CN213119453U - Indoor thermal environment control system - Google Patents
Indoor thermal environment control system Download PDFInfo
- Publication number
- CN213119453U CN213119453U CN202022274390.5U CN202022274390U CN213119453U CN 213119453 U CN213119453 U CN 213119453U CN 202022274390 U CN202022274390 U CN 202022274390U CN 213119453 U CN213119453 U CN 213119453U
- Authority
- CN
- China
- Prior art keywords
- indoor
- module
- temperature
- signal
- environment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Landscapes
- Air Conditioning Control Device (AREA)
Abstract
The utility model relates to an indoor thermal environment control system relates to building environmental control technical field. The system comprises: the voice recognition device and the detection device are connected with the control device; obtaining and identifying indoor sound through a voice identification device to obtain a comfortable sensation information signal; detecting the indoor environment temperature through a detection device and generating a temperature signal; the control device obtains a target temperature domain for the user according to the comfort information signal and the temperature signal by using an online learning algorithm, and generates an environment control signal; the control device is also used for adjusting the set temperature of the indoor environment adjusting equipment according to the environment control signal. The utility model discloses can provide comfortable indoor thermal environment for the user, improve the hot comfortable accuracy of portrayal personnel through speech recognition device, detection device and controlling means.
Description
Technical Field
The utility model relates to a building environmental control technical field especially relates to an indoor thermal environment control system.
Background
The control strategies of the existing air conditioning systems can be roughly divided into three categories: control based on set values and control based on the person PMV and control based on human thermal sensation.
1) Control based on set value
In practical applications, the most common control logic is set-point based control, i.e. the control of environmental parameters (e.g. temperature, humidity) within the set-point and its reasonable fluctuations by the operation of the air conditioning equipment. This control strategy is derived from the process air conditioning used in industrial production, and environmental parameters are often controlled in a small range in order to ensure production needs.
However, this conventional control strategy has several drawbacks that are difficult to overcome: first, when the building manager sets the indoor temperature uniformly, the control strategy can only meet the thermal comfort of most people (for example, the indoor temperature is set according to the regulation of national standard "air conditioning and ventilation system operation management standard" GB 50365-2019), the difference of personal thermal comfort is not considered, and if a user with special thermal comfort occurs, the control easily causes the thermal sensation complaint of the person. Second, long-term manual adjustment by a user or administrator may create tired emotions, and frequent switching and adjustment may also adversely affect the life of the device. Thirdly, in actual operation, people usually cannot accurately know the comfortable temperature, and the set temperature value often deviates from the comfortable interval, so that the result of neither comfort nor energy saving is caused.
2) Control based on Predictive Mean Volume (PMV)
The PMV based control of the personnel quantifies the thermal comfort of the personnel, namely, a numerical value representing the thermal comfort of the personnel is calculated through related parameters representing the environment and the characteristics of the personnel, so as to guide the control of the air conditioning system. The earliest quantitative research on thermal comfort introduced Fanger et al, who introduced a predictive mean opinion voting (PMV) model and a Predictive Percentage Dissatisfaction (PPD) model, taking into account wind speed, human activity, clothing, temperature, humidity, and mean radiant temperature, among other relevant factors, to quantify the thermal comfort of building occupants. The PMV model and PPD model propose a possibility to control the air conditioning system based on thermal comfort, but this strategy does not take into account the personalization and dynamic changes of the thermal comfort of the user (such as the change of thermal preference of the person in different seasons, the adaptability of the person to the environment, etc.). de Dear and brair et al propose an adaptive model to try to solve this problem, and study and establish the relationship between indoor comfort temperature and outdoor ambient temperature. Although the adaptive model is considered more comprehensively, the general rule of the thermal comfort of the user is summarized from a large amount of sample data on the basis of a statistical model similar to the PMV model and the PDD model in nature, and the subjective nature of the concept of the personal characteristics of the resident and the thermal comfort (such as individuals with special preference for cold and hot, personal adaptability to the environment, changes caused by age increase and the like) is not considered, so that the real-time thermal comfort of the person cannot be accurately depicted.
