Electronic device, cleaning method thereof and judgment method of fan state judgment model of electronic device

Abstract

The electronic device comprises a fan, an embedded controller, a fan state judgment model and a processor. The fan has operating parameters. The embedded controller is used for controlling the fan to operate according to the control signal. The fan state judgment model is used for judging whether the fan is in a normal state or an abnormal state according to the control signal and the operation parameters. The processor is used for inputting the control signals and the operation parameters into the fan state judgment model. The embedded controller is further configured to control the fan to reverse based on the fan being in an abnormal condition.

Description

Technical Field

[0001] The invention relates to an electronic device, a cleaning method thereof and a determination method of a fan status determination model thereof. Background Art

[0002] Electronic devices are usually equipped with fans for heat dissipation. However, when the fan is covered with dirt or dust, the heat dissipation function of the fan will be reduced, and at the same time, problems such as fan temperature rise and noise increase will occur. Therefore, proposing a technology that can improve the above problems is one of the goals of the industry in this field. Summary of the invention

[0003] The embodiments of the present invention are related to an electronic device, a cleaning method thereof and a determination method of a fan status determination model thereof, which can improve the above-mentioned conventional problems.

[0004] According to one embodiment of the present invention, an electronic device is provided. The electronic device includes a fan, an embedded controller, a fan state judgment model and a processor. The fan has an operating parameter. The embedded controller is used to: control the operation of the fan with a control signal. The fan state judgment model is used to: determine whether the fan is in a normal state or an abnormal state based on the control signal and the operating parameter. The processor is used to: input the control signal and the operating parameter into the fan state judgment model. The embedded controller is further used to: control the fan to reverse based on the fan being in an abnormal state.

[0005] According to another embodiment of the present invention, a cleaning method is provided. The cleaning method is used for a fan. The cleaning method comprises the following steps: controlling a fan to operate with a control signal, wherein the fan has an operating parameter; inputting the control signal and the operating parameter into a fan state judgment model; the fan state judgment model judges whether the fan is in a normal state or an abnormal state according to the control signal and the operating parameter; and, based on the fan being in an abnormal state, controlling the fan to reverse.

[0006] According to another embodiment of the present invention, a method for determining a fan status model is provided, which comprises the following steps: receiving a control signal and an operating parameter of a fan; determining a cleanliness status of the fan according to the control signal and the operating parameter; and outputting the cleanliness status.

[0007] In order to better understand the above and other aspects of the present invention, embodiments are given below and described in detail with reference to the accompanying drawings: BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A functional block diagram of an electronic device according to an embodiment of the present invention is shown.

[0009] Figure 2 Draw Figure 1Schematic diagram of the training of the fan status judgment model.

[0010] Figure 3 FIG. 4 is a functional block diagram of an electronic device according to another embodiment of the present invention.

[0011] Figure 4 Draw Figure 1 Flow chart of a method for cleaning an electronic device.

[0012] Description of Reference Numerals

[0013] 100,200: Electronic devices

[0014] 110: Fan

[0015] 120:Embedded Controller

[0016] 130: Fan status judgment model

[0017] 140: Processor

[0018] 150: Operating System

[0019] C1: First current value

[0020] C2: Second current value

[0021] P: Operation parameters

[0022] P1: Actual speed

[0023] P2: Operation noise

[0024] P3: Temperature

[0025] P4: Frequency of use

[0026] P5: Usage time

[0027] P6: Cleaning cycle

[0028] S1: Control signal

[0029] S110~S150: Steps DETAILED DESCRIPTION

[0030] Please refer to Figure 1 , which illustrates a functional block diagram of an electronic device 100 according to an embodiment of the present invention. The electronic device 100 is, for example, a desktop computer, a notebook computer or other electronic devices equipped with a fan.

