CN111238005A

CN111238005A - Automatic control system and method of air conditioner sleep mode based on sleep staging

Info

Publication number: CN111238005A
Application number: CN202010044131.5A
Authority: CN
Inventors: 连之伟; 曹婷
Original assignee: Shanghai Jiaotong University
Current assignee: Shanghai Jiaotong University
Priority date: 2020-01-15
Filing date: 2020-01-15
Publication date: 2020-06-05

Abstract

The invention provides an automatic control system and method of an air conditioner sleep mode based on sleep staging, which comprises the following steps: a sleep monitoring module: monitoring and reflecting physiological parameters of a sleep stage of a user; a learning module: performing machine learning on the sleep stage condition of the user according to the physiological parameters, and sending a temperature adjusting instruction; a regulation module: and regulating and controlling the environmental parameters according to the temperature regulating instruction. The invention combines a scientific objective sleep staging method, and objectively improves the sleep quality of the user in terms of mechanism; the invention improves the control logic, and makes the whole automatic control strategy have higher precision and application value by means of an AI model machine learning tool; the invention considers the characteristics of individual difference and the like of the sleeping conditions of the users, and the invention can be popularized to more users.

Description

Automatic control system and method of air conditioner sleep mode based on sleep staging

Technical Field

The invention relates to the technical field of sleep medicine, in particular to an automatic control system and method of an air conditioner sleep mode based on sleep staging.

Background

Sleep is an indispensable physiological phenomenon for human beings, and can keep the health of the human body, supplement physical strength and improve the working capacity. The sleep of the human body consists of two alternating different sleep periods, one non-rapid eye movement sleep (NREM) and the other rapid eye movement sleep (REM). Wherein the non-rapid eye movement sleep is divided into three stages, which are respectively: sleep stage 1 (N1), sleep stage 2 (N2), sleep stage 3 (N3). A typical sleep process for healthy adults is roughly as follows: arousal-non-rapid eye movement sleep, i.e. NREM phase (sleep 1 phase-sleep 2 phase-sleep 3 phase-sleep 2 phase) -rapid eye movement phase, i.e. REM phase-non-rapid eye movement sleep (sleep 1 phase-sleep 2 phase-sleep 3 phase-sleep 2 phase) -rapid eye movement phase-non-rapid eye movement phase-arousal. The quality of sleep is closely related to the health of human body, and the quality of good sleep is very important. It has been found that people with poor sleep quality are now more and more widespread. Insomnia can lead to mental fatigue, poor decision making, learning disabilities and metabolic disorders, and in severe cases, chronic disease. The quality of human sleep is affected by many factors, such as health, emotional state, bedding system and objective environmental factors. The health condition and the emotional state belong to subjective factors influencing sleep and are uncontrollable factors, and the bedding and physical environment factors belong to objective factors and are objectively controllable factors. Studies have shown a close and inseparable link between human thermal regulation and sleep rhythms. Therefore, the invention focuses the eyesight to the thermal environment in the objective physical environment, and hopefully improves the sleep quality of the sleeping healthy people by improving the hot and humid environment during the sleep.

In recent years, especially the study on the influence of summer air-conditioning room temperature on sleep quality is very popular, some researchers find the optimal temperature value of summer air-conditioning room for promoting sleep quality, and other researchers think that the influence on sleep quality when the overnight room temperature is changed is tried to be found because the core temperature and the skin temperature of the whole night of the human body are continuously changed along with the sleep period. However, at present, the problem of how to change the hot and humid environment of a room to achieve the purpose of improving the sleep quality and reducing the energy consumption of the air conditioner still exists.

The invention aims to provide a new idea for the automatic control strategy of the sleep mode of the air conditioner which is widely researched at present. The environmental parameters such as room temperature and humidity are changed to improve the sleep quality, so that for the purpose, because the sleep of the users has inter-individual difference, the sleep data of each user needs to be learned by a machine, and the sleep stage conversion characteristics of the users throughout the night are memorized. Meanwhile, the objective thermal environment is adjusted according to the thermal environment required by the physiology of the user in different sleep stages, so that the REM time length and the slow wave sleep (N3 period) time length are prolonged, and the sleep quality of the user is further improved.

