CN116734389B - Intelligent air conditioner control system based on wearable equipment and convolutional neural network - Google Patents
Intelligent air conditioner control system based on wearable equipment and convolutional neural network Download PDFInfo
- Publication number
- CN116734389B CN116734389B CN202311011802.8A CN202311011802A CN116734389B CN 116734389 B CN116734389 B CN 116734389B CN 202311011802 A CN202311011802 A CN 202311011802A CN 116734389 B CN116734389 B CN 116734389B
- Authority
- CN
- China
- Prior art keywords
- thermal comfort
- refrigeration
- data
- target
- subarea
- 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
- 238000013527 convolutional neural network Methods 0.000 title claims abstract description 22
- 238000005057 refrigeration Methods 0.000 claims abstract description 153
- 238000007405 data analysis Methods 0.000 claims abstract description 42
- 239000000178 monomer Substances 0.000 claims abstract description 13
- 238000012216 screening Methods 0.000 claims abstract description 4
- 238000012549 training Methods 0.000 claims description 20
- 238000001514 detection method Methods 0.000 claims description 17
- 230000009467 reduction Effects 0.000 claims description 14
- 230000002776 aggregation Effects 0.000 claims description 12
- 238000004220 aggregation Methods 0.000 claims description 12
- 238000000034 method Methods 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 210000004556 brain Anatomy 0.000 claims description 6
- 238000012795 verification Methods 0.000 claims description 6
- 238000012806 monitoring device Methods 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 claims description 3
- 238000004378 air conditioning Methods 0.000 claims description 2
- 238000010606 normalization Methods 0.000 description 9
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000011176 pooling Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000003238 somatosensory effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/30—Control or safety arrangements for purposes related to the operation of the system, e.g. for safety or monitoring
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/64—Electronic processing using pre-stored data
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/70—Control systems characterised by their outputs; Constructional details thereof
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/88—Electrical aspects, e.g. circuits
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2120/00—Control inputs relating to users or occupants
- F24F2120/20—Feedback from users
Landscapes
- Engineering & Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention relates to the field of air conditioner control, in particular to an intelligent air conditioner control system based on wearable equipment and a convolutional neural network, which comprises the following components: the data acquisition unit is used for acquiring human physiological data and refrigeration area environment information of the target acquisition body; the data analysis unit is used for determining the refrigeration mode of the refrigeration subarea according to the number of target acquisition bodies in the refrigeration subarea; a thermal comfort model unit to determine effective physiological signal data and generate a thermal comfort index model from the effective physiological signal; the data screening unit is used for determining whether the target acquisition body is a thermal comfort reference target body according to the leaving frequency of the target acquisition body in the refrigeration subarea, and determining a temperature regulation mode of the refrigeration subarea according to the effective physiological signal data of the thermal comfort reference target body; the refrigeration control unit calculates a target thermal comfort value average value according to each effective monomer thermal comfort value and correspondingly adjusts the temperature of the refrigeration subarea; the invention reduces the secondary adjustment rate of the air conditioner temperature.
Description
Technical Field
The invention relates to the field of air conditioner control, in particular to an intelligent air conditioner control system based on wearable equipment and a convolutional neural network.
Background
Along with the increasing popularity of intelligent home, in the age of Internet of things and everything, equipment in daily life is improved gradually through big data and artificial intelligence, and the performance of products is improved. For indoor environments, air conditioners are an indispensable part for improving indoor environments, but for the current air conditioners, the temperature of the air conditioners can be adjusted by the user only according to the comfort level felt by the user, the temperature of the air conditioners can be adjusted only according to experience or body feeling, and the temperature of the air conditioners cannot be adjusted once to enable the environment temperature to meet the comfort requirement of the human body. Especially when people are on an air conditioner and are at rest, the phenomenon of hot or cold awakening is commonly caused, and the discomfort is caused when the people are serious. Therefore, how to automatically adjust the air conditioner to reduce the adjustment times of the air conditioner and improve the adjustment accuracy is a problem to be solved.
Chinese patent publication No. CN106054658B discloses a coordinated control method, a coordinated control system and an intelligent wearable device, comprising: the intelligent wearable device is provided with a master mode and a slave mode, and controls the air conditioner based on the master mode and the slave mode; the linkage control method of the intelligent wearable device comprises the following steps: the intelligent wearable device alternately switches between the master mode and the slave mode to control the air conditioner. It can be seen that the above technical solution has the following problems: the problem that the number of times of air conditioner adjustment is large is caused by the fact that the refrigeration end cannot be adaptively adjusted according to the actual working environment information of the refrigeration area and the physiological information of the human body.
