Electric fan wind direction dynamic adjusting method and electric fan

Abstract

The invention relates to the technical field of intelligent control, in particular to an electric fan wind direction dynamic adjusting method and an electric fan. The method comprises the steps that environment sensing data and human body position information are obtained, the data are input into a preset deep learning model, and wind direction adjusting characteristics including human body thermal comfort demand characteristics and airflow coverage deviation characteristics are generated; determining a fan head-swinging angle adjusting parameter and a head-swinging speed compensation coefficient according to the wind direction adjusting characteristics; and dynamically adjusting the head swinging angle and the head swinging speed of the fan based on the head swinging angle adjusting parameter, the head swinging speed compensation coefficient and the current rotating speed signal of the fan so as to realize self-adaptive adjustment of the wind direction. According to the method, environment perception and human body position information are fused, a deep learning model is combined, user requirements are accurately recognized, and a fan head swinging strategy is dynamically optimized, so that the problem that the air supply range is not matched with the human body requirements due to a traditional fan fixed head swinging mode is solved, and the air supply efficiency and the user experience are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a method for dynamically adjusting the wind direction of an electric fan and the electric fan. Background Art

[0002] The wind direction adjustment of traditional electric fans mostly relies on a fixed swing head angle or manual setting, and can only achieve mechanical reciprocating swing. It can neither dynamically adjust the swing head strategy according to real-time environmental parameters, nor can it perceive the position distribution and movement trend of the human body in the fan coverage area, resulting in a mismatch between the air supply range and the human body's thermal comfort needs. For example, in a high temperature and high humidity environment, the fan still swings its head at a fixed speed, which may reduce the cooling efficiency due to airflow coverage blind spots or unreasonable swing head rhythm; when the human body moves, the traditional swing head mode cannot actively track, and it is easy to have the contradiction of excessive air supply or lack of air supply, resulting in energy waste and reduced user experience.

[0003] The above content is only used to assist in understanding the technical solution of the present invention and does not constitute an admission that the above content is prior art. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method for dynamically adjusting the wind direction of an electric fan and an electric fan, aiming to solve the technical problem that the wind direction adjustment of traditional electric fans mostly relies on a fixed swing head angle or manual setting, and is unable to sense the position distribution and movement trend of the human body in the fan coverage area, resulting in a mismatch between the air supply range and the thermal comfort needs of the human body.

[0005] To achieve the above object, the present invention provides a method for dynamically adjusting the wind direction of an electric fan, the method comprising: Acquire environmental perception data and human position information, the environmental perception data including real-time temperature, humidity, and current fan speed signals, and the human position information including coordinate distribution and movement trends of the human body within the fan coverage area; Inputting the environmental perception data and human position information into a preset deep learning model to generate wind direction adjustment features, wherein the wind direction adjustment features include human thermal comfort demand features and airflow coverage deviation features; Determine the fan swing angle adjustment parameter and the swing speed compensation coefficient according to the wind direction adjustment feature; Based on the swing head angle adjustment parameter, the swing head speed compensation coefficient and the current speed signal of the fan, the swing head angle and the swing head speed of the fan are dynamically adjusted to achieve adaptive adjustment of the wind direction.

[0006] Optionally, determining the fan swing angle adjustment parameter and the swing speed compensation coefficient according to the wind direction adjustment feature includes: Decoupling the human thermal comfort demand feature and the airflow coverage deviation feature in the wind direction adjustment feature is performed to extract the rate of change of human surface temperature, the change in the relative distance between the human body and the fan, and the offset angle between the current swing head angle and the center position of the human body; Determining the airflow regulation requirement intensity using a preset regulation intensity calculation formula based on the human body surface temperature change rate, the relative distance change, and the offset angle; The fan historical adjustment data is obtained, and the swing head angle adjustment parameter and the swing head speed compensation coefficient are obtained by fitting according to the airflow adjustment demand intensity and the fan historical adjustment data.

[0007] Optionally, the preset adjustment intensity calculation formula is:

[0008] Where, To adjust the required intensity of airflow, is the rate of change of human body surface temperature, To preset the maximum temperature change rate threshold, is the temperature change weight coefficient; is the change in the relative distance between the human body and the fan, To preset the effective coverage distance of the fan, is the distance change weight coefficient; is the current head swing angle, is the head swing angle corresponding to the center position of the human body, To preset the adjustment range of the swing head angle, is the angle offset weight coefficient.

[0009] Optionally, inputting the environmental perception data and human body position information into a preset deep learning model to generate wind direction adjustment features includes: Normalizing the environmental perception data, encoding the human body position information by spatial coordinates, and generating a multidimensional feature vector, wherein the multidimensional feature vector includes spatial distribution characteristics of temperature and humidity distribution and time series characteristics of human body motion trajectory; Inputting the multidimensional feature vector into a preset deep learning model, the preset deep learning model includes a convolutional layer, a long short-term memory network layer, and a fully connected layer, the convolutional layer is used to process the spatial distribution characteristics of the temperature and humidity distribution, and the long short-term memory network layer is used to capture the time series characteristics of the human body motion trajectory; The spatial distribution features and time series features are extracted through the convolutional layer and long short-term memory network layer of the preset deep learning model. The extracted spatial distribution features and time series features are classified and regressed through the fully connected layer to output the wind direction adjustment features.

