Method, system and equipment for regulating and controlling output electric power of fan of control cabinet

Abstract

The invention discloses a control cabinet fan output electric power regulation and control method, system and device, and relates to the technical field of electric power regulation and control, and the method comprises the following steps: obtaining multiple groups of internal data and external data of a control cabinet based on time sequence data, the internal data comprises an internal temperature value, an internal humidity value and internal fan output electric power in the internal environment of the control cabinet, and the external data comprises an external temperature value and an external humidity value in the external environment; generating an inner fan output electric power prediction model of the control cabinet based on the multiple groups of internal data and external data; an inner fan output electric power prediction value is calibrated based on the electric power error parameter, an inner fan output electric power calibration value of the control cabinet is generated, the accuracy of the inner fan output electric power prediction value predicted by the inner fan output electric power prediction model is guaranteed, and it is guaranteed that the inner fan output electric power of the regulated and controlled fan in the control cabinet is accurate; and the regulation and control accuracy of the remote control cabinet is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric power regulation, and particularly to a method, a system, and a device for regulating the output electric power of a control cabinet fan. Background Art

[0002] Generally, in order to ensure that the electronic devices inside the control cabinet are in a stable working environment, both the temperature value and the humidity value inside the control cabinet need to be controlled within a certain range (the specific range is adaptively determined by those skilled in the art according to the requirements of the actual electronic devices installed inside the control cabinet). Excessive temperature or excessive humidity will affect the stable operation of the electronic devices installed inside the control cabinet. Therefore, it is necessary to regulate the working electric power of the fan, that is, the output electric power, to ensure that the internal environment of the control cabinet is in a stable state.

[0003] However, as a mechanical device, the fan in the control cabinet will inevitably develop mechanical fatigue or other factors after long-term operation, resulting in a deviation between the preset working electric power during its operation and the working electric power that actually produces the effect (regulating the temperature and humidity inside the control cabinet). In this way, when regulating the output electric power of the fan through the background, inaccurate situations will occur. Summary of the Invention

[0004] In order to overcome the above technical problems, the purpose of the present invention is to provide a method and a device for regulating the output electric power of a control cabinet fan, so as to solve the problem in the prior art that after the fan works for a long time, mechanical fatigue or other factors will inevitably cause a deviation between the preset working electric power during its operation and the working electric power that actually produces the effect (regulating the temperature and humidity inside the control cabinet), resulting in inaccurate regulation of the output electric power of the fan through the background.

[0005] The purpose of the present invention can be achieved by the following technical solutions:

[0006] Specifically, the first aspect of the present invention provides a method for regulating the output electric power of a control cabinet fan, including the following steps:

[0007] S1: Obtain multiple sets of internal data and external data of the control cabinet based on time series data, where the internal data includes the internal temperature value, the internal humidity value, and the internal fan output electric power in the internal environment of the control cabinet, and the external data includes the external temperature value and the external humidity value in the external environment.

[0008] S2: Generate an internal fan output electric power prediction model for the control cabinet based on multiple sets of internal data and external data.

[0009] S3: Generate an internal fan output electric power change model for the control cabinet based on multiple sets of internal data and external data.

[0010] S4: Collect the internal data and external data of the control cabinet in real time, where the internally collected data in real time includes the internal temperature value, the internal humidity value, and the output electric power of the first internal fan;

[0011] S5: Based on the internal data and external data of the control cabinet collected in real time, use the internal fan output electric power change model to generate the output electric power of the second internal fan;

[0012] S6: Generate an electric power error parameter based on the output electric power of the first internal fan and the output electric power of the second internal fan;

[0013] S7: Based on the internal data and external data collected in real time, use the internal fan output electric power prediction model to generate the predicted value of the output electric power of the internal fan of the control cabinet;

[0014] S8: Calibrate the predicted value of the output electric power of the internal fan based on the electric power error parameter to generate the calibrated value of the output electric power of the internal fan of the control cabinet.

[0015] As a further solution of the present invention: the internal temperature value is the internal average temperature value of the control cabinet, and the internal humidity value is the internal average humidity value of the control cabinet.

[0016] As a further solution of the present invention: the S2 includes the following steps:

[0017] S2.1: Pre-collect multiple groups of internal data and external data;

[0018] S2.2: Generate a two-dimensional data set of multiple groups of internal data and external data based on the time series data;

[0019] S2.3: Train the two-dimensional data set using a recurrent neural network model to generate an internal fan output electric power prediction model;

[0020] S2.4: Based on the time series data, use the internal fan output electric power prediction model to predict the output electric power of the internal fan in the internal environment of the control cabinet.