In view of the fact that the research on the automatic control strategy of the environmental parameters such as the temperature and the humidity of the air-conditioning room is very popular and has certain application requirements, a plurality of temperature-adjusting control logic strategies combined with the air conditioner are already appeared in the market at present. Through the search and discovery of the prior patent documents, the invention patent of application number CN201910857143.7 discloses an air conditioner sleep control method, device and air conditioner, comprising: acquiring a sleep mode operation signal of the air conditioner; in the sleep mode operation process, recording an adjusting instruction sent by a user to an air conditioner and the sending time of the adjusting instruction, and generating and storing a self-adjusting control instruction according to the adjusting instruction and the sending time; and when the air conditioner operates the sleep mode again, operating the self-adjusting control instruction before the first preset time of the sending time. The method only records the air conditioner adjusting behavior of the user during the sleeping period, and does not start from a sleeping staging mechanism to improve the sleeping effect of the user. The invention patent with the application number of CN201910902521.9 discloses an environment temperature self-adaptive adjusting system based on sleep physiological signal monitoring, which aims to solve the problems that the prior art is inconvenient to use and cannot realize environment adjustment based on physiological signals, but the system excessively discusses the realization problems of the system in the application process, including equivalent resistance and control voltage corresponding to a field effect tube in a calculation filtering circuit and a gain circuit, and obtaining environment temperature adjustment quantity and the like according to environment temperature signals, physiological signals and physiological characteristic signals. Therefore, the automatic control strategy of the sleep mode of the air conditioner based on the sleep stage of the human body needs to be researched and innovated more deeply.

Therefore, the automatic control strategy idea of the sleep mode of the air conditioner is to fundamentally make clear the sleep stage mechanism of the human body and combine objective indexes and scientific indexes for judging the sleep stage so as to realize reasonable and automatic control of the sleep mode of environmental parameters such as the temperature, the humidity and the like of an air-conditioned room.

Patent document CN109059214A (application number: 201810932323.2) discloses an automatic sleep control method of an air conditioner, which includes the steps of: the method comprises the steps that the air conditioner is powered on to operate, and detection data of the air conditioner are obtained, wherein the detection data comprise current time, current environment brightness and current environment temperature; and when the detection data meet the automatic sleep mode entering condition, controlling the air conditioner to operate the automatic sleep mode. The invention also discloses an air conditioner and a computer readable storage medium.

Disclosure of Invention

Aiming at the defects in the prior art, the invention aims to provide an automatic control system and method of a sleep mode of an air conditioner based on sleep staging.

The automatic control system of the sleep mode of the air conditioner based on the sleep stage provided by the invention comprises:

a sleep monitoring module: monitoring and reflecting physiological parameters of a sleep stage of a user;

a learning module: performing machine learning on the sleep stage condition of the user according to the physiological parameters, and sending a temperature adjusting instruction;

a regulation module: and regulating and controlling the environmental parameters according to the temperature regulating instruction.

Preferably, the sleep monitoring module comprises: the duration of the transition segment of the user's N2-N3 phase, the brain waveform near the end of the user's N3 phase, and the brain waveform near the end of the user's REM phase are monitored.

Preferably, the learning module comprises:

a machine recording module: the recording includes: the gender, age, and type of bedding in the bedroom of the user;

a machine learning module: the learning includes: the method comprises the steps of carrying out electroencephalogram stage division on a user, prolonging physiological temperatures of the N3 stage and the REM stage, and actually sensing the time length for setting environmental parameters by the user.

Preferably, the learning module learns the time length required for the air conditioner to adjust to the environmental parameter set by the user, namely the time difference between a certain sleep time and the start of the period N3 or the end of the period REM;

and determining the time required for adjusting the environmental parameters set by the user according to the air conditioner, and sending an adjusting instruction to the air conditioner adjusting and controlling center and the sleep stage time of the user.