[0010] Optionally, dynamically adjusting the fan's swing angle and swing speed based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal includes: Determining a pulse width modulation signal for fan adjustment based on the swing head angle adjustment parameter, the swing head speed compensation coefficient and the current fan speed signal; The fan's swing angle and speed are dynamically adjusted based on the pulse width modulation signal.

[0011] Optionally, before dynamically adjusting the fan's swing angle and speed based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal, the method further includes: Obtaining a fluctuation curve of a speed signal during fan head swinging, and calculating a slope change rate of the fluctuation curve; Comparing the slope change rate with a preset dynamic adjustment trigger threshold to determine a dynamic adjustment period for wind direction compensation; Accordingly, the fan's swing angle and swing speed are dynamically adjusted based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal, including: Based on the dynamic adjustment period, the swing head angle adjustment parameter, the swing head speed compensation coefficient and the current speed signal of the fan, the swing head angle and the swing head speed of the fan are dynamically adjusted.

[0012] Optionally, the preset dynamic adjustment trigger threshold includes an acceleration trigger threshold, a deceleration trigger threshold, and a stability threshold; the slope change rate is obtained by calculating a first slope and a second slope of the speed signal at two adjacent moments before and after the current moment; and comparing the slope change rate with the preset dynamic adjustment trigger threshold to determine the dynamic adjustment period of wind direction compensation includes: Comparing the slope change rate with a preset dynamic adjustment trigger threshold, and if the slope change rate is greater than the acceleration trigger threshold and the second slope is greater than the stability threshold, determining that an acceleration adjustment period has begun; If the slope change rate is greater than the deceleration trigger threshold and the second slope is less than the stability threshold, it is determined to enter the deceleration adjustment period; The dynamic adjustment period of wind direction compensation is determined according to the start and end times of the acceleration adjustment period and the deceleration adjustment period.

[0013] In addition, to achieve the above-mentioned purpose, the present invention further provides a device for dynamically adjusting the wind direction of an electric fan, the device comprising: A data acquisition module is used to obtain environmental perception data and human position information. The environmental perception data includes real-time temperature, humidity, and the current fan speed signal. The human position information includes the coordinate distribution and movement trend of the human body within the fan coverage area. a model generation module, configured to input the environmental perception data and human position information into a preset deep learning model to generate wind direction adjustment features, wherein the wind direction adjustment features include human thermal comfort demand features and airflow coverage deviation features; a parameter determination module, configured to determine a fan swing angle adjustment parameter and a swing speed compensation coefficient based on the wind direction adjustment feature; The dynamic adjustment module is used to dynamically adjust the fan's swing angle and swing speed based on the swing angle adjustment parameter, the swing speed compensation coefficient and the fan's current speed signal to achieve adaptive adjustment of the wind direction.

[0014] In addition, to achieve the above-mentioned purpose, the present invention also provides an electric fan, which includes: a memory, a processor, and an electric fan wind direction dynamic adjustment program stored in the memory and runnable on the processor, and the electric fan wind direction dynamic adjustment program is configured to implement the steps of the electric fan wind direction dynamic adjustment method as described in any one of the above.

[0015] In addition, to achieve the above-mentioned purpose, the present invention also provides a storage medium, on which a dynamic adjustment program for the wind direction of an electric fan is stored. When the dynamic adjustment program for the wind direction of an electric fan is executed by a processor, the steps of the dynamic adjustment method for the wind direction of an electric fan as described in any one of the above descriptions are implemented.

[0016] The present invention provides a method for dynamically adjusting the wind direction of an electric fan. The method combines real-time temperature, humidity, and human coordinate distribution and movement trend to dynamically generate human thermal comfort demand characteristics, so that the fan head swing angle and speed can accurately match the user's actual physical needs, avoiding the air supply blind area or excessive air supply problems in the traditional fixed swing head mode, and significantly improving the cooling efficiency and comfort. At the same time, a deep learning model is used to analyze the airflow coverage deviation characteristics, which can respond to environmental changes and human movement trajectories in real time, and automatically adjust the swing head strategy without manual intervention. It solves the problems of uneven air supply in multi-person scenarios and delayed air supply when the human body moves, enhances the fan's adaptability to complex scenarios, and reduces air supply in invalid areas by dynamically compensating the swing head speed and combining it with the fan speed signal, avoiding energy waste in the fixed mode, and achieving energy-saving optimization while improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 Schematic diagram of the electric fan structure of the hardware operating environment involved in the embodiment of the present invention; Figure 2 This is a flow chart of an embodiment of a method for dynamically adjusting the wind direction of an electric fan according to the present invention; Figure 3 This is a structural block diagram of an embodiment of a device for dynamically adjusting wind direction of an electric fan according to the present invention.