[0021] As a further solution of the present invention: the S2.3 includes the following steps:

[0022] S2.31: Pre-collect N groups of training data, where N is a positive integer, and each group of training data includes feature data and label data, where the feature data is the internal temperature value and internal humidity value in the internal environment of the control cabinet and the external temperature value and external humidity value in the external environment, and the label data is the output electric power of the internal fan;

[0023] S2.32: Convert the feature data into feature vectors. The feature vectors are used as the input of the prediction model for the output electric power of the internal fan. The predicted output electric power of the internal fan by the prediction model for the output electric power of the internal fan with respect to the feature data is used as the output, with the output electric power of the internal fan in the label data as the prediction target and minimizing the sum of all prediction accuracies as the training target.

[0024] S2.33: Train the prediction model for the output electric power of the internal fan until the training stops when the prediction accuracy reaches convergence.

[0025] As a further solution of the present invention: The calculation formula for the prediction accuracy is:

[0026]

[0027] where ui is the prediction accuracy, ai is the output electric power of the internal fan of the prediction model for the output electric power of the internal fan corresponding to the i-th group of feature data, and yi is the output electric power of the internal fan in the label data corresponding to the i-th group of feature data.

[0028] As a further solution of the present invention: The said S6 includes the following steps:

[0029] S6.1: Input the output electric power of the first internal fan and the output electric power of the second internal fan.

[0030] S6.2: Generate an electric power error parameter based on the output electric power of the first internal fan and the output electric power of the second internal fan.

[0031] S6.3: Conduct a sliding statistics on the electric power error parameter to generate an average error.

[0032] S6.4: Use time series data attenuation correction based on the average error.

[0033] As a further solution of the present invention: The said S7 includes the following steps:

[0034] S7.1: Collect the internal data and external data of the control cabinet in real time.

[0035] S7.2: Based on the internal data and external data, use the prediction model for the output electric power of the internal fan to predict the predicted value of the output electric power of the internal fan of the control cabinet based on time series data.

[0036] S7.3: Output the predicted value of the output electric power of the internal fan of the control cabinet based on time series data.

[0037] As a further solution of the present invention: The said S8 includes the following steps:

[0038] S8.1: Input the predicted value of the output electric power of the internal fan of the control cabinet based on time series data.

[0039] S8.2: Generate the calibration value of the output electric power of the internal fan of the control cabinet based on the electric power error parameter;

[0040] S8.3: Regulate the air cooling of the control cabinet based on the calibration value of the output electric power of the internal fan.

[0041] Specifically, the second aspect of the present invention provides a control system for regulating the output electric power of the fan of the control cabinet, including:

[0042] A temperature acquisition module, configured to acquire the internal temperature value and the external temperature value of the control cabinet and transmit them to the processing module;

[0043] A humidity acquisition module, configured to acquire the internal humidity value and the external humidity value of the control cabinet and transmit them to the processing module;

[0044] A fan module, configured to acquire the output electric power of the internal fan of the control cabinet and transmit it to the processing module;

[0045] A processing module, which is configured to execute the method for regulating the output electric power of the fan of the control cabinet according to any one of claims 1-8 based on the characteristics acquired by the temperature acquisition module, the humidity acquisition module, and the fan module.

[0046] Specifically, the third aspect of the present invention provides a control device for regulating the output electric power of the fan of the control cabinet, including:

[0047] A processor;

[0048] A memory, which is used to store an executable program, wherein the processor is configured to read the executable program in the memory and execute the executable program to implement the method for regulating the output electric power of the fan of the control cabinet according to any one of claims 1-8.

[0049] Advantages of the present invention:

[0050] In the present invention, the electric power error parameter can be generated through the internal fan output electric power prediction model and the internal fan output electric power change model. Based on the electric power error parameter, the predicted value of the internal air cooling electric power is calibrated to generate the calibrated value of the internal air cooling electric power of the control cabinet. By calibrating the predicted value of the internal air cooling electric power with the electric power error parameter, the accuracy of the predicted value of the internal air cooling electric power predicted by the internal fan output electric power prediction model can be ensured, and the accuracy of the output electric power of the internal fan of the fan inside the control cabinet to be regulated can be guaranteed, thereby improving the regulation accuracy of the remote control cabinet. Description of the Drawings

[0051] The following further describes the present invention with reference to the drawings.