Preferably, the regulatory module comprises: determining physiological environment parameters required by different sleep stages according to the thermal regulation intensity and rate; and inputting the physiological environment parameters required by different sleep periods into a learning module for machine learning to obtain the set sleep heat environment parameters required to be adjusted by the air conditioner.

Preferably, the physiological environment parameters required for the different sleep stages include: prolonging physiological temperature of N3 phase and REM phase.

Preferably, the thermal environment parameters include temperature and humidity.

Preferably, the air conditioner is automatically adjusted according to the thermal environment parameters and the sleep stage condition of the user.

Preferably, the sleep monitoring module comprises: detecting the period from lying on the bed to before entering the sleep stage N1 period, namely the time when the user lies on the bed is used as the time when the air conditioning equipment is switched to the sleep mode automatic control; according to the change of the thermal environment requirement and the user requirement, the physiological signals of the user, including electroencephalogram, blood pressure, heart rate and skin temperature, are learned, and the thermal environment parameters before the user enters sleep are regulated and controlled.

The automatic control method of the sleep mode of the air conditioner based on the sleep stage provided by the invention comprises the following steps:

a sleep monitoring step: monitoring and reflecting physiological parameters of a sleep stage of a user;

a learning step: performing machine learning on the sleep stage condition of the user according to the physiological parameters, and sending a temperature adjusting instruction;

regulating and controlling: and regulating and controlling the environmental parameters according to the temperature regulating instruction.

Compared with the prior art, the invention has the following beneficial effects:

1. the invention combines a scientific objective sleep staging method, and objectively improves the sleep quality of the user in terms of mechanism;

2. the learning content of the AI model in the invention considers the important factors in the current sleep heat environment research, improves the control logic compared with other strategies, and makes the whole automatic control strategy have higher precision and application value by means of the tool of machine learning of the AI model;

3. the invention considers the characteristics of individual difference and the like of the sleeping conditions of the users through the AI model machine learning module, so that the invention can be popularized to more users.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments with reference to the following drawings:

fig. 1 is a frame diagram of an automatic control strategy of an air conditioner sleep mode based on human sleep stages.

The module part is represented by a solid line frame, the instruction or judgment content is represented by a dashed line frame, and lines with arrows between the line frames represent the route direction of the whole logic framework.

Detailed Description

The present invention will be described in detail with reference to specific examples. The following examples will assist those skilled in the art in further understanding the invention, but are not intended to limit the invention in any way. It should be noted that it would be obvious to those skilled in the art that various changes and modifications can be made without departing from the spirit of the invention. All falling within the scope of the present invention.

The new method for the automatic control strategy of the sleep mode of the air conditioner provided by the invention is mainly based on objective sleep staging data, and establishes a relation between the sleep condition of a user and the automatic control process of the air conditioner by means of the current machine learning method. From the mechanism, the automatic control logic framework for adjusting the thermal environment of the air-conditioning room can really achieve the purpose of improving the sleep quality of the user. See in particular the following examples:

the present embodiment will explain the flow of the whole logical framework in terms of the user experience process: the logical framework is shown in fig. 1, and the module part represented by the solid line box should be a real object or abstract concept representing certain functions in the application. The instruction or judgment content indicated by the dashed box should be executed by the module in each solid box or issued to other modules for execution, and the whole flow direction of the logical framework needs to refer to the flow line with the arrow. The entire logical framework is integrated with the user experience process as follows:

day one of user experience:

A1. since the AI model is the first machine learning for the user, the user is first required to input the sex, age, and type of bed system to the machine, and set the sensors of the air conditioner near the bed.

B1. The user needs to wear the instrument for monitoring sleep, and the points to be noted here are: the precision of the instrument is close to that of the current scientific sleep staging judging instrument, so that the sleep staging precision of a user can be ensured.

C1. When the user is in bed ready to start falling asleep, an instruction to "start sleeping" is issued to the system, which means that the sleep monitoring apparatus is about to start transmitting the electroencephalogram data of the user to the AI model.