[0031] like Figure 1As shown, the electronic device 100 includes a fan 110, an embedded controller (EC) 120, a fan state determination model 130, a processor 140, and an operating system (OS) 150. The fan 110 has an operating parameter P. The embedded controller 120 is used to control the operation of the fan 110 with a control signal S1. The fan state determination model 130 is used to determine whether the fan 110 is in a normal state or an abnormal state according to the control signal S1 and the operating parameter P. The processor 140 is used to input the control signal S1 and the operating parameter P into the fan state determination model 130. The embedded controller 120 is further used to control the fan 110 to reverse based on the abnormal state of the fan 110 to clean the fan 110.

[0032] In one embodiment, the control signal S1 and the operating parameter P are used as inputs of the fan state judgment model 130, and the fan state judgment model 130 outputs a judgment result accordingly. When the judgment result is 0, the fan 110 is in a normal state, and when the judgment result is 1, the fan 110 is in an abnormal state. In this embodiment, the "abnormal state" herein is, for example, that the cleanliness of the fan is low, so it needs to be cleaned, and the "normal state" herein is, for example, that the fan is in a state where it can be operated normally (that is, the cleanliness of the fan meets expectations), so it does not need to be cleaned.

[0033] In one embodiment, the operating parameter P includes, for example, at least one of the actual rotation speed P1, the operating noise P2, the temperature P3, the operating frequency P4, the operating time P5, and the cleaning cycle P6 of the fan 110. The control signal S1 is pulse width modulation (PWM). The embedded controller 120 is electrically connected to the fan 110 and outputs the control signal S1 to the fan 110 to control the rotation speed and direction of the fan 110. The embedded controller 120 can detect or obtain at least one operating parameter from the fan 110, such as the actual rotation speed P1, the operating frequency P4, and / or the operating time P5 of the fan 110. The operating frequency P4 and the operating time P5 are, for example, the recorded time of the actual operation of the fan 110. In addition, the operating noise P2 is, for example, measured by a microphone (not shown) in the electronic device 100. When the operating noise P2 becomes louder, it means that the fan 110 may have too much dust or be blocked by dirt. The temperature P3 is, for example, obtained by the basic input / output system (BIOS) of the electronic device 100. When the temperature P3 increases, it indicates that the friction loss of the fan 110 may increase due to dust. The cleaning cycle P6 is, for example, a time record of the user cleaning the fan 110, which can be recorded in the embedded controller 120 or the processor 130. In an embodiment, all operating parameters P can be uniformly transmitted to the embedded controller 120, and then transmitted to the processor 140 by the embedded controller 120. However, according to the actual hardware planning, one or some of all operating parameters can be transmitted to the processor 140 through the embedded controller 120, and the rest of all operating parameters can be transmitted to the processor 140 without passing through the embedded controller 120.

[0034] The embedded controller 120 is used to: control the fan 110 to rotate around a first rotation direction with a first current value C1 based on the fan 110 being in a normal state; and control the fan 110 to rotate around a second rotation direction with a second current value C2 based on the fan 110 being in an abnormal state. The second current value C2 is less than the first current value C1, for example, the second current value is 3 amperes, and the first current value is 5 amperes. The first rotation direction is, for example, a preset rotation direction (for example, forward rotation), and the second rotation direction is opposite to the first rotation direction, and the second rotation direction is also called reverse rotation.

[0035] In this embodiment, the fan status determination model 130 can be embedded in the operating system 150. The operating system 150 is a set of interrelated system software programs that manage and control computer operations, use and execute hardware and software resources, and provide public services to organize user interactions. It is also the core and cornerstone of the computer system. The operating system needs to handle basic tasks such as managing and configuring memory, determining the priority of system resource supply and demand, controlling input and output devices, operating networks, and managing file systems. The operating system also provides an operating interface for users to interact with the system.

[0036] The processor 140 is further configured to: after the electronic device 100 is turned on and the operating system 150 is loaded, call (or load) the fan status determination model 130 to determine whether the fan needs to be cleaned, and if so, start a cleaning method for the fan 110 .