[0018] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0019] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0020] Reference Figure 1 , Figure 1 The figure is a schematic diagram of the structure of an electric fan in the hardware operating environment involved in the embodiment of the present invention.

[0021] like Figure 1 As shown, the electric fan may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen. Optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. In the present invention, the wired interface of the user interface 1003 may be a USB interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a wireless fidelity (WI-FI) interface). The memory 1005 may be a high-speed random access memory (RAM) or a non-volatile memory (NVM), such as a disk drive. The memory 1005 may also be a storage device independent of the processor 1001.

[0022] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation to the electric fan, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0023] like Figure 1 As shown, the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a fan wind direction dynamic adjustment program.

[0024] exist Figure 1 In the electric fan shown, the network interface 1004 is mainly used to connect to the background server and communicate data with the background server; the user interface 1003 is mainly used to connect to peripheral devices; the electric fan calls the electric fan wind direction dynamic adjustment program stored in the memory 1005 through the processor 1001, and executes the electric fan wind direction dynamic adjustment method provided by the embodiment of the present invention.

[0025] Based on the above hardware structure, an embodiment of the electric fan wind direction dynamic adjustment method is proposed.

[0026] Referring to Figure 2 , Figure 2 The flowchart of an embodiment of the electric fan wind direction dynamic adjustment method is shown in the figure, and an embodiment of the electric fan wind direction dynamic adjustment method is proposed.

[0027] In an embodiment, the electric fan wind direction dynamic adjustment method comprises the following steps: S10: Obtain environmental perception data and human body position information, wherein the environmental perception data includes real-time temperature, humidity and fan current speed signal, and the human body position information includes the coordinate distribution and movement trend of the human body in the fan coverage area.

[0028] It should be noted that the environmental perception data is quantitative information reflecting the current environmental state and equipment operating parameters collected by sensors in real time. The real-time temperature is the ambient air temperature value obtained by the temperature sensor, which directly affects the human body thermal comfort demand judgment. The higher the temperature, the stronger the human body's demand for airflow. The humidity is the relative humidity of the environment collected by the humidity sensor, which determines the human body's feeling temperature together with the temperature. High humidity will reduce the evaporation efficiency of sweat, and stronger airflow is needed to enhance heat dissipation. The fan current speed signal is the real-time speed of the fan obtained by the motor encoder or current monitoring, which reflects the current fan air supply intensity and is used for dynamic matching of the swing speed, such as adjusting the swing speed when the speed is high to avoid uneven airflow coverage.

[0029] The human body position information is the information obtained by sensing devices such as cameras, millimeter wave radars, infrared sensors, etc., which describes the spatial position and motion state of the human body in the fan coverage area. The human body coordinate distribution is to establish a coordinate system with the fan as the origin, and mark the specific position coordinates of each human body in the fan coverage area, which is used to identify the air supply key area. The movement trend is the human body motion parameter calculated by the continuous frame coordinate data, including moving speed, motion direction and trajectory prediction, based on the future position trend fitted by the historical coordinates, which is used to dynamically adjust the swing strategy, such as following the moving human body to preferentially cover its path of travel. The fan coverage area refers to the space range that can be effectively covered by the airflow when the fan is working normally, which is the spatial boundary for defining the position of the human body. The sensor completes human body detection and positioning in this area to ensure that the air supply strategy is only for the human body demand in the actual effective area.

[0030] It should be understood that environmental perception data is acquired through physical sensors integrated into the fan, while human position information relies on external or built-in visual / radar perception modules, such as depth cameras and UWB positioning systems. Both types of data require clock synchronization and calibration before being input into the algorithm model. Environmental data reflects the intensity of thermal comfort needs, while human position information clarifies the spatial distribution and dynamics of these needs. The combination of these two enables fans to transition from fixed-mode air delivery to user-centric intelligent air delivery.

[0031] S20: Inputting the environmental perception data and human body position information into a preset deep learning model to generate wind direction adjustment features, wherein the wind direction adjustment features include human body thermal comfort demand features and airflow coverage deviation features.

[0032] It should be noted that the preset deep learning model is a neural network model trained on a large amount of historical data before the fan leaves the factory or is used. It is used to extract abstract features from environmental perception data and human position information and is the core algorithm carrier for realizing intelligent decision-making. The training data includes scenarios with different combinations of temperature, humidity, fan speed, number of people, and movement trajectories. The labeling target is the optimal wind direction adjustment parameter, and the label is generated through simulated or measured thermal comfort experiments. Through nonlinear mapping, the raw input data is converted into high-level semantic features that have direct guiding significance for wind direction adjustment, namely wind direction adjustment features, replacing the manual experience modeling of traditional rule engines.