[0052] Figure 1 It is a flowchart of the method for regulating the output electric power of the fan of the control cabinet of the present invention;

[0053] Figure 2 It is a schematic diagram of the generation process of the prediction model of the output electric power of the internal fan in the regulation method of the output electric power of the control cabinet fan of the present invention;

[0054] Figure 3 It is a schematic diagram of the training process of the recurrent neural network model in the regulation method of the output electric power of the control cabinet fan of the present invention;

[0055] Figure 4 It is a schematic diagram of the generation process of the electric power error parameter in the regulation method of the output electric power of the control cabinet fan of the present invention;

[0056] Figure 5 It is a system structure block diagram of the regulation system of the output electric power of the control cabinet fan of the present invention. Specific embodiments

[0057] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0058] As an embodiment of the present invention, as Figures 1 - 5 shown, a regulation method for the output electric power of the control cabinet fan is disclosed. When the present invention monitors the humidity inside the control cabinet, it is carried out according to the following steps, specifically:

[0059] S1: First, it is necessary to obtain multiple sets of internal data and external data of the control cabinet based on time-series data. Among them, the internal data includes the internal temperature value, internal humidity value, and output electric power of the internal fan in the control cabinet environment, and the external data includes the external temperature value and external humidity value in the external control environment. It should be noted that the internal temperature value, internal humidity value, and output electric power of the internal fan in the control cabinet environment refer to the temperature, humidity inside the control cabinet, and the working electric power of the internal air-cooled motor. Since the working electric powers of air-cooled motors of different specifications are different, the output electric power of the internal fan needs to be adaptively determined according to the working electric power of the air-cooled motor installed inside the control cabinet. Temperature sensors, humidity sensors, and fans (air-cooled motors) need to be installed inside the control cabinet. The installation positions and data of the temperature sensors and humidity sensors can be adaptively adjusted by those skilled in the art according to the space size inside the control cabinet and the positions of the internal equipment, as long as the temperature and humidity inside the control cabinet can be accurately collected. The fan is usually installed at the end or side of the control cabinet to provide a cooling air source for the internal environment of the control cabinet. During the operation of the fan, the fan can bring the air in the external environment of the control cabinet into the interior of the control cabinet. The air in the external environment includes its temperature and humidity (i.e., the external temperature value and external humidity value in the external data). Therefore, the temperature and humidity inside the control cabinet are closely related to the temperature and humidity in the air of the external environment and the working electric power of the fan (output electric power of the internal fan). Usually, in order for the control cabinet to ensure that its internal electronic equipment is in a stable working environment, the temperature value and humidity value inside the control cabinet need to be controlled within a certain range (the specific range is adaptively determined by those skilled in the art according to the requirements of the actual electronic equipment installed inside the control cabinet). Excessive temperature or humidity will affect the stable operation of the electronic equipment installed inside the control cabinet. Therefore, by adjusting the working electric power of the fan, that is, the output electric power of the internal fan, the internal environment of the control cabinet can be ensured to be in a stable state (i.e., the internal temperature value and internal humidity value are within a controllable range).

[0060] In order to ensure that the internal environment of the control cabinet can finally be within a controllable range, it is necessary to collect the internal data and external data of the control cabinet (adaptively set by those skilled in the art according to needs). The meaning of "based on time-series data" is that the internal data and external data of the control cabinet need to be marked with the corresponding time points when collected. In order to reduce the collection cost and collection efficiency of the internal data and external data of the control cabinet, the internal data and external data of different control cabinets can be collected simultaneously to ensure the quantity of the internal data and external data.

[0061] S2: Generate a prediction model for the output electric power of the internal fan of the control cabinet based on multiple sets of internal data and external data. It should be noted that the quantity of the multiple sets of internal data and external data is adaptively adjusted by those skilled in the art according to the training situation of the specific prediction model for the output electric power of the internal fan, as long as the training of the prediction model for the output electric power of the internal fan is successful;

[0062] S3: Generate a change model for the output electric power of the internal fan of the control cabinet based on multiple sets of internal data and external data. The quantity of the multiple sets of internal data and external data is adaptively adjusted by those skilled in the art according to the training situation of the specific change model for the output electric power of the internal fan, as long as the training of the change model for the output electric power of the internal fan is successful;

[0063] The generation method of the change model for the output electric power of the internal fan is to perform fitting using a computer, specifically, multiple regression fitting can be adopted or a machine learning model (random forest or neural network) can be used:

[0064] Obtain the change relationship between the internal humidity value, internal temperature value, external humidity value, and external temperature value and the output electric power of the internal fan, thereby obtaining the change model for the output electric power of the internal fan;

