Air conditioner indoor unit control method, device, equipment and storage medium

Abstract

The application relates to the technical field of air conditioner control, and discloses an air conditioner indoor unit control method, device, equipment and storage medium, the method realizes the cooperative control of the air conditioner indoor unit and the lighting mechanism in the night scene, and a complete logic chain of scene recognition, information collection, state determination and accurate control is constructed; through the combination of a millimeter wave radar sensor and a human state detection model, the accurate perception of the human state is realized, non-target interference can be effectively filtered, and three states, i.e., frequent rotation, sitting up and getting out of bed, can be distinguished, so that the false touch rate is greatly reduced; the air conditioner parameter adjustment or the lighting control is respectively performed according to different states, the problems of high false touch rate and unsuitable brightness of a traditional night light can be solved, the defects of the fixed parameter operation of a traditional air conditioner can be made up, and the user's night sleep comfort and night convenience are significantly improved; meanwhile, a high-frequency scanning mode is adopted, the real-time performance and the accuracy of state determination are ensured, and a reliable data basis is provided for subsequent parameter adjustment and lighting control.

Description

Technical Field

[0001] This invention relates to the field of air conditioning control technology, and in particular to an air conditioning indoor unit control method, device, equipment and storage medium. Background Technology

[0002] Currently, users have higher requirements for intelligent and user-friendly control of the bedroom environment. However, current technologies still have significant shortcomings in terms of coordinated control of lighting and air conditioning systems, environmental adaptability, and user experience optimization, making it difficult to meet users' integrated needs for sleep comfort and convenience of nighttime activities.

[0003] In existing technologies, traditional nightlights typically employ a stand-alone installation design, which not only occupies limited bedroom space and disrupts the overall integrity and aesthetics of the interior decoration, but also has significant drawbacks in its sensing triggering mechanism: the infrared sensing technology used is easily interfered with by non-target factors such as pet activity, ambient temperature fluctuations, and airflow disturbances, resulting in a high false trigger rate; moreover, the lighting brightness is mostly set with fixed parameters, making it impossible to personalize the adjustment according to different users' age, eye sensitivity, and usage scenarios; in addition, traditional nightlights lack a coordinated control mechanism with core bedroom appliances such as air conditioners, requiring independent wiring and power supply, which not only increases installation complexity and maintenance costs, but also fails to achieve linkage adjustment between lighting and temperature and humidity environment, making it difficult to meet users' integrated control needs for the nighttime environment.

[0004] Meanwhile, existing nighttime lighting devices generally adopt an instantaneous on / off working mode. Direct strong light can significantly inhibit the normal secretion of melatonin in the human body, leading to a prolonged latency period for users to fall back asleep and affecting sleep quality. Furthermore, they lack a smooth brightness gradual adjustment mechanism, and sudden changes in brightness can easily stimulate the visual system of sensitive groups such as the elderly and children, posing a safety hazard.

[0005] It is evident that existing technologies still need improvement and enhancement. Summary of the Invention

[0006] In order to overcome the shortcomings of the existing technology, the purpose of this invention is to provide an air conditioner indoor unit control method, which achieves the technical effects of improved sleep comfort, reduced energy consumption, and system integration through a complete logic chain of precise sensing, intelligent judgment and scenario-based control.

[0007] The first aspect of this invention provides a method for controlling an indoor unit of an air conditioner. The indoor unit includes a main control system, a millimeter-wave radar sensor electrically connected to the main control system, and a lighting mechanism. The method includes: when it is determined that it is nighttime, controlling the millimeter-wave radar sensor to switch to a high-frequency scanning mode; acquiring human activity information based on real-time sensing signals collected by the millimeter-wave radar sensor, the human activity information including movement speed, limb movement characteristics, and spatial position information; pre-constructing and training a human state detection model, inputting the human activity information into the human state detection model for processing, and obtaining a state determination result, the state determination result including frequent turning, sitting up, and getting out of bed; when the state determination result is frequent turning or sitting up, adjusting the operating parameters of the indoor unit of the air conditioner based on the state determination result and the human activity information; when the state determination result is getting out of bed, adjusting the working state of the lighting mechanism based on the human activity information.

[0008] Optionally, in a first implementation of the first aspect of the present invention, the step of controlling the millimeter-wave radar sensor to switch to a high-frequency scanning mode when it is determined to be a nighttime period includes: acquiring real-time ambient light intensity; determining that the current period is nighttime when the real-time ambient light intensity is ≤5 lux; and controlling the millimeter-wave radar sensor to switch to a high-frequency scanning mode, thereby increasing the scanning frequency of the millimeter-wave radar sensor to 10 times / second.

