A multi-zone dynamic sleep temperature control method based on human thermal physiological rhythm

Abstract

This invention relates to the field of temperature control technology, specifically to a multi-zone dynamic sleep temperature control method based on human thermophysiological rhythms. It involves dividing the sleep support surface along the longitudinal direction of the body into multiple temperature control zones, acquiring the user's sleep stage information in real time, and then executing dynamic sleep temperature control. On one hand, during the sleep onset stage, the increased temperature in the feet triggers reflexive vasodilation of peripheral blood vessels to open heat dissipation pathways; the decreased temperature in the head lowers the hypothalamic temperature set point to enhance heat dissipation drive; and the slightly cooler torso provides a temperature gradient for core heat dissipation, thus significantly improving the rate of core body temperature decrease and significantly shortening the sleep latency. On the other hand, after entering the deep sleep maintenance stage, the method maintains the peripheral vasodilation state established during the sleep onset stage with minimal energy consumption, preventing the feet from cooling back and causing vasoconstriction, and avoiding micro-awakening triggered by torso temperature fluctuations, achieving a balance between thermal comfort and energy conservation throughout the night.

Description

Technical Field

[0001] This invention relates to the field of temperature control technology, and more specifically, to a multi-zone dynamic sleep temperature control method based on human thermophysiological rhythms. Background Technology

[0002] Sleep quality directly affects human health. The rhythmic changes in the body's core temperature are one of the key physiological mechanisms regulating the sleep-wake cycle. When falling asleep, the body needs to lower its core temperature through peripheral vasodilation, thereby inducing sleepiness and entering a deep sleep state. However, traditional mattresses, blankets, and other devices that are heated or cooled as a whole cannot accurately match the different temperature needs of different parts of the body at different stages of sleep, often resulting in "hot head and cold feet" or "local overheating," which actually interferes with sleep. In existing technologies, such as Chinese patent application CN115227066A, a zoned temperature-controlled mattress based on sleeping posture and the number of people is disclosed. This solution sets target temperatures for different body parts by detecting the number of people and their sleeping posture. However, this solution mainly focuses on adapting to sleeping posture and does not deeply integrate the core body temperature change patterns of the human sleep cycle (falling asleep, deep sleep, and the second half of the night) for dynamic and proactive temperature control intervention. It fails to specifically design for the specific physiological need of "actively warming the feet to accelerate the decrease of core body temperature during the fall-off stage" and also lacks an active head cooling strategy linked to the sleep stage. Therefore, there is still significant room for improvement in shortening the time to fall asleep, prolonging the duration of deep sleep, and reducing awakenings in the second half of the night.

[0003] In view of this, we propose a multi-zone dynamic sleep temperature control method based on the human body's thermophysiological rhythm. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-zone dynamic sleep temperature control method based on human thermophysiological rhythms, in order to solve the problem of how to integrate dynamic sleep staging, human zone thermophysiological rhythms, differentiated actuators, and sleep cycle linkage control strategies mentioned in the background art, thereby solving sleep problems such as slow sleep onset, light deep sleep, and easy awakening in the second half of the night. This invention uses the human sleep cycle as the core time axis and the thermophysiological rhythms of various body surface areas as spatial coordinates. It solves the problems raised in the background technology by actively inducing sleep during the onset of sleep, maintaining stable deep sleep, and coordinating differentiated actuators for rapid foot warming, slow head cooling, and constant trunk temperature.

[0005] To achieve the above objectives, this invention provides a multi-zone dynamic sleep temperature control method based on human thermophysiological rhythms, comprising the following steps: S1. Divide the sleep support surface along the longitudinal direction of the human body into multiple temperature control zones, including at least a head temperature control zone, a trunk core temperature control zone, and a foot temperature control zone. Configure an active heating actuator for the foot temperature control zone, an active heat dissipation actuator for the head temperature control zone, and a constant temperature actuator for the trunk core temperature control zone. S2. Acquire the user's sleep stage information in real time, wherein the sleep stage includes at least the sleep onset stage and the deep sleep maintenance stage; S3. Perform dynamic sleep temperature control, including: During the sleep stage, the active heating actuator of the foot temperature control area is activated simultaneously for active heating and the active heat dissipation actuator of the head temperature control area for active heat dissipation, so that the temperature of the foot temperature control area is 2℃-4℃ higher than the temperature of the core torso temperature control area, and the temperature of the head temperature control area is stabilized in the range of 22.0-23.5℃. At the same time, the constant temperature actuator of the core torso temperature control area is controlled to actively adjust in both directions to maintain a reference slightly cool temperature. During the deep sleep maintenance phase, the active heating actuator stops active heating and switches to low-power heat preservation, the active cooling actuator stops active cooling and switches to low-power cooling, and the constant temperature actuator controls the temperature fluctuation of the torso core temperature control area within ±0.3℃ and assigns it the highest control priority.

[0006] As a further improvement to this technical solution, a longitudinally extending partition line is sewn along the transverse direction of the human body in the middle of the sleep support surface, dividing the mattress into multiple temperature control zones, including at least a head temperature control zone, a core torso temperature control zone, and a foot temperature control zone. Each temperature control zone is an independent chamber, and the internal structure of each independent chamber, from top to bottom, consists of: a fabric layer, a temperature control conduction layer, a comfort layer, a sensor layer, a zoned support layer, a heat insulation layer, a bottom actuator layer, and a bottom reinforcement layer, wherein: The fabric layer is made of a composite fabric of Tencel and antibacterial knitted fabric, which is intended to come into direct contact with human skin. The active heating actuator, the constant temperature actuator, and the active heat dissipation actuator are pre-installed at the bottom layer of the actuator and correspond to the foot temperature control area, the torso core temperature control area, and the head temperature control area, respectively. The temperature control conduction layer is based on the comfort layer to meet the needs of different temperature control zones; The sensor layer is a PVDF piezoelectric thin film array, which is laid between the comfort layer and the partitioned support layer; The zoned support layer uses independent pocket springs to provide differentiated support that matches the weight distribution and physiological curves of different parts of the body, so that the spine can maintain a natural neutral position when lying supine or on the side, while providing a base for the temperature control conduction layer; The bottom reinforcement layer uses high-density rigid foam to provide structural support for the mattress, while also providing a flat mounting base for the actuator.