[0033] The wind direction adjustment feature is an abstract feature vector output by the deep learning model that reflects the wind direction adjustment demand in the current scenario. It serves as an intermediate semantic representation connecting input data with control parameters and consists of two core sub-features. The human thermal comfort demand feature characterizes the intensity and priority of airflow demand within the fan coverage area. It combines environmental thermophysical parameters with the spatial distribution of human bodies to quantify the thermal comfort gap for individuals at different locations. For example, based on human thermal comfort models such as the PMV-PPD model, temperature and humidity can be converted into a balance between metabolic heat production and heat dissipation. If the current environment results in insufficient heat dissipation, such as high temperature and humidity, the demand weight at the corresponding location is increased. Combined with the human coordinate distribution, a regional demand heat map is generated. For example, if high-demand areas have higher demand feature values, the fan is guided to prioritize increasing the fan's swing coverage duration or angle towards high-demand areas. The airflow coverage deviation feature characterizes the difference between the actual airflow coverage and the ideal demand coverage under the fan's swing mode. This is used to identify airflow blind spots or areas of excessive coverage. For example, the target area that needs to be covered theoretically can be calculated based on the human body's position coordinates and movement trends, such as the current position of the human body and the predicted position within the next 3 seconds; by comparing the current head swing angle coverage range with the spatial overlap of the target area, a deviation value is generated, such as the proportion of uncovered area, coverage delay time, etc. If the deviation value is large, such as the human body is in the blind spot of the head swing, the head swing angle or speed adjustment is triggered.

[0034] Specifically, the step of inputting the environmental perception data and the human body position information into a preset deep learning model to generate wind direction adjustment features includes: Normalizing the environmental perception data, encoding the human body position information by spatial coordinates, and generating a multidimensional feature vector, wherein the multidimensional feature vector includes spatial distribution characteristics of temperature and humidity distribution and time series characteristics of human body motion trajectory; Inputting the multidimensional feature vector into a preset deep learning model, the preset deep learning model includes a convolutional layer, a long short-term memory network layer, and a fully connected layer, the convolutional layer is used to process the spatial distribution characteristics of the temperature and humidity distribution, and the long short-term memory network layer is used to capture the time series characteristics of the human body motion trajectory; The spatial distribution features and time series features are extracted through the convolutional layer and long short-term memory network layer of the preset deep learning model. The extracted spatial distribution features and time series features are classified and regressed through the fully connected layer to output the wind direction adjustment features.

[0035] It should be understood that normalization is a preprocessing operation that converts environmental sensing data of different dimensions and ranges into a unified numerical range. This avoids model training bias caused by large differences in feature values. For example, if the speed signal value is much larger than the temperature, it may dominate the gradient descent direction. It also accelerates the convergence of deep learning models and improves training efficiency. Spatial coordinate encoding is the process of converting the coordinates of human positions within the fan coverage area into numerical feature vectors that can be processed by the deep learning model. The multidimensional feature vector is the high-dimensional input data that integrates environmental sensing data with human position information. It is the intermediate representation that connects the raw sensor signal with the deep learning model. The convolutional layer is a layer in the deep learning model that specifically processes spatial features. It extracts features from the input spatial distribution data using learnable convolution kernels. For example, it identifies the spatial overlap between high-temperature and high-humidity areas and human gathering areas, such as the temperature >30°C and the humidity >70% within 3 meters of coordinates (2, 3), and the presence of two people. The convolution operation outputs a thermal comfort demand density feature map, marking the air supply priority of different areas, such as red areas with high demand and blue areas with low demand. The Long Short-Term Memory (LSTM) layer is an improved version of the Recurrent Neural Network (RNN). It uses a forget gate, input gate, and output gate mechanism to capture long-term dependencies in time series data. This makes it suitable for processing dynamically changing data such as human motion trajectories. For example, based on a person's coordinate sequence over the past N frames, such as position changes over the past 5 seconds, it can predict future movement direction and speed, such as moving toward the right front of the fan at 0.5 m / s. It also distinguishes between stationary and moving people. While stationary people do not need to dynamically adjust their head swing, moving people require pre-adjustment of their head swing direction to cover their path, avoiding air supply delays caused by their instantaneous movement.

[0036] It should be noted that the fully connected layer (FCL) is the terminal layer of the deep learning model, fusing the spatial features extracted by the convolutional layer with the temporal features extracted by the LSTM layer, and generating the final output through weighted connections. In this embodiment, the spatial heat map features output by the convolutional layer are concatenated with the motion trend features output by the LSTM, and then compressed into a low-dimensional feature vector using the weight matrix of the FCL. This layer outputs continuous-value wind direction adjustment features, such as the intensity of human thermal comfort demand and the airflow coverage deviation value. This layer indirectly distinguishes scene types, such as multi-person stationary scenes or single-person fast-moving scenes, to guide different adjustment strategies. Classification and regression are the tasks of the output layer of the deep learning model. Classification and regression are combined here, with classification representing scene recognition and regression representing continuous-value prediction. The regression layer outputs specific numerical values ​​of the wind direction adjustment features, such as a thermal comfort demand characteristic value of 0.8 and an airflow coverage deviation value of 0.3, which are used to quantify the adjustment intensity.