[0009] Optionally, in a second implementation of the first aspect of the present invention, the step of acquiring human activity information based on real-time sensing signals collected by the millimeter-wave radar sensor, wherein the human activity information includes movement speed, limb movement characteristics, and spatial position information, includes: calculating the frequency difference between the transmitted signal and the echo signal of the millimeter-wave radar sensor based on the Doppler effect principle, and calculating the movement speed of the human target in combination with the operating frequency band parameters of the millimeter-wave radar sensor; extracting the time-domain amplitude variation characteristics and phase variation period of the echo signal of the millimeter-wave radar sensor to generate limb movement characteristics, wherein the limb movement characteristics are movement feature vectors; calculating the corresponding time delay based on the frequency difference, and calculating the straight-line distance between the human target and the millimeter-wave radar sensor according to the time delay and the electromagnetic wave propagation speed; using the MUSIC algorithm to perform angle estimation on the echo signal collected by the millimeter-wave radar sensor to obtain the horizontal and vertical angles of the human target; and determining the spatial position information of the human target based on the straight-line distance, horizontal angle, and vertical angle, wherein the spatial position information is three-dimensional spatial position coordinates.

[0010] Optionally, in a third implementation of the first aspect of the present invention, the pre-construction and training of the human state detection model, and the input of the human activity information into the human state detection model for processing to obtain a state determination result, includes: constructing a CNN-LSTM hybrid model and obtaining sample datasets in multiple scenarios, wherein the sample datasets include samples of users of different ages getting up at night, turning over, sitting up, and walking, samples of pets of different body types, and samples of environmental interference signals; performing data cleaning, feature extraction, and normalization preprocessing on the sample datasets, and performing multiple rounds of iterative training on the constructed CNN-LSTM hybrid model based on the preprocessed sample data to train the human state detection model; performing feature standardization and dimensional reconstruction preprocessing on the human activity information to generate a feature matrix; and inputting the feature matrix into the human state detection model to obtain a state determination result.

[0011] Optionally, in a fourth implementation of the first aspect of the present invention, the step of adjusting the operating parameters of the indoor air conditioner unit based on the state determination result and human activity information when the state determination result is frequent rotation or sitting up includes: when the state determination result is frequent rotation or sitting up, acquiring real-time environmental parameters and real-time air conditioning parameters; mapping the state determination result, human activity information, real-time environmental parameters, and real-time air conditioning parameters into fuzzy linguistic variables; retrieving and matching a pre-built fuzzy rule library according to the fuzzy linguistic variables to generate preliminary adjustment parameters; and adjusting the operating parameters of the indoor air conditioner unit using a PID algorithm based on the preliminary adjustment parameters.

[0012] Optionally, in a fifth implementation of the first aspect of the present invention, adjusting the working state of the lighting mechanism based on the human activity information when the state determination result is getting out of bed includes: when the state determination result is getting out of bed, acquiring a preset linear change curve of brightness, a target brightness value, and a gradual change duration; based on the linear change curve of brightness, decomposing the target brightness value into multiple equal time steps according to the gradual change duration; calculating the PWM duty cycle corresponding to the equal time steps, and adjusting the working state of the lighting mechanism based on the PWM duty cycle.

[0013] Optionally, in the sixth implementation of the first aspect of the present invention, after adjusting the working state of the lighting mechanism based on the human activity information when the state determination result is getting out of bed, the method further includes: determining whether the human target is in a continuous activity state based on the real-time sensing signal collected by the millimeter-wave radar sensor; if it is determined to be in a continuous activity state, calculating and determining the spatial position information of the human target based on the real-time sensing signal collected by the millimeter-wave radar sensor; obtaining real-time air conditioning parameters, and adjusting the operating parameters of the indoor unit of the air conditioner based on the real-time air conditioning parameters and the spatial position information using a fuzzy PID control algorithm.

[0014] A second aspect of the present invention provides an air conditioner indoor unit control device, comprising: a switching module, configured to control a millimeter-wave radar sensor to switch to a high-frequency scanning mode when it is determined that the period is nighttime; an acquisition module, configured to acquire human activity information based on real-time sensing signals collected by the millimeter-wave radar sensor, the human activity information including movement speed, limb movement characteristics, and spatial position information; a determination module, configured to pre-build and train a human state detection model, input the human activity information into the human state detection model for processing, and obtain a state determination result, the state determination result including frequent turning, sitting up, and getting out of bed; a first adjustment module, configured to adjust the operating parameters of the air conditioner indoor unit based on the state determination result and the human activity information when the state determination result is frequent turning or sitting up; and a second adjustment module, configured to adjust the working state of the lighting mechanism based on the human activity information when the state determination result is getting out of bed.

[0015] A third aspect of the present invention provides an air conditioner indoor unit control device, the air conditioner indoor unit control device comprising: a memory and at least one processor, the memory storing instructions; at least one processor calling the instructions in the memory to cause the air conditioner indoor unit control device to execute each step of the air conditioner indoor unit control method described above.

[0016] A fourth aspect of the present invention provides a computer-readable storage medium storing instructions that, when executed by a processor, implement the steps of the air conditioner indoor unit control method described in any of the preceding claims.