[0065] S4: Collect the internal data and external data of the control cabinet in real time. The internal data collected in real time includes the internal temperature value, the output electric power of the first internal fan, and the output electric power of the internal fan. It should be noted that the internal data of the control cabinet is determined through the temperature sensor, humidity sensor, and fan (the working electric power of the fan) inside the control cabinet, and the external data of the control cabinet is determined through the temperature sensor and humidity sensor outside the control cabinet. Although the internal data of the control cabinet, namely the internal temperature value, internal humidity value, and the output electric power of the first internal fan, can be quickly determined through the temperature sensor, humidity sensor, and fan inside the control cabinet, after the fan has been working for a long time, due to its own mechanical fatigue or other factors, it will cause a certain error, that is, the working electric power of the controlled fan is inconsistent with the actual working electric power of the fan. In this way, the output electric power of the first internal fan will not be accurate enough, and the control of the internal temperature value and internal humidity value in the internal environment of the control cabinet will also be inaccurate, thereby affecting the operation of the electronic devices installed inside the control cabinet;

[0066] S5: Based on the internally collected data and externally collected data of the control cabinet in real time, use the internal fan output electric power change model to generate the second internal fan output electric power of the control cabinet. It should be noted that at this time, based on the internally collected data and externally collected data of the control cabinet in real time, use the internal fan output electric power change model to generate the second internal fan output electric power of the control cabinet. This second internal fan output electric power is generated based on the change of the internal fan output electric power change model, avoiding the errors generated by the fan itself. Because by using the internally collected data and externally collected data of the control cabinet in real time and using the internal fan output electric power change model to generate the second internal fan output electric power of the control cabinet, it is the actual working electric power of the fan;

[0067] S6: Generate an electric power error parameter based on the first internal fan output electric power and the second internal fan output electric power. By generating an electric power error parameter through the first internal fan output electric power and the second internal fan output electric power, the working electric power of the fan inside the control cabinet can be calibrated, ensuring that the internal fan output electric power of the controlled control cabinet is accurate, and improving the control accuracy of the remote control cabinet;

[0068] S7: Based on the internally collected data and externally collected data in real time, use the internal fan output electric power prediction model to predict the predicted value of the internal air-cooling electric power of the control cabinet. By using the internally collected data and externally collected data in real time and using the internal fan output electric power prediction model, the internal air-cooling electric power of the fan in the future internal environment of the control cabinet can be predicted. Those skilled in the art can preset the humidity threshold (temperature threshold) in the internal environment of the control cabinet. If the internal humidity value (temperature value) in the control cabinet regulated by the predicted internal air-cooling electric power predicted value is greater than or equal to the humidity threshold (temperature threshold), it means that the control cabinet will be in an unfavorable working environment. The internal air-cooling electric power of the fan can be controlled in advance to avoid the control cabinet being in an unfavorable working environment and keep the control cabinet in a stable working environment all the time;

[0069] S8: Calibrate the predicted value of the internal air-cooling electric power based on the electric power error parameter to generate the calibrated value of the internal air-cooling electric power of the control cabinet. By calibrating the predicted value of the internal air-cooling electric power with the electric power error parameter, the accuracy of the predicted value of the internal air-cooling electric power predicted by the internal fan output electric power prediction model can be ensured, ensuring that the internal fan output electric power of the fan inside the controlled control cabinet is accurate, and improving the control accuracy of the remote control cabinet.

[0070] As an embodiment of the present invention, such as Figures 1 - 5As shown, the internal temperature value is the average internal temperature value of the control cabinet, and the internal humidity value is the average internal humidity value of the control cabinet. Multiple temperature sensors and humidity sensors need to be set inside and outside the control cabinet. The installation positions and data of the temperature sensors and humidity sensors are adaptively adjusted by those skilled in the art according to the internal space size and the positions of the internal devices of the control cabinet, as long as the temperature and humidity inside the control cabinet can be accurately collected. Specifically, the average temperature collected by multiple temperature sensors in the internal environment of the control cabinet is used as the internal temperature value of the control cabinet, and the average humidity collected by multiple humidity sensors in the internal environment of the control cabinet is used as the internal humidity value of the control cabinet. Additionally, it should be noted that the values collected by multiple temperature sensors or humidity sensors can also be differentiated by coefficients according to the installation positions of the devices inside the control cabinet. For example, the temperature sensors or humidity sensors close to the positions of the devices inside the control cabinet (core positions) give high coefficients, such as 1.2 - 1.5, and the temperature sensors or humidity sensors close to the edge positions inside the control cabinet give low coefficients, such as 0.5 - 0.8. Specifically, it is adaptively adjusted by those skilled in the art according to the on-site situation of the control cabinet and the internal space situation.

[0071] As an embodiment of the present invention, as Figures 1 - 5 shown, S2 includes the following steps:

[0072] S2.1: Pre-collect multiple groups of internal data and external data. It should be noted that the number of internal data and external data collected is adaptively adjusted by those skilled in the art according to the specific training situation of the internal fan output electric power prediction model. In order to reduce the collection cost and collection efficiency of the internal data and external data of the control cabinet, the internal data and external data of different control cabinets can be collected simultaneously to ensure the quantity of the internal data and external data.

