trains the algorithm model using feature data to obtain the target algorithm model.

[0174] like Figure 4 As shown, after feature processing of the collected historical data, algorithm selection is performed, and the optimal algorithm for the target algorithm model is determined based on the training set, and a trained target algorithm model is established.

[0175] In one embodiment, training the algorithm model with feature data to obtain the target algorithm model includes: training the algorithm model with derived feature data and another part of the feature data to obtain the target algorithm model.

[0176] In one embodiment, the feature data includes training data and test data. For example, each set of input and output data can be either a set of training data or a set of test data, where 80% of the data in the training dataset can be used as training data and 20% as test data.

[0177] In one embodiment, training the algorithm model using feature data to obtain the target algorithm model includes: training an initial data model using training data and testing the trained initial data model using test data to obtain the target algorithm model. Optionally, the algorithm model includes at least one of a support vector machine model, a random forest model, a gradient boosting decision tree algorithm model, and a ridge regression model.

[0178] In one embodiment, training the algorithm model using training data includes: inputting the input and output data of each set of training data into the algorithm model, adjusting the parameters of the algorithm model, until the algorithm model reaches the convergence condition.

[0179] In one embodiment, testing the trained algorithm model with test data to obtain the target algorithm model includes: inputting the test data into the trained algorithm model to obtain a predicted value; calculating the error of the predicted value based on the predicted value and the true value in the test data; and determining the trained algorithm model as the target algorithm model if the error of the predicted value obtained based on the trained algorithm model is less than or equal to a preset value.

[0180] In one embodiment, each set of test data includes input data and output data. The input data consists of feature data other than the actual outlet water temperature, and the output data is the actual outlet water temperature, i.e., the true value in the test data. The input data of each set of test data is input into the algorithm model to obtain a predicted value (predicted outlet water temperature). The prediction error is calculated based on the predicted value and the true value of each set of test data. The error k = |predicted value - true value| / |true value|. Optionally, the preset value is 3%. If the error is less than or equal to the preset value, the trained algorithm model is determined as the target algorithm model. If the error is greater than the preset value, supplementary feature data is added to continue training the algorithm model.

[0181] In the above embodiments, training data and test data are input into any one of the support vector machine model, random forest model, gradient boosting decision tree algorithm model, and ridge regression model to establish the target algorithm model. In another embodiment, training data and test data can be input into multiple of the support vector machine model, random forest model, gradient boosting decision tree algorithm model, and ridge regression model to establish multiple candidate models, and then the optimal model is determined from the multiple candidate models.

[0182] Please see Figures 15A-15D The figures show the prediction results of the Random Forest model, Ridge Regression model, Gradient Boosting Decision Tree algorithm model, and Support Vector Machine model, respectively. The horizontal axis (X-axis) represents the target value in the test data (i.e., the actual outlet water temperature of the cooling tower), and the vertical axis (Y-axis) represents the model prediction result obtained by each algorithm model (i.e., the predicted outlet water temperature of the cooling tower). Dashed line A is the curve Y = X, and dashed line B is the center curve of the scatter points (predicted outlet water temperature of the cooling tower). If the prediction result of the algorithm model is accurate, the two dashed lines will be infinitely close.

[0183] In another embodiment, the support vector machine algorithm, random forest algorithm, gradient boosting decision tree algorithm and ridge regression algorithm are trained using feature data to obtain multiple candidate models. The determination coefficient, mean squared error and mean absolute error of each candidate model are calculated based on the predicted value and the true value of each candidate model. The target algorithm model is determined based on the determination coefficient, mean squared error and mean absolute error of multiple candidate models.

[0184] In another embodiment, after establishing multiple candidate models, an evaluation index is calculated for each candidate model after training. The evaluation index includes at least one of the coefficient of determination, mean squared error, and mean absolute error. Based on the evaluation index, the optimal model among the candidate models after training is determined, and the target algorithm model is obtained.

[0185] In another embodiment, assume the predicted values ​​of the trained candidate model. The true value y = {y1, y2, ..., y n The predicted and true values ​​of the candidate model are generated by inputting test data from the training dataset into the candidate model, and / or by inputting new test data into the candidate model.