3) Thermal sensation based control
A control model based on human body thermal sensation is used for acquiring the thermal sensation of a person in a questionnaire mode, the mode can accurately acquire the real-time thermal sensation of the person, the subjectivity of the index of 'comfort' is better reflected, and compared with a control strategy based on a set value, the control model can save 15% of energy consumption. However, this control strategy also has drawbacks: on one hand, the method does not consider the influence caused by the change of the characteristics of the people in a short time (such as the change of the activity amount and the sudden change of the clothing amount), and does not consider the factor when filling out the questionnaire; on the other hand, a person can only reflect his own thermal complaints by filling out a questionnaire, which may give the person a tired mood, affect the stability of the results of the thermal questionnaire, and also give rise to abnormal values.
Therefore, the control strategy of the existing air conditioning system cannot accurately depict the thermal comfort of the personnel.
SUMMERY OF THE UTILITY MODEL
The utility model aims at providing an indoor hot environmental control system improves the hot comfortable accuracy of portrayal personnel.
In order to achieve the above object, the utility model provides a following scheme:
an indoor thermal environment control system comprising: the system comprises a voice recognition device, a detection device, a control device and indoor environment adjusting equipment;
the voice recognition device, the detection device and the indoor environment adjusting equipment are all connected with the control device;
the voice recognition device is used for acquiring and recording indoor sound, recognizing the indoor sound to obtain comfortable sensation information of a user to an indoor environment, and generating a comfortable sensation information signal according to the comfortable sensation information;
the detection device is used for detecting the temperature of the indoor environment and generating a temperature signal according to the temperature;
the control device is used for receiving the comfortable sensation information signal and the temperature signal, obtaining a target temperature domain of the indoor environment for the user by utilizing an online learning algorithm according to the comfortable sensation information signal and the temperature signal, and generating an environment control signal according to the target temperature domain;
the control device is further used for adjusting the set temperature of the indoor environment adjusting equipment according to the environment control signal, so that the environment temperature of the indoor environment is in the target temperature range.
Optionally, the speech recognition apparatus specifically includes: the system comprises a microphone, a monitoring recording module, an ASRT voice recognition system identification module and a keyword extraction module;
the microphone is connected with the monitoring recording module, the monitoring recording module is connected with the ASRT voice recognition system recognition module, and the keyword extraction module is connected with the control device;
the microphone is used for acquiring indoor sound;
the monitoring and recording module is used for recording the indoor sound;
the ASRT voice recognition system recognition module is used for carrying out voice recognition on the recorded indoor sound to obtain a voice text;
the keyword extraction module is used for extracting keywords from the voice text to obtain keywords expressing thermal sensation; the keyword expressing the thermal sensation is the comfort sensation information.
Optionally, the ASRT speech recognition system recognition module specifically includes: the device comprises a feature extraction sub-module, an acoustic model identification sub-module, a decoding sub-module and a statistical language model text conversion sub-module;
the feature extraction submodule is respectively connected with the monitoring recording module and the acoustic model identification submodule, the acoustic model identification submodule is connected with the decoding submodule, the decoding submodule is connected with the statistical language model text conversion submodule, and the statistical language model text conversion submodule is connected with the keyword extraction module;
the feature extraction submodule is used for extracting features of the recorded indoor sound to obtain a spectrogram;
the acoustic model identification submodule is used for identifying the spectrogram by using a convolutional neural network acoustic model of an ASRT (asynchronous serial transcription) voice identification system to obtain a noise pinyin sequence;
the decoding submodule is used for carrying out connectivity time sequence classification decoding on the noise pinyin sequence to obtain an actual pinyin sequence;
and the statistical language model text conversion submodule is used for obtaining a voice text by using a statistical language model according to the actual pinyin sequence.
Optionally, the detection device is a temperature sensor.
Optionally, the control device specifically includes: the online learning system comprises a data transmission sub-device and an online learning module;
the data transmission sub-device and the online learning module are both connected with a database;
the data transmission sub-device is also respectively connected with the voice recognition device and the detection device, and is used for receiving the comfortable sensation information signal and the temperature signal and transmitting the comfortable sensation information signal and the temperature signal to the database;
the online learning module is used for reading the comfort information signal and the temperature signal in the database, obtaining a target temperature domain of the indoor environment for the user by utilizing an online learning algorithm according to the comfort information signal and the temperature signal, and generating an environment control signal according to the target temperature domain.