[0007] As a further improvement to this technical solution, the temperature control conductive layer, based on the comfort layer, meets the needs of different temperature control zones, including: The temperature control conduction layer in the head temperature control area is composed of 3D mesh fabric and air guide channel. The active heat dissipation actuator adopts a fan, which draws in ambient air and sends it into the air guide channel through a flexible air duct. The air diffuses horizontally and vertically in the grid of the air guide channel, and then passes upward through the 3D mesh fabric, so that the airflow is distributed throughout the entire head area. The temperature control conduction layer of the core temperature control zone of the torso adopts a silicone water circulation pipeline embedded in the comfort layer. The silicone tubes are arranged in a serpentine manner. The constant temperature actuator includes a micro water pump, a semiconductor cooling chip and a thermistor heater. The semiconductor cooling chip performs the cooling operation, and the thermistor heater performs the heating operation to generate constant temperature water. The micro water pump drives the constant temperature water to circulate in the pipeline. The heat or cold is conducted through the pipe wall to the outer wall of the silicone tube, and then transferred to the comfort layer through thermally conductive silicone grease. Finally, it is radiated and convectioned to the human skin through the fabric layer. The temperature control conduction layer of the foot temperature control zone uses a flexible carbon fiber heating film, which is directly attached to the lower surface of the comfort layer with the heating surface facing upwards. An insulation layer is set below. After the active heating actuator controls the flexible carbon fiber heating film to be powered on, the heat is conducted from the flexible carbon fiber heating film to the sponge layer, and then through the fabric layer to the foot skin.

[0008] As a further improvement to this technical solution, step S2, which involves acquiring the user's sleep stage information in real time, includes the following steps: S2.1 When mechanical stress is sensed by the PVDF piezoelectric thin film array of the sensor layer, the internal dipoles undergo orientation changes, generating a charge signal proportional to the stress at both ends of the electrode. The static pressure signal, body motion signal and weak vibration signal are separated by a digital bandpass filter. S2.2 The signal is converted into a voltage signal by a multi-channel charge amplifier, and then digitized by an analog-to-digital converter. Features are extracted based on static pressure signal, body motion signal and weak vibration signal. S2.3 Calculate the probabilities of wakefulness, light sleep, deep sleep and REM sleep states based on the characteristics, and determine the sleep onset stage and deep sleep maintenance stage according to the time sequence rules.

[0009] As a further improvement to this technical solution, the feature extraction based on static pressure signals, body motion signals, and weak vibration signals includes: The static pressure signal is low-pass filtered to obtain a pressure distribution map that changes slowly over time. The sum of the static pressure values ​​is calculated. If the sum exceeds a preset threshold, it is determined that the person is in bed; otherwise, it is determined that the person is out of bed. After the person is in bed, the position and shape of the center of gravity of the pressure distribution map are analyzed. When the center of gravity is close to the center of the mattress and the pressure distribution is relatively uniform, the person is lying on their back. When the center of gravity is biased to the left or right and the pressure distribution is in a strip pattern, the person is lying on their side. When the pressure distribution is concentrated in the chest and abdomen area, the person is lying on their stomach. The 30-second window is divided into 60 sub-windows of 0.5 seconds each. The root mean square value of the body motion signal in each sub-window is calculated to determine whether it exceeds the threshold. The number of sub-windows that exceed the threshold is counted and divided by 60 to obtain the body motion index. Peak detection is performed on weak vibration signals, the time interval between adjacent peaks is calculated, the reciprocal is taken to obtain the breathing frequency, and the ratio of the standard deviation to the mean of the envelope within a 30-second window is calculated.

[0010] As a further improvement to this technical solution, the probabilities of wakefulness, light sleep, deep sleep, and REM sleep states are calculated based on characteristics, and the sleep onset stage and deep sleep maintenance stage are determined according to temporal rules, including the following steps: The system automatically learns features at different time scales using a one-dimensional convolutional neural network model, and outputs the probabilities of wakefulness, light sleep, deep sleep, and REM sleep, where: The start time of the sleep onset stage is defined as the moment when the probability of wakefulness is below 50% for two consecutive windows, the body movement index is below 0.2 for two consecutive windows, and the coefficient of variation of respiratory amplitude is below 0.25 for two consecutive windows. The end time of the sleep onset stage is defined as the moment when the probability of deep sleep is above 60% for three consecutive windows. The start time of the deep sleep maintenance phase is the end time of the sleep onset phase. The end time of the deep sleep maintenance phase is defined as the moment when the probability of deep sleep is below 30% for five consecutive windows after more than 5 hours of sleep onset or more than 3 hours of sleep onset.