The AI model needs to automatically regulate and control the thermal environment before the user enters sleep according to the individual thermal environment requirements of the user, and simultaneously learns the physiological signals of the user in the period, such as electroencephalogram, blood pressure, heart rate, skin temperature and the like.

E1. When the user formally enters the sleep state, the sleep monitoring instrument needs to pay attention to the following key points: the duration of the transition period of the user's N2-N3 phase, the brain waveform of the user towards the end of the N3 phase, and the brain waveform of the user towards the end of the REM phase.

F1. Since the AI model is not familiar with the characteristics of the user's sleep stage, which is the first night of the user's experience, the entire AI model will only instruct the air conditioner to adjust when the user is in stages N3 and REM.

G1. When the air conditioner sends an instruction, the air conditioner sensor feeds back the thermal environment parameters around the user at that moment to the air conditioner control center, and then the air conditioner starts to adjust the environment parameters such as temperature, humidity and the like, and at the moment, the air conditioner sensor feeds back the real-time environment data around the user to the air conditioner control center in time. When the environmental data around the user reaches the thermal environment state required by the period N3 and the period REM, the time length is recorded and learned by the AI model.

H1. Meanwhile, data of sleep stages of the user also need to be fed back to the AI model in real time during the air conditioning regulation, the electroencephalogram condition of the user is observed when the air conditioning is adjusted to the temperature required by the physiology of the user, if the sleep stage of the user is still in the N3 stage and the REM stage, the environment state adjusted by the air conditioning is considered to be appropriate and can be recorded and learned by the AI model, if the sleep stage of the user is converted into other stages except the N3 stage and the REM stage, the environment state is considered to be inappropriate, the environment parameters need to be adjusted individually, and the parameter values after adjustment can be continuously tested in the next N3 stage or the REM stage, so that whether the sleep stage of the user is changed due to the adjustment of the environment parameters is observed again.

The next and following day of the user experience:

A2. through machine learning on the first night, the user needs to wear the instrument for monitoring sleep.

B2. When a user is ready to fall asleep in bed, the system sends a sleep start command, the air conditioning system starts automatic control, and continuously adjusts the environmental parameters to a sleep environment capable of promoting the user to fall asleep according to the thermal environment requirement of the user.

C2. When the user formally enters the sleep state, the AI model should be aware of the state of the user's sleep stage and transition period due to the machine learning of the first night. In conjunction with the learning result of "the time period during which the environmental parameters around the user reach the N3 phase and the REM phase require the thermal environment" of the first night, the AI model needs to determine the time at which an instruction is given to the air conditioner, that is, a certain time from the start (or end) of the N3 phase (or REM phase) of sleep of the user.

E2. When the sleep monitoring equipment detects the waveform of the time when the air conditioner needs to be adjusted, an instruction is sent to the AI model, and the AI model sends an air conditioner adjusting command to the air conditioner control center.

F2. The air conditioner sets thermal environment parameter values needed by the user in different sleep states after machine learning in the first day as environment parameter set values which can be sensed by the user, automatically regulates and controls the air conditioner, and simultaneously feeds back data around the user monitored by the air conditioner sensor to the AI model in real time. In addition, real-time electroencephalogram data detected by the sleep monitoring equipment are also timely transmitted back to the AI model to be used as a judgment basis for further adjustment of the air conditioner.