Optionally, the data transmission sub-apparatus specifically includes: the device comprises a communication module and an analog input channel;
the communication module and the analog input channel are connected with a database;
the communication module is also connected with the voice recognition device and is used for receiving the comfortable sensation information signal and transmitting the comfortable sensation information signal to the database;
the analog quantity input channel is also connected with the detection device and used for receiving the temperature signal and transmitting the temperature signal to the database.
According to the utility model provides a concrete embodiment, the utility model discloses a following technological effect:
the utility model provides an indoor thermal environment control system. The system comprises: the system comprises a voice recognition device, a detection device, a control device and indoor environment adjusting equipment; the voice recognition device, the detection device and the indoor environment adjusting equipment are all connected with the control device; the voice recognition device is used for acquiring and recording indoor sound, recognizing the indoor sound to obtain comfortable sensation information of a user to the indoor environment, and generating a comfortable sensation information signal according to the comfortable sensation information; the detection device is used for detecting the temperature of the indoor environment and generating a temperature signal according to the temperature; the control device is used for receiving the comfortable sensation information signal and the temperature signal, obtaining a target temperature domain of the indoor environment aiming at the user by utilizing an online learning algorithm according to the comfortable sensation information signal and the temperature signal, and generating an environment control signal according to the target temperature domain; the control device is also used for adjusting the set temperature of the indoor environment adjusting equipment according to the environment control signal so as to enable the environment temperature of the indoor environment to be in the target temperature range. The utility model discloses a speech recognition device discerns the subjective impression of the comfort of user's language expression, obtains comfort information, and controlling means adjusts the settlement temperature of indoor environment regulating equipment according to the temperature of comfort information and indoor environment, provides comfortable, satisfied indoor hot environment for the user, avoids indoor supercooling or overheated, has improved the hot comfortable accuracy of portrayal personnel.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without inventive labor.
Fig. 1 is a structural diagram of an indoor thermal environment control system according to an embodiment of the present invention.
Description of the symbols: 1. a voice recognition device; 2. a detection device; 3. a control device; 4. an air conditioning apparatus.
Detailed Description
The technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments in the present invention, all other embodiments obtained by a person skilled in the art without creative work belong to the protection scope of the present invention.
The utility model aims at providing an indoor hot environmental control system improves the hot comfortable accuracy of portrayal personnel.
In order to make the above objects, features and advantages of the present invention more comprehensible, the present invention is described in detail with reference to the accompanying drawings and the detailed description.
This embodiment provides an indoor thermal environment control system, and fig. 1 is the utility model discloses the embodiment provides an indoor thermal environment control system's structure chart, see fig. 1, indoor thermal environment control system includes: speech recognition device 1, detection device 2, control device 3 and indoor environment adjusting equipment.
The voice recognition device 1, the detection device 2 and the indoor environment adjusting device are all connected with the control device 3.
The voice recognition device is used for acquiring and recording indoor sound (namely natural language of a user), recognizing the indoor sound to obtain comfort information of the user to the indoor environment, and generating a comfort information signal according to the comfort information.
The speech recognition device specifically includes: the system comprises a microphone, a monitoring recording module, an ASRT voice recognition system recognition module and a keyword extraction module.
The utility model discloses a Speech Recognition device uses ASRT (auto Speech Recognition tool) Speech Recognition system as the basis, and ASRT Speech Recognition system's acoustic model has mainly adopted the algorithm of degree of depth convolutional network (CNN) and the decoding technique of connectivity time sequence classification (CTC), can realize converting the audio file of wav format into this function of chinese text. On ASRT speech recognition system's basis, the utility model discloses added "monitoring recording" and "keyword extraction" two modules to extended former ASRT speech recognition system's corpus, added the commonly used sentence that uses the air conditioner in-process promptly in former ASRT speech recognition system's corpus, be used for improving natural language identification's accuracy, so constitute complete speech recognition device, can realize automatic recording user's speech expression and output and the hot sense relevant judged result, transmit the controlling means to next floor.