[0073] S2.2: Generate two-dimensional data sets of multiple groups of internal data and external data based on the time-series data. It should be noted that the internal temperature value, internal humidity value, internal fan output electric power, external temperature value, and external humidity value in the internal data and external data are marked with time points, and the time points should include year, month, day, and specific time points, which can be accurate to seconds. The internal temperature value, internal humidity value, internal fan output electric power, external temperature value, and external humidity value marked by time points are grouped together to obtain the two-dimensional data set.

[0074] S2.3: Train the two-dimensional data set using a recurrent neural network model to generate a predicted model for the output electric power of the internal fan. It should be noted that a recurrent neural network (RNN) is a deep learning model specifically designed to process sequential data. Its core function is to capture the temporal dependencies and dynamic changes in the data. Different from traditional neural networks, RNNs introduce a recurrent structure that enables the network to pass the hidden state from the previous moment to the current moment, thereby retaining historical information and influencing the current output. This feature makes it perform well in fields such as natural language processing, speech recognition, and time series prediction. The architecture of the RNN encodes and transmits information in the time dimension through the recurrent connection of the hidden state. The input at each time step includes not only the current data but also the hidden state from the previous moment, thereby dynamically updating the memory content of the network. This mechanism enables the RNN to simulate the way humans process information - gradually understanding the current situation based on historical experience.

[0075] Therefore, through the recurrent neural network model, the operating electric power (output electric power of the internal fan) of the fan in the control cabinet based on the time series data can be obtained.

[0076] S2.4: Based on the time series data, use the predicted model for the output electric power of the internal fan to predict the internal humidity value in the control cabinet environment. It should be noted that those skilled in the art can train the output electric power of the internal fan in the control cabinet environment based on the time series data according to the specific training situation of the predicted model for the output electric power of the internal fan. To ensure the accuracy of the predicted output electric power of the internal fan in the control cabinet environment, a threshold should be set for the time length of the predicted time series data, such as 24 hours to 36 hours, that is, predict the output electric power of the internal fan in the control cabinet environment 24 hours to 36 hours in the future. When the internal temperature value (internal humidity value) of the control cabinet corresponding to the predicted output electric power of the internal fan reaches the preset temperature threshold (humidity threshold) in the art, early warning can be given in advance through the monitoring background, improving the safety supervision of the control cabinet.

[0077] As an implementation manner of the present invention, as Figures 1 - 5 shown, S2.3 includes the following steps:

[0078] S2.31: Pre-collect N sets of training data, where N is a positive integer. Each set of training data includes feature data and label data. The feature data is the internal temperature value, internal humidity value in the control cabinet environment, external temperature value, and external humidity value in the external environment, and the label data is the output electric power of the internal fan.

[0079] It should be noted that the internal temperature value is the internal temperature value in the control cabinet environment collected by the control cabinet when each set of training data is collected.

[0080] The internal humidity value is the internal humidity value in the internal environment collected by the control cabinet when each group of training data is collected;

[0081] The external temperature value is the external temperature value in the external environment collected by the control cabinet when each group of training data is collected;

[0082] The external humidity value is the external humidity value in the external environment collected by the control cabinet when each group of training data is collected;

[0083] The label data is the output electric power of the internal fan in the internal environment collected by the control cabinet when each group of training data is collected;

[0084] S2.32: Convert the feature data into a feature vector. The feature vector is used as the input of the internal fan output electric power change model, and the internal fan output electric power predicted by the internal fan output electric power change model for the feature data is used as the output. The output electric power of the internal fan in the label data is used as the prediction target, and minimizing the sum of all prediction accuracies is used as the training target. The calculation formula for the prediction accuracy is: where ui is the prediction accuracy, ai is the internal fan output electric power predicted by the internal fan output electric power change model corresponding to the i-th group of feature data, and yi is the internal fan output electric power in the label data corresponding to the i-th group of feature data;

[0085] S2.33: Train the internal fan output electric power prediction model until the prediction accuracy reaches convergence and then stop training. It should be noted that the convergence criterion is adaptively adjusted by those skilled in the art according to the specific model training situation. For example, it can be set that ui≦0.01, that is, when the sum of the prediction accuracies is 0.01, the model training is considered completed.