[0017] The air conditioner indoor unit control method of this invention realizes intelligent collaborative control of the air conditioner indoor unit and lighting mechanism in nighttime scenarios, constructing a complete logical chain of scene recognition, information collection, state determination, and precise control. By combining millimeter-wave radar sensors with human body state detection models, it achieves precise perception of human body state, effectively filtering non-target interference and distinguishing between three states: frequent turning, sitting up, and getting out of bed, significantly reducing the false touch rate. It executes air conditioner parameter adjustments or lighting control separately for different states, solving the problems of high false touch rate and unsuitable brightness of traditional nightlights, and compensating for the shortcomings of traditional air conditioners operating with fixed parameters, significantly improving users' nighttime sleep comfort and convenience of getting up at night. Simultaneously, the high-frequency scanning mode ensures the real-time and accuracy of state determination, providing a reliable data foundation for subsequent parameter adjustments and lighting control. Attached Figure Description

[0018] Figure 1 A logic flowchart of an air conditioner indoor unit control method provided in an embodiment of the present invention; Figure 2This is a schematic diagram of the structure of the air conditioner indoor unit control device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of the air conditioner indoor unit control device provided in an embodiment of the present invention. Detailed Implementation

[0019] This invention provides a method, apparatus, device, and storage medium for controlling an indoor unit of an air conditioner. In this invention, the terms "first," "second," "third," "fourth," etc. (if present)," in the specification, claims, and accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0020] This application discloses a control method for an indoor unit of an air conditioner. The applicable indoor unit includes a main control system and a millimeter-wave radar sensor, a lighting mechanism, an ambient light sensor, a temperature sensor, and a storage module electrically connected to the main control system. The main control system receives signals from each sensor, executes control algorithms, and outputs control commands. The lighting mechanism consists of six 0.2W LED beads, a frosted diffuser plate, and a PWM driver chip, integrated into a light trough along the lower edge of the indoor unit, supporting linear brightness adjustment. The light trough is a hidden light trough located in the middle of the lower edge of the indoor unit, and is physically isolated from the air outlet grille of the indoor unit to avoid airflow disturbance. The movement causes the lights to flicker; the millimeter-wave radar sensor uses 60-64GHz FMCW frequency-modulated continuous wave technology to collect sensor signals related to human activity; the millimeter-wave radar sensor is installed on the left side of the light slot, maintaining a certain distance (e.g., 5cm) from the LED beads to avoid electromagnetic interference; and the installation position of the millimeter-wave radar sensor is offset from the evaporator air outlet to prevent airflow from affecting the detection accuracy; the ambient light sensor is used to detect the ambient light intensity in real time, the temperature sensor is used to collect the real-time indoor ambient temperature, and the storage module is used to store preset parameters, model data, and fuzzy rule base information to ensure the stable operation of the main control system.

[0021] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of the air conditioner indoor unit control method in this invention includes: 101. When it is determined that it is nighttime, control the millimeter-wave radar sensor to switch to high-frequency scanning mode; In this embodiment, the determination of nighttime hours is achieved by an ambient light sensor integrated into the air conditioner indoor unit. The ambient light intensity is used to accurately pinpoint the nighttime usage scenario, enabling the switching of the millimeter-wave radar sensor's operating mode. This avoids energy waste caused by high-frequency scanning during non-nighttime hours, while ensuring the millimeter-wave radar sensor's sensitivity to human activity during nighttime hours, laying the foundation for subsequent signal acquisition and status determination.

[0022] 102. Human activity information is obtained based on the real-time sensing signals collected by the millimeter-wave radar sensor, wherein the human activity information includes movement speed, limb movement characteristics and spatial location information; In this embodiment, the millimeter-wave radar sensor transmits a 60-64GHz continuously modulated signal in real time. After the signal is reflected by the human body, it forms an echo signal. The main control system analyzes the difference between the transmitted signal and the echo signal to extract human activity information in layers. By acquiring three types of human activity information, comprehensive and reliable data support is provided for subsequent human status determination, avoiding control decision deviations due to missing information or errors.

[0023] 103. Pre-build and train a human state detection model, input the human activity information into the human state detection model for processing, and obtain state determination results, the state determination results including frequent turning, sitting up and getting out of bed; In this embodiment, by constructing and training a human state detection model, high-precision classification of human state is achieved, effectively filtering irrelevant signals such as pet activities and environmental interference, improving the accuracy of state determination, avoiding ineffective control caused by misjudgment, and providing accurate decision-making basis for subsequent air conditioning parameter adjustment and lighting control.

[0024] 104. When the state determination result is frequent turning or sitting up, adjust the operating parameters of the indoor unit of the air conditioner based on the state determination result and human activity information; In this embodiment, to address the issue of users frequently turning or sitting up due to an uncomfortable sleeping environment, the air conditioner's operating parameters are adaptively and precisely adjusted to quickly improve the indoor temperature environment, avoid discomfort caused by sudden parameter changes, and enhance the user's sleep comfort.

[0025] 105. When the status determination result is getting out of bed, adjust the working status of the lighting mechanism based on the human activity information; In this embodiment, when the status determination result is getting out of bed, the lighting is turned on as needed, which satisfies the need for safe lighting when getting up at night and avoids energy waste.