[0011] As a further improvement to this technical solution, in step S3, a controller is used to perform dynamic sleep temperature control; When the decision logic confirms the start of the sleep phase, the controller synchronously executes the following control actions: A PWM signal with a duty cycle of 80% to 100% is output to the active heating actuator, causing the foot temperature to rise to 28.5 to 30.5°C, which is 2 to 4°C higher than the current core temperature of the torso, within 5 to 10 minutes. The rapid temperature rise of the foot triggers the reflexive dilation of blood vessels in the foot, and the core heat is conducted to the foot. The active cooling actuator outputs a PWM signal at 60% to 80% speed to stabilize the head temperature at 22.0 to 23.5℃. The continuous cooling of the head causes the hypothalamus to lower the body temperature set point and enhance the heat dissipation command. The thermostatic actuator is controlled to maintain a baseline slightly cool temperature of 25.0 to 26.0°C, and active heating is prohibited to provide a heat dissipation gradient for the core heat.

[0012] As a further improvement to this technical solution, during the deep sleep maintenance phase, the low-power heat preservation is as follows: the controller detects the foot temperature every 3-10 minutes, and when the foot temperature is below 27℃, the active heating actuator is activated for 15-60 seconds to maintain the foot temperature in the range of 27.5℃-28.5℃. The low-power heat dissipation is achieved by reducing the rotation speed of the active heat dissipation actuator to 10%-20% of its maximum speed, thereby raising the head temperature to 23.5℃-24.5℃, and ensuring that the head temperature control area is 1.5℃-2.0℃ lower than the torso core temperature control area.

[0013] As a further improvement to this technical solution, in the deep sleep maintenance phase, the highest control priority is specifically configured as follows: When the measured temperature of the core temperature control zone of the torso deviates from the target temperature by more than ±0.3℃, the controller will first call the constant temperature actuator for correction. At this time, the temperature of the foot temperature control zone is allowed to fluctuate within ±1.0℃ without active intervention. After the temperature deviation of the core temperature control zone of the torso returns to within ±0.3℃, the temperature adjustment needs of the head temperature control zone and the foot temperature control zone will be processed in turn.

[0014] As a further improvement to this technical solution, the multiple temperature control zones also include a left shoulder and neck temperature control zone, a right shoulder and neck temperature control zone, and a waist and hip temperature control zone. The temperature control strategy is adjusted according to the sleeping posture information. During the deep sleep maintenance phase, the left shoulder and neck temperature control zone, the right shoulder and neck temperature control zone, and the waist and hip temperature control zone are controlled to maintain a constant temperature and comfort state. The temperature range of the constant temperature and comfort state is 26.0℃-27.5℃.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: This multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm divides the sleep support surface longitudinally into head, core trunk, and foot temperature control zones, and equips each zone with active heat dissipation, constant temperature, and active heating actuators. During the sleep onset stage, the rapid heating of the feet and active cooling of the head are simultaneously activated according to the sleep rhythm, while maintaining a slightly cool baseline temperature for the trunk. During the deep sleep maintenance stage, active heating / heat dissipation is stopped and switched to low-power heat preservation / heat dissipation, giving the trunk constant temperature the highest control priority. On the one hand, during the sleep onset stage, the simultaneous stimulation of active heating of the feet, active heat dissipation of the head, and maintaining a slightly cool baseline temperature for the core trunk causes the feet to heat up, triggering peripheral vasodilation to open heat dissipation pathways. The cooling of the head lowers the hypothalamic temperature set point to enhance the heat dissipation driving force, and the slightly cool trunk provides a temperature gradient for core heat dissipation. The three work together to improve the overall rate of core temperature decrease, thereby significantly shortening the sleep latency. On the other hand, once deep sleep is maintained, the strong heating of the feet and the strong heat dissipation of the head cease, switching to a low-power insulation and slightly cooling state, while giving the core temperature control area of ​​the torso the highest priority for constant temperature control. This synergistic mode maintains the peripheral vasodilation state established during the sleep onset stage with minimal energy consumption, preventing the feet from cooling back and causing vasoconstriction, and avoiding micro-awakening triggered by torso temperature fluctuations. Thus, while ensuring the stability of deep sleep, a balance between thermal comfort and energy saving throughout the night is achieved. Attached Figure Description

[0016] Figure 1 This is an overall flowchart of the present invention; Figure 2 This is a cross-sectional schematic diagram of the sleep support surface of the present invention; Figure 3 This is a schematic diagram illustrating the principle of dividing the sleep support surface according to the present invention. Detailed Implementation

[0017] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1 When the human body is lying supine or on its side, there are significant differences in the thermophysiological rhythms of different parts along the longitudinal direction (from head to toe). Specifically: Head: The brain has a high metabolic rate and produces about 15-20% of the body's heat. However, the scalp has a large blood flow and is the body's main heat dissipation window. If the head temperature is too high during sleep, it will directly inhibit the hypothalamus's sleep initiation signal, so it is necessary to actively cool it down.

[0019] The core of the torso includes core heat-generating organs such as the heart and liver. Temperature fluctuations in this area are rapidly transmitted to the brain via the autonomic nervous system, triggering micro-arousals. Therefore, highly stable thermoregulation is required.

[0020] The feet: a peripheral region far from the core, where vascular smooth muscle is rich in α-adrenergic receptors and is sensitive to temperature changes. Moderate warming can induce vasodilation and open up the core's heat dissipation pathways.

[0021] Please see Figures 1-3 As shown, this embodiment provides a multi-zone dynamic sleep temperature control method based on human thermophysiological rhythms, including the following steps: S1. Divide the sleep support surface along the longitudinal direction of the human body into multiple temperature control zones, including at least a head temperature control zone, a trunk core temperature control zone, and a foot temperature control zone. Configure an active heating actuator for the foot temperature control zone, an active heat dissipation actuator for the head temperature control zone, and a constant temperature actuator for the trunk core temperature control zone.