The idea of the method is mainly presented by a logic frame diagram: the whole logic framework starts from a sleep monitoring device, which is used for monitoring physiological parameters (taking electroencephalogram as an example) reflecting the sleep stages of the users, and meanwhile, the sleep monitoring device needs to output the sleep stage condition of each user to an AI model. The sleep detection device needs to monitor the following three conditions: a. duration of user N2-N3 phase transition; b. brain waveforms towards the end of user phase N3; c. brain waveforms near the end of the user REM period. For the case a, because there is great difference in electroencephalograms during the sleep of the user, some users may have a transition period of N2-N3 with a long time, but some users have a short transition period of N2-N3, or even have almost no such transition period, so that the sleep monitoring device is to output the conditions of such a transition period (including the electroencephalogram conditions of the transition period, etc.) to the AI model, so that the machine learning can deeply grasp the characteristics of the user when the user transitions to the N3 period, so as to give an instruction to the air conditioner in advance, thereby achieving the purpose of prediction. For cases b and c, since objective index assessment of sleep quality is mainly seen in current scientific studies, sleep latency, REM duration, and duration of slow wave sleep (period N3). The shorter the time when falling asleep or the longer the duration of REM and slow wave sleep (period N3), the better the quality of sleep. Therefore, for improving the index of "time to sleep", the AI model mainly needs to learn that the time from lying in bed to entering the sleep stage N1 is changed due to the change of the core temperature and the skin temperature, which results in the change of the demand of the thermal environment. For improving the two indexes of the REM time length and the slow wave sleep (N3 period) time length, the AI model needs to learn the brain waveforms of the REM and the N3 periods at the end of the time, because when the model receives the brain waveform signals from the sleep monitoring equipment, namely the REM and the N3 periods at the end of the time, the air conditioner can be instructed in advance to regulate and control the thermal environment in advance so as to prolong the time lengths of the two sleep stages and achieve the purpose of improving the sleep quality.

The most important part of the whole logic framework diagram is the AI model, the function of which is mainly responsible for machine learning, and besides learning the sleep stage of each user, the model also needs to record the bedding system type of the user. Because the user and the objective physical environment are separated by the bedding system, the user directly feels a bedding microenvironment formed by a human body, clothes, bedding and the like, and the microenvironment is influenced by the room thermal environment. Therefore, the AI model needs to consider the user bedding system, i.e. the bedding microenvironment as a part of machine learning, in order to improve the accuracy of the user's perception of the thermal environment and the efficiency of air conditioning control.

The control strategy idea of the air conditioner side follows. The air conditioner sensor which can detect the surrounding thermal environment of a user can transmit environmental parameter data such as temperature, humidity and the like to an air conditioner control center in time, and the air conditioner control center has an input from an AI model. Therefore, the other part of the AI model that needs machine learning is the spatial relationship between the room air conditioner and the user, and the time length required for the air conditioner to adjust the actual thermal environment parameters monitored by the sensor to the set environmental parameters.

After the control strategy of the user sleep monitoring device side and the control strategy of the air conditioner side are discussed respectively, the two control strategies of the user and the air conditioner need to be combined, and the combination of the two control strategies needs a plurality of parameters as a bridge. The first is the "time" bridge: when the AI model learns the time length that the air conditioner adjusts to the user-set (required) thermal environment parameter, the time length is equal to the time difference between a certain sleep time and the beginning (or end) of the N3 period (or REM period). Can be considered visually as: in a one-dimensional coordinate system, when the length of a line segment and the end point of the line segment are known, the start point of the line segment can be obtained. This starting point is the moment at which the machine learns to issue an "adjustment" command to the air conditioning control center.

Secondly, a bridge of thermal environment parameters: the above-mentioned user-set (required) thermal environment parameters (including temperature, humidity, etc.) are also set environment parameters that the air conditioner needs to adjust, and a series of parameter values should be derived from the user, but it should be excluded that the thermal environment parameters are derived from parameter values that the user subjectively perceives to be comfortable at first, because the user is in a sleep state, the current thermal and wet perception states cannot be reflected in time, and because of the problems of labor intensity, metabolic rate and core temperature between the sleep state and the waking state, the thermal environment that promotes the sleep quality of the user is not necessarily the thermal environment that the subject feels to be comfortable when being awake, and therefore, the corresponding parameter values of temperature, humidity, etc. in the thermal environment should be associated with objective physiological parameter indexes. The three cases of sleep staging classified in the first part of machine learning described above, the N3 and REM stages of sleep cause the physiological environment parameters required by the person to be different due to different intensity and rate of thermal modulation. Therefore, the environment parameters required by physiology, particularly the temperature required by physiology, are found according to the different sleep periods, the physiological temperature capable of prolonging the N3 period and the REM period is found, and the physiological temperature is further input into an AI model for machine learning and is used as a set sleep heat environment required to be adjusted by an air conditioner.