[0086] As an implementation manner of the present invention, as Figures 1 - 5 shown, S6 includes the following steps:

[0087] S6.1: Input the first internal fan output electric power and the second internal fan output electric power, then there is:

[0088]

[0089] where △W t is the instantaneous electric power error at the t-th time point, is the internal fan output electric power predicted by the internal fan output electric power prediction model at the t-th time point, is the internal fan output electric power of the internal fan in the control cabinet at the t-th time point;

[0090] S6.2: Generate an electric power error parameter based on the output electric power of the first internal fan and the output electric power of the second internal fan, where △W can be the electric power error parameter of the output electric power of the first internal fan and the output electric power of the second internal fan;

[0091] S6.3: Conduct a sliding statistics on the electric power error parameter to generate an average error. Specifically:

[0092]

[0093] where M is the length of the sliding window. For example, when M = 3, it means calculating the average error of the instantaneous electric power errors at 3 consecutive time points each time. t is greater than or equal to M - 1, μ △W is the average error of the window period, σ △W is the standard deviation of the error, △W k is the instantaneous electric power error at time point t = k;

[0094] S6.4: Use time series data decay correction based on the average error. Specifically:

[0095]

[0096] α = e -△t / τ;

[0097] where, is the corrected error at time point t, α is the forgetting factor, e is the natural constant, △t is the data acquisition interval, τ is the time constant, for example, a time (3600 seconds), and τ is specifically set arbitrarily by those skilled in the art;

[0098] Then, based on the corrected error calibrate

[0099]

[0100] is the calibrated value of the output electric power of the internal fan after calibration.

[0101] As an embodiment of the present invention, as Figures 1 - 5 shown, S7 includes the following steps:

[0102] S7.1: Real-time collect the internal data and external data of the control cabinet. The internal temperature value, internal humidity value, and the output electric power of the internal fan in the internal environment of the control cabinet are collected in real time through the temperature sensor, humidity sensor, and fan set inside the control cabinet. The external temperature value and external humidity value in the external environment of the control cabinet are collected in real time through the temperature sensor and humidity sensor set outside;

[0103] S7.2: Based on internal data and external data, use the internal fan output electric power prediction model to predict the predicted value of the internal air-cooling electric power of the control cabinet based on time-series data. Specifically, mark the internal temperature value, internal humidity value, internal fan output electric power, external temperature value, and external humidity value in the internal data and external data with time points, which should include year, month, day, and specific time points, and can be accurate to seconds. Combine the internal temperature value, internal humidity value, internal fan output electric power, external temperature value, and external humidity value marked by time points to obtain a two-dimensional data set. Train the two-dimensional data set using a recurrent neural network model to generate an internal fan output electric power prediction model;

[0104] S7.3: Output the predicted value of the internal air-cooling electric power of the control cabinet based on time-series data. Based on time-series data, use the internal fan output electric power prediction model to predict the predicted value of the internal air-cooling electric power in the internal environment of the control cabinet.

[0105] As an implementation manner of the present invention, as Figures 1 - 5 shown, S8 includes the following steps:

[0106] S8.1: Input the predicted value of the internal air-cooling electric power of the control cabinet based on time-series data, that is, input the predicted value of the internal air-cooling electric power predicted by the internal fan output electric power prediction model into the calibration system;

[0107] S8.2: Generate the calibration value of the internal air-cooling electric power of the control cabinet based on the electric power error parameter. Specifically: Input the first internal fan output electric power and the second internal fan output electric power, then there is:

[0108]

[0109] where △W t is the instantaneous electric power error at the t-th time point, is the internal fan output electric power predicted by the internal fan output electric power prediction model at the t-th time point, is the internal fan output electric power of the internal fan of the control cabinet at the t-th time point;

[0110] Generate an error parameter based on the first internal fan output electric power and the second internal fan output electric power, where △W can be the electric power error parameter of the first internal fan output electric power and the second internal fan output electric power;

[0111] Perform sliding statistics on the electric power error parameter to generate an average error. Specifically:

[0112]

[0113] Where M is the length of the sliding window. For example, when M = 5, it means calculating the average error of the instantaneous electric power error at 5 consecutive time points each time. t is greater than or equal to M - 1, μ △W is the average error in the window period, σ △W is the standard deviation of the error, △W k is the instantaneous electric power error at time point t when k;

[0114] Based on the average error, use time series data attenuation correction. Specifically:

[0115]

[0116] α = e -△t / τ;

[0117] Among them, is the corrected error at time point t, α is the forgetting factor, and its value range is (0, 1), e is the natural constant, △t is the data acquisition interval, τ is the time constant, such as a time (3600 seconds), and τ is specifically set arbitrarily by those skilled in the art;

[0118] Then based on the corrected error Calibrate

[0119]

[0120] is the calibrated output electric power calibration value of the internal fan of the control cabinet;

[0121] S8.3: Output the calibrated output electric power calibration value of the internal fan of the control cabinet, and use the calibrated internal air-cooled electric power value as the actual working electric power of the fan of the control cabinet. By calibrating the predicted value of the internal air-cooled electric power through the electric power error parameter, the accuracy of the predicted value of the internal air-cooled electric power predicted by the output electric power prediction model of the internal fan can be ensured. Then, the calibrated internal air-cooled electric power value is transmitted to the monitoring background, and the monitoring background can precisely control the fan to adjust the internal environment of the control cabinet through the calibrated internal air-cooled electric power value, ensuring the normal operation of the control cabinet.