[0186] Among them, the coefficient of determination R 2 The calculation formula is:

[0187]

[0188] The optimal model among the trained candidate models is determined based on evaluation metrics, resulting in the target algorithm model. This includes identifying the candidate model with the highest determination coefficient as the optimal model. (See also...) Figures 15A-15D , Figure 16 As shown, the RandomForest algorithm model has the largest determination coefficient. Therefore, the trained RandomForest algorithm model is determined to be the optimal model, that is, the target algorithm model is the trained RandomForest algorithm model.

[0189] The formula for calculating the mean squared error (MSE) is as follows:

[0190]

[0191] The optimal model among the candidate models after training is determined based on evaluation metrics, resulting in the target algorithm model, which includes identifying the candidate model with the minimum mean square error as the optimal model. For example... Figures 15A-15D , Figure 16 As shown, the RandomForest algorithm model has the smallest mean squared error. Therefore, the trained RandomForest algorithm model is determined to be the optimal model, that is, the target algorithm model is the trained RandomForest algorithm model.

[0192] The formula for calculating the Mean Absolute Error (MAE) is as follows:

[0193]

[0194] The optimal model among the candidate models after training is determined based on evaluation metrics, resulting in the target algorithm model, which includes identifying the candidate model with the smallest mean absolute error as the optimal model. For example... Figures 15A-15D , Figure 16 As shown, the RandomForest algorithm model has the smallest mean absolute error. Therefore, the trained RandomForest algorithm model is determined to be the optimal model, that is, the target algorithm model is the trained RandomForest algorithm model.

[0195] Please see Figure 17 The diagram shown is a schematic of the intelligent control device for a cooling tower provided in a preferred embodiment of this application.

[0196] The intelligent control device 2 is communicatively connected to the cooling tower 10 and is used for intelligent control of the cooling tower 10. The intelligent control device 2 includes at least a first processor 201, which is communicatively connected to a thermometer 15 and a data acquisition unit 16. The first processor 201 is used to determine the target cooling parameters of the cooling tower 10 from the thermometer 15 and the real-time operating parameters of the cooling tower 10 from the data acquisition unit 16. The target cooling parameters include the sum of the ambient wet-bulb temperature and the preset temperature. The real-time operating parameters include at least the operating status parameters of the multiple fans 110 of the cooling tower 10. The first processor 201 is also used to input the real-time operating parameters of the cooling tower 10 into a target algorithm model to obtain the current cooling parameters of the cooling tower 10; and to adjust the real-time operating parameters of the cooling tower 10 to the target operating parameters based on the current cooling parameters and the target cooling parameters. The target operating parameters are the operating parameters of the cooling tower 10 that save the most electricity among all operating parameters that make the current cooling parameters meet the target cooling parameters.

[0197] To further illustrate the execution of this intelligent control, in one embodiment, the intelligent control device 2 may include, but is not limited to, a first processor 201, a first memory 202, and a first computer program 203 stored in the first memory 202 and executable on the first processor 201. For example, the first computer program 203 is a cooling tower energy-saving intelligent control program. When the first processor 201 executes the first computer program 203, it implements the steps in the cooling tower energy-saving intelligent control method, for example... Figure 3 The steps S101 to S103 shown are as follows: Figure 5 As shown in S201 to S203, Figure 7 S301 to S303 are shown. Figure 9 As shown in S401 to S402, Figure 11 As shown in S501 to S503, Figure 12 S5011 to S5014 are shown. Figure 14S5021 to S5024 are shown.

[0198] For example, the first computer program 203 can be divided into one or more modules / units, one or more modules / units are stored in the first memory 202 and executed by the first processor 201 to complete this application. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the first computer program 203 in the intelligent control device 2.

[0199] Those skilled in the art will understand that the schematic diagram is merely an example of the intelligent control device 2 and does not constitute a limitation on the intelligent control device 2. It may include more or fewer components than shown in the diagram, or combine certain components, or different components. For example, the intelligent control device 2 may also include input / output devices, network access devices, buses, etc.

[0200] The first processor 201 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the first processor 201 can be any conventional processor. The first processor 201 is the control center of the intelligent control device 2, connecting all parts of the intelligent control device 2 via various interfaces and lines.