[0037] The formula of the preset deep learning model is:

[0038] Where, is the environmental perception data vector, Encode vectors for human body position and motion trajectory; and are the convolutional layer weights and biases, is the activation function; Output features for the convolutional layer; is the long short-term memory network layer, is the hidden state at the previous moment, Hide the state for the current moment; and are the weight and bias of the fully connected layer respectively; is the head angle adjustment parameter, is the compensation coefficient of the swing head speed.

[0039] It should be noted that It is the environmental perception data vector, which contains the normalized numerical feature vector of real-time temperature, humidity, and current fan speed. It is the encoding vector of the human body position and motion trajectory, including the encoding vector of the human body coordinate distribution and movement trajectory in the fan coverage area. The convolution layer weights refer to the learnable convolution kernel parameters, which are used to extract the spatial distribution characteristics of environmental data, such as the spatial correlation of high temperature and high humidity areas. The bias vector of the convolution layer is used to add a learnable bias term to the convolution operation result to prevent the output value from approaching zero and improve the model fitting ability. The activation function can be ReLU or Sigmoid, which introduces nonlinearity for the output of the convolutional layer, enabling the model to learn complex spatial feature combinations. The output of the convolutional layer is a feature vector representing the joint characteristics of the environment data and the spatial distribution of the human body, indicating the intensity of the thermal comfort demand in different areas, such as high values corresponding to areas in urgent need of air supply. The LSTM layer is used to process time series data of human motion trajectories, capturing past-present-future position dependencies, such as predicting the direction of human movement. As the recurrent input of the LSTM layer, it carries historical motion information and calculates the current hidden state together with the current input, enabling cumulative learning of time series features, such as predicting the next time step to continue moving right if the past 5 seconds have been moving right. The concatenated feature vector is obtained by concatenating the output of the convolutional layer (spatial features) and the output of the LSTM (temporal features) along the feature dimension, for example, combining the information of where to supply air and where the person will go to form a complete basis for wind direction adjustment decision. The fully connected layer is used to map the high-dimensional feature vector after concatenation to the output space, i.e., the fan swing angle adjustment parameter and the speed compensation coefficient . The bias is added to the linear transformation result of the fully connected layer to adjust the baseline value of the output parameters, such as defaulting the swing angle to 0° and the speed compensation to 1.0 when there is no demand.

[0040] S30: Determine the fan swing angle adjustment parameter and swing speed compensation coefficient based on the wind direction adjustment feature.

[0041] It should be noted that, is the offset relative to the current swing angle, used to expand or reduce the swing range, or adjust the center angle. is a proportional adjustment factor for the current swing speed, used to dynamically match the speed of human movement, accelerating the swing when moving quickly to avoid air supply lag.

[0042] Specifically, the determination of the fan swing angle adjustment parameter and the swing speed compensation coefficient based on the wind direction adjustment feature includes: Decouple the human thermal comfort demand feature and the airflow coverage deviation feature in the wind direction adjustment feature, extract the human body surface temperature change rate, the relative distance change between the human body and the fan, and the offset angle between the current swing angle and the center position of the human body; Based on the human body surface temperature change rate, the relative distance change, and the offset angle, determine the airflow adjustment demand intensity through a pre-set adjustment intensity calculation formula. Obtain fan historical adjustment data, and fit a swing angle adjustment parameter and a swing speed compensation coefficient according to the air flow adjustment demand intensity and the fan historical adjustment data.

[0043] It should be noted that the human thermal comfort demand feature is a core feature of the wind direction adjustment feature representing the urgency of the air flow demand of the human body, reflecting the physiological comfort state of whether the human body needs stronger / more accurate air supply, and can be determined by detecting the human body surface temperature distribution through an infrared thermal imager. The air flow coverage deviation feature is a deviation index of the wind direction adjustment feature representing the matching degree of the current fan air supply range and the actual demand area of the human body, reflecting the spatial error of whether the air supply accurately covers the target area. Feature decoupling is used to decompose the high-dimensional abstract wind direction adjustment feature into independent interpretable physical quantity indexes, facilitating the establishment of clear control logic. The human body surface temperature change rate is the change amount of the average temperature of the human body surface per unit time, representing the dynamic change trend of the human body thermal state. The relative distance change amount between the human body and the fan is the difference between the distance from the human body position to the fan center at the current time and the distance at the previous time, representing the movement trend of the human body approaching or moving away from the fan. The offset angle between the current swing angle and the human body center position is the included angle between the current swing center angle of the fan and the direction of the human body group center position, representing the alignment deviation of the air supply direction and the actual position of the human body. The preset adjustment intensity calculation formula is a mathematical model for quantifying the urgency of air flow adjustment in the current scene based on the decoupled physical indexes, which is usually a weighted linear combination or a nonlinear function. The air flow adjustment demand intensity is a 0-1 interval scalar value calculated by the preset formula, directly reflecting the urgency of the fan swing angle and speed adjustment in the current scene, and the larger the value, the more intense the demand. The fan historical adjustment data is a data set storing the adjustment parameters and corresponding scene data of the fan in the past period of time, used for model parameter fitting and strategy optimization. For example, a mapping relationship is established by linear regression, or the adjustment action-effect feedback in the historical data is used, such as a temperature drop of 0.5℃ / min after adjustment is considered as positive feedback, and the adjustment strategy is optimized by Q-Learning.