The microphone is connected with the monitoring recording module, the monitoring recording module is connected with the ASRT voice recognition system recognition module, and the keyword extraction module is connected with the control device.
The microphone is used for acquiring indoor sound.
The monitoring and recording module is used for recording indoor sound. And the monitoring recording module records the indoor sound and stores the indoor sound as a wav file.
The ASRT voice recognition system recognition module is used for carrying out voice recognition on the recorded indoor sound to obtain a voice text.
The ASRT speech recognition system recognition module specifically includes: the device comprises a feature extraction sub-module, an acoustic model identification sub-module, a decoding sub-module and a statistical language model text conversion sub-module.
The feature extraction submodule is respectively connected with the monitoring recording module and the acoustic model identification submodule, the acoustic model identification submodule is connected with the decoding submodule, the decoding submodule is connected with the statistical language model text conversion submodule, and the statistical language model text conversion submodule is connected with the keyword extraction module.
And the feature extraction submodule is used for extracting features of the recorded indoor sound to obtain a spectrogram. The characteristic extraction submodule is used for extracting the characteristics of the wav file to obtain a spectrogram; specifically, the method takes a wav voice signal of a wav file as input, realizes frequency spectrum conversion through operations such as framing, windowing and the like, and finally outputs a two-dimensional frequency spectrum image, namely a spectrogram.
And the acoustic model identification submodule is used for identifying the spectrogram by using a convolutional neural network acoustic model of the ASRT voice identification system to obtain a noise pinyin sequence. The acoustic model identification submodule identifies the spectrogram by using a convolutional neural network acoustic model trained in advance to obtain a noise pinyin sequence, and particularly performs convolution and pooling operations on the spectrogram by using the convolutional neural network acoustic model to obtain characteristic information of voice, namely a rough pinyin sequence (noise pinyin sequence). And the convolutional neural network acoustic model is obtained by training the acoustic model by adopting a python program of an open source on a GitHub network according to the pre-acquired training data.
The decoding submodule is used for decoding a connectivity time sequence Classification (CTC) of the noise pinyin sequence to obtain an actual pinyin sequence.
And the statistical language model text conversion submodule is used for obtaining the voice text by using the statistical language model according to the actual pinyin sequence.
The keyword extraction module is used for extracting keywords from the voice text to obtain keywords expressing thermal sensation; the keyword expressing the thermal sensation is comfort information. The keyword extraction module searches and compares the recognized characters in the corpus to find out whether keywords related to heat sensation exist, such as 'cold' or 'hot' and the like. The comfort information includes: warm, hot, cool, cold or cold.
The monitoring and recording module monitors and records indoor sound (natural language) through a microphone and stores the indoor sound as a wav file; the characteristic extraction submodule is used for extracting the characteristics of the wav file to obtain a spectrogram; the acoustic model identification submodule identifies the spectrogram by utilizing a previously trained convolutional neural network acoustic model to obtain a noise pinyin sequence; performing CTC decoding on the noise pinyin sequence through a decoding submodule to obtain an actual pinyin sequence; then the statistical language model text conversion submodule obtains a text recognized from the natural language through the statistical language model; and finally, the keyword extraction module extracts keywords from the text to find out the keywords expressing the heat feeling, such as cold or hot.
The detection device is used for detecting the temperature of the indoor environment and generating a temperature signal according to the temperature. The detection device is a temperature sensor.
The control device is used for receiving the comfortable sensation information signal and the temperature signal, obtaining a target temperature domain of the indoor environment aiming at the user by utilizing an online learning algorithm according to the comfortable sensation information signal and the temperature signal, and generating an environment control signal according to the target temperature domain.
The control device specifically includes: the online learning device comprises a data transmission sub-device and an online learning module.
The data transmission sub-device and the online learning module are both connected with the database.
The data transmission sub-device is also respectively connected with the voice recognition device and the detection device, and is used for receiving the comfort information signal and the temperature signal and transmitting the comfort information signal and the temperature signal to the database.
The data transmission sub-device specifically comprises: the device comprises a communication module and an analog input channel.