[0201] The first memory 202 can be used to store the first computer program 203 and / or modules / units. The first processor 201 realizes various functions of the intelligent control device 2 by running or executing the computer program and / or modules / units stored in the first memory 202 and calling the data stored in the first memory 202. The first memory 202 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function, etc.; the data storage area may store data created according to the use of the intelligent control device 2, etc. In addition, the first memory 202 may include volatile and non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other storage devices.

[0202] Please see Figure 18 The diagram shown is a structural schematic of an electronic device provided in a preferred embodiment of this application.

[0203] Electronic device 3 may be a personal computer, server, etc. Electronic device 3 includes, but is not limited to, a second processor 301, a second memory 302, and a second computer program 303 stored in the second memory 302 and executable on the second processor 301. For example, the second computer program 303 is a cooling tower energy-saving intelligent control program. When the second processor 301 executes the second computer program 303, it implements the steps in the cooling tower energy-saving intelligent control method, for example... Figure 3 The steps S101 to S103 shown are as follows: Figure 5 As shown in S201 to S203, Figure 7 S301 to S303 are shown. Figure 9 As shown in S401 to S402, Figure 11 As shown in S501 to S503, Figure 12 S5011 to S5014 are shown. Figure 14 S5021 to S5024 are shown.

[0204] For example, the second computer program 303 may be divided into one or more modules / units, one or more of which are stored in the second memory 302 and executed by the second processor 301 to complete this application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the second computer program 303 in the electronic device 3.

[0205] Those skilled in the art will understand that the schematic diagram is merely an example of the electronic device 3 and does not constitute a limitation on the electronic device 3. It may include more or fewer components than shown, or combine certain components, or different components. For example, the electronic device 3 may also include input / output devices, network access devices, buses, etc.

[0206] The second processor 301 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the second processor 301 can be any conventional processor. The second processor 301 is the control center of the electronic device 3, connecting various parts of the electronic device 3 through various interfaces and lines.

[0207] The second memory 302 can be used to store the second computer program 303 and / or modules / units. The second processor 301 implements various functions of the electronic device 3 by running or executing the computer program and / or modules / units stored in the second memory 302 and calling the data stored in the second memory 302. The second memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device 3, etc. In addition, the second memory 302 may include volatile and non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other storage devices.

[0208] If the modules / units integrated in electronic device 3 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), and random access memory (RAM).

[0209] This application provides a cooling tower energy-saving intelligent control method, device, electronic device, and medium that can predict the cooling tower outlet water temperature based on the real-time operating parameters of the cooling tower. When the cooling tower of this application is used to provide cooling water for cooling equipment, the cooling tower outlet water temperature is avoided from being too low while meeting the cooling demand. This is conducive to the evaporation of heat from the cooling tower, thereby reducing energy consumption, improving energy efficiency, and promoting energy conservation and environmental protection, which is in line with the concepts of carbon neutrality and carbon peaking.

[0210] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from the spirit or essential characteristics of this application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of this application is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within this application. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in the apparatus claims may also be implemented by the same unit or device in software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any particular order.

[0211] The above embodiments are only used to illustrate the technical solutions of this application and are not intended to limit it. Although this application has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of this application without departing from the spirit and scope of the technical solutions of this application.

Claims

1. A cooling tower energy-saving intelligent control method, characterized in that, The method comprises: obtaining a target cooling parameter and a real-time operation parameter of a cooling tower; inputting the real-time operation parameter of the cooling tower into a target algorithm model to obtain a current cooling parameter of the cooling tower; adjusting the real-time operation parameter of the cooling tower to a target operation parameter according to the current cooling parameter and the target cooling parameter of the cooling tower, the target operation parameter being an operation parameter of the cooling tower that makes the current cooling parameter meet the target cooling parameter and saves the power consumption of the cooling tower most, the target cooling parameter comprising a target outlet water temperature of the cooling tower, the target outlet water temperature being the sum of an ambient wet-bulb temperature and a preset temperature, the current cooling parameter comprising a predicted outlet water temperature of the cooling tower, the real-time operation parameter and the target operation parameter of the cooling tower both comprising a fan operation state parameter of the cooling tower; if the fan is a combination of a variable frequency fan and a constant frequency fan, the operation state parameter of the fan comprising the number of the constant frequency fans turned on and the working frequency of the variable frequency fan, adjusting the real-time operation parameter of the cooling tower according to the current cooling parameter and the target cooling parameter of the cooling tower comprising: determining the absolute value of a temperature difference value according to the predicted outlet water temperature and the target outlet water temperature, comparing the absolute value of the temperature difference value with a preset difference value; when the temperature difference value is less than 0 and the absolute value of the temperature difference value is greater than the preset difference value, reducing the number of the constant frequency fans turned on and / or the working frequency of the variable frequency fan; or when the temperature difference value is greater than 0 and the absolute value of the temperature difference value is greater than the preset difference value, increasing the number of the constant frequency fans turned on and / or the working frequency of the variable frequency fan.