[0022] The sleep support surface can be a smart mattress, a smart mattress cover, or a temperature-controlled pad that can be laid on an existing mattress. Taking a smart mattress as an example, a longitudinally extending dividing line is sewn along the horizontal (left-right) direction of the human body in the middle of the sleep support surface, dividing the mattress into multiple temperature-controlled zones, including at least a head temperature-controlled zone, a core torso temperature-controlled zone, and a foot temperature-controlled zone. Each temperature-controlled zone is an independent chamber, and its internal structure from top to bottom consists of: a fabric layer, a temperature-conducting layer, a comfort layer, a sensor layer, a zoned support layer, an insulation layer, a bottom actuator layer, and a bottom reinforcement layer. The fabric layer uses a composite fabric of Tencel and antibacterial knitted fabric, with a weight of 280 grams per square meter. Tencel is a lyocell fiber extracted from eucalyptus pulp, which has natural moisture-wicking properties and a silky feel. The antibacterial knitted fabric is treated with silver ions to inhibit the growth of common bacteria such as Staphylococcus aureus and Escherichia coli. It is used in direct contact with human skin, providing a skin-friendly and soft touch and improving comfort. Through the capillary effect of the fibers, it quickly absorbs and diffuses sweat, keeping the skin dry. The silver ion antibacterial treatment prevents the growth of bacteria and mites during long-term use. In addition, its high breathability ensures that the hot air generated by the head's active heat dissipation and the moisture emitted by the torso can be discharged in time, avoiding a stuffy feeling.

[0023] The comfort layer uses memory foam, a viscoelastic polyurethane foam. Its molecular chains have a long relaxation time at room temperature and slowly rebound after being compressed. When a person lies down, the pressure of their body weight causes the memory foam to creep, gradually conforming to the body's curves and distributing the pressure to a larger contact area, thereby reducing peak pressure. At the same time, the low thermal conductivity of the foam gives it a certain heat diffusion capacity, which can evenly distribute local heat and avoid local hot spots.

[0024] The active heating actuator, the constant temperature actuator, and the active heat dissipation actuator are pre-installed at the bottom of the actuator and correspond to the foot temperature control area, the torso core temperature control area, and the head temperature control area, respectively.

[0025] The temperature-controlled conductive layer, based on the comfort layer, meets the needs of different temperature control zones, including: The temperature control conduction layer in the head temperature control area is composed of 3D mesh fabric and air guide channels. 3D mesh fabric is a three-dimensional woven fabric consisting of two layers of mesh fabric and a middle monofilament support layer. It is 8 mm thick and has a compression resilience of ≥95%. The air guide channels are horizontal and vertical grooves pre-molded on the surface of the comfort sponge layer. The grooves are 10 mm wide, 8 mm deep, and 30 mm apart, forming a grid-like air channel. The active heat dissipation actuator uses a fan to draw in ambient air (room temperature 22 to 26 degrees Celsius) and send it into the air guide channels through flexible air ducts. The air diffuses horizontally and vertically in the grid of the air guide channels and then passes upward through the 3D mesh fabric, so that the airflow is distributed throughout the entire head area. At the same time, it avoids the airflow directly impacting the skin and causing discomfort. The three-dimensional open structure of the 3D mesh fabric changes the airflow from a concentrated flow to a large-area, low-speed uniform penetration. When the airflow reaches the lower surface of the fabric layer, it has diffused from the initial concentrated jet (wind speed of about 2 to 3 meters per second) into a uniform permeable flow (wind speed of no more than 0.3 meters per second), eliminating the feeling of wind while ensuring heat dissipation efficiency. The temperature control conduction layer of the core temperature control zone of the torso uses silicone water circulation pipes embedded in the comfort layer. The silicone pipes are arranged in a serpentine (S-shape). The pipes and the comfort layer are filled with thermally conductive silicone grease to eliminate air gaps. The thermostatic actuator includes a miniature water pump, a semiconductor cooling chip, and a thermistor heater. The semiconductor cooling chip performs cooling operation, and the thermistor heater performs heating operation to produce constant temperature water. The miniature water pump drives the constant temperature water (the temperature is regulated by the semiconductor cooling chip or heater) to circulate in the pipes. Heat (or cold) is conducted through the pipe wall to the outer wall of the silicone pipe, and then transferred to the comfort layer through the thermally conductive silicone grease. Finally, it is radiated and convectioned to the human skin through the fabric layer. Because the spacing between the silicone water circulation pipes is only 35 mm, the heat-affected zones of adjacent silicone water circulation pipes overlap, forming a continuous and uniform temperature field with a temperature difference of no more than ±0.3 degrees Celsius. This ensures large-area coverage without dead corners and meets the uniformity requirements of the torso area. The temperature control conduction layer of the foot temperature control zone uses a flexible carbon fiber heating film, which is directly attached to the lower surface of the comfort layer with the heating side facing upwards. An insulation layer is set below. After the active heating actuator controls the carbon fiber flexible heating film to be powered on, the heat is conducted from the carbon fiber flexible heating film to the sponge layer, and then through the fabric layer to the skin of the foot. The dilation of blood vessels in the foot requires "rapid and large-amplitude" temperature stimulation. Slow temperature rise (such as 0.5 degrees Celsius per minute) or small temperature rise (such as less than 2 degrees Celsius) cannot effectively trigger the reflexive dilation of blood vessels.

[0026] The sensor layer uses a PVDF piezoelectric thin film array, which is laid between the comfort layer and the partition support layer. The resolution is 16 by 8 points, that is, 16 sensing units in the horizontal direction and 8 sensing units in the vertical direction, for a total of 128 independent sensing points.