The communication module and the analog input channel are connected with the database.
The communication module is also connected with the voice recognition device and is used for receiving the comfortable sensation information signal and transmitting the comfortable sensation information signal to the database. The communication module is specifically used for sending the comfort information signal obtained by the voice recognition device to the database for storage. The communication module is connected with an indoor environmental parameter sensor (namely, a detection device). When the temperature sensor transmits the temperature information in a communication mode, the temperature information detected by the temperature sensor is sent to a database through the communication module to be stored.
The analog quantity input channel is also connected with the detection device and used for receiving the temperature signal and transmitting the temperature signal to the database. The analog input channel is also used for storing the received temperature signal in a database. When the temperature sensor transmits the temperature information in the form of analog quantity, the temperature information detected by the temperature sensor is sent to a database for storage through an analog quantity input channel. The online learning module is not directly related to the analog input channel, and the online learning module and the analog input channel perform data interaction through the database.
The online learning module is used for reading the comfortable sensation information signal and the temperature signal in the database, obtaining a target temperature domain of the indoor environment for the user by utilizing an online learning algorithm according to the comfortable sensation information signal and the temperature signal, and generating an environment control signal according to the target temperature domain. The online learning module reads the comfort information signal and the temperature signal in the database, learns the personalized comfortable temperature domain of the user through an online learning algorithm, specifically resolves the temperature corresponding to the temperature signal according to the temperature signal, sets a target temperature domain according to the comfort information and the temperature, and generates an environment control signal corresponding to the target temperature domain.
The utility model discloses an online learning module can combine the ambient temperature that temperature sensor detected to synthesize the optimal indoor temperature setting value of decision according to the comfort information of speech recognition device output. The online learning method of the online learning module comprises the following steps: obtaining an initial temperature set value Ts, wherein the initial temperature set value Ts of an indoor thermal environment control system is a set temperature of 26 ℃ recommended by national standard air conditioning and ventilation system operation management standard GB 50365-2019; receiving comfort information identified by a voice recognition device; judging whether a user generates a heat sensation expression or not, namely whether the on-line learning module receives comfort information identified by the voice identification device or not; after receiving the comfort information, the online learning module immediately monitors the environmental temperature Te of the current indoor environment, updates a user comfort domain (namely a target temperature domain) by combining the comfort information of the user, calculates a new temperature set value Ts according to the updated user comfort domain, generates an environmental control signal according to the new temperature set value Ts, returns 'judges whether the comfort information identified by the voice recognition device is received', and adjusts the set temperature of the indoor environment adjusting equipment according to the environmental control signal by the control device. Tup is the upper temperature limit of the target temperature range and Tdown is the lower temperature limit. The online learning principle of the target temperature domain adopts a fuzzy algorithm, the target temperature domain is calculated by adopting different adjusting ranges according to the thermal sensation feedback degree (namely, comfort information) of a user, and the specific fuzzy algorithm is shown in table 1.
TABLE 1 fuzzy Algorithm
| Target temperature update | |
| Warm (a bit hot) | Tup＝Te；Tdown＝Tup-2δ；Ts＝(Tup+Tdown)/2 |
| Heat generation | Tup＝Te-1℃；Tdown＝Tup-2δ；Ts＝(Tup+Tdown)/2 |
| Inflammation heat (very hot) | Tup＝Te-2℃；Tdown＝Tup-2δ；Ts＝(Tup+Tdown)/2 |
| Cool (somewhat cold) | Tdown＝Te；Tup＝Tdown+2δ；Ts＝(Tup+Tdown)/2 |
| Cold | Tdown＝Te+1℃；Tup＝Tdown+2δ；Ts＝(Tup+Tdown)/2 |
| Cold (very cold) | Tdown＝Te+2℃；Tup＝Tdown+2δ；Ts＝(Tup+Tdown)/2 |
Delta represents the preset indoor temperature control return difference which can be set independently and is usually 0.5 ℃ or 1 ℃. The online learning module identifies six user thermal sensation states (comfort information) according to natural language: and the current target temperature domain is corrected according to each heat sensation state to obtain the personalized target temperature domain of each user.