[0026] The air conditioner indoor unit control method disclosed in this application realizes intelligent collaborative control of the air conditioner indoor unit and lighting mechanism in nighttime scenarios, constructing a complete logical chain of scene recognition, information collection, state determination, and precise control. By combining millimeter-wave radar sensors with human body state detection models, it achieves precise perception of human body state, effectively filters non-target interference, and distinguishes between three states: frequent turning, sitting up, and getting out of bed, significantly reducing the false touch rate. It performs air conditioner parameter adjustment or lighting control separately for different states, which can solve the problems of high false touch rate and uncomfortable brightness of traditional night lights, and can also make up for the defects of fixed parameter operation of traditional air conditioners, significantly improving the user's nighttime sleep comfort and convenience of getting up at night. At the same time, the high-frequency scanning mode can ensure the real-time and accuracy of state determination, providing a reliable data foundation for subsequent parameter adjustment and lighting control.

[0027] Furthermore, in this embodiment of the invention, controlling the millimeter-wave radar sensor to switch to high-frequency scanning mode when it is determined to be during nighttime includes: 201. Obtain real-time ambient light intensity; 202. When the real-time ambient light intensity is ≤5 lux, it is determined that the current period is nighttime; In this embodiment, 5 lux is a preset threshold for determining the nighttime period. This threshold is determined based on the light perception characteristics of human body at night and the requirements for indoor ambient light at night in GB / T50033-2013 "Standard for Daylighting Design of Buildings". When the real-time ambient light intensity detected by the ambient light sensor is ≤5 lux, it is determined that the current period is nighttime.

[0028] 203. Control the millimeter-wave radar sensor to switch to high-frequency scanning mode, so that the scanning frequency of the millimeter-wave radar sensor is increased to 10 times / second; In this embodiment, the millimeter-wave radar sensor is in low-power scanning mode by default. In this mode, the scanning frequency is once every 2 seconds and the standby power consumption is ≤10mW. When switching to high-frequency scanning mode, the scanning frequency is increased to 10 times / second. This frequency setting is based on the needs of capturing human activity signals at night. It can ensure accurate capture of human movement signals in the range of 0.1-1m / s without increasing energy consumption due to excessive scanning.

[0029] Furthermore, in this embodiment of the invention, the acquisition of human activity information based on real-time sensing signals collected by the millimeter-wave radar sensor includes movement speed, limb movement characteristics, and spatial location information, including: 301. Based on the Doppler effect principle, calculate the frequency difference between the transmitted signal and the echo signal of the millimeter-wave radar sensor, and combine it with the operating frequency band parameters of the millimeter-wave radar sensor to calculate the moving speed of the human target. In this embodiment, based on the Doppler effect principle, the millimeter-wave radar sensor emits a 60-64GHz continuously modulated signal. The movement of the human target causes a frequency difference between the echo signal and the transmitted signal. The main control system calculates this frequency difference through fast Fourier transform and, combined with the electromagnetic wave propagation speed and the operating frequency band parameters (radar carrier frequency) of the millimeter-wave radar sensor, calculates the moving speed of the human target. Specifically: moving speed = (Doppler frequency difference × electromagnetic wave propagation speed) / (2 × radar carrier frequency).

[0030] 302. Extract the time-domain amplitude variation characteristics and phase variation period of the echo signal from the millimeter-wave radar sensor to generate limb movement features, wherein the limb movement features are movement feature vectors; In this embodiment, the extraction of limb movement features is achieved through signal processing technology, focusing on capturing the temporal amplitude variation characteristics and phase variation period of the echo signal. The temporal amplitude variation characteristics include peak value, mean, variance, rising slope, and falling slope. These feature parameters are integrated to generate a 1×5 dimension movement feature vector. The feature vectors corresponding to different limb movements have unique distribution patterns. For example, the peak amplitude of the standing up movement is 2-3 times that of the rolling over movement, and the phase variation period is 1.5-2 times that of the rolling over movement, thereby distinguishing different types of limb movements.

[0031] 303. Calculate the corresponding time delay based on the frequency difference, and calculate the straight-line distance between the human target and the millimeter-wave radar sensor according to the time delay and the electromagnetic wave propagation speed; 304. The MUSIC algorithm is used to estimate the angle of the echo signal collected by the millimeter-wave radar sensor to obtain the horizontal and vertical angles of the human target. 305. Based on the straight-line distance, horizontal angle, and vertical angle, determine the spatial position information of the human target, wherein the spatial position information is three-dimensional spatial coordinates; In this embodiment, the acquisition of spatial location information is divided into three steps. First, the time delay between the transmitted signal and the echo signal is calculated by the frequency difference. Combined with the formula: straight-line distance = (electromagnetic wave propagation speed × time delay) / 2, the straight-line distance between the human body and the radar sensor is obtained. Second, the covariance matrix of the echo signals received from multiple channels is constructed using the MUSIC algorithm. The signal subspace and noise subspace are obtained through eigenvalue decomposition. The horizontal and vertical angles of the human target are calculated using spatial spectrum estimation. The horizontal detection angle covers 0-120°, and the vertical detection angle covers 0-60°. Finally, the straight-line distance, horizontal angle, and vertical angle are converted into three-dimensional spatial coordinates (x, y, z) using the polar coordinate to rectangular coordinate conversion formula to complete the human body location.