[0027] The zoned support layer uses individually pocketed springs to provide differentiated support that matches the weight distribution and physiological curves of different parts of the body, keeping the spine in a natural neutral position when lying supine or on its side. It also provides a base for the temperature-controlled conductive layer. The head area has a hardness of 25 to 35 (slightly soft), the shoulder and neck area has a hardness of 20 to 30 (softest), the back area has a hardness of 40 to 50 (medium-firm), the waist area has a hardness of 55 to 70 (firmest), the hip area has a hardness of 35 to 45 (medium-soft), the thigh area has a hardness of 35 to 45 (medium-soft), and the calf and foot area has a hardness of 30 to 40 (slightly soft). Each area is separated by non-woven fabric partitions to maintain independent hardness. Each spring is individually wrapped in a non-woven fabric bag, and adjacent springs can extend and contract independently without interfering with each other. When a person is lying down, the springs corresponding to the shoulders and hips (where the body weight is greater) have a greater compression, while the springs corresponding to the waist and head (where the body weight is less) have a smaller compression.

[0028] The insulation layer uses aerogel felt, which is 5 mm thick and has a thermal conductivity of no more than 0.02 watts per meter Kelvin. Aerogel felt is a flexible felt-like composite material with silica aerogel as the filling material and glass fiber as the reinforcing skeleton, which prevents heat from being lost to the bottom of the mattress.

[0029] The bottom reinforcement layer uses high-density rigid foam with an IFD indentation hardness of 100 and a thickness of 15 mm. It is used to provide structural support for the mattress, prevent deformation, and provide a flat mounting base for the actuator.

[0030] S2. Obtain the user's sleep stage information in real time. The sleep stage includes at least the sleep onset stage and the deep sleep maintenance stage.

[0031] The real-time acquisition of users' sleep stage information includes the following steps: S2.1 When mechanical stress (such as pressure changes caused by body movement, weak vibrations caused by respiration and heart rate) is sensed by the PVDF piezoelectric thin film array in the sensor layer, the internal dipoles undergo an orientation change, generating a charge signal proportional to the stress at both ends of the electrode. The signal is then separated into three frequency bands by a digital bandpass filter, including: The static component of the user's weight causes the PVDF piezoelectric film array to generate a static pressure signal with constant compressive deformation. When a user rolls over, adjusts their posture, or moves their limbs, the pressure acting on the PVDF piezoelectric film array changes, outputting a body movement signal. The rhythmic vibrations of the human body caused by heart rate and breathing are transmitted to the mattress surface through bones, muscles and soft tissues, causing the PVDF film to generate periodic charge output and form a weak vibration signal. S2.2 The signal is converted into a voltage signal by a multi-channel charge amplifier, and then digitized by an analog-to-digital converter. Features are extracted based on static pressure signal, body motion signal and weak vibration signal. S2.3 Calculate the probabilities of wakefulness, light sleep, deep sleep and REM sleep states based on the characteristics, and determine the sleep onset stage and deep sleep maintenance stage according to the time sequence rules.

[0032] It is worth noting that the features extracted based on static pressure signals, body motion signals, and weak vibration signals include: The static pressure signal is low-pass filtered to obtain a pressure distribution map that changes slowly over time. The sum of the static pressure values ​​is calculated. If the sum exceeds a preset threshold (e.g., 5 kg), it is determined that the person is in bed; otherwise, it is determined that the person is out of bed. After the person is in bed, the position and shape of the center of gravity of the pressure distribution map are analyzed. When the center of gravity is close to the center of the mattress and the pressure distribution is relatively uniform, the person is lying on their back. When the center of gravity is biased to the left or right and the pressure distribution is in a strip pattern, the person is lying on their side. When the pressure distribution is concentrated in the chest and abdomen area, the person is lying on their stomach. The 30-second window is divided into 60 sub-windows of 0.5 seconds each. The root mean square value of the body movement signal in each sub-window is calculated to determine whether it exceeds the threshold. The number of sub-windows that exceed the threshold is counted and divided by 60 to obtain the body movement index. The body movement index ranges from 0 to 1. The body movement index during deep sleep is less than 0.05, the body movement index during light sleep is between 0.05 and 0.2, the body movement index during REM sleep is between 0.1 and 0.3, and the body movement index during wakefulness is usually higher than 0.3. Peak detection was performed on weak vibration signals, and the time interval between adjacent peaks was calculated. The reciprocal of the time interval was used to obtain the respiratory rate. The normal resting respiratory rate of adults is 12 to 20 breaths per minute, which may drop to 10 to 14 breaths per minute during deep sleep. At the same time, the envelope of the respiratory signal (i.e., the instantaneous amplitude of the respiratory waveform) was extracted by Hilbert transform, and the ratio of the standard deviation to the mean of the envelope within a 30-second window was calculated. During deep sleep, breathing is regular and uniform, and the ratio is less than 0.1. During light sleep and REM sleep, respiratory variability increases, and the ratio is between 0.1 and 0.3. During wakefulness or the early stage of sleep, breathing may be irregular, and the ratio exceeds 0.3.

[0033] Furthermore, based on the characteristics, the probabilities of wakefulness, light sleep, deep sleep, and REM sleep states are calculated. According to the timing rules, the sleep onset stage and the deep sleep maintenance stage are determined, including the following steps: This system automatically learns features at different time scales using a one-dimensional convolutional neural network (CNN) model. The CNN model consists of an input layer, three convolutional layers (each containing 64, 128, and 256 1x3 convolutional kernels respectively), two fully connected layers, and a softmax output layer. The input feature vector consists of 128 features extracted within a 30-second time window. The four nodes of the output layer correspond to the probabilities of wakefulness, light sleep, deep sleep, and REM sleep, respectively, and output the probabilities of wakefulness, light sleep, deep sleep, and REM sleep. The start time of the sleep onset stage is defined as the moment when the probability of wakefulness is below 50% for two consecutive windows, the body movement index is below 0.2 for two consecutive windows, and the coefficient of variation of respiratory amplitude is below 0.25 for two consecutive windows. The end time of the sleep onset stage is defined as the moment when the probability of deep sleep is above 60% for three consecutive windows. The start time of the deep sleep maintenance phase is the end time of the sleep onset phase. The end time of the deep sleep maintenance phase is defined as the moment when the probability of deep sleep is below 30% for five consecutive windows after more than 5 hours of sleep onset or more than 3 hours of sleep onset.

[0034] S3. Perform dynamic sleep temperature control, including: Firstly, during the sleep-onset phase, the active heating mechanism in the foot temperature control zone and the active cooling mechanism in the head temperature control zone are activated simultaneously to actively heat the body. This ensures that the temperature in the foot temperature control zone is 2-4°C higher than the temperature in the core torso temperature control zone, while the temperature in the head temperature control zone remains stable between 22.0-23.5°C. Simultaneously, the constant temperature mechanism in the core torso temperature control zone actively regulates bidirectionally to maintain a slightly cooler baseline temperature (25.0-26.0°C), primarily focusing on cooling or heat dissipation. This prevents the torso from passively warming due to foot heating, ensuring a stable temperature gradient for core heat dissipation. This creates a physiological synchronicity between peripheral vasodilation when the foot temperature control zone rises and central cooling when the head temperature control zone cools, synergistically accelerating the decrease in core body temperature. Maintaining a slightly cooler baseline temperature in the torso prevents passive warming caused by foot heating, ensuring that core heat is effectively dissipated rather than retained. The combined effect of these three mechanisms increases the rate of core body temperature decrease compared to a single method, thereby shortening the time to fall asleep.

[0035] Secondly, during the deep sleep maintenance phase, the active heating actuator stops active heating and switches to low-power heat preservation, and the active cooling actuator stops active cooling and switches to low-power cooling. At the same time, the constant temperature actuator controls the temperature fluctuation of the core temperature control area of ​​the torso within ±0.3℃ and assigns it the highest control priority. When the temperature requirements of multiple temperature control areas conflict, the temperature stability of the core temperature control area of ​​the torso is guaranteed first, so as to avoid the temperature fluctuation of the core area of ​​the torso causing awakening during deep sleep.

[0036] In S3, a controller is used to perform dynamic sleep temperature control; When the decision logic confirms the start of the sleep phase, the controller synchronously executes the following control actions: A PWM signal with a duty cycle of 80% to 100% is output to the active heating actuator, causing the foot temperature to rise to 28.5 to 30.5°C, which is 2 to 4°C higher than the current core temperature of the torso, within 5 to 10 minutes. The rapid temperature rise of the foot triggers the reflexive dilation of blood vessels in the foot, and the core heat is conducted to the foot. The active cooling actuator outputs a PWM signal at 60% to 80% speed to stabilize the head temperature at 22.0 to 23.5℃. The continuous cooling of the head causes the hypothalamus to lower the body temperature set point and enhance the heat dissipation command. The thermostatic actuator is controlled to maintain a reference slightly cool temperature of 25.0 to 26.0°C, and active heating is prohibited to provide a heat dissipation gradient for the core heat. Therefore, the three control actions are initiated simultaneously within 5 milliseconds, forming a positive feedback amplification loop. The rapid warming of the feet triggers peripheral vasodilation, opening the heat dissipation pathway. The continuous cooling of the head lowers the body temperature set point, enhancing the heat dissipation driving force. The torso remains relatively cool, providing a heat dissipation gradient. The three work together to increase the rate at which the core body temperature decreases.

[0037] Furthermore, during deep sleep, the feet do not require the same strong heat stimulation as during the sleep onset phase. Maintaining a slightly warm temperature is sufficient to prevent vasoconstriction. If the foot temperature is too low (below 26°C), blood vessels may constrict again, leading to increased peripheral resistance and core temperature fluctuations, potentially triggering micro-arousals. A slightly warm range of 27.5°C to 28.5°C is sufficient to maintain a mildly dilated state of blood vessels in the feet without causing overheating. Therefore, during the deep sleep maintenance phase, low-power warming is achieved by the controller detecting foot temperature every 3-10 minutes. When the foot temperature drops below 27°C, the active heating actuator is activated for 15-60 seconds to maintain the foot temperature within the 27.5°C-28.5°C range. This facilitates the transition from strong heat stimulation to weak warming, maintaining the peripheral vasodilation established during the sleep onset phase with minimal energy consumption and preventing rebound vasoconstriction caused by foot cooling. Low-power heat dissipation involves reducing the speed of the active cooling actuator to 10%-20% of its maximum speed, allowing the head temperature to rise to 23.5℃-24.5℃. Furthermore, the head temperature control zone is 1.5℃-2.0℃ lower than the torso core temperature control zone. This helps shift the goal of head cooling from induction to maintenance. The purpose of low-power heat dissipation is to keep the head slightly cool with minimal energy consumption, while avoiding interference with deep sleep due to fan noise or airflow disturbances.