The indoor environment conditioning apparatus of the present embodiment is an air conditioning apparatus 4.
The control device is also used for adjusting the set temperature of the indoor environment adjusting equipment according to the environment control signal so as to enable the environment temperature of the indoor environment to be in the target temperature range. The control device is also used for executing the adjusting action according to the environment control signal so as to adjust the environment temperature to the corresponding target temperature domain.
The utility model provides an indoor thermal environment control system based on natural language processing discernment heat sensation can adjust the settlement temperature of indoor environment conditioning equipment based on the subjective feeling of comfort (comfort information) that the user expressed through the language, provides comfortable, satisfied indoor thermal environment for the user, avoids indoor supercooling or overheated; and the energy consumption minimization of the building indoor environment adjusting equipment is realized by integrating and optimizing the control of each indoor environment parameter.
With the development of scientific technology, methods for acquiring the heat sensation of people are more abundant, and some technologies enable an air conditioning system to acquire the heat sensation of the people without interfering with the daily activities of users. If the infrared thermal imager is used for detecting the skin temperature of the human body, the cold and hot feeling of the human body is predicted; the cold and heat feeling is predicted through the temperature measuring bracelet, so that the cold and heat complaining times of a user are effectively reduced, and the energy consumption of an air conditioning system is reduced; the skin temperature of the user is identified and acquired by using an artificial intelligence computer vision technology, and the skin temperature comfort interval of the user is determined by using an artificial intelligence machine learning method. However, these methods generally suffer from problems such as excessive cost and limited range of user's activities. How to accurately, cheaply and conveniently acquire the thermal comfort of people is still the focus of attention of the students. The development of artificial intelligence technology provides a new idea for solving the problem, namely an indoor thermal environment control system for recognizing thermal sensation based on natural language processing.
Natural language processing technology, an important branch of artificial intelligence, refers to the processing of natural language information by computers, which is colloquially understood to be the "manipulation" and "processing" of text. Speech recognition is an important sub-field of natural language processing. In the 21 st century, the great improvement of the recognition accuracy of the speech recognition technology depends on the introduction of the deep learning technology, and therefore, the speech recognition technology is more widely applied. In 2009, Hinton used Deep Neural Networks (DNN) technology for the first time in acoustic modeling of speech, with the best results at that time being achieved in testing of the timmit data set. In 2011, Shu dong, Denli and other people used DNN technology on a large vocabulary continuous voice recognition task, and the accuracy rate of voice recognition is greatly improved. Speech recognition has thus far entered the DNN-HMM (hidden markov model) era.
The development of the deep learning technology greatly improves the recognition precision of natural language processing. Since 2006, deep learning techniques have begun to receive intense academic attention and have been developed. To date, deep learning techniques have provided efficient solutions for application scenarios such as the internet. The deep learning simulates human brain and establishes a model structure, key features are collected from input data, and then a mapping relation between bottom signal features and high-level semantics is established, so that the deep learning model plays an irreplaceable role in the fields of voice, image recognition and the like. Zhang Jianhua combines deep learning and speech recognition technology, and explores the application of deep neural networks in acoustic feature extraction, initial and final attribute extraction and acoustic modeling. The improvement of the accuracy rate of the deep learning model depends on the increase of the data volume of the training data set on one hand and the optimization of the model algorithm on the other hand. With the progress of research, various deep Neural network models are proposed, the most representative of which is the Current Neural Network (CNN).
In the building field, the artificial intelligence technology can play a great role in the fields of building environment control, building operation optimization and the like, on one hand, the comfort and satisfaction of users can be improved, on the other hand, the realization of energy conservation and emission reduction of buildings is facilitated, and the artificial intelligence technology has great application potential. Relevant research of the kyoto university indicates that the cooperative cooperation of an Artificial Intelligence (AI) technology and a Big Data (Big Data, BD) technology is a trend in the field of building energy conservation, and a deep learning algorithm is proposed to pay more attention to prediction accuracy than a traditional technology in predicting building energy consumption rather than prediction model accuracy (the traditional prediction technology always tries to find a more reasonable physical model to simulate the real state of a building).
In conclusion, the air conditioning system control based on the thermal sensation has the dual advantages of energy conservation and comfort, and the development of the artificial intelligence technology can enable the thermal sensation to be recognized to achieve a better effect. Therefore the utility model provides a thermal environment control system based on hot sensation of natural language processing discernment has huge practical value, when guaranteeing that the personnel are comfortable, has the huge potentiality that reduces air conditioning system energy consumption again.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The principle and the implementation of the present invention are explained herein by using specific examples, and the above description of the embodiments is only used to help understand the method and the core idea of the present invention; meanwhile, for the general technical personnel in the field, according to the idea of the present invention, there are changes in the concrete implementation and the application scope. In summary, the content of the present specification should not be construed as a limitation of the present invention.
Claims (6)
1. An indoor thermal environment control system, comprising: the system comprises a voice recognition device, a detection device, a control device and indoor environment adjusting equipment;
the voice recognition device, the detection device and the indoor environment adjusting equipment are all connected with the control device;
the voice recognition device is used for acquiring and recording indoor sound, recognizing the indoor sound to obtain comfortable sensation information of a user to an indoor environment, and generating a comfortable sensation information signal according to the comfortable sensation information;
the detection device is used for detecting the temperature of the indoor environment and generating a temperature signal according to the temperature;
the control device is used for receiving the comfortable sensation information signal and the temperature signal, obtaining a target temperature domain of the indoor environment for the user by utilizing an online learning algorithm according to the comfortable sensation information signal and the temperature signal, and generating an environment control signal according to the target temperature domain;
the control device is further used for adjusting the set temperature of the indoor environment adjusting equipment according to the environment control signal, so that the environment temperature of the indoor environment is in the target temperature range.
2. An indoor thermal environment control system according to claim 1, wherein the speech recognition device specifically comprises: the system comprises a microphone, a monitoring recording module, an ASRT voice recognition system identification module and a keyword extraction module;
the microphone is connected with the monitoring recording module, the monitoring recording module is connected with the ASRT voice recognition system recognition module, and the keyword extraction module is connected with the control device;
the microphone is used for acquiring indoor sound;
the monitoring and recording module is used for recording the indoor sound;
the ASRT voice recognition system recognition module is used for carrying out voice recognition on the recorded indoor sound to obtain a voice text;
the keyword extraction module is used for extracting keywords from the voice text to obtain keywords expressing thermal sensation; the keyword expressing the thermal sensation is the comfort sensation information.
3. The indoor thermal environment control system of claim 2, wherein the ASRT speech recognition system recognition module specifically comprises: the device comprises a feature extraction sub-module, an acoustic model identification sub-module, a decoding sub-module and a statistical language model text conversion sub-module;
the feature extraction submodule is respectively connected with the monitoring recording module and the acoustic model identification submodule, the acoustic model identification submodule is connected with the decoding submodule, the decoding submodule is connected with the statistical language model text conversion submodule, and the statistical language model text conversion submodule is connected with the keyword extraction module;
the feature extraction submodule is used for extracting features of the recorded indoor sound to obtain a spectrogram;
the acoustic model identification submodule is used for identifying the spectrogram by using a convolutional neural network acoustic model of an ASRT (asynchronous serial transcription) voice identification system to obtain a noise pinyin sequence;
the decoding submodule is used for carrying out connectivity time sequence classification decoding on the noise pinyin sequence to obtain an actual pinyin sequence;
and the statistical language model text conversion submodule is used for obtaining a voice text by using a statistical language model according to the actual pinyin sequence.
4. An indoor thermal environment control system according to claim 1, wherein the detection means is a temperature sensor.
5. An indoor thermal environment control system according to claim 1, wherein the control device specifically comprises: the online learning system comprises a data transmission sub-device and an online learning module;
the data transmission sub-device and the online learning module are both connected with a database;
the data transmission sub-device is also respectively connected with the voice recognition device and the detection device, and is used for receiving the comfortable sensation information signal and the temperature signal and transmitting the comfortable sensation information signal and the temperature signal to the database;
the online learning module is used for reading the comfort information signal and the temperature signal in the database, obtaining a target temperature domain of the indoor environment for the user by utilizing an online learning algorithm according to the comfort information signal and the temperature signal, and generating an environment control signal according to the target temperature domain.
6. An indoor thermal environment control system according to claim 5, wherein the data transmission sub-apparatus specifically comprises: the device comprises a communication module and an analog input channel;
the communication module and the analog input channel are connected with a database;
the communication module is also connected with the voice recognition device and is used for receiving the comfortable sensation information signal and transmitting the comfortable sensation information signal to the database;
the analog quantity input channel is also connected with the detection device and used for receiving the temperature signal and transmitting the temperature signal to the database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202022274390.5U CN213119453U (en) | 2020-10-14 | 2020-10-14 | Indoor thermal environment control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202022274390.5U CN213119453U (en) | 2020-10-14 | 2020-10-14 | Indoor thermal environment control system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN213119453U true CN213119453U (en) | 2021-05-04 |
Family
ID=75666406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202022274390.5U Active CN213119453U (en) | 2020-10-14 | 2020-10-14 | Indoor thermal environment control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN213119453U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112113317A (en) * | 2020-10-14 | 2020-12-22 | 清华大学 | Indoor thermal environment control system and method |
| CN117077854A (en) * | 2023-08-15 | 2023-11-17 | 广州视声智能科技有限公司 | Building energy consumption monitoring method and system based on sensor network |
-
2020
- 2020-10-14 CN CN202022274390.5U patent/CN213119453U/en active Active
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112113317A (en) * | 2020-10-14 | 2020-12-22 | 清华大学 | Indoor thermal environment control system and method |
| CN112113317B (en) * | 2020-10-14 | 2024-05-24 | 清华大学 | Indoor thermal environment control system and method |
| CN117077854A (en) * | 2023-08-15 | 2023-11-17 | 广州视声智能科技有限公司 | Building energy consumption monitoring method and system based on sensor network |
| CN117077854B (en) * | 2023-08-15 | 2024-04-16 | 广州视声智能科技有限公司 | Building energy consumption monitoring method and system based on sensor network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112113317B (en) | Indoor thermal environment control system and method | |
| CN111459290B (en) | Interactive intention determining method and device, computer equipment and storage medium | |
| JP6730843B2 (en) | Communication support system | |
| CN112051743A (en) | Device control method, conflict processing method, corresponding devices and electronic device | |
| CN213119453U (en) | Indoor thermal environment control system | |
| US20170270922A1 (en) | Smart home control method based on emotion recognition and the system thereof | |
| CN108413588B (en) | Personalized air conditioner control system and method based on thermal imaging and BP neural network | |
| KR102309682B1 (en) | Method and platform for providing ai entities being evolved through reinforcement machine learning | |
| CN110726222B (en) | Air conditioner control method and device, storage medium and processor | |
| CN106815321A (en) | Chat method and device based on intelligent chat robots | |
| CN111447124B (en) | Intelligent household control method and intelligent control equipment based on biological feature recognition | |
| CN109429415A (en) | Illumination control method, apparatus and system | |
| US11222638B2 (en) | Communication device and control program for communication device | |
| CN110286600B (en) | Scene setting method and device of intelligent household operating system | |
| KR102305177B1 (en) | Platform for gathering information for ai entities and method by using the same | |
| CN117637134A (en) | Health management system, method, electronic device, and computer-readable storage medium | |
| WO2018023523A1 (en) | Motion and emotion recognizing home control system | |
| Liu et al. | Research on general model of intelligence level for smart home | |
| Karpouzis et al. | Induction, recording and recognition of natural emotions from facial expressions and speech prosody | |
| CN116105307A (en) | Air conditioner control method, device, electronic equipment and storage medium | |
| CN111245690A (en) | Shortcut control system based on voice control | |
| TWM580715U (en) | Smart voice output device for semantic creation | |
| Chong CHENa et al. | Research on Optimization of Interior Space Design Based on Machine Learning Algorithm | |
| CN117889531A (en) | Air conditioning method, device, equipment and storage medium | |
| CN117289803A (en) | Household intelligent kitchen robot and control method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| GR01 | Patent grant | ||
| GR01 | Patent grant |