[0122] As an embodiment of the present invention, as Figures 1 - 5 shown, a control system for regulating the output electric power of the fan of the control cabinet is disclosed, including:

[0123] A temperature acquisition module for acquiring the internal temperature value and the external temperature value of the control cabinet and transmitting them to the processing module. It should be noted that the temperature acquisition module is a temperature sensor arranged in the internal environment and the external environment of the control cabinet;

[0124] The humidity acquisition module is used to acquire the internal humidity value and external humidity value of the control cabinet and transmit them to the processing module. The humidity acquisition module is the humidity sensor set in the internal environment and external environment of the control cabinet. The installation positions and data of the temperature sensor and humidity sensor are adaptively adjusted by those skilled in the art according to the internal space size and internal equipment position of the control cabinet to ensure that the temperature and humidity inside the control cabinet can be accurately acquired. Specifically, the average temperature collected by multiple temperature sensors in the internal environment of the control cabinet is used as the internal temperature value of the control cabinet, and the average humidity collected by multiple humidity sensors in the internal environment of the control cabinet is used as the internal humidity value of the control cabinet. Additionally, it should be noted that the coefficients of the values collected by multiple temperature sensors or humidity sensors can also be differentiated according to the equipment setting positions inside the control cabinet. For example, the temperature sensor or humidity sensor close to the internal equipment position (core position) of the control cabinet gives a high coefficient, such as 1.2 - 1.5, and the temperature sensor or humidity sensor close to the internal edge position of the control cabinet gives a low coefficient, such as 0.5 - 0.8. Specifically, it is adaptively adjusted by those skilled in the art according to the on-site situation of the control cabinet and the internal space situation;

[0125] The fan module is used to acquire the output electric power of the internal fan of the control cabinet and transmit it to the processing module. It should be noted that the fan module is the fan set inside the control cabinet. The fan is usually set at the end or side of the control cabinet to provide a cooling air source for the internal environment of the control cabinet. During the operation of the fan, the fan can bring the air in the external environment of the control cabinet into the interior of the control cabinet;

[0126] The processing module is used to execute the aforementioned control method for the output electric power of the control cabinet fan based on the characteristics collected by the temperature acquisition module, humidity acquisition module, and fan module, that is, based on the internal data and external data of the control cabinet in the time-series data, to achieve the calibration of the predicted value of the internal air-cooling electric power through the electric power error parameter, which can ensure the accuracy of the predicted value of the internal air-cooling electric power predicted by the internal fan output electric power prediction model. Then, the calibrated value of the internal air-cooling electric power is transmitted to the monitoring background, and the monitoring background can precisely control the fan to regulate the internal environment of the control cabinet through the calibrated value of the internal air-cooling electric power to ensure the normal operation of the control cabinet.

[0127] As an embodiment of the present invention, as Figures 1 - 5 shown, a control device for the output electric power of the control cabinet fan is disclosed, including a processor and a storage. The storage includes, but is not limited to, USB flash drives, hard disks, and other devices that can carry executable programs. The executable program executes the control method for the output electric power of the control cabinet fan in the above embodiment. The processor includes, but is not limited to, a microprocessor or a PLC, etc., which can read and run the executable program in the storage.

[0128] The above has described in detail an embodiment of the present invention, but the above content is only a preferred embodiment of the present invention and cannot be considered as defining the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention shall still fall within the scope covered by the patent of the present invention.

Claims

1. A method for regulating the output electric power of a control cabinet fan, characterized in that, Air cooling applied to a control cabinet includes the following steps: S1: Obtain multiple sets of internal data and external data of the control cabinet based on time-series data, where the internal data includes the internal temperature value, internal humidity value, and internal fan output electric power in the internal environment of the control cabinet, and the external data includes the external temperature value and external humidity value in the external environment; S2: Generate an internal fan output electric power prediction model for the control cabinet based on multiple sets of internal data and external data; S3: Generate an internal fan output electric power change model for the control cabinet based on multiple sets of internal data and external data; S4: Real-time collect the internal data and external data of the control cabinet, where the real-time collected internal data includes the internal temperature value, internal humidity value, and the first internal fan output electric power; S5: Based on the real-time collected internal data and external data of the control cabinet, use the internal fan output electric power change model to generate the second internal fan output electric power; S6: Generate an electric power error parameter based on the first internal fan output electric power and the second internal fan output electric power; S7: Based on the real-time collected internal data and external data, use the internal fan output electric power prediction model to generate a predicted value of the internal fan output electric power of the control cabinet; S8: Calibrate the predicted value of the internal fan output electric power based on the electric power error parameter to generate a calibrated value of the internal fan output electric power of the control cabinet.