[0044] The preset adjustment intensity calculation formula is:

[0045] In the formula, is the air flow adjustment demand intensity, is the human body surface temperature change rate, is a preset maximum temperature change rate threshold, is a temperature change weight coefficient; is the relative distance change amount between the human body and the fan, is a preset fan effective coverage distance, is a distance change weight coefficient; is the current swing angle, is the head swing angle corresponding to the center position of the human body, To preset the adjustment range of the swing head angle, is the angle offset weight coefficient.

[0046] It should be noted that the preset maximum temperature change rate threshold Defines the extreme temperature change boundary that the human body can accept for thermal comfort, and is used for normalization processing to avoid a single parameter dominating the demand intensity. Temperature change weight coefficient Adjust the contribution of temperature change to demand intensity, reflecting the system's priority setting for thermal comfort needs. Preset fan effective coverage distance The maximum distance at which the fan's airflow can effectively affect the thermal comfort of the human body. If the wind speed is lower than 0.1m / s beyond this distance, it is considered invalid coverage. To adjust the contribution of distance change to demand intensity, and reflect the system's priority setting for human body dynamic tracking. To define the maximum acceptable range of the head angle deviation, used to normalize the angle offset. Angle offset weight coefficient To adjust the contribution of angle offset to demand intensity, and reflect the system's priority setting for the accuracy of air supply direction.

[0047] S40: Based on the swing angle adjustment parameter, the swing speed compensation coefficient and the current speed signal of the fan, the swing angle and the swing speed of the fan are dynamically adjusted to achieve adaptive adjustment of the wind direction.

[0048] It should be noted that the head swing angle adjustment parameters include the target head swing angle, angle adjustment step, and head swing angle range. The target head swing angle is the final control target angle calculated based on the head swing angle corresponding to the human body center position and the current adjustment requirements. The angle adjustment step is the minimum angle change for each head swing action, which is used to avoid frequent micro-adjustments. The head swing angle range is the angle interval of the fan's back-and-forth swing. The fan's steering angle can be adjusted according to the head swing angle adjustment parameters. The head swing speed compensation coefficient is used to correct the head swing motor speed to match the head swing speed with the fan's air supply speed, where the fan's air supply speed is obtained through the fan's current speed signal.

[0049] Specifically, the dynamically adjusting the fan's swing angle and swing speed based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal includes: Determining a pulse width modulation signal for fan regulation based on the swing head angle adjustment parameter, the swing head speed compensation coefficient, and the current fan speed signal; The fan's swing angle and speed are dynamically adjusted based on the pulse width modulation signal.

[0050] It should be noted that the pulse width modulation signal is a digital signal that transmits control instructions through the duty cycle of periodic rectangular pulses. The rotation speed is controlled by the pulse frequency of the PWM signal, and the rotation angle is controlled by the number of pulses. The power supply voltage is adjusted by the PWM duty cycle, thereby controlling the rotation speed.

[0051] In another embodiment, before dynamically adjusting the fan's swing angle and swing speed based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal, the method further includes: Obtaining a fluctuation curve of a speed signal during fan head swinging, and calculating a slope change rate of the fluctuation curve; Comparing the slope change rate with a preset dynamic adjustment trigger threshold to determine a dynamic adjustment period for wind direction compensation; Accordingly, the fan's swing angle and swing speed are dynamically adjusted based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal, including: The fan's swing angle and speed are dynamically adjusted based on the dynamic adjustment period, the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal.

[0052] It should be noted that the speed signal fluctuation curve represents a real-time curve of the fan's main motor speed changes over time during the swinging process, reflecting the dynamic impact of the swing mechanism's movement on the main motor load. The slope change rate represents the rate of speed change. A preset dynamic adjustment trigger threshold serves as the switching condition for determining whether to initiate dynamic adjustment, distinguishing normal fluctuations from abnormal load changes that require compensation. The dynamic adjustment period is the time period from the moment the slope change rate exceeds the limit to the time the system completes adjustment and returns to stability.

[0053] Specifically, the preset dynamic adjustment trigger threshold includes an acceleration trigger threshold, a deceleration trigger threshold, and a stability threshold. The slope change rate is obtained by calculating the first slope and the second slope of the speed signal at two adjacent moments before and after the current moment. The slope change rate is compared with the preset dynamic adjustment trigger threshold to determine the dynamic adjustment period of wind direction compensation, including: Comparing the slope change rate with a preset dynamic adjustment trigger threshold, and if the slope change rate is greater than the acceleration trigger threshold and the second slope is greater than the stability threshold, determining that an acceleration adjustment period has begun; If the slope change rate is greater than the deceleration trigger threshold and the second slope is less than the stability threshold, it is determined to enter the deceleration adjustment period; The dynamic adjustment period of wind direction compensation is determined according to the start and end times of the acceleration adjustment period and the deceleration adjustment period.