[0038] During the deep sleep maintenance phase, the highest control priority is specifically configured as follows: When the measured temperature of the core torso temperature control zone deviates from the target temperature by more than ±0.3℃, the controller prioritizes calling the thermostatic actuator for correction. At this time, the temperature of the foot temperature control zone is allowed to fluctuate within ±1.0℃ without active intervention. After the temperature deviation of the core torso temperature control zone returns to within ±0.3℃, the temperature adjustment needs of the head temperature control zone and the foot temperature control zone are processed in turn. This is beneficial when resources are limited (such as a maximum power output of 150W) or when there are conflicts in the needs of multiple zones, sacrificing the foot temperature stability which has the least impact on sleep, and ensuring the torso temperature stability which has the greatest impact on sleep.

[0039] Multiple temperature control zones include a left neck and shoulder temperature control zone, a right neck and shoulder temperature control zone, and a waist and hip temperature control zone. The temperature control strategy is adjusted according to the sleeping posture information. During the deep sleep maintenance stage, the temperature control zones of the left neck and shoulder, right neck and shoulder, and waist and hip are kept at a constant and comfortable temperature. The temperature range of the constant and comfortable temperature is 26.0℃-27.5℃. The sleeping posture adaptively adjusts the temperature difference between the left and right neck and shoulder, further improving the comfort of side-lying users.

[0040] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples and are not intended to limit the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A multi-zone dynamic sleep temperature control method based on human thermophysiological rhythms, characterized in that, Includes the following steps: S1. Divide the sleep support surface along the longitudinal direction of the human body into multiple temperature control zones, including at least a head temperature control zone, a trunk core temperature control zone, and a foot temperature control zone. Configure an active heating actuator for the foot temperature control zone, an active heat dissipation actuator for the head temperature control zone, and a constant temperature actuator for the trunk core temperature control zone. S2. Acquire the user's sleep stage information in real time, wherein the sleep stage includes at least the sleep onset stage and the deep sleep maintenance stage; S3. Perform dynamic sleep temperature control, including: During the sleep stage, the active heating actuator of the foot temperature control area is activated simultaneously for active heating and the active heat dissipation actuator of the head temperature control area for active heat dissipation, so that the temperature of the foot temperature control area is 2℃-4℃ higher than the temperature of the core torso temperature control area, and the temperature of the head temperature control area is stabilized in the range of 22.0-23.5℃. At the same time, the constant temperature actuator of the core torso temperature control area is controlled to actively adjust in both directions to maintain a reference slightly cool temperature. During the deep sleep maintenance phase, the active heating actuator stops active heating and switches to low-power heat preservation, the active cooling actuator stops active cooling and switches to low-power cooling, and the constant temperature actuator controls the temperature fluctuation of the torso core temperature control area within ±0.3℃ and assigns it the highest control priority.

2. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 1, characterized in that: The sleep support surface has a longitudinally extending dividing line sewn along the horizontal direction of the human body, dividing the mattress into multiple temperature control zones, including at least a head temperature control zone, a core torso temperature control zone, and a foot temperature control zone. Each temperature control zone is an independent chamber, and the internal structure of each independent chamber, from top to bottom, consists of: a fabric layer, a temperature control conduction layer, a comfort layer, a sensor layer, a zoned support layer, an insulation layer, a bottom actuator layer, and a bottom reinforcement layer. The fabric layer is made of a composite fabric of Tencel and antibacterial knitted fabric, which is intended to come into direct contact with human skin. The active heating actuator, the constant temperature actuator, and the active heat dissipation actuator are pre-installed at the bottom layer of the actuator and correspond to the foot temperature control area, the torso core temperature control area, and the head temperature control area, respectively. The temperature control conductive layer is based on the comfort layer to meet the needs of different temperature control zones; The sensor layer is a PVDF piezoelectric thin film array, which is laid between the comfort layer and the partitioned support layer; The zoned support layer uses independent pocket springs to provide differentiated support that matches the weight distribution and physiological curves of different parts of the body, so that the spine can maintain a natural neutral position when lying supine or on the side, while providing a base for the temperature control conduction layer; The bottom reinforcement layer uses high-density rigid foam to provide structural support for the mattress, while also providing a flat mounting base for the actuator.

3. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 2, characterized in that: The temperature-controlled conductive layer, based on the comfort layer, meets the needs of different temperature control zones, including: The temperature control conduction layer in the head temperature control area is composed of 3D mesh fabric and air guide channel. The active heat dissipation actuator adopts a fan, which draws in ambient air and sends it into the air guide channel through a flexible air duct. The air diffuses horizontally and vertically in the grid of the air guide channel, and then passes upward through the 3D mesh fabric, so that the airflow is distributed throughout the entire head area. The temperature control conduction layer of the core temperature control zone of the torso adopts a silicone water circulation pipeline embedded in the comfort layer. The silicone tubes are arranged in a serpentine manner. The constant temperature actuator includes a micro water pump, a semiconductor cooling chip and a thermistor heater. The semiconductor cooling chip performs the cooling operation, and the thermistor heater performs the heating operation to generate constant temperature water. The micro water pump drives the constant temperature water to circulate in the pipeline. The heat or cold is conducted through the pipe wall to the outer wall of the silicone tube, and then transferred to the comfort layer through thermally conductive silicone grease. Finally, it is radiated and convectioned to the human skin through the fabric layer. The temperature control conduction layer of the foot temperature control zone uses a flexible carbon fiber heating film, which is directly attached to the lower surface of the comfort layer with the heating surface facing upwards. An insulation layer is set below. After the active heating actuator controls the flexible carbon fiber heating film to be powered on, the heat is conducted from the flexible carbon fiber heating film to the sponge layer, and then through the fabric layer to the foot skin.

4. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 3, characterized in that: In step S2, the user's sleep stage information is acquired in real time, including the following steps: S2.1 When mechanical stress is sensed by the PVDF piezoelectric thin film array of the sensor layer, the internal dipoles undergo orientation changes, generating a charge signal proportional to the stress at both ends of the electrode. The static pressure signal, body motion signal and weak vibration signal are separated by a digital bandpass filter. S2.2 The signal is converted into a voltage signal by a multi-channel charge amplifier, and then digitized by an analog-to-digital converter. Features are extracted based on static pressure signal, body motion signal and weak vibration signal. S2.3 Calculate the probabilities of wakefulness, light sleep, deep sleep and REM sleep states based on the characteristics, and determine the sleep onset stage and deep sleep maintenance stage according to the time sequence rules.

5. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 4, characterized in that: The feature extraction based on static pressure signals, body motion signals, and weak vibration signals includes: The static pressure signal is low-pass filtered to obtain a pressure distribution map that changes slowly over time. The sum of the static pressure values ​​is calculated. If the sum exceeds a preset threshold, it is determined that the person is in bed; otherwise, it is determined that the person is out of bed. After the person is in bed, the position and shape of the center of gravity of the pressure distribution map are analyzed. When the center of gravity is close to the center of the mattress and the pressure distribution is relatively uniform, the person is lying on their back. When the center of gravity is biased to the left or right and the pressure distribution is in a strip pattern, the person is lying on their side. When the pressure distribution is concentrated in the chest and abdomen area, the person is lying on their stomach. The 30-second window is divided into 60 sub-windows of 0.5 seconds each. The root mean square value of the body motion signal in each sub-window is calculated to determine whether it exceeds the threshold. The number of sub-windows that exceed the threshold is counted and divided by 60 to obtain the body motion index. Peak detection is performed on weak vibration signals, the time interval between adjacent peaks is calculated, the reciprocal is taken to obtain the breathing frequency, and the ratio of the standard deviation to the mean of the envelope within a 30-second window is calculated.

6. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 5, characterized in that: The probabilities of wakefulness, light sleep, deep sleep, and REM sleep states are calculated based on characteristics. The sleep onset and deep sleep maintenance stages are determined according to temporal rules, including the following steps: The system automatically learns features at different time scales using a one-dimensional convolutional neural network model, and outputs the probabilities of wakefulness, light sleep, deep sleep, and REM sleep, where: The start time of the sleep onset stage is defined as the moment when the probability of wakefulness is below 50% for two consecutive windows, the body movement index is below 0.2 for two consecutive windows, and the coefficient of variation of respiratory amplitude is below 0.25 for two consecutive windows. The end time of the sleep onset stage is defined as the moment when the probability of deep sleep is above 60% for three consecutive windows. The start time of the deep sleep maintenance phase is the end time of the sleep onset phase. The end time of the deep sleep maintenance phase is defined as the moment when the probability of deep sleep is below 30% for five consecutive windows after more than 5 hours of sleep onset or more than 3 hours of sleep onset.

7. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 1, characterized in that: In S3, a controller is used to perform dynamic sleep temperature control; When the decision logic confirms the start of the sleep phase, the controller synchronously executes the following control actions: A PWM signal with a duty cycle of 80% to 100% is output to the active heating actuator, causing the foot temperature to rise to 28.5 to 30.5°C, which is 2 to 4°C higher than the current core temperature of the torso, within 5 to 10 minutes. The rapid temperature rise of the foot triggers the reflexive dilation of blood vessels in the foot, and the core heat is conducted to the foot. The active cooling actuator outputs a PWM signal at 60% to 80% speed to stabilize the head temperature at 22.0 to 23.5℃. The continuous cooling of the head causes the hypothalamus to lower the body temperature set point and enhance the heat dissipation command. The thermostatic actuator is controlled to maintain a baseline slightly cool temperature of 25.0 to 26.0°C, and active heating is prohibited to provide a heat dissipation gradient for the core heat.

8. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 7, characterized in that: During the deep sleep maintenance phase, the low-power heat preservation is as follows: the controller detects the foot temperature every 3-10 minutes. When the foot temperature is below 27°C, the active heating actuator is activated for 15-60 seconds to maintain the foot temperature in the range of 27.5°C-28.5°C. The low-power heat dissipation is achieved by reducing the rotation speed of the active heat dissipation actuator to 10%-20% of its maximum speed, thereby raising the head temperature to 23.5℃-24.5℃, and ensuring that the head temperature control area is 1.5℃-2.0℃ lower than the torso core temperature control area.

9. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 8, characterized in that: During the deep sleep maintenance phase, the highest control priority is specifically configured as follows: When the measured temperature of the core temperature control zone of the torso deviates from the target temperature by more than ±0.3℃, the controller will first call the constant temperature actuator for correction. At this time, the temperature of the foot temperature control zone is allowed to fluctuate within ±1.0℃ without active intervention. After the temperature deviation of the core temperature control zone of the torso returns to within ±0.3℃, the temperature adjustment needs of the head temperature control zone and the foot temperature control zone will be processed in turn.

10. The multi-zone dynamic sleep temperature control method based on human thermophysiological rhythm according to claim 5, characterized in that: The multiple temperature control zones also include a left shoulder and neck temperature control zone, a right shoulder and neck temperature control zone, and a waist and hip temperature control zone. The temperature control strategy is adjusted according to the sleeping posture information. During the deep sleep maintenance phase, the left shoulder and neck temperature control zone, the right shoulder and neck temperature control zone, and the waist and hip temperature control zone are controlled to maintain a constant temperature and comfort state. The temperature range of the constant temperature and comfort state is 26.0℃-27.5℃.

Images