2. The cooling tower energy-saving intelligent control method of claim 1, wherein, if the fan is a constant frequency fan, the operation state parameter of the fan comprising the number of the constant frequency fans turned on, adjusting the real-time operation parameter of the cooling tower according to the current cooling parameter and the target cooling parameter of the cooling tower comprising: determining the absolute value of a temperature difference value according to the predicted outlet water temperature and the target outlet water temperature, comparing the absolute value of the temperature difference value with a preset difference value; when the temperature difference value is less than 0 and the absolute value of the temperature difference value is greater than the preset difference value, reducing the number of the constant frequency fans turned on; or when the temperature difference value is greater than 0 and the absolute value of the temperature difference value is greater than the preset difference value, increasing the number of the constant frequency fans turned on.

3. The cooling tower energy-saving intelligent control method of claim 1, wherein, adjusting the real-time operation parameter of the cooling tower according to the current cooling parameter and the target cooling parameter of the cooling tower further comprises: if multiple adjustable fan operation state parameters are obtained according to the predicted outlet water temperature and the target outlet water temperature, determining the total power consumption of the fan in each fan operation state parameter; determining the fan operation state parameter with the lowest total power consumption, and adjusting the number of the constant frequency fans turned on and / or the working frequency of the variable frequency fan according to the fan operation state parameter with the lowest total power consumption.

4. The cooling tower energy-saving intelligent control method of claim 1, wherein, The method further comprises: obtaining historical data of the cooling tower, the historical data comprising historical target cooling parameters and multiple historical operation parameters of the cooling tower; Preprocess the historical data to obtain a preset number of feature data; Train an algorithm model through the feature data to obtain the target algorithm model.

5. The cooling tower energy-saving intelligent control method of claim 4, wherein, The historical data of the cooling tower includes: Using a first server to collect bottom equipment data of the cooling tower through a reading gateway, and archiving the bottom equipment data to a second server; Using a third server to collect data from the second server in the form of incremental extraction, and saving the collected data to the raw data layer; Preprocessing the data of the raw data layer and saving the preprocessed data to the data warehouse; Performing row-to-column processing on the data in the data warehouse to obtain a data wide table, and saving the data wide table to a data mart as historical data to be preprocessed.

6. The cooling tower energy-saving intelligent control method according to claim 4 or 5, characterized in that, The preprocessing of the historical data to obtain a preset number of feature data includes: Copying the historical data to a local database corresponding to the device to which the target algorithm model belongs through data warehouse technology; Performing structured query language integration on the historical data in the local database; Preprocessing the integrated historical data to obtain a feature database, the feature database including a plurality of feature data; Selecting the preset number of feature data from the plurality of feature data in the feature database.

7. The cooling tower energy-saving intelligent control method of claim 6, wherein, The preprocessing of the historical data to obtain a preset number of feature data further includes: Inputting the plurality of feature data included in the feature database into an automatic machine learning algorithm to obtain a weight order of the plurality of feature data; Selecting the preset number of feature data from the plurality of feature data according to the weight order.

8. The cooling tower energy-saving intelligent control method of claim 6, wherein, The preprocessing of the historical data of the cooling tower to obtain a preset number of feature data further includes: Inputting the plurality of feature data included in the feature database into a chi-square verification algorithm to output the preset number of feature data.

9. The cooling tower energy-saving intelligent control method of claim 6, wherein, The preprocessing of the historical data of the cooling tower to obtain a preset number of feature data further includes: Deriving a part of the preset number of feature data to obtain derived feature data to replace the part of the feature data, and the derived feature data and another part of the feature data are used to train an algorithm model to obtain the target algorithm model.

10. The cooling tower energy-saving intelligent control method of claim 9, wherein, The cooling tower includes at least one fixed-frequency fan, and the part of the feature data includes at least one fixed-frequency fan operation state parameter obtained by preprocessing a fan operation state parameter of the at least one fixed-frequency fan, Deriving the at least one fixed-frequency fan operation state parameter as one fixed-frequency fan opening number feature data as the derived feature data.

11. The cooling tower energy-saving intelligent control method of claim 4, wherein, The algorithm model is a regression algorithm model, and the regression algorithm model is any one of a support vector machine algorithm, a random forest algorithm, a gradient boosting decision tree algorithm, and a ridge regression algorithm.

12. The cooling tower energy-saving intelligent control method of claim 4, wherein, The training of the algorithm model through the feature data to obtain the target algorithm model includes: The characteristic data are used to train a support vector machine algorithm, a random forest algorithm, a gradient boosting decision tree algorithm and a ridge regression algorithm respectively to obtain a plurality of candidate models; A determination coefficient, a mean square error and a mean absolute error of each candidate model are calculated based on a predicted value and an actual value of each candidate model; A determination coefficient, a mean square error and a mean absolute error of each candidate model are calculated based on a predicted value and an actual value of each candidate model; 13. An energy-saving intelligent control device for a cooling tower, the cooling tower comprising a water tower, a thermometer, a collector and a plurality of fans, the thermometer being arranged in an external environment where the water tower is located to sense an ambient wet-bulb temperature, the collector being electrically connected to at least the plurality of fans to collect operating state parameters of the plurality of fans, characterized in that The device comprises: a first processor in communication connection with the thermometer and the collector, configured to determine a target cooling parameter of the cooling tower from the thermometer and determine real-time operating parameters of the cooling tower from the collector, the target cooling parameter comprising a sum of the ambient wet-bulb temperature and a preset temperature, and the real-time operating parameters comprising at least operating state parameters of the plurality of fans of the cooling tower; the first processor is further configured to input the real-time operating parameters of the cooling tower into a target algorithm model to obtain a current cooling parameter of the cooling tower, and adjust the real-time operating parameters of the cooling tower to target operating parameters according to the current cooling parameter and the target cooling parameter, the target operating parameters being operating parameters of the cooling tower that make the current cooling parameter meet the target cooling parameter and save the most power consumption of the cooling tower, the target cooling parameter comprising a target outlet water temperature of the cooling tower, the target outlet water temperature being a sum of the ambient wet-bulb temperature and the preset temperature, and the current cooling parameter comprising a predicted outlet water temperature of the cooling tower, the real-time operating parameters and the target operating parameters of the cooling tower both comprising fan operating state parameters of the cooling tower; if the fans are a combination of variable-frequency fans and constant-frequency fans, the operating state parameters of the fans comprising a number of the constant-frequency fans turned on and working frequencies of the variable-frequency fans, and adjusting the real-time operating parameters of the cooling tower according to the current cooling parameter and the target cooling parameter comprises: determining an absolute value of a temperature difference according to the predicted outlet water temperature and the target outlet water temperature, and comparing the absolute value of the temperature difference with a preset difference value; when the temperature difference is less than 0 and the absolute value of the temperature difference is greater than the preset difference value, reducing the number of the constant-frequency fans turned on and / or the working frequencies of the variable-frequency fans; or when the temperature difference is greater than 0 and the absolute value of the temperature difference is greater than the preset difference value, increasing the number of the constant-frequency fans turned on and / or the working frequencies of the variable-frequency fans.

14. An electronic device, characterized by The electronic device comprises: a second processor; and a second memory having instructions stored therein, the instructions being loaded and executed by the second processor to perform the energy-saving intelligent control method for the cooling tower according to any one of claims 1 to 12.

15. A computer readable storage medium having stored thereon at least one computer instruction, wherein, The instructions are loaded and executed by the second processor to perform the cooling tower energy-saving intelligent control method as claimed in any one of claims 1 to 12.

Images