[0037] The fan state judgment model 130 may execute a judgment method, such as a judgment method for fan cleanliness. The judgment method may include: receiving the control signal S1 and the operating parameter P of the fan 110 (that is, the control signal S1 and the operating parameter P are the inputs of the fan state judgment model 130); judging the cleanliness state of the fan 110 according to the control signal S1 and the operating parameter P; and outputting the cleanliness state (in other words, the cleanliness state is the output of the fan state judgment model 130) to the embedded controller 120 or the processor 140. The "cleanliness state" here is, for example, one of the aforementioned "abnormal state" and "normal state".

[0038] The fan status determination model 130 is obtained by using a ReLU activation function in a deep neural network (DNN), but the embodiment of the present invention does not limit the technology for obtaining the fan status determination model 130 .

[0039] Please refer to Figure 2 , which shows Figure 1The training diagram of the fan state judgment model 130 is shown in FIG. 1 . In the process of training the fan state judgment model 130, the ReLU linear regression model is used to train the machine learning algorithm, and the control signal S1 and at least one of all the operating parameters are used as input variables (i.e., X), and the fan cleanliness state (normal state is defined as 0, and abnormal state is defined as 1) is used as the output variable (i.e., Y). The linear regression model will learn a linear function to map the input variable to the output variable, thereby generating or training the fan state judgment model 130. In addition, during the training stage, the Scikit-learn function library can be used to generate the fan state judgment model 130 with 80% of the N samples as the training set, and the remaining 20% ​​of the N samples can be used as a verification set to verify the accuracy of the fan state judgment model 130. The embodiment of the present invention does not limit the value of N. It has been verified that the accuracy of the cleaning method using the fan state judgment model 130 can reach more than 97%.

[0040] The use of activation functions in neural networks is mainly to use nonlinear equations to solve nonlinear problems. If the activation function is not used, the neural network will combine operations in a linear way, because the hidden layer and the output layer both input the results of the upper layer and calculate them in linear combination as the output of this layer, so that the output and input only have a linear relationship. In reality, all problems are nonlinear problems. Therefore, if a nonlinear activation function is not used, the model trained by the neural network will lose its meaning. Rectified Linear Unit (ReLU), also known as the corrected linear unit, is a commonly used activation function in neural networks, usually referring to nonlinear functions represented by ramp functions and their variants.

[0041] Please refer to Figure 3 , which illustrates a functional block diagram of an electronic device 200 according to another embodiment of the present invention. The electronic device 200 includes a fan 110, an embedded controller 120, and a fan status judgment model 130. The fan 110 has an operating parameter P. The embedded controller 120 is used to: control the operation of the fan 110 with a control signal S1. The fan status judgment model 130 is used to: determine whether the fan 110 is in a normal state or an abnormal state based on the control signal S1 and the operating parameter P. The electronic device 200 of the embodiment of the present invention has the same or similar technical features as the aforementioned electronic device 100. One of the differences is that the fan status judgment model 130 can be called or loaded by the embedded controller 120. The embedded controller 120 can execute the cleaning method without waiting until the operating system stage is entered.

[0042] like Figure 3As shown, the embedded controller 120 is used to: input the control signal S1 and the operating parameter P into the fan state determination model 130, and control the fan 110 to reverse based on the abnormal state of the fan 110 to clean the fan 110. In this embodiment, the fan state determination model 130 is, for example, firmware, which can be stored in a read-only memory (ROM).

[0043] Please refer to Figure 4 , which shows Figure 1 Flow chart of a cleaning method for the electronic device 100 .

[0044] In step S110 , the embedded controller 120 controls the fan 110 to operate with a control signal S1 , wherein the fan 110 has an operating parameter P.

[0045] In step S120 , the processor 140 inputs the control signal S1 and the operating parameter P into the fan status determination model 130 .

[0046] In step S130, the fan status determination model 130 determines whether the fan 120 is in a normal state or an abnormal state according to the control signal S1 and the operating parameter P. If the fan 120 is in a normal state, the process proceeds to step S140; if the fan 120 is in an abnormal state, the process proceeds to step S150.