[0054] It should be noted that the acceleration trigger threshold determines the critical slope change rate of the speed signal entering the acceleration phase, serving as the dividing line between normal fluctuations and acceleration load changes requiring compensation. The deceleration trigger threshold determines the critical slope change rate of the speed signal entering the deceleration phase. The stability threshold determines whether the speed signal is in the stable phase, serving as the dividing line between dynamic changes and steady-state operation. The first slope reflects the speed change rate from the previous moment to the current moment; the second slope reflects the speed change rate from the current moment to the next moment. By comparing the two adjacent slopes, the slope change rate is calculated to determine the acceleration trend of the speed change. The start time is the sampling moment when the trigger conditions are first met simultaneously. A buffering period of two sampling cycles is used to avoid missing previous trends. The end time is when the trigger conditions are no longer met, e.g., when the slope change rate falls back within the threshold and the second slope enters the stable range. The regulation duration is extended by five sampling cycles to ensure the completion of the control action.

[0055] In addition, an embodiment of the present invention further provides a storage medium storing a program for dynamically adjusting the wind direction of an electric fan. When the program is executed by a processor, the steps of the method for dynamically adjusting the wind direction of an electric fan as described above are implemented.

[0056] In addition, refer to Figure 3 The embodiment of the present invention further provides a device for dynamically adjusting the wind direction of an electric fan, the device comprising: The data acquisition module 10 is used to obtain environmental perception data and human position information. The environmental perception data includes real-time temperature, humidity, and the current fan speed signal. The human position information includes the coordinate distribution and movement trend of the human body in the fan coverage area. A model generation module 20 is configured to input the environmental perception data and human position information into a preset deep learning model to generate wind direction adjustment features, wherein the wind direction adjustment features include human thermal comfort demand features and airflow coverage deviation features; a parameter determination module 30, configured to determine a fan swing angle adjustment parameter and a swing speed compensation coefficient according to the wind direction adjustment characteristics; The dynamic adjustment module 40 is used to dynamically adjust the fan's swing angle and swing speed based on the swing angle adjustment parameter, the swing speed compensation coefficient and the fan's current speed signal to achieve adaptive adjustment of the wind direction.

[0057] Other embodiments or specific implementations of the electric fan wind direction dynamic adjustment device of the present invention can refer to the above-mentioned method embodiments and will not be repeated here.

[0058] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0059] The serial numbers of the embodiments of the present invention are for descriptive purposes only and do not represent superiority or inferiority of the embodiments. In a unit claim that enumerates several means, several of these means may be embodied by the same item of hardware. The use of the terms first, second, and third, etc., does not denote any order; these terms should be interpreted as designations.

[0060] Through the above description of the embodiments, those skilled in the art will clearly understand that the above-mentioned embodiments and methods can be implemented using software plus the necessary general hardware platform. Of course, hardware can also be used, but in many cases, the former is a more preferred embodiment. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as a read-only memory image (ROM) / random access memory (RAM), a magnetic disk, or an optical disk) and includes a number of instructions for enabling an end-user device (such as a mobile phone, computer, server, air conditioner, or network user device, etc.) to execute the methods described in various embodiments of the present invention.

[0061] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for dynamically adjusting the wind direction of an electric fan, characterized in that: The method comprises: Acquire environmental perception data and human position information, the environmental perception data including real-time temperature, humidity, and current fan speed signals, and the human position information including coordinate distribution and movement trends of the human body within the fan coverage area; Inputting the environmental perception data and human position information into a preset deep learning model to generate wind direction adjustment features, wherein the wind direction adjustment features include human thermal comfort demand features and airflow coverage deviation features; Determine the fan swing angle adjustment parameter and the swing speed compensation coefficient according to the wind direction adjustment feature; Based on the swing head angle adjustment parameter, the swing head speed compensation coefficient and the current speed signal of the fan, the swing head angle and the swing head speed of the fan are dynamically adjusted to achieve adaptive adjustment of the wind direction.

2. The method for dynamically adjusting the wind direction of an electric fan according to claim 1, wherein: The determining of the fan swing angle adjustment parameter and the swing speed compensation coefficient according to the wind direction adjustment feature includes: Decoupling the human thermal comfort demand feature and the airflow coverage deviation feature in the wind direction adjustment feature is performed to extract the rate of change of human surface temperature, the change in the relative distance between the human body and the fan, and the offset angle between the current swing head angle and the center position of the human body; Determining the airflow regulation requirement intensity using a preset regulation intensity calculation formula based on the human body surface temperature change rate, the relative distance change, and the offset angle; The fan historical adjustment data is obtained, and the swing head angle adjustment parameter and the swing head speed compensation coefficient are obtained by fitting according to the airflow adjustment demand intensity and the fan historical adjustment data.

3. The method for dynamically adjusting the wind direction of an electric fan according to claim 2, wherein: The preset adjustment intensity calculation formula is: Where, To adjust the required intensity of airflow, is the rate of change of human body surface temperature, To preset the maximum temperature change rate threshold, is the temperature change weight coefficient; is the change in the relative distance between the human body and the fan, To preset the effective coverage distance of the fan, is the distance change weight coefficient; is the current head swing angle, is the head swing angle corresponding to the center position of the human body, To preset the adjustment range of the swing head angle, is the angle offset weight coefficient.

4. The method for dynamically adjusting the wind direction of an electric fan according to claim 1, wherein: The step of inputting the environmental perception data and the human body position information into a preset deep learning model to generate wind direction adjustment features includes: Normalizing the environmental perception data, encoding the human body position information by spatial coordinates, and generating a multidimensional feature vector, wherein the multidimensional feature vector includes spatial distribution characteristics of temperature and humidity distribution and time series characteristics of human body motion trajectory; Inputting the multidimensional feature vector into a preset deep learning model, the preset deep learning model includes a convolutional layer, a long short-term memory network layer, and a fully connected layer, the convolutional layer is used to process the spatial distribution characteristics of the temperature and humidity distribution, and the long short-term memory network layer is used to capture the time series characteristics of the human body motion trajectory; The spatial distribution features and time series features are extracted through the convolutional layer and long short-term memory network layer of the preset deep learning model. The extracted spatial distribution features and time series features are classified and regressed through the fully connected layer to output the wind direction adjustment features.

5. The method for dynamically adjusting the wind direction of an electric fan according to any one of claims 1 to 4, characterized in that: The method of dynamically adjusting the fan's swing angle and speed based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal includes: Determining a pulse width modulation signal for fan regulation based on the swing head angle adjustment parameter, the swing head speed compensation coefficient, and the current fan speed signal; The fan's swing angle and speed are dynamically adjusted based on the pulse width modulation signal.

6. The method for dynamically adjusting the wind direction of an electric fan according to claim 1, wherein: Before dynamically adjusting the fan's swing angle and speed based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal, the method further includes: Obtaining a fluctuation curve of a speed signal during fan head swinging, and calculating a slope change rate of the fluctuation curve; Comparing the slope change rate with a preset dynamic adjustment trigger threshold to determine a dynamic adjustment period for wind direction compensation; Accordingly, the fan's swing angle and swing speed are dynamically adjusted based on the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal, including: The fan's swing angle and speed are dynamically adjusted based on the dynamic adjustment period, the swing angle adjustment parameter, the swing speed compensation coefficient, and the fan's current speed signal.

7. The method for dynamically adjusting the wind direction of an electric fan according to claim 6, wherein: The preset dynamic adjustment trigger threshold includes an acceleration trigger threshold, a deceleration trigger threshold, and a stability threshold. The slope change rate is obtained by calculating the first slope and the second slope of the speed signal at two adjacent moments before and after the current moment. The slope change rate is compared with the preset dynamic adjustment trigger threshold to determine the dynamic adjustment period of wind direction compensation, including: Comparing the slope change rate with a preset dynamic adjustment trigger threshold, and if the slope change rate is greater than the acceleration trigger threshold and the second slope is greater than the stability threshold, determining that an acceleration adjustment period has begun; If the slope change rate is greater than the deceleration trigger threshold and the second slope is less than the stability threshold, it is determined to enter the deceleration adjustment period; The dynamic adjustment period of wind direction compensation is determined according to the start and end times of the acceleration adjustment period and the deceleration adjustment period.

8. A device for dynamically adjusting the wind direction of an electric fan, characterized in that: The electric fan wind direction dynamic adjustment device comprises: A data acquisition module is used to obtain environmental perception data and human position information. The environmental perception data includes real-time temperature, humidity, and the current fan speed signal. The human position information includes the coordinate distribution and movement trend of the human body within the fan coverage area. a model generation module, configured to input the environmental perception data and human position information into a preset deep learning model to generate wind direction adjustment features, wherein the wind direction adjustment features include human thermal comfort demand features and airflow coverage deviation features; a parameter determination module, configured to determine a fan swing angle adjustment parameter and a swing speed compensation coefficient based on the wind direction adjustment feature; The dynamic adjustment module is used to dynamically adjust the fan's swing angle and swing speed based on the swing angle adjustment parameter, the swing speed compensation coefficient and the fan's current speed signal to achieve adaptive adjustment of the wind direction.

9. An electric fan, characterized in that: The electric fan includes: a memory, a processor, and an electric fan wind direction dynamic adjustment program stored in the memory and executable on the processor, wherein the electric fan wind direction dynamic adjustment program is configured to implement the steps of the electric fan wind direction dynamic adjustment method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores a program for dynamically adjusting the wind direction of an electric fan. When the program is executed by the processor, the steps of the method for dynamically adjusting the wind direction of an electric fan according to any one of claims 1 to 7 are implemented.

Images