2. The method for regulating the output electric power of the control cabinet fan according to claim 1, wherein The internal temperature value is the internal average temperature value of the control cabinet, and the internal humidity value is the internal average humidity value of the control cabinet.

3. The method for regulating the output electric power of the control cabinet fan according to claim 1, characterized in that The S2 includes the following steps: S2.1: Pre-collect multiple sets of internal data and external data; S2.2: Generate a two-dimensional data set of multiple sets of internal data and external data based on time-series data; S2.3: Train the two-dimensional data set using a recurrent neural network model to generate an internal fan output electric power prediction model; S2.4: Based on time-series data, use the internal fan output electric power prediction model to predict the internal fan output electric power in the internal environment of the control cabinet.

4. The method for regulating the output electric power of the control cabinet fan according to claim 3, characterized in that, The S2.3 includes the following steps: S2.31: Pre-collect N sets of training data, where N is a positive integer, and each set of training data includes feature data and label data. The feature data is the internal temperature value and internal humidity value in the internal environment of the control cabinet and the external temperature value and external humidity value in the external environment, and the label data is the internal fan output electric power; S2.32: Convert the feature data into feature vectors. The feature vectors are used as the input of the internal fan output electric power change model, and the internal fan output electric power predicted by the internal fan output electric power prediction model for the feature data is used as the output. The internal fan output electric power in the label data is used as the prediction target, and minimizing the sum of all prediction accuracies is used as the training target; S3.33: Train the internal fan output electric power prediction model until the prediction accuracy reaches convergence and then stop training.

5. The control method for regulating the output electric power of the control cabinet fan according to claim 4, wherein, The calculation formula for the prediction accuracy is: where ui is the prediction accuracy, ai is the internal fan output electric power of the internal fan output electric power prediction model corresponding to the i-th group of feature data, and yi is the internal fan output electric power in the label data corresponding to the i-th group of feature data.

6. The regulation method of the output electric power of the control cabinet fan according to claim 1, characterized in that, The S6 includes the following steps: S6.1: Input the output electric power of the first internal fan and the output electric power of the second internal fan; S6.2: Generate an electric power error parameter based on the output electric power of the first internal fan and the output electric power of the second internal fan; S6.3: Conduct a sliding statistics on the electric power error parameter to generate an average error; S6.4: Use time series data attenuation correction based on the average error.

7. The regulation method of the output electric power of the control cabinet fan according to claim 1, characterized in that, The said S7 includes the following steps: S7.1: Real-time collect the internal data and external data of the control cabinet; S7.2: Based on the internal data and external data, use the internal fan output electric power prediction model to predict the predicted value of the internal fan output electric power of the control cabinet based on time series data; S7.3: Output the predicted value of the internal fan output electric power of the control cabinet based on time series data.

8. The regulation method for the output electric power of the control cabinet fan according to claim 7, characterized in that, The said S8 includes the following steps: S8.1: Input the predicted value of the internal fan output electric power of the control cabinet based on time series data; S8.2: Generate a calibration value of the internal fan output electric power of the control cabinet based on the electric power error parameter; S8.3: Regulate the air cooling of the control cabinet based on the calibration value of the internal fan output electric power.

9. A control system for regulating the output electric power of a control cabinet fan, characterized in that, Comprising: A temperature acquisition module, which is used to acquire the internal temperature value and the external temperature value of the control cabinet and transmit them to the processing module; A humidity acquisition module, which is used to acquire the internal humidity value and the external humidity value of the control cabinet and transmit them to the processing module; A fan module, which is used to acquire the output electric power of the internal fan of the control cabinet and transmit it to the processing module; A processing module, which is used to execute the control method for the output electric power of the control cabinet fan described in any one of claims 1-8 based on the features collected by the temperature acquisition module, the humidity acquisition module and the fan module.

10. A control device for regulating the output electric power of a control cabinet fan, characterized in that, Comprising: A processor; A storage, which is used to store executable programs, wherein the processor is used to read the executable programs in the storage and execute the executable programs to implement the control method for the output electric power of the control cabinet fan described in any one of claims 1-8.

Images