The automatic air-conditioning control strategy in the period from lying on the bed to before entering the sleep stage N1 also needs two bridges of 'time' and 'thermal environment parameter'. Firstly, the time bridge is relatively simple, namely the moment when the user lies in bed is marked as the sleep mode starting moment, and the air conditioning equipment is switched to the automatically controlled sleep mode at the moment. The method is characterized in that a thermal environment parameter bridge is arranged on the basis of the traditional method, the physiological state of a user is in a process of changing from a waking state to a sleeping state in the stage, so that the requirement on the thermal environment is changed in the whole process due to the change of core temperature and skin temperature, an AI (artificial intelligence) model needs to regulate and control the thermal environment before the user enters the sleep according to the individual requirement of the user, and physiological signals of the user in the stage, such as electroencephalogram, blood pressure, heart rate, skin temperature and the like, are learned.

Those skilled in the art will appreciate that, in addition to implementing the systems, apparatus, and various modules thereof provided by the present invention in purely computer readable program code, the same procedures can be implemented entirely by logically programming method steps such that the systems, apparatus, and various modules thereof are provided in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Therefore, the system, the device and the modules thereof provided by the present invention can be considered as a hardware component, and the modules included in the system, the device and the modules thereof for implementing various programs can also be considered as structures in the hardware component; modules for performing various functions may also be considered to be both software programs for performing the methods and structures within hardware components.

The foregoing description of specific embodiments of the present invention has been presented. It is to be understood that the present invention is not limited to the specific embodiments described above, and that various changes or modifications may be made by one skilled in the art within the scope of the appended claims without departing from the spirit of the invention. The embodiments and features of the embodiments of the present application may be combined with each other arbitrarily without conflict.

Claims

1. An automatic control system of a sleep mode of an air conditioner based on sleep staging is characterized by comprising:

2. The automatic control system for sleep mode of air conditioner based on sleep staging as claimed in claim 1, wherein the sleep monitoring module includes: the duration of the transition segment of the user's N2-N3 phase, the brain waveform near the end of the user's N3 phase, and the brain waveform near the end of the user's REM phase are monitored.

3. The automatic control system for sleep mode of air conditioner based on sleep staging as claimed in claim 1, wherein the learning module includes:

4. The automatic control system for sleep mode of air conditioner based on sleep stages as claimed in claim 1, wherein said learning module learns the time length required for the air conditioner to adjust to the user-set environmental parameter, i.e. the time difference between a certain sleep time and the beginning of period N3 or the end of REM period;

5. The automatic control system for sleep mode of air conditioner based on sleep staging as claimed in claim 1, wherein the regulating module comprises: determining physiological environment parameters required by different sleep stages according to the thermal regulation intensity and rate; and inputting the physiological environment parameters required by different sleep periods into a learning module for machine learning to obtain the set sleep heat environment parameters required to be adjusted by the air conditioner.

6. The automatic control system for sleep mode of air conditioner based on sleep staging as claimed in claim 5, wherein the physiological environment parameters required for different sleep stages include: prolonging physiological temperature of N3 phase and REM phase.

7. The automatic control system for sleep mode of a sleep staging based air conditioner according to claim 5, wherein the thermal environment parameters include temperature and humidity.

8. The automatic control system for sleep mode of air conditioner based on sleep staging as claimed in claim 5, wherein the air conditioner is automatically adjusted according to thermal environment parameters and user sleep stage condition.

9. The automatic control system for sleep mode of air conditioner based on sleep staging as claimed in claim 1, wherein the sleep monitoring module includes: detecting the period from lying on the bed to before entering the sleep stage N1 period, namely the time when the user lies on the bed is used as the time when the air conditioning equipment is switched to the sleep mode automatic control; according to the change of the thermal environment requirement and the user requirement, the physiological signals of the user, including electroencephalogram, blood pressure, heart rate and skin temperature, are learned, and the thermal environment parameters before the user enters sleep are regulated and controlled.

10. An automatic control method of a sleep mode of an air conditioner based on sleep staging is characterized by comprising the following steps: