Cooling tower fan current prediction method, fault detection method and related device

Abstract

This disclosure presents a cooling tower fan current prediction method, a fault detection method, and related apparatus. A method for detecting a cooling tower fault includes: acquiring interrelated power, frequency, and current values ​​measured for the cooling tower fan; predicting the cooling tower fan current value using a prediction model based on the measured power and frequency values, the prediction model being a function of current with respect to power and frequency; and issuing an alarm to indicate a detected cooling tower fault if the difference between the predicted current value and the measured current value exceeds a difference threshold.

Description

Technical Field

[0001] This disclosure relates to the field of cooling tower technology, and more specifically, to a method for predicting the current of a cooling tower fan, a method and apparatus for detecting cooling tower faults, a computing device, a non-transient storage medium, and a computer program product. Background Technology

[0002] Cooling towers are an important component of refrigeration systems, often used in the HVAC architecture of data centers to dissipate heat from chillers or plate heat exchangers. Their core principle is to transfer heat from the cooling water to the air through direct or indirect contact between the cooling water and the air, utilizing evaporative heat transfer, convective heat transfer, and conductive heat transfer, thereby lowering the water temperature.

[0003] Cooling towers typically consist of a tower body, packing, and a fan. The packing increases the contact area between the cooling water and the air, improving heat exchange efficiency. The fan accelerates airflow through forced ventilation, enhancing the cooling effect. A fan usually includes a fan and a motor that drives it. Due to the relatively large size of cooling tower fans, currently operational cooling tower fans typically use asynchronous motors, which, along with a speed reducer and transmission belt, indirectly drive the fan. Summary of the Invention

[0004] A brief overview of this disclosure is given below to provide a basic understanding of some aspects of it. However, it should be understood that this overview is not an exhaustive summary of this disclosure. It is not intended to identify key or essential parts of this disclosure, nor is it intended to limit the scope of this disclosure. Its purpose is merely to present certain concepts of this disclosure in a simplified form as a prelude to the more detailed description that follows.

[0005] According to a first aspect of this disclosure, a method for predicting the current of a fan in a cooling tower is provided, comprising: acquiring a plurality of data points measured for the fan of the cooling tower under normal operating conditions, each data point including a measured power value, a measured frequency value, and a measured current value that are correlated with each other; fitting a prediction model using a first subset of the plurality of data points to obtain fitting parameters for the prediction model, the prediction model being a function of current with respect to power and frequency; and using the prediction model having the fitting parameters to predict the current value of the fan in the cooling tower based on the correlated power and frequency values ​​measured for the fan in the cooling tower.

[0006] In some embodiments, the prediction model determines the current based on the product of a power term of the power and a power term of the frequency.

[0007] In some embodiments, the prediction model is represented as: I = CPm f n Where I represents the current, C represents a constant parameter, P represents the power, m represents the power power parameter, f represents the frequency, and n represents the frequency power parameter.

[0008] In some embodiments, the method includes: validating the prediction model having the fitting parameters using a second subset of the plurality of data points, which is different from the first subset, wherein the validation includes: for each data point in the second subset, using the prediction model to determine a predicted current value for the data point based on the measured power value and the measured frequency value of the data point, and determining an error value between the predicted current value and the measured current value of the data point, wherein the validation passes if the ratio between the number of data points in the second subset whose error value exceeds a first error threshold and the total number of data points in the second subset does not exceed a first ratio threshold; and using the validated prediction model to predict the current value of the fan of the cooling tower based on mutually correlated power values ​​and frequency values ​​measured for the fan of the cooling tower.

[0009] In some embodiments, the verification includes: the verification is passed when the ratio between the number of data points in the second subset whose error value exceeds the first error threshold and the total number of data points in the second subset does not exceed the first ratio threshold, and the ratio between the number of data points in the second subset whose error value exceeds the second error threshold and the total number of data points in the second subset does not exceed the second ratio threshold, wherein the second error threshold is less than the first error threshold, and the second ratio threshold is greater than the first ratio threshold.

[0010] In some embodiments, the verification includes: the verification is passed when the ratio between the number of data points in the second subset whose error value exceeds the first error threshold and the total number of data points in the second subset does not exceed the first ratio threshold, and the data points in the second subset whose error value exceeds the first error threshold are continuously distributed.

[0011] In some embodiments, the verification includes: the verification is passed if the ratio between the number of data points in the second subset whose error value exceeds the first error threshold and the total number of data points in the second subset does not exceed the first ratio threshold, and the data points in the first subset are continuously distributed.

[0012] In some embodiments, the measured power value, measured frequency value, and measured current value at each data point correspond to the same time. In some embodiments, the measured power value, measured frequency value, and measured current value at each data point correspond to a first time, a second time, and a third time, respectively, wherein the third time is offset relative to the first time and / or the second time.

[0013] According to a second aspect of this disclosure, a method for detecting a fault in a cooling tower is provided, comprising: acquiring mutually correlated power, frequency, and current values ​​measured for a fan of the cooling tower; predicting a current value of the fan of the cooling tower based on the measured power and frequency values ​​using a prediction model, the prediction model being a function of current with respect to power and frequency; and issuing an alarm to indicate that a fault in the cooling tower has been detected if the difference between the predicted current value and the measured current value is greater than a difference threshold.

[0014] In some embodiments, an alarm is triggered to indicate that a fault in the cooling tower has been detected if the difference between the predicted current value and the measured current value is greater than the difference threshold for a duration threshold.

[0015] In some embodiments, the prediction model determines the current based on the product of a power term of the power and a power term of the frequency.

[0016] In some embodiments, the prediction model is represented as: I = CP m f n Where I represents the current, C represents a constant parameter, P represents the power, m represents the power power parameter, f represents the frequency, and n represents the frequency power parameter.

[0017] In some embodiments, predicting the current value of the fan of the cooling tower based on the measured power value and the measured frequency value using a predictive model is performed by the method according to any embodiment of the first aspect of this disclosure.

[0018] In some embodiments, the method further includes: if the difference between the predicted current value and the measured current value is greater than the difference threshold and reaches the duration threshold, determining the cause of the fault based on one or more of the following: the magnitude relationship between the predicted current value and the measured current value, the time-varying relationship of the current value measured for the fan of the cooling tower over a past time period, and the ambient temperature.

[0019] In some embodiments, determining the cause of the fault includes: if the predicted current value is greater than the measured current value and the change shows a step-down pattern, determining the cause of the fault as a broken or fallen belt of the fan used to drive the cooling tower.

[0020] In some embodiments, determining the cause of the fault includes: if the predicted current value is greater than the measured current value, the change relationship shows a gradual decrease, and the ambient temperature is not higher than the freezing point of the cooling tower packing, determining the cause of the fault is that the cooling tower packing is icing; or if the predicted current value is greater than the measured current value, the change relationship shows a gradual decrease, and the ambient temperature is higher than the freezing point of the cooling tower packing, determining the cause of the fault is that the cooling tower packing is clogged.

[0021] In some embodiments, the severity of icing or clogging of the packing material in the cooling tower is determined based on the gradually decreasing slope.

[0022] In some embodiments, determining the cause of the fault includes: if the predicted current value is less than the measured current value, determining the cause of the fault as a change in the blade angle of the fan of the cooling tower's blower.

[0023] In some embodiments, the difference threshold is based on one or more of the following: grid voltage fluctuations, outdoor wind speed, and cooling tower operating conditions.

[0024] In some embodiments, the measured power value, the measured frequency value, and the measured current value correspond to a first time moment, a second time moment, and a third time moment, respectively, and wherein the difference threshold is set based on the offset of the third time moment relative to the first time moment and / or the second time moment.

[0025] In some embodiments, the duration threshold is set based on the time interval for acquiring interrelated power, frequency, and current values ​​measured for the fan of the cooling tower.

[0026] According to a third aspect of this disclosure, an apparatus for detecting a fault in a cooling tower is provided, comprising: a data acquisition unit configured to acquire interrelated power, frequency, and current values ​​measured for a fan of the cooling tower; a controller coupled to the data acquisition unit and configured to detect a fault in the cooling tower using a method according to any embodiment of a second aspect of this disclosure; and a user interface via which the controller issues an alarm to indicate that a fault in the cooling tower has been detected.

[0027] In some embodiments, the apparatus includes at least one of the following: a data storage device coupled to the data acquisition device and the controller and configured to store data acquired by the data acquisition device and the calculation results of the controller; a building automation (BA) system, the BA system including one or more of the user interface and the controller.

[0028] In some embodiments, the cooling tower includes a fan and a frequency converter that performs frequency conversion control on the fan, the data acquisition unit is configured to acquire fan measurement data for the cooling tower from the frequency converter, and wherein the prediction model is built based on the frequency converter.

[0029] According to a fourth aspect of this disclosure, a computing device is provided, comprising: one or more processors; and a memory storing computer-executable instructions, which, when executed by the one or more processors, cause the one or more processors to perform the method described according to any embodiment of the first or second aspect of this disclosure.

[0030] According to a fifth aspect of this disclosure, a non-transient storage medium having computer-executable instructions stored thereon is provided, which, when executed by a computer, cause the computer to perform the method described according to any embodiment of the first or second aspect of this disclosure.

[0031] According to a sixth aspect of this disclosure, a computer program product is provided, the computer program product including instructions that, when executed by a processor, implement the method according to any embodiment of the first or second aspect of this disclosure. Attached Figure Description

[0032] The foregoing and other features and advantages of this disclosure will become clear from the following description of embodiments illustrated in conjunction with the accompanying drawings. The drawings, incorporated herein and forming a part of the specification, are further used to explain the principles of this disclosure and to enable those skilled in the art to make and use it. Wherein:

[0033] Figure 1 This is a flowchart illustrating a method for detecting a cooling tower malfunction according to some embodiments of the present disclosure;

[0034] Figure 2 This is a flowchart illustrating a method for predicting the current of a cooling tower fan according to some embodiments of the present disclosure;

[0035] Figure 3 This is an exemplary 3D plot showing data points for a first subset used to fit a prediction model and a prediction model fitted using the first subset;

[0036] Figure 4 This is a flowchart illustrating a method for predicting the current of a cooling tower fan according to some embodiments of the present disclosure;

[0037] Figure 5 This is an exemplary 3D diagram showing data points for a second subset used to validate a prediction model and a prediction model validated using the second subset;

[0038] Figure 6 This is a flowchart illustrating a method for detecting a cooling tower malfunction according to some embodiments of the present disclosure;

[0039] Figure 7 This is an exemplary flowchart illustrating a process for determining the cause of a failure, in which a method for detecting a failure in a cooling tower according to some embodiments of this disclosure is applied.

[0040] Figure 8 It is a graph showing the predicted current value and the measured current value obtained by using the prediction model in an example application scenario, which change over time.

[0041] Figure 9 This is a schematic block diagram illustrating a fault detection apparatus according to some embodiments of the present disclosure;

[0042] Figure 10 This is a schematic block diagram illustrating a fault detection device and a cooling tower according to some embodiments of the present disclosure;

[0043] Figure 11 This is a schematic block diagram illustrating a fault detection device and a cooling tower according to some embodiments of the present disclosure;

[0044] Figure 12 This is a schematic block diagram illustrating a computing device according to some embodiments of the present disclosure;

[0045] Figure 13 This is a schematic block diagram illustrating a computer system on which embodiments of the present disclosure may be implemented.

[0046] Note that in the embodiments described below, the same reference numerals are sometimes used across different figures to denote the same parts or parts having the same function, and repeated descriptions are omitted. In this specification, similar reference numerals and letters are used to denote similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0047] For ease of understanding, the positions, dimensions, and extents of the structures shown in the accompanying drawings and other materials may not represent actual positions, dimensions, and extents. Therefore, the disclosed invention is not limited to the positions, dimensions, and extents disclosed in the accompanying drawings and other materials. Furthermore, the drawings are not necessarily drawn to scale, and some features may be enlarged to show details of specific components. Detailed Implementation

[0048] Various exemplary embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0049] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use. Those skilled in the art will understand that they merely illustrate exemplary ways that can be used to implement this disclosure, and are not exhaustive.

[0050] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0051] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0052] Additionally, when the concept of “value” is mentioned in this article, it usually refers to the absolute value, unless otherwise stated.

[0053] A typical cooling tower consists of multiple belts (e.g., 5-6) to drive the fan. Over long-term operation, these belts can loosen and wear, and in severe cases, may even break or fall off. This affects the cooling tower's operating efficiency, leading to high temperatures in the cooling water output, causing the cooling tower to alarm and shut down, and consequently impacting the data center's cooling system and terminal equipment. Because data centers have high requirements for the safe and stable operation of their HVAC systems, cooling tower belts that show signs of loosening, wear, breakage, or falling off must be promptly maintained or repaired.

[0054] Currently, the cooling tower belts are maintained through routine manual inspections. This means that the detection of belt drops or breaks is not timely, and there may be a situation where no one responds for a long time after a belt drops or breaks, leading to high temperature alarms in the data center cooling system.

[0055] Besides belt breakage or detachment, cooling tower malfunctions also include changes in the fan blade angle and packing icing or blockage. During operation, the fan's fastening nuts may loosen, causing the fan blades to tilt at an increased angle. This increases resistance to the blades during operation, reducing the cooling tower's efficiency. If changes in blade angle are not detected promptly, the blades may strike the cooling tower body, ultimately damaging or rendering them unusable. Additionally, when outdoor temperatures are low, such as in winter, ice may form inside the packing. Alternatively, the packing may accumulate dirt, impurities, or microorganisms over time, clogging its pores. Icing or blockage in the packing increases airflow resistance and reduces air volume, further lowering cooling tower efficiency and hindering energy-efficient system operation.

[0056] To promptly detect cooling tower malfunctions, an improved method for cooling tower fault detection was desired. Through extensive research, the inventors discovered that the current value of the cooling tower fan can be used to monitor whether a malfunction has occurred, and even to determine the cause of the malfunction. For example, in the case of a broken or fallen belt, the motor cannot transmit some or all of its power to the fan, causing a short-term reduction in motor load and resulting in a sudden drop in fan current. Another example is when the fan blade angle changes, increasing the resistance and torque experienced by the fan, thus increasing the fan current. Yet another example is when the packing material freezes or becomes clogged, gradually blocking the cooling tower's air inlet, leading to increased airflow resistance and reduced air volume. This gradually reduces the airflow to the fan, thus decreasing the resistance and consequently reducing the fan current.

[0057] Based on the above, this disclosure provides a method for detecting cooling tower faults, which can identify whether a cooling tower has malfunctioned by the deviation between the actual current value and the theoretical current value of the cooling tower fan.

[0058] The methods for detecting malfunctions in cooling towers according to various embodiments of the present disclosure will now be described in detail with reference to the accompanying drawings. It will be understood that actual methods may include other steps, which are not shown in the drawings and will not be discussed herein in order to avoid obscuring the essential points of the disclosure.

[0059] Figure 1 A flowchart of a method 10 for detecting faults in a cooling tower according to some embodiments of the present disclosure is shown. Figure 1 As shown, method 10 includes steps S12 to S16.

[0060] In step S12, the power, frequency, and current values ​​of the fan for the cooling tower are obtained in relation to each other.

[0061] Specifically, the power value and current value refer to the power and current consumed by the fan, respectively, and the frequency value refers to the rotational frequency of the fan.

[0062] In some embodiments, the cooling tower fan is controlled by a frequency converter. For example, the power, frequency, and current values ​​of the fan can be obtained via the frequency converter. In some embodiments, additional sensors or measuring devices may also be installed to collect the power, frequency, and current values ​​of the fan. This disclosure does not impose any particular limitations on the specific methods for measuring the power, frequency, and current values ​​of the fan or the methods for obtaining the measurement results.

[0063] In step S14, a prediction model is used to predict the current value of the cooling tower fan based on the measured power value and the measured frequency value. The prediction model is a function of current with respect to power and frequency.

[0064] The predictive model and the method for using the predictive model to predict the current value of the cooling tower fan will be explained in more detail below.

[0065] In step S16, if the difference between the predicted current value and the measured current value is greater than the difference threshold, an alarm is triggered to indicate that a fault has been detected in the cooling tower.

[0066] The predicted current value indicates the theoretical current value under normal operating conditions of the cooling tower. When the measured current value deviates significantly from the predicted current value, the cooling tower may be operating abnormally, indicating a malfunction.

[0067] In this article, the normal operating condition of a cooling tower can include meeting the heat exchange efficiency target, matching water flow and air volume, stable equipment operation, good packing condition, and / or energy consumption within a reasonable range. More specifically, the normal operating condition of a cooling tower can include, for example, that the fan belt is not broken or fallen, the fan blade angle is normal, and / or the packing is not iced or clogged.

[0068] Compared to manual inspection or monitoring of cooling tower water temperature, fault detection methods based on monitoring current can detect sudden faults more promptly, immediately triggering alarms to instruct staff to resolve issues and prevent more serious consequences (e.g., avoiding water pressure fluctuations in the cooling water system caused by high-temperature alarm shutdowns of cooling towers), thus ensuring the safe and stable operation of the data center system. Furthermore, manual inspections rely on experience-based judgment, and water temperature monitoring is often a reactive measure; these methods struggle to detect early, latent faults. The fault detection method based on monitoring current disclosed in this invention effectively improves the timeliness, objectivity, and accuracy of fault detection. Additionally, this fault detection method eliminates the need for system shutdown for inspection, reducing production interruptions. It also avoids the risks associated with manual high-altitude or confined space operations.

[0069] Furthermore, this disclosure uses a predictive model to predict the current of the cooling tower fan. The predictive model is capable of predicting the cooling tower fan current based on other operating parameters of the cooling tower fan measured under corresponding operating conditions. Specifically, the predictive model of this disclosure is a function of current with respect to power and frequency. Therefore, this disclosure can use the predictive model to predict the current value of the cooling tower fan based on the measured power value and the measured frequency value.

[0070] For example, the prediction model can be a mathematical model built based on the principles of frequency converters. In some embodiments, the prediction model may include the product of a power term for power and a power term for frequency. Such a model can take into account the instability of the mathematical relationship between the current, power, and frequency of the cooling tower fan caused by voltage fluctuations, power factor fluctuations, and magnetic flux fluctuations during motor operation, making the model's predictions more accurate.

[0071] In some embodiments, the prediction model can be represented as: I = CP m f n In this model, I represents current, C represents a constant parameter, P represents power, m represents a power parameter of power, f represents frequency, and n represents a power parameter of frequency. This predictive model has fewer parameters, ensuring accurate predictions while reducing the difficulty of fitting the data.

[0072] The parameters C, m, and n can be set according to the specific circumstances. For example, they can be determined using the fitting method described later, or based on testing experience. In some examples, m is constrained to be positive while n is constrained to be negative, i.e. In some examples, the values ​​of m and n are further set to 1, that is...

[0073] It is understandable that parameters m and n are related to fluctuations, so their sign can be left unconstrained during the fitting process. During the fitting process, there may be cases where m is positive and n is negative, or m is negative and n is positive, or both m and n are positive, or both m and n are negative. The specific case depends on whether the fitting result meets the accuracy and reliability requirements.

[0074] In some embodiments, the prediction model can be represented as: I = CP m f n +A, where A represents a constant parameter. Compared to the previous example, this prediction model has an adjustable baseline level, allowing it to flexibly adapt to different cooling tower operating conditions.

[0075] In some embodiments, the prediction model can be in polynomial form. In some examples, the prediction model can be represented as: Among them, a iThe coefficient parameter representing power, b j The coefficients represent the frequency parameters. Such prediction models are highly parameterized and scalable, making them suitable for cooling tower operations that generate complex data points. In some examples, the prediction model can be represented as: Such a predictive model can strike a balance between scalability and fitting difficulty.

[0076] The following will combine Figures 2 to 5 The method for determining the parameters of the prediction model and the method for predicting the current of the cooling tower fan are described in detail.

[0077] Figure 2 A flowchart of a method 20 for predicting the current of a cooling tower fan, according to some embodiments of the present disclosure, is shown. Figure 2 As shown, method 20 includes steps S22 to S26.

[0078] In step S22, multiple data points for the cooling tower fan are acquired under normal operating conditions. Each data point includes a measured power value, a measured frequency value, and a measured current value that are associated with each other.

[0079] In step S24, the prediction model is fitted using a first subset of the multiple data points to obtain the fitting parameters of the prediction model. The prediction model is a function of current with respect to power and frequency.

[0080] In step S26, a prediction model with fitting parameters is used to predict the current value of the cooling tower fan based on the interrelated power and frequency values ​​measured for the cooling tower fan.

[0081] For example, the least squares method can be used to fit the prediction model. Of course, other model fitting methods such as maximum likelihood estimation, Bayesian estimation, and random sampling consistency are also feasible, and this disclosure does not impose any particular limitation on the specific fitting method. The fitting method can be implemented using a computer program, such as a program written in a programming language like Python.

[0082] During the model fitting process, the model's coefficient of determination R can be used. 2 The fitting is considered complete when the coefficient of determination threshold is reached or exceeded. The coefficient of determination threshold can also be considered as the accuracy requirement for the prediction model, and it can be set according to the operating conditions of the cooling tower and on-site requirements. In some embodiments, the coefficient of determination threshold can be set between 0.90 and 0.99, or between 0.95 and 0.99, for example, it can be 0.98.

[0083] For the purpose of non-restrictive illustrative purposes, Figure 3The diagram schematically illustrates the data points of the first subset used to fit the prediction model and the prediction model I=CP fitted using the first subset. m f n . Figure 3 The 3D graph shown includes three coordinate axes: the P-axis represents power as the independent variable, the f-axis represents frequency as the independent variable, and I represents current as the dependent variable. Each point represents a data point in the first subset. The surface is the fitted surface obtained using the first subset and fitted based on the least squares method, which graphically represents the prediction model with fitted parameters. This prediction model is fitted based on the operating data of a data center's cooling tower over six months (the first subset), with specific fitted parameters C = 55.1797, m = 0.5985, and n = -0.5312. The coefficient of determination R of this prediction model is... 2 The value is 0.9880, and it can be clearly seen that the data points are distributed almost continuously near the fitted surface, indicating that the model fits the data well. Using such a prediction model can obtain more accurate predicted current values, and further, can reduce the false alarm rate of fault detection.

[0084] In addition, the fitted prediction model can be validated to evaluate its prediction accuracy. For example... Figure 4 As shown, the difference between method 20' and method 20 is that step S25 is added and step S26' is replaced by step S26'.

[0085] In step S25, the prediction model with fitting parameters is validated using a second subset of multiple data points, which is different from the first subset.

[0086] In step S26', using a validated prediction model, the current value of the cooling tower's fan is predicted based on the interrelated power and frequency values ​​measured for the cooling tower's fan.

[0087] The second subset may include some or all of the other data points in the plurality of data points besides the first subset. In some cases, the second subset may also include some or all of the data points in the first subset.

[0088] For example, verification may include, for each data point in the second subset, using a prediction model to determine the predicted current value for that data point based on the measured power value and the measured frequency value, and determining the error between the predicted current value and the measured current value for that data point.

[0089] The conditions for successful validation can vary. In some embodiments, validation is successful if the ratio between the number of data points in the second subset whose error values ​​exceed a first error threshold and the total number of data points in the second subset does not exceed a first proportional threshold. The first error threshold and the first proportional threshold, combined, can reflect the minimum acceptable prediction accuracy and can be set separately depending on the specific circumstances. In some embodiments, the first error threshold can be between 2.5 amperes (A) and 4 A, for example, 3 A. In some embodiments, the first proportional threshold can be between 0.1% and 1%, or between 0.2% and 0.5%, for example, 0.25%.

[0090] In some embodiments, validation is successful if the ratio between the number of data points in the second subset whose error values ​​exceed the first error threshold and the total number of data points in the second subset does not exceed the first proportional threshold, and the ratio between the number of data points in the second subset whose error values ​​exceed the second error threshold and the total number of data points in the second subset does not exceed the second proportional threshold. The second error threshold is less than the first error threshold, and the second proportional threshold is greater than the first proportional threshold. The combination of the first error threshold and the first proportional threshold reflects the minimum acceptable prediction accuracy, while the combination of the second error threshold and the second proportional threshold reflects the achievable prediction accuracy, which can be set separately according to specific circumstances. This gradient evaluation provides a more granular basis for evaluating the model's prediction performance. In some embodiments, the second error threshold can be between 0.5A and 2.5A, for example, 2A. In some embodiments, the second proportional threshold can be between 0.5% and 2%, or between 0.5% and 1%, for example, 0.75%.

[0091] by Figure 3 Taking the predicted model shown as an example, we will verify the model. Substitute the data points from the first subset into the prediction model I = 55.1797P. 0.5985 f -0.5312 The data showed that 0.60% of the data had a current error value above 2A (the second error threshold of 2A) (not exceeding the second proportion threshold of 0.75%), and 0.09% had a current error value above 3A (the first error threshold of 3A) (not exceeding the first proportion threshold of 0.25%). Using the data from another six months of operation of the data center as a second subset to validate the prediction model, the data showed that 0.64% of the data had a current error value above 2A, and 0.16% had a current error value above 3A. This demonstrates that the prediction model has a good fit.

[0092] It is understood that the above embodiments of setting the two combinations of error threshold and proportion threshold are merely exemplary and not restrictive, and more combinations of error threshold and proportion threshold can be set according to actual needs.

[0093] Additionally or alternatively, in addition to evaluating using a combination of error threshold and proportion threshold, the distribution of data points can be considered to assess whether the data points used for fitting and / or validation are appropriate.

[0094] In some embodiments, the verification passes when the ratio between the number of data points in the second subset whose error values ​​exceed the first error threshold and the total number of data points in the second subset does not exceed a first ratio threshold, and the data points in the second subset whose error values ​​exceed the first error threshold are continuously distributed. Here, continuous distribution means that in a three-dimensional graph with power, frequency, and current as axes, the data points are continuously distributed rather than discretely. That is, the distance between adjacent data points with error values ​​greater than the first error threshold in the three-dimensional space defined by power, frequency, and current is small. The inventors have noted that due to various reasons such as grid voltage fluctuations, outdoor wind speed effects, and open operation of cooling towers, there may be various situations such as a sudden drop in the measured power value leading to a lower predicted current value, a sudden drop in the measured frequency value leading to a higher predicted current value, and a sudden increase in the measured current value. Data points with error values ​​greater than the first error threshold caused by these reasons usually do not show a significant continuous distribution, and using such data points to verify the prediction model may be inaccurate. Therefore, in such cases, the verification can be failed, and the verification can be re-performed by changing a batch of data points.

[0095] In some embodiments, validation is successful when the ratio between the number of data points in the second subset whose error values ​​exceed a first error threshold and the total number of data points in the second subset does not exceed a first ratio threshold, and the data points in the first subset are continuously distributed. Here, continuous distribution means that the data points are distributed continuously rather than discretely in a three-dimensional graph with power, frequency, and current as axes. That is, the distance between adjacent data points in the first subset is small in the three-dimensional space defined by power, frequency, and current. A continuous distribution of data points in the first subset, especially along the fitting surface near the fitting surface, indicates higher fitting effectiveness, and alarms generated by a prediction model validated in this way are more valuable. From another perspective, this is also reflected in the coefficient of determination R of the fitting results. 2 Regarding the fluctuations in power grid voltage, outdoor wind speed, and the open operation of cooling towers, various situations may arise, such as a sudden drop in measured power leading to a lower predicted current value, a sudden drop in measured frequency leading to a higher predicted current value, and a sudden increase in measured current value. These factors may cause instability in the mathematical relationship between the current value, power value, and frequency value of the data points, making the fitting of the prediction model using such data points inaccurate. Therefore, in such cases, the validation can fail, and the fitting can be re-performed by replacing a batch of data points.

[0096] For the purpose of non-restrictive illustrative purposes, Figure 5 The diagram schematically illustrates the data points of a second subset used to validate the prediction model, and the prediction model validated using this second subset, which includes the data points of the first subset. It can be observed that these data points are discretely distributed, and many fail to fall on the fitting surface characterizing the prediction model. Furthermore, the coefficient of determination R0 of the fitting results... 2 It is only 0.8408. Therefore, as Figure 5 The predicted model shown cannot be validated.

[0097] There are several ways to adjust the prediction model if validation fails. In some embodiments, the fitting and validation process can be repeated, continuously updating the fitting parameters until the validation conditions are met. In some embodiments, the data used for model fitting can be adjusted, selecting data that better reflects the normal operating conditions of the cooling tower. In some embodiments, the model function can be adjusted, such as increasing / decreasing fitting parameters. In some embodiments, the fitting method can be adjusted to obtain a fitting result that better matches the distribution of the data points.

[0098] The following is for reference. Figure 6 The discussion continues with methods for detecting malfunctions in cooling towers according to some embodiments of this disclosure. For example... Figure 6 As shown, the difference between method 10' and method 10 is that step S16 is replaced by step S16', and step S18 is optionally added.

[0099] In step S16', an alarm is triggered to indicate a detected cooling tower malfunction if the difference between the predicted current value and the measured current value exceeds a difference threshold and reaches a duration threshold. In other words, compared to step S16 of method 10, the alarm condition in step S16' of method 10' requires not only that the difference exceeds the difference threshold but also that the duration of this phenomenon exceeds the duration threshold. This can advantageously reduce false alarms.

[0100] In step S18, the cause of the fault is determined. It is understood that step S18 may also be included in... Figure 1 In method 10 shown.

[0101] In some embodiments, the cause of the fault can be determined based on one or more of the following: the relationship between the predicted current value and the measured current value, the relationship between the measured current value of the cooling tower fan and the change over time in the past period, and the ambient temperature.

[0102] For example, the past time period may include a period preceding the moment when the difference between the predicted current value and the measured current value begins to exceed a difference threshold, and in some cases may also include a period following the moment when the difference between the predicted current value and the measured current value begins to exceed the difference threshold. In some embodiments, the past time period may be 5 to 25 minutes, or 10 to 20 minutes, such as 15 minutes. This disclosure does not impose a particular limitation on the length of the past time period, as long as the past time period includes the period in which the current change triggers the alarm and its vicinity.

[0103] Specifically, in some examples, when the predicted current value is greater than the measured current value, and the measured current value for the cooling tower fan shows a step-down relationship over a past period, the cause of the failure is determined to be a broken or fallen belt of the fan used to drive the cooling tower.

[0104] In some examples, when the predicted current value is greater than the measured current value, and the measured current value for the cooling tower fan shows a gradual decrease over a past period, the cause of the failure is determined to be icing or blockage of the cooling tower packing. Furthermore, if the ambient temperature is not higher than the freezing point of the cooling tower packing, the cause of the failure can be determined to be icing of the cooling tower packing; if the ambient temperature is higher than the freezing point of the cooling tower packing, the cause of the failure is determined to be blockage of the cooling tower packing. For example, the severity of icing or blockage of the cooling tower packing can be determined based on the slope of the gradual decrease. The greater the slope (the faster the decrease), the more severe the icing or blockage. This provides an objective indicator for prioritizing maintenance, allowing cooling towers with more severe icing or blockage to receive maintenance earlier, improving the reliability of cooling tower operation. In some examples, when the predicted current value is less than the measured current value, the cause of the failure is determined to be a change in the fan blade angle of the cooling tower fan.

[0105] For purposes of non-restrictive description, Figure 7 An exemplary flowchart is shown for a process used to determine the cause of a failure. For example... Figure 7 As shown, step S18 includes steps S182 to S1814.

[0106] After executing step S16 or step S16', proceed to step S182. In step S182, determine whether the predicted current value is greater than the measured current value. Alternatively, determine the sign of the difference between the predicted current value and the measured current value. If the determination is yes (the difference is positive), proceed to step S184. If the determination is no (the difference is negative), proceed to step S1814, that is, determine the cause of the fault as a change in the fan blade angle of the cooling tower's fan.

[0107] In step S184, the change in the measured current value of the cooling tower fan over time is determined. If the determination result of step S184 is a step-down, then proceed to step S186, that is, determine the cause of the fault as a broken or fallen belt of the fan driving the cooling tower. If the determination result of step S184 is a gradual decrease, then proceed to step S188. In this disclosure, a step-down can refer to a sudden drop in current over a short period of time, which may exhibit a step shape on the current-time curve; a gradual decrease can refer to a sloped decrease in current over time, which may exhibit a sloping shape on the current-time curve. For example, a slope threshold can be set to distinguish between step-down and gradual decrease.

[0108] It is important to note that when the cause of the malfunction is a belt detachment or breakage, a typical pattern is that the cooling tower fan current does not change significantly when the number of detached or broken belts does not exceed a certain percentage (e.g., half), and only when the number of detached or broken belts exceeds this percentage does the cooling tower fan current change significantly. Therefore, when step S186 is reached, the cooling tower fan belt has already detached or broken more than the aforementioned percentage. The specific number of detached or broken belts that causes a significant change in current depends on the cooling tower model and the usage of the cooling tower fan, among other factors.

[0109] by Figure 3Taking the predicted model shown as an example, this model was applied to a fault detection method to detect a cooling tower with 5 belts. The difference threshold was set to 2A, and the duration threshold was set to 15 minutes. When one belt fell, the measured current value did not change significantly, and the system did not alarm. When two belts fell, the measured current value of the cooling tower fan changed from 39.9A to 30.4A, a difference of 9.5A. This difference from the predicted current value was higher than 2A, and after 15 minutes, an alarm was issued to indicate a fault was detected. Furthermore, based on the fact that the predicted current value was higher than the measured current value and that the measured current value decreased in a stepwise manner, the cause of the fault was determined to be a belt falling or breaking. Continuing to observe the changes in the measured current value, the measured current value of the cooling tower fan remained at around 30A for 3 hours and 15 minutes. Then, the third belt fell off, and the measured current value changed from 28.4A to 23.2A, a variation of 5.2A. This difference from the predicted current value exceeded 2A. After 15 minutes, an alarm was triggered, confirming the fault as a belt falling off or breaking. Continued observation of the measured current value revealed that for the next hour, the measured current value repeatedly dropped to 0A and then fluctuated to around 20A, triggering an alarm. After one hour, the measured current value returned to 0A. With only two belts remaining, the cooling tower could no longer drive the fan, resulting in drastic fluctuations in the measured current value. According to the above embodiment, when two or more cooling tower belts fall off or break, the measured current value of the cooling tower fan changes significantly, triggering an alarm and correctly identifying the fault after meeting the alarm logic. This demonstrates the reliability of the fault detection method disclosed herein.

[0110] In step S188, it is determined whether the ambient temperature is higher than the freezing point of the cooling tower packing. The freezing point of the packing can be, for example, 0 degrees Celsius. Alternatively or additionally, it can be determined whether the current time is a non-winter period. A winter period can be, for example, November to February of the following year, and the specific value can be based on the local climate where the cooling tower is located. If the determination is no, proceed to step S1810, that is, determine the cause of the fault as icing of the cooling tower packing. If the determination is yes, proceed to step S1812, that is, determine the cause of the fault as dirt blockage of the cooling tower packing.

[0111] It is understood that although the embodiments described herein primarily use examples of belt slippage or breakage, packing icing or blockage, and fan blade angle changes, this disclosure is not limited to these. Any fault that can be identified through changes in the fan current can be included. Figure 7 The process is shown.

[0112] In some embodiments, after the cause of the failure is determined, any operation that puts the malfunctioning cooling tower into a repairable state, such as cooling tower shutdown or shutdown, can be performed to facilitate timely maintenance by staff and ensure the safe and stable operation of the data center.

[0113] The difference threshold in steps S16 and S16' can be determined by considering multiple factors. For example, these factors include, but are not limited to, fluctuations in the measured power, frequency, and current values, and delays between the recording times of power, frequency, and current.

[0114] For purposes of non-restrictive description, Figure 8 The diagram shows how the predicted and measured current values ​​obtained using the prediction model change over time in an example application scenario. Figure 8 The light-colored solid line represents the measured current value, and the dark dashed line represents the predicted current value. It can be seen that in most areas, the predicted current value curve and the measured current value curve highly overlap, indicating that the prediction model proposed in this disclosure has high accuracy in predicting wind turbine current. In some areas, such as the area circled by the ellipse, the measured current value exhibits fluctuations. These fluctuations may be caused by grid voltage fluctuations, outdoor wind speed influences, and adjustments to cooling tower operating conditions. Ideally, such current value fluctuations should not trigger alarms. However, as analyzed above, the prediction model's prediction or fitting effect for this data is poor, leading to undesirable alarms. Therefore, a difference threshold can be set based on one or more of the following: grid voltage fluctuations, outdoor wind speed, and cooling tower operating conditions. Specifically, the current fluctuation can be considered when setting the difference threshold. For example, assuming the maximum fluctuation amplitude is I... max Therefore, the difference threshold can be set to be greater than I. max The value of is adjusted to reduce the false alarm rate.

[0115] Ideally, the power, frequency, and current values ​​described in the fault detection methods 10, 10' and current prediction methods 20, 20' of this disclosure are correlated to each other if the measurements of these parameters for each data point occur at the same time. Current prediction models obtained using such data points have better accuracy. Fault detection methods using such data points as input also have a lower false alarm rate.

[0116] However, in reality, there is a time delay in the recording of data from equipment (e.g., frequency inverters). Therefore, power, frequency, and current values ​​that are related to each other may not be acquired at the same time. For example, power, frequency, and current values ​​measured essentially simultaneously may be acquired at a first, second, and third time point, respectively, often with an offset between these points. This can affect data accuracy: for example, the current value at the same power level may differ by approximately 1-2A. To address this, on one hand, if this offset is known, the data can be preprocessed to reconstruct the data acquired at each time point using the time offset, generating corresponding data points that ensure accurate correspondence between frequency, power, and current values. On the other hand, a difference threshold can be set based on this offset or the current deviation caused by it. For example, the difference threshold could be set to 2A. The specific value of this difference threshold can be further determined based on field data in actual applications, thereby reducing the possibility of false alarms.

[0117] Therefore, for methods 10 and 10', in some embodiments, the measured power value, measured frequency value, and measured current value correspond to a first time point, a second time point, and a third time point, respectively, wherein the difference threshold can be set based on the offset of the third time point relative to the first and / or second time points. Similarly, for methods 20 and 20', in some embodiments, the measured power value, measured frequency value, and measured current value in each data point correspond to the same time point; or, the measured power value, measured frequency value, and measured current value in each data point correspond to a first time point, a second time point, and a third time point, respectively, wherein the third time point can have an offset relative to the first and / or second time points. Here, "time" can refer to the data acquisition time.

[0118] In some embodiments, reference Figure 6 The duration threshold in step S16' is set based on the time interval between acquiring the correlated power, frequency, and current values ​​measured for the cooling tower's fan. This time interval can also be referred to as the data acquisition step size. When the data acquisition step size is short (e.g., several seconds), the duration threshold can be set longer compared to the data acquisition step size (e.g., several minutes or several times the data acquisition step size). This prevents false alarms due to erroneous data. Furthermore, considering that belt breakage occurs instantaneously and the time interval between belt drops is not excessively long, the duration threshold should not be set too long. When the data acquisition step size is long (e.g., greater than 10 minutes), the duration threshold can be set shorter, or even zero, meaning an alarm can be triggered immediately upon exceeding the difference threshold without waiting. This improves the immediacy of the alarm.

[0119] On the other hand, this disclosure also provides a fault detection device for cooling towers. Figure 9 A schematic block diagram of a fault detection apparatus 300 according to some embodiments of the present disclosure is shown. Figure 9 As shown, the fault detection device 300 includes a data acquisition unit 320, a controller 340, and a user interface 360. The data acquisition unit 320 is configured to acquire correlated power, frequency, and current values ​​measured for the fan of the cooling tower. The controller 340 is coupled to the data acquisition unit 320 and configured to detect a fault in the cooling tower using any embodiment of methods 10, 10'. The controller 340 issues an alarm via the user interface 360 ​​to indicate that a fault in the cooling tower has been detected. For example, a pop-up alarm may be displayed on the interactive page of the user interface 360. Alternatively, an alarm sound may be emitted via the user interface 360. In some embodiments, the user interface 360 ​​may display or announce the cause of the fault.

[0120] In some embodiments, the fault detection device 300 includes at least one of the following: a data storage device coupled to the data acquisition device 320 and the controller 340 and configured to store data acquired by the data acquisition device 320 and the calculation results of the controller 340; and a building automation (BA) system including one or more of the user interface 360 ​​and the controller 340.

[0121] Figure 10 A non-limiting implementation 300' of the fault detection device 300 and a cooling tower 400 are shown. For example... Figure 10 As shown, the fault detection device 300' may further include a data storage device 380 and a BA system 350. The data storage device 380 is coupled to the data acquisition device 320 and the controller 340, and is configured to store the data acquired by the data acquisition device 320 and the calculation results of the controller 340. The BA system 350 includes a user interface 360 ​​and a controller 340. In some embodiments, when the BA system 350 determines that the measured current value is abnormal, it records the duration of the abnormality. If the accumulated duration reaches a duration threshold, the BA system 350 issues an alarm and stores the alarm record in the data storage device 380. In some embodiments, after manual operation or automatic logic identification, the BA system 350 can cause the controller 340 to send a control signal to implement fault maintenance.

[0122] Cooling tower 400 includes a frequency converter 420 and a fan 450. The fan 450 may include a motor 452 and a fan 454. The motor 452 drives the fan 454 via belt drive. The frequency converter 420 can perform frequency conversion control on the fan 450. A data acquisition unit 320 can acquire measured data for the fan 450 via the frequency converter 420. A predictive model can be built based on the principle of the frequency converter 420. A controller 340 can send parameter control commands to the frequency converter 420, causing the frequency converter 420 to control the operating state of the fan 450 (e.g., fan rotation frequency) according to the received parameter control commands. Cooling tower 400 may also include a start-stop module (not shown). The controller 340 can send shutdown operation commands, fault trip commands, etc., to the start-stop module, causing the cooling tower to stop working and enter a maintainable state.

[0123] Figure 11 Another non-limiting implementation of the fault detection device 300” is shown in relation to the cooling tower 400. (e.g.) Figure 11 As shown, in the fault detection device 300", the BA system 350 only provides a user interface 360 ​​and does not provide a controller 340. That is, the controller 340 is a control component separate from the BA system 350. The settings of the other components are as follows: Figure 10 The fault detection device 300' shown is the same as that of the cooling tower 400, and will not be described again here.

[0124] This disclosure also provides a computing device that may include one or more processors and a memory storing computer-executable instructions, which, when executed by the one or more processors, cause the one or more processors to perform the methods described according to any of the foregoing embodiments of this disclosure. Figure 12As shown, computing device 500 may include one or more processors 520 and memory 540 storing computer-executable instructions that, when executed by the one or more processors 520, cause the one or more processors 520 to perform the methods described according to any of the foregoing embodiments of this disclosure. The one or more processors 520 may be, for example, a central processing unit (CPU) of computing device 500. The one or more processors 520 may be any type of general-purpose processor, or may be a processor specifically designed for cooling tower fault detection and / or fan current prediction, such as an application-specific integrated circuit (“ASIC”). Memory 540 may be coupled to the one or more processors 520 and may include various computer-readable media accessible by the one or more processors 520. In various embodiments, memory 540 described herein may include volatile and non-volatile media, removable and non-removable media. For example, memory 540 may include any combination of random access memory (“RAM”), dynamic RAM (“DRAM”), static RAM (“SRAM”), read-only memory (“ROM”), flash memory, cache memory, and / or any other type of non-transient computer-readable media. The memory 540 may store instructions that, when executed by the processor 520, cause the processor 520 to perform the method described according to any of the foregoing embodiments of the present disclosure.

[0125] This disclosure also provides a non-transient storage medium having computer-executable instructions stored thereon, which, when executed by a computer, cause the computer to perform the methods described according to any of the foregoing embodiments of this disclosure.

[0126] This disclosure also provides a computer program product that may include instructions that, when executed by a processor, can implement the methods described according to any of the foregoing embodiments of this disclosure. The instructions may be any set of instructions that will be executed directly by one or more processors, such as machine code, or any set of instructions that will be executed indirectly, such as a script. The instructions may be stored in an object code format for direct processing by one or more processors, or stored in any other computer language, including scripts or sets of independent source code modules that are interpreted on demand or compiled in advance.

[0127] Figure 13This is a schematic block diagram illustrating a computer system 600 on which embodiments of the present disclosure may be implemented. The computer system 600 includes a bus 602 or other communication mechanism for transmitting information, and a processing device 604 coupled to the bus 602 for processing information. The computer system 600 also includes a memory 606 coupled to the bus 602 for storing instructions to be executed by the processing device 604; the memory 606 may be random access memory (RAM) or other dynamic storage device. The memory 606 may also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by the processing device 604. The computer system 600 also includes a read-only memory (ROM) 608 or other static storage device coupled to the bus 602 for storing static information and instructions for the processing device 604. A storage device 610, such as a magnetic disk or optical disk, is provided and coupled to the bus 602 for storing information and instructions. Computer system 600 may be coupled via bus 602 to output device 612 for providing output to a user, such as, but not limited to, a display (such as a cathode ray tube (CRT) or liquid crystal display (LCD)), speakers, etc. Input device 614, such as a keyboard, mouse, microphone, etc., is coupled to bus 602 for transmitting information and command selections to processing device 604. Computer system 600 may perform embodiments of this disclosure. Consistent with certain implementations of this disclosure, results are provided by computer system 600 in response to processing device 604 executing one or more sequences of one or more instructions contained in memory 606. Such instructions may be read into memory 606 from another computer-readable medium, such as storage device 610. Execution of the sequence of instructions contained in memory 606 causes processing device 604 to perform the methods described herein. Alternatively, the teachings may be implemented using hard-wired circuitry instead of or in combination with software instructions. Therefore, implementations of this disclosure are not limited to any particular combination of hardware circuitry and software. In various embodiments, computer system 600 can be connected across a network to one or more other computer systems, such as computer system 600, to form a networked system via network interface 616. This network may include a private network or a public network such as the Internet. In a networked system, one or more computer systems can store data and supply data to other computer systems. As used herein, the term "computer-readable medium" refers to any medium that participates in providing instructions to processing device 604 for execution. Such media can take many forms, including but not limited to non-volatile media, volatile media, and transmission media. Non-volatile media include, for example, optical discs or magnetic disks such as storage device 610. Volatile media include dynamic memory such as memory 606. Transmission media include coaxial cables, copper wires, and optical fibers, including wiring that includes bus 602.Common forms of computer-readable media or computer program products include, for example, floppy disks, flexible disks, hard disks, magnetic tapes, or any other magnetic media, CD-ROMs, digital video discs (DVDs), Blu-ray discs, any other optical media, thumb drives, memory cards, RAM, PROMs and EPROMs, fast EPROMs, any other memory chips or cartridges, or any other tangible media from which a computer can read. Various forms of computer-readable media may be involved when carrying one or more sequences of one or more instructions to processing device 604 for execution. For example, instructions may initially be carried on a disk of a remote computer. The remote computer may load the instructions into its dynamic memory and transmit the instructions over a telephone line using a modem. A modem local to computer system 600 may receive data over a telephone line and convert the data into an infrared signal using an infrared transmitter. An infrared detector coupled to bus 602 may receive the data carried in the infrared signal and place the data on bus 602. Bus 602 carries the data to memory 606, from which processing device 604 retrieves and executes the instructions. Optionally, instructions received by memory 606 may be stored on storage device 610 before or after execution by processing device 604.

[0128] According to various embodiments, instructions configured to be executed by a processing device to perform a method are stored on a computer-readable medium. The computer-readable medium may be a device for storing digital information. For example, a computer-readable medium includes a compact disc read-only memory (CD-ROM) as known in the art for storing software. The computer-readable medium is accessed by a processor adapted to execute the instructions configured to be executed.

[0129] The foregoing has described one or more exemplary embodiments of this disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0130] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or physical entities, or by products with certain functions. A typical implementation device is a server system. Of course, this disclosure does not exclude the possibility that, with the future development of computer technology, the computer implementing the functions of the above embodiments may be, for example, a personal computer, a laptop computer, an in-vehicle human-machine interaction device, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a game console, a tablet computer, a wearable device, or any combination thereof.

[0131] The terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, product, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, product, or apparatus. Without further limitation, the presence of other identical or equivalent elements in the process, method, product, or apparatus that includes said elements is not excluded. For example, the use of terms such as "first" or "second" to denote names does not indicate any particular order.

[0132] For ease of description, the above devices are described in terms of function, divided into various modules. Of course, when implementing one or more embodiments of this disclosure, the functions of each module can be implemented in one or more software and / or hardware, or a module that performs the same function can be implemented by a combination of multiple sub-modules or sub-units. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0133] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams.

[0134] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0135] Those skilled in the art will understand that one or more embodiments of this disclosure may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of this disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] One or more embodiments of this disclosure can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a particular task or implement a particular abstract data type. One or more embodiments of this disclosure can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0137] The same or similar parts between the various embodiments of this disclosure can be referred to mutually, and each embodiment focuses on describing the differences from other embodiments. In particular, for the apparatus embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to the description of the method embodiments. In the description of this disclosure, the reference to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," "exemplary," etc., means that the specific feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of this disclosure. In this disclosure, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this disclosure and the features of different embodiments or examples.

[0138] Additionally, when used in this disclosure, the terms “here,” “above,” “below,” “below,” “in the following,” “overall,” and similar terms should refer to the entirety of this disclosure and not any particular part thereof. Furthermore, unless expressly stated otherwise or otherwise understood in the context in which they are used, conditional language used herein, such as “may,” “possibly,” “for example,” “like,” etc., is generally intended to express that certain embodiments include, while other embodiments do not, certain features, elements, and / or states. Therefore, such conditional language is not generally intended to imply that one or more embodiments require features, elements, and / or states in any way, or whether such features, elements, and / or states are included or performed in any particular embodiment.

[0139] The above description is merely an embodiment of one or more embodiments of this disclosure and is not intended to limit the scope of the one or more embodiments of this disclosure. Various modifications and variations can be made to the one or more embodiments of this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of the claims.

Claims

1. A method for predicting the current of a cooling tower fan, comprising: Acquire multiple data points for the fan of the cooling tower under normal operating conditions, each data point including a measured power value, a measured frequency value and a measured current value that are correlated with each other; The prediction model is fitted using a first subset of the plurality of data points to obtain the fitting parameters of the prediction model, wherein the prediction model is a function of current with respect to power and frequency; as well as Using the prediction model with the aforementioned fitting parameters, the current value of the fan for the cooling tower is predicted based on the correlated power and frequency values ​​measured for the fan of the cooling tower. The prediction model determines the current based on the product of the power term and the frequency term.

2. The method according to claim 1, wherein, The prediction model is represented as follows: I=CP m f n Wherein, I represents the current, C represents a constant parameter, P represents the power, m represents the power power parameter, f represents the frequency, and n represents the frequency power parameter.

3. The method according to claim 1, wherein, The method includes: The prediction model with the fitted parameters is validated using a second subset of the plurality of data points, which is different from the first subset, wherein the validation includes: For each data point in the second subset, the prediction model is used to determine the predicted current value of that data point based on its measured power and frequency values, and the error value between the predicted current value and the measured current value is also determined. The verification is successful if the ratio between the number of data points in the second subset whose error values ​​exceed the first error threshold and the total number of data points in the second subset does not exceed the first ratio threshold; and Using the validated prediction model, the current value of the fan in the cooling tower is predicted based on the correlated power and frequency values ​​measured for the fan in the cooling tower.

4. The method according to claim 3, wherein, The verification includes: The verification is successful if the ratio between the number of data points in the second subset whose error value exceeds the first error threshold and the total number of data points in the second subset does not exceed the first ratio threshold, and the ratio between the number of data points in the second subset whose error value exceeds the second error threshold and the total number of data points in the second subset does not exceed the second ratio threshold. Wherein, the second error threshold is less than the first error threshold, and the second ratio threshold is greater than the first ratio threshold.

5. The method according to claim 3, wherein, The verification includes: The verification is successful if the ratio between the number of data points in the second subset whose error value exceeds the first error threshold and the total number of data points in the second subset does not exceed the first ratio threshold, and if the data points in the second subset whose error value exceeds the first error threshold are continuously distributed; or The verification is successful if the ratio between the number of data points in the second subset whose error value exceeds the first error threshold and the total number of data points in the second subset does not exceed the first ratio threshold, and the data points in the first subset are continuously distributed.

6. The method according to claim 1, wherein: The measured power value, measured frequency value, and measured current value in each data point correspond to the same time. or The measured power value, measured frequency value, and measured current value in each data point correspond to a first time point, a second time point, and a third time point, respectively, wherein the third time point has an offset relative to the first time point and / or the second time point.

7. A method for detecting faults in a cooling tower, comprising: Obtain the correlated power, frequency, and current values ​​measured for the fan of the cooling tower; The current value of the fan in the cooling tower is predicted using a predictive model based on measured power and frequency values; the predictive model is a function of current with respect to power and frequency. An alarm is triggered if the difference between the predicted and measured current values ​​exceeds a threshold value, indicating a detected fault in the cooling tower. The prediction model determines the current based on the product of the power term and the frequency term.

8. The method according to claim 7, wherein, An alarm is triggered if the difference between the predicted current value and the measured current value exceeds a duration threshold, indicating that a fault has been detected in the cooling tower.

9. The method according to claim 7, wherein, The prediction model is represented as follows: I=CP m f n Wherein, I represents the current, C represents a constant parameter, P represents the power, m represents the power power parameter, f represents the frequency, and n represents the frequency power parameter.

10. The method according to claim 7, wherein, Predicting the current value of the fan of the cooling tower based on the measured power value and the measured frequency value using a predictive model is performed by the method according to any one of claims 1 to 6.

11. The method of claim 8, further comprising: If the difference between the predicted current value and the measured current value is greater than the difference threshold and reaches the duration threshold, the cause of the fault is determined based on one or more of the following: the relationship between the predicted current value and the measured current value, the relationship between the measured current value of the fan of the cooling tower and the change over time in the past time period, and the ambient temperature.

12. The method according to claim 11, wherein, The causes of the fault include: If the predicted current value is greater than the measured current value and the change shows a step-down pattern, the cause of the fault is determined to be a broken or fallen belt of the fan used to drive the cooling tower.

13. The method according to claim 11, wherein, The causes of the fault include: If the predicted current value is greater than the measured current value, the change shows a gradual decrease, and the ambient temperature is not higher than the freezing point of the cooling tower packing, the cause of the fault is determined to be icing of the cooling tower packing; or If the predicted current value is greater than the measured current value, the change shows a gradual decrease, and the ambient temperature is higher than the freezing point of the packing material of the cooling tower, the cause of the fault is determined to be that the packing material of the cooling tower is clogged.

14. The method according to claim 13, wherein, The severity of icing or clogging of the packing material in the cooling tower is determined based on the gradually decreasing slope.

15. The method according to claim 11, wherein, The causes of the fault include: If the predicted current value is less than the measured current value, the cause of the fault is determined to be a change in the blade angle of the fan of the cooling tower.

16. The method according to claim 7, wherein, The difference threshold is set based on one or more of the following: grid voltage fluctuation, outdoor wind speed, and cooling tower operating conditions.

17. The method according to claim 7, wherein, The measured power value, the measured frequency value, and the measured current value correspond to a first time moment, a second time moment, and a third time moment, respectively, and wherein the difference threshold is set based on the offset of the third time moment relative to the first time moment and / or the second time moment.

18. The method according to claim 8, wherein, The duration threshold is set based on the time interval between the acquisition of interrelated power, frequency, and current values ​​measured for the fan of the cooling tower.

19. An apparatus for detecting faults in a cooling tower, comprising: The data acquisition unit is configured to acquire interrelated power, frequency, and current values ​​measured for the fan of the cooling tower. A controller, coupled to the data acquisition unit and configured to detect a fault in the cooling tower using the method according to any one of claims 7-18; as well as The controller uses a user interface to issue alarms to indicate that a fault has been detected in the cooling tower.

20. The apparatus of claim 19, comprising at least one of the following: A data storage device, coupled to the data acquisition device and the controller, and configured to store the data acquired by the data acquisition device and the calculation results of the controller; A building automation (BA) system, wherein the BA system includes one or more of the user interface and the controller.

21. The apparatus according to claim 19, wherein, The cooling tower includes a fan and a frequency converter that performs frequency conversion control on the fan, the data acquisition unit is configured to acquire fan measurement data for the cooling tower from the frequency converter, and the prediction model is built based on the frequency converter.

22. A computing device, comprising: One or more processors; as well as A memory storing computer-executable instructions, which, when executed by the one or more processors, cause the one or more processors to perform the method according to any one of claims 1 to 18.

23. A non-transient storage medium having stored thereon computer-executable instructions, which, when executed by a computer, cause the computer to perform the method according to any one of claims 1-18.

24. A computer program product comprising instructions that, when executed by a processor, implement the method according to any one of claims 1-18.

Images