[0047] In step S140 , based on the fact that the fan 110 is in a normal state, indicating that the fan 110 does not need to be cleaned, the embedded controller 120 may maintain the current operation of the fan 110 or control the fan 110 to rotate forward.

[0048] In step S150 , based on the fact that the fan 110 is in an abnormal state, indicating that the fan 110 needs to be cleaned, the embedded controller 120 controls the fan 110 to rotate in reverse to clean the fan 110 .

[0049] In summary, the present invention provides an electronic device, a cleaning method thereof and a fan status judgment model thereof, which can automatically or actively judge the cleanliness of the fan. When the cleanliness of the fan is low and needs to be cleaned, the embedded controller controls the fan to reverse to clean the fan.

[0050] In summary, although the present invention has been disclosed as above by the embodiments, it is not intended to limit the present invention. A person with ordinary knowledge in the technical field to which the present invention belongs can make various changes and modifications without departing from the spirit and scope of the present invention, which are not limited by the embodiments of the present invention and are still within the protection scope of the present invention. Therefore, the protection scope of the present invention shall be determined by the definition of the appended claims.

Claims

1. An electronic device, characterized in that: include: a fan having an operating parameter; An embedded controller for: Controlling the operation of the fan with a control signal; A fan status judgment model is used to: According to the control signal and the operating parameter, determining whether the fan is in a normal state or an abnormal state; A processor for: Inputting the control signal and the operating parameter into the fan state determination model; The embedded controller is further used for: Based on the fact that the fan is in the abnormal state, the fan is controlled to rotate in reverse.

2. The electronic device according to claim 1, wherein: The fan status determination model is embedded in an operating system; the processor is further used to: After the electronic device is turned on and the operating system is loaded, the fan status determination model is called.

3. The electronic device according to claim 1, wherein: The operating parameter includes at least one of an actual rotation speed, an operating noise, a temperature, a use frequency, a use time and a cleaning cycle of the fan.

4. The electronic device according to claim 1, wherein: The control signal is pulse width modulated.

5. The electronic device according to claim 1, wherein: The fan status judgment model is obtained by using the ReLU excitation function in a deep neural network.

6. The electronic device as claimed in claim 1, wherein: The embedded controller is also used to: Based on the fan being in the normal state, controlling the fan to rotate around a first rotation direction with a first current value; and Based on the fan being in the abnormal state, controlling the fan to rotate around a second rotation direction with a second current value; The second current value is smaller than the first current value, and the second rotation direction is opposite to the first rotation direction.

7. A cleaning method for a fan, characterized in that: The cleaning method includes: Controlling the operation of a fan with a control signal, wherein the fan has an operation parameter; Inputting the control signal and the operating parameter into a fan status determination model; The fan state determination model determines whether the fan is in a normal state or an abnormal state according to the control signal and the operating parameter; and Based on the fact that the fan is in the abnormal state, the fan is controlled to rotate in reverse.

8. The cleaning method according to claim 7, characterized in that: The fan status determination model is embedded in an operating system; the fan cleaning method further includes: After the electronic device is turned on and the operating system is loaded, the fan status determination model is called.

9. The cleaning method according to claim 7, characterized in that: The step of inputting the control signal and the operating parameter into the fan state determination model comprises: The control signal and at least one of an actual rotation speed, an operation noise, a temperature, a use frequency, a use time and a cleaning cycle of the fan are input into the fan state judgment model.

10. The cleaning method according to claim 7, characterized in that: The steps of controlling the fan to run with the control signal include: The fan is controlled to operate by a pulse width modulation.

11. The cleaning method according to claim 7, characterized in that: Also includes: Based on the fan being in the normal state, controlling the fan to rotate around a first rotation direction with a first current value; and Based on the fan being in the abnormal state, the fan is controlled to rotate around a second rotation direction with a second current value.

12. A method for determining a fan status model, characterized in that: include: Receiving a control signal and an operating parameter of a fan; Determining a cleanliness state of the fan according to the control signal and the operating parameter; as well as The cleanliness status is output.

Images