[0032] Furthermore, in this embodiment of the invention, the pre-built and trained human state detection model, which processes the human activity information by inputting it into the human state detection model to obtain a state determination result, includes: 401. Construct a CNN-LSTM hybrid model and obtain sample datasets in multiple scenarios. The sample datasets include samples of users of different ages getting up at night, turning over, sitting up, and walking, as well as activity samples of pets of different sizes and environmental interference signal samples. In this embodiment, the human state detection model adopts a CNN-LSTM hybrid model. The connection relationship between the layers of the model is as follows: The input layer receives a feature matrix with a dimension of N×4, where N is the number of sampling points. In high-frequency mode, 5 seconds of data are taken, i.e., N=50. The four features are the standardized data of distance, velocity, amplitude, and phase, respectively. The output of the input layer is directly connected to the first convolutional layer of the CNN. The first convolutional layer uses 3×3 convolutional kernels, with a number of 32, a stride of 1, a "same" padding method, and a ReLU activation function. Its output is connected to the first max pooling layer. The first max pooling layer uses 2×2 pooling kernels, with a stride of 2. Its output is connected to the second convolutional layer. The second convolutional layer uses 3×3 convolutional kernels, with a number of 64, and a stride of 1. The padding method is "same," the activation function is ReLU, and the output is connected to the second max pooling layer. The second max pooling layer uses a 2×2 pooling kernel with a stride of 2 and an output dimension of 13×1×64. This output is directly input into the LSTM temporal analysis layer. The LSTM layer has 64 hidden units, a dropout rate of 0.2, and an output dimension of 64. Its output is connected to the first fully connected layer. The first fully connected layer has 32 neurons and uses the ReLU activation function. Its output is connected to the second fully connected layer. The second fully connected layer has 16 neurons and uses the ReLU activation function. Its output is connected to the output layer. The output layer uses the Softmax activation function and outputs three types of state probability values, forming a complete hierarchical connection chain.

[0033] 402. Perform data cleaning, feature extraction and normalization preprocessing on the sample dataset, and conduct multiple rounds of iterative training on the constructed CNN-LSTM hybrid model based on the preprocessed sample data to obtain the human state detection model. In this embodiment, before training the CNN-LSTM hybrid model, it is necessary to obtain a multi-scene sample dataset. The samples cover the actions of users of different ages, such as children, youth, and the elderly, including getting up at night, turning over, sitting up, and walking; the activities of pets of different sizes, such as small, medium, and large; and environmental interference signals such as shaking doors and windows, airflow disturbances, and temperature fluctuations. The total number of samples is no less than 100,000 sets to ensure sample diversity.

[0034] 403. Perform feature standardization and dimension reconstruction preprocessing on the human activity information to generate a feature matrix; In this embodiment, the sample dataset is first cleaned by removing outliers using the 3σ criterion, and then key features such as distance, speed, amplitude, and phase are extracted to obtain feature data. Next, the Min-Max normalization method is used to compress the feature data to the 0-1 range to eliminate dimensional differences. Based on the preprocessed sample data, the training set, validation set, and test set are divided in a 7:2:1 ratio. The model is trained iteratively for 200 rounds using the training and test sets, employing the Adam optimizer with an initial learning rate of 0.001, which decays by 50% every 50 rounds. Training stops when the validation set classification accuracy is ≥99% for 10 consecutive rounds, resulting in the human state detection model.

[0035] 404. Input the feature matrix into the human body state detection model to obtain the state determination result; In this embodiment, the real-time collected human activity information is normalized using the same Min-Max normalization method as the training samples, and then reconstructed into an N×4 feature matrix. After being input into the human state detection model, the model outputs the probability values ​​of three states: frequent turning, sitting up, and getting out of bed. The category corresponding to the maximum probability is the state determination result.

[0036] Furthermore, in this embodiment of the invention, when the state determination result is frequent turning or sitting up, adjusting the operating parameters of the indoor unit of the air conditioner based on the state determination result and human activity information includes: 501. When the state determination result is frequent rotation or sitting up, obtain real-time environmental parameters and real-time air conditioning parameters; In this embodiment, when the state determination result is frequent rotation or sitting up, the main control system first starts the parameter acquisition process, and collects the real-time indoor ambient temperature through the temperature sensor; at the same time, it reads the real-time air conditioning parameters from the main control system, including the current set temperature, wind speed level and air outlet angle; specifically, when the state determination result is frequent rotation, the action frequency must meet ≥3 times / minute; when the state determination result is sitting up, the action duration must meet ≥2 seconds.

[0037] 502. Map the state determination results, human activity information, real-time environmental parameters, and real-time air conditioning parameters into fuzzy linguistic variables; In this embodiment, the state determination result, human activity information, real-time ambient temperature, and real-time air conditioning parameters are then mapped to fuzzy linguistic variables: the state determination result is mapped to "frequent rotation" and "sitting up"; the action frequency is mapped to "high frequency" (≥3 times / minute) and "medium frequency" (1-2 times / minute); the ambient temperature is mapped to "high" (>24℃), "suitable" (22-24℃), and "low" (<22℃); the wind speed level is mapped to "low", "low medium", and "medium"; and the air outlet angle is mapped to "small angle" (0-20°), "medium angle" (21-40°), and "large angle" (41-60°).

[0038] 503. Retrieve and match the pre-built fuzzy rule base based on the fuzzy linguistic variables to generate preliminary adjustment parameters; In this embodiment, the pre-built fuzzy rule base is constructed based on 500 sets of user comfort test data and air conditioner operating characteristics. It contains clear input-output correspondences. For example, "frequent rotation, high ambient temperature, low current wind speed" corresponds to "low temperature by 1°C, low wind speed, and 10° deflection of the air outlet". "sitting up, suitable ambient temperature, low current wind speed" corresponds to "temperature remains unchanged, low wind speed, and 5° deflection of the air outlet". Preliminary adjustment parameters are generated by searching and matching the fuzzy rule base. In the preliminary adjustment parameters, the temperature adjustment range is -2°C to 0°C with a step size of 0.5°C; the wind speed level is limited to low or low-medium wind; and the air outlet angle adjustment range is 5-15°.

[0039] 504. Based on the aforementioned preliminary adjustment parameters, the operating parameters of the indoor unit of the air conditioner are adjusted using a PID algorithm; In this embodiment, based on the initial adjustment parameters, a PID algorithm is used for dynamic optimization. The PID algorithm has a proportional coefficient Kp=0.8, an integral coefficient Ki=0.2, and a derivative coefficient Kd=0.1. The proportional element quickly responds to parameter deviations, the integral element eliminates static deviations, and the derivative element suppresses overshoot. Finally, the target operating parameters of the indoor unit of the air conditioner are determined and adjusted. During the adjustment process, the parameter change rate is ≤1℃ / minute and the angle change rate is ≤5° / second to ensure smooth parameter transition, achieve rapid improvement of the indoor temperature environment, enhance user sleep comfort, and ensure the stability of air conditioner operation.

[0040] Furthermore, in this embodiment of the invention, adjusting the working state of the lighting mechanism based on the human activity information when the state determination result is getting out of bed includes: 601. When the state determination result is getting out of bed, obtain the preset linear change curve of brightness, the target brightness value and the gradual change duration; In this embodiment, the gradual change duration includes the on-light gradual change duration and the off-light gradual change duration. When the status determination result is getting out of bed, the main control system reads the preset parameters related to lighting adjustment from the storage module. These preset parameters are personalized by the user through the air conditioner remote control or smart home APP. Specifically, the target brightness value is set in the range of 3-10 lux, with a default of 8 lux; the on-light gradual change duration is set in the range of 3-5 seconds, with a default of 3 seconds; the off-light gradual change duration is set in the range of 5-8 seconds, with a default of 5 seconds. The brightness linear change curve is a linear rising or linear falling curve. The parameter settings must comply with the EN 62471 photobiological safety standard to avoid blue light hazards.

[0041] 602. Based on the linear change curve of brightness, the target brightness value is decomposed into multiple equal time steps according to the gradual change duration; In this embodiment, based on a preset linear brightness change curve, the target brightness value is uniformly decomposed into multiple equal time steps according to the gradual change duration, and the time step is set to 10 milliseconds; for example, when the target brightness is 8 lux and the starting brightness gradual change duration is 3 seconds, it is decomposed into 300 time steps of 10 milliseconds each, and the brightness increment corresponding to each time step is 8 lux / 300≈0.027 lux.

[0042] 603. Calculate the PWM duty cycle corresponding to the equal time step, and adjust the working state of the lighting mechanism based on the PWM duty cycle; In this embodiment, the dimming accuracy of the PWM driver chip of the lighting mechanism is 1%, and the brightness is linearly mapped to the PWM duty cycle. The mapping formula is D=(L / Lmax)×100%, where D is the duty cycle, L is the current target brightness, Lmax is the maximum brightness, and Lmax=30lux. Through this mapping formula, the duty cycle corresponding to each time step is calculated. The main control system transmits the duty cycle signal to the PWM driver chip, which drives the LED beads to adjust the brightness according to the preset curve, so as to achieve a smooth linear change in lighting brightness, avoid strong light stimulation caused by instantaneous on and off, protect the user's eye health, and conform to the characteristics of human nighttime visual perception.

[0043] Furthermore, in this embodiment of the invention, after adjusting the working state of the lighting mechanism based on the human activity information when the state determination result is getting out of bed, the method further includes: 701. Based on the real-time sensing signals collected by the millimeter-wave radar sensor, determine whether the human target is in a state of continuous activity; In this embodiment, the millimeter-wave radar sensor maintains a high-frequency scanning frequency of 10 times / second. Based on the collected real-time sensing signals, it determines whether the human body is in a continuous activity state from three dimensions: movement speed, spatial position, and time. When the human body's movement speed is continuously detected to be in the range of 0.1-1m / s, and the horizontal displacement of the human body is greater than 0.3m every 5 seconds; if the duration of the above two conditions being met is ≥5 seconds, and there is no stationary state exceeding 5 seconds during the period, it is determined to be a continuous activity state.

[0044] 702. If it is determined that the target is in a state of continuous activity, the spatial position information of the human target is calculated and determined based on the real-time sensing signal collected by the millimeter-wave radar sensor. In this embodiment, if it is determined that the person is in a continuous activity state, the millimeter-wave radar sensor continues to collect sensing signals at a frequency of 10 times / second, and updates the three-dimensional spatial position coordinates of the human body in real time through the distance calculation, angle estimation and coordinate conversion methods described in steps 301-305.

[0045] 703. Obtain real-time air conditioning parameters, and based on the real-time air conditioning parameters and the spatial location information, use a fuzzy PID control algorithm to adjust the operating parameters of the indoor unit of the air conditioner; In this embodiment, the main control system reads real-time air conditioning parameters, including the current set temperature, fan speed level, and air outlet angle. Combined with the real-time updated spatial location information of the human target, a fuzzy PID control algorithm is used to adjust the air conditioning operating parameters. For example, when the horizontal distance between the human target and the indoor unit is less than 2 meters, the fan speed is set to low; when the horizontal distance is 2-3 meters, the fan speed is set to low-medium; and when the horizontal distance is greater than 3 meters, the fan speed is set to medium. The air outlet angle is adjusted according to the horizontal angle of the human body to ensure the airflow direction is aligned with... The system positions the user at a 30-45° angle to avoid direct airflow. When the ambient temperature is above 24°C, the indoor unit temperature is lowered by 0.5-1°C. When the ambient temperature is below 22°C, the indoor unit temperature remains unchanged. Through multi-dimensional continuous activity status determination and real-time location updates, the system dynamically optimizes air conditioning parameters during continuous user activity, avoiding localized temperature discomfort or direct airflow issues caused by user movement, ensuring a comfortable temperature environment for users throughout their nighttime activities. Simultaneously, it achieves coordinated control of lighting and air conditioning, further enhancing the system's intelligence and user experience.

[0046] The air conditioner indoor unit control method in the embodiments of the present invention has been described above. The air conditioner indoor unit control device in the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 2 One embodiment of the air conditioner indoor unit control device in this invention includes: The switching module 801 is used to control the millimeter-wave radar sensor to switch to high-frequency scanning mode when it is determined that it is nighttime. The acquisition module 802 is used to acquire human activity information based on the real-time sensing signals collected by the millimeter-wave radar sensor. The human activity information includes movement speed, limb movement characteristics and spatial location information. The determination module 803 is used to pre-build and train a human state detection model, input the human activity information into the human state detection model for processing, and obtain a state determination result. The state determination result includes frequent turning, sitting up and getting out of bed. The first adjustment module 804 is used to adjust the operating parameters of the indoor unit of the air conditioner based on the state determination result and human activity information when the state determination result is frequent rotation or sitting up. The second adjustment module 805 is used to adjust the working state of the lighting mechanism based on the human activity information when the state determination result is getting out of bed.

[0047] Based on the same ideas as the methods in the above embodiments, the apparatus provided in this application can implement the methods in the above embodiments.

[0048] above Figure 2 The air conditioner indoor unit control device in this embodiment of the invention will be described in detail from the perspective of modular functional entities. The air conditioner indoor unit control device in this embodiment of the invention will be described in detail from the perspective of hardware processing.

[0049] Figure 3 This is a schematic diagram of the structure of an air conditioner indoor unit control device 900 provided in an embodiment of the present invention. The air conditioner indoor unit control device 900 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 910 and memory 920, and one or more storage media 930 (e.g., one or more mass storage devices) storing application programs 933 or data 932. The memory 920 and storage media 930 can be temporary or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the diagram), each module may include a series of instruction operations on the air conditioner indoor unit control device 900. Furthermore, the processor 910 may be configured to communicate with the storage media 930 and execute the series of instruction operations in the storage media 930 on the air conditioner indoor unit control device 900 to implement the steps of the air conditioner indoor unit control method provided in the above-described method embodiments.

[0050] The air conditioner indoor unit control device 900 may also include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 3 The structure of the air conditioner indoor unit control device shown does not constitute a limitation on the air conditioner indoor unit control device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0051] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of an air conditioner indoor unit control method.

[0052] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0053] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0054] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling an indoor unit of an air conditioner, characterized in that, The air conditioner indoor unit includes a main control system and a millimeter-wave radar sensor and a lighting mechanism electrically connected to the main control system. The air conditioner indoor unit control method includes: When it is determined that it is nighttime, the millimeter-wave radar sensor is switched to high-frequency scanning mode; Human activity information is obtained based on real-time sensing signals collected by the millimeter-wave radar sensor, including movement speed, limb movement characteristics and spatial location information. A human state detection model is pre-built and trained. The human activity information is input into the human state detection model for processing to obtain a state determination result. The state determination result includes frequent turning, sitting up, and getting out of bed. When the state determination result is frequent turning or sitting up, the operating parameters of the indoor unit of the air conditioner are adjusted based on the state determination result and human activity information. When the status determination result is getting out of bed, the working status of the lighting mechanism is adjusted based on the human activity information.

2. The air conditioner indoor unit control method according to claim 1, characterized in that, When it is determined that it is nighttime, controlling the millimeter-wave radar sensor to switch to high-frequency scanning mode includes: Obtain real-time ambient light intensity; When the real-time ambient light intensity is ≤5 lux, it is determined that the current period is nighttime; Control the millimeter-wave radar sensor to switch to high-frequency scanning mode, thereby increasing the scanning frequency of the millimeter-wave radar sensor to 10 times / second.

3. The air conditioner indoor unit control method according to claim 1, characterized in that, The method acquires human activity information based on real-time sensing signals collected by the millimeter-wave radar sensor. This human activity information includes movement speed, limb movement characteristics, and spatial location information, including: Based on the Doppler effect principle, the frequency difference between the transmitted signal and the echo signal of the millimeter-wave radar sensor is calculated, and combined with the operating frequency band parameters of the millimeter-wave radar sensor, the moving speed of the human target is calculated. The temporal amplitude variation characteristics and phase variation period of the millimeter-wave radar sensor echo signal are extracted to generate limb movement features, which are movement feature vectors. The corresponding time delay is calculated based on the frequency difference, and the straight-line distance between the human target and the millimeter-wave radar sensor is calculated based on the time delay and the electromagnetic wave propagation speed. The MUSIC algorithm is used to estimate the angle of the echo signal collected by the millimeter-wave radar sensor to obtain the horizontal and vertical angles of the human target. Based on the straight-line distance, horizontal angle, and vertical angle, the spatial position information of the human target is determined, and the spatial position information is three-dimensional spatial coordinates.

4. The air conditioner indoor unit control method according to claim 1, characterized in that, The pre-built and trained human state detection model processes the human activity information input into the human state detection model to obtain a state determination result, including: A CNN-LSTM hybrid model was constructed, and sample datasets were obtained in multiple scenarios. The sample datasets included samples of users of different ages getting up at night, turning over, sitting up, and walking, as well as activity samples of pets of different sizes and environmental interference signal samples. Data cleaning, feature extraction and normalization preprocessing are performed on the sample dataset, and the constructed CNN-LSTM hybrid model is trained in multiple rounds based on the preprocessed sample data to obtain the human state detection model. The human activity information is preprocessed by feature standardization and dimension reconstruction to generate a feature matrix; The feature matrix is ​​input into the human state detection model to obtain the state determination result.

5. The air conditioner indoor unit control method according to claim 1, characterized in that, When the state determination result indicates frequent turning or sitting up, adjusting the operating parameters of the indoor air conditioner unit based on the state determination result and human activity information includes: When the state determination result is frequent rotation or sitting up, real-time environmental parameters and real-time air conditioning parameters are obtained. The state determination results, human activity information, real-time environmental parameters, and real-time air conditioning parameters are mapped into fuzzy linguistic variables. Based on the fuzzy linguistic variables, a pre-built fuzzy rule base is retrieved and matched to generate preliminary adjustment parameters; Based on the initial adjustment parameters, the operating parameters of the indoor unit of the air conditioner are adjusted using a PID algorithm.

6. The air conditioner indoor unit control method according to claim 1, characterized in that, When the status determination result is getting out of bed, adjusting the working state of the lighting mechanism based on the human activity information includes: When the state determination result is getting out of bed, obtain the preset linear change curve of brightness, the target brightness value and the gradual change duration; Based on the linear change curve of brightness, the target brightness value is decomposed into multiple equal time steps according to the gradual change duration; Calculate the PWM duty cycle corresponding to the equal time step, and adjust the working state of the lighting mechanism based on the PWM duty cycle.

7. The air conditioner indoor unit control method according to claim 1, characterized in that, When the state determination result is "getting out of bed", after adjusting the working state of the lighting mechanism based on the human activity information, the method further includes: Based on the real-time sensing signals collected by the millimeter-wave radar sensor, it is determined whether the human target is in a state of continuous activity. If it is determined that the human target is in a state of continuous activity, the spatial location information of the human target is calculated and determined based on the real-time sensing signals collected by the millimeter-wave radar sensor. Real-time air conditioning parameters are obtained, and based on the real-time air conditioning parameters and the spatial location information, the operating parameters of the indoor unit of the air conditioner are adjusted using a fuzzy PID control algorithm.

8. An air conditioner indoor unit control device, characterized in that, include: The switching module is used to control the millimeter-wave radar sensor to switch to high-frequency scanning mode when it is determined that it is nighttime. The acquisition module is used to acquire human activity information based on the real-time sensing signals collected by the millimeter-wave radar sensor. The human activity information includes movement speed, limb movement characteristics and spatial location information. The determination module is used to pre-build and train a human state detection model, input the human activity information into the human state detection model for processing, and obtain the state determination result, which includes frequent turning, sitting up and getting out of bed. The first adjustment module is used to adjust the operating parameters of the indoor unit of the air conditioner based on the state determination result and human activity information when the state determination result is frequent rotation or sitting up. The second adjustment module is used to adjust the working state of the lighting mechanism based on the human activity information when the state determination result is getting out of bed.

9. An air conditioner indoor unit control device, characterized in that, The air conditioner indoor unit control device includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the air conditioner indoor unit control device to perform the steps of the air conditioner indoor unit control method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the various steps of the air conditioner indoor unit control method as described in any one of claims 1-7.

Images