Disclosure of Invention
Therefore, the invention provides an intelligent air conditioner control system based on a wearable device and a convolutional neural network, which is used for solving the problem that in the prior art, the secondary adjustment rate of the environment temperature after the air conditioner refrigeration adjustment is high due to the fact that the refrigeration end cannot be intelligently adjusted according to the actual working environment information of a refrigeration area and the physiological information of a human body.
To achieve the above object, the present invention provides an intelligent air conditioner control system based on a wearable device and a convolutional neural network, comprising:
the data acquisition unit comprises a wearable device for acquiring human physiological data of a target acquisition body and a monitoring device for acquiring environment information of a refrigeration area;
the data analysis unit is connected with the data acquisition unit and is used for determining the refrigeration mode of the refrigeration subarea according to the number of target acquisition bodies in the refrigeration subarea and determining the initial temperature of the refrigeration subarea according to the Euclidean distance between the target acquisition bodies or the number of target acquisition bodies;
the thermal comfort model unit is connected with the data analysis unit and comprises a data noise reduction sub-module, a Z-score module and a model generation module, wherein the data noise reduction sub-module is used for carrying out noise reduction treatment on the physiological signal data which is required to be reduced by a user so as to determine effective physiological signal data, the Z-score module is used for processing the effective physiological signal data into a training set, and the model generation module is used for generating a thermal comfort index model according to the data of the training set and the corresponding environment temperature;
the data screening unit is connected with the data acquisition unit and the thermal comfort model unit, and is used for determining whether a target acquisition body is a thermal comfort reference target body according to the departure frequency of the target acquisition body in the refrigeration subarea, and determining the temperature regulation mode of the refrigeration subarea according to the effective physiological signal data of the thermal comfort reference target body, wherein the thermal comfort model unit is used for respectively generating a single thermal comfort value of each effective physiological signal data through a thermal comfort index model, and determining the effective single thermal comfort value according to a thermal comfort value mean value and a thermal comfort standard deviation;
the refrigeration control unit is connected with the thermal comfort model unit and is used for calculating a target thermal comfort value average value according to each effective monomer thermal comfort value and correspondingly adjusting the temperature of the refrigeration subarea;
wherein the human physiological data comprises the human skin temperature, the brain electrical signal and the electrocardiosignal of the target acquisition body; the physiological signal data to be noise-reduced are brain electrical signals and electrocardiosignals, and the skin temperature of the human body is effective physiological signal data; the refrigeration area environment information comprises the number of target acquisition bodies, the environment temperature and the positions of the target acquisition bodies;
the refrigeration end comprises a plurality of refrigeration subareas with refrigeration function.
Further, the data denoising submodule is connected with the data acquisition unit and used for denoising the physiological signal data to be denoised of the target acquisition body through a Butterworth filter so as to generate effective physiological signal data;
the Z-score module is connected with the data denoising submodule and is used for combining effective physiological signal data into an input set, normalizing the input set by using a Z-score model to obtain a normal input set, and randomly distributing the data of the normal input set to a training set and a verification set;
the model generation module is used for training the CNN model by using the training set to generate a thermal comfort index model which meets the requirements of users;
the thermal comfort value is a difference between a current ambient temperature and a target ambient temperature corresponding to a human comfort state.
Further, the data analysis unit determines a refrigeration mode of the refrigeration subarea according to the target collection body quantity in the refrigeration subarea under a first data analysis condition;
if the number of the target acquisition bodies is smaller than the reference value of the preset signal number, the data analysis unit judges that the refrigeration subarea starts circulating refrigeration, and determines the initial temperature of the refrigeration subarea according to the Euclidean distance between the target acquisition bodies in the refrigeration subarea;
if the number of the target acquisition bodies is greater than or equal to the preset signal number reference value, the data analysis unit judges that the refrigeration subarea is started to continuously refrigerate, and determines the initial temperature of the refrigeration subarea according to the number difference delta N between the number N of the target acquisition bodies and the preset signal number reference value N0;
wherein Δn=n—n0, and the first data analysis condition is that the data acquisition unit detects that a target acquisition body exists in the refrigeration subarea.
Further, the data analysis unit calculates Euclidean distances between a single target acquisition body and other target acquisition bodies according to the plane coordinate positions of the target acquisition bodies under the second data analysis condition;
the data analysis unit is provided with a preset Euclidean distance threshold value;
if the Euclidean distance between the two target acquisition bodies is smaller than a preset Euclidean distance threshold, the data analysis unit judges that the two target acquisition bodies are in an aggregation relationship;
the data analysis unit determines the initial temperature of the refrigeration subarea according to the aggregation relation quantity of the single target acquisition body with the largest aggregation relation;
the second data analysis condition is that the number of target acquisition bodies of the single refrigeration subarea is smaller than a preset signal number reference value.
Further, the data denoising sub-module performs denoising treatment on the physiological signal data to be denoised of the user, calculates a peak detection parameter F in the physiological signal data to be denoised after denoising, and determines whether the physiological signal data to be denoised after denoising is effective physiological signal data according to the peak detection parameter F;
if F is smaller than the preset peak detection parameter, the data denoising sub-module judges that the denoising physiological signal data to be denoised is effective physiological signal data;
wherein, the calculation formula of F is:
F＝FPR＋FNR
wherein FPR is false positive rate, FNR is false negative rate, wherein electrocardiosignals and electroencephalogram signals respectively correspond to different FPRs and FNRs,
the calculation formula of the FPR and FNR corresponding to the electroencephalogram signals is FPR= (FP/TN) multiplied by 100 percent, FNR= (FN/TP) multiplied by 100 percent,
the calculation formula of the FPR and FNR corresponding to the electrocardiosignal is FPR= [ FP/(FP+TN) ]. Times.100%, FNR= [ FN/(TP+FN) ]. Times.100%,
here, FP represents the number of times classified as intended but not actually intended, FN represents the number of times classified as not intended but not actually intended, TP represents the number of times classified as intended and actually intended, and TN represents the number of times classified as not intended and not actually intended.
Further, the thermal comfort model unit determines whether the target acquisition body is a thermal comfort reference target body according to the leaving frequency of the target acquisition body in the refrigeration subarea, and determines a plurality of temperature regulation modes of the refrigeration subarea according to the effective physiological signal data of the thermal comfort reference target body;
and if the departure frequency of the target acquisition body is in a first preset departure frequency state, the thermal comfort model unit judges that the target acquisition body is a thermal comfort reference target body.
Further, the thermal comfort model unit generates a plurality of single thermal comfort values Xi through the thermal comfort index model based on the effective physiological signal data of the thermal comfort reference target body of the refrigeration subarea under the first thermal comfort judging condition, wherein i=1, 2,3, … …, n, n is the total number of the thermal comfort reference target bodies;
wherein the first thermal comfort determination condition is that thermal comfort reference target determination is completed.
Further, the thermal comfort model unit is configured to, under a second thermal comfort determination conditionCalculating a thermal comfort value mean value according to each single thermal comfort valueAnd a thermal comfort standard deviation delta, and the individual thermal comfort value and the thermal comfort value mean +.>And a comparison of the thermal comfort standard deviation delta, determining whether the individual thermal comfort value is a valid individual thermal comfort value;
if it isThe thermal comfort model unit determines the individual thermal comfort value as a valid individual thermal comfort value;
wherein, setting，/>The method comprises the steps of carrying out a first treatment on the surface of the u is a range compensation coefficient, and u is more than or equal to 1;
the second thermal comfort determination condition is that the generation of the individual thermal comfort values is completed.
Further, the refrigeration control unit calculates a target thermal comfort value average value according to each effective monomer thermal comfort valueAnd according to->Determining the environmental temperature to be reached by the air conditioner;
setting upXc is the c-th effective monomer thermal comfort value, and cmax is the total number of effective monomer thermal comfort values.
Further, the refrigeration control unit controls the refrigeration capacity of the air conditioner according to the target thermal comfort value average value so as to enable the ambient temperature to reach the target ambient temperature;
if the target thermal comfort value average value is a negative value, the refrigeration control unit regulates down the refrigeration capacity so as to improve the temperature of the refrigeration subarea;
if the target thermal comfort value average value is a positive value, the refrigeration control unit adjusts the refrigeration capacity to reduce the temperature of the refrigeration subarea;
the target thermal comfort value average value is the difference value of the current ambient temperature minus the target ambient temperature of the comfort state of the target acquisition body.
Compared with the prior art, the method has the advantages that the refrigeration area is uniformly divided into the plurality of refrigeration subareas, so that information acquisition and temperature control are more accurate, and for a single refrigeration subarea, the initial temperature of the refrigeration subarea is more in accordance with an actual scene by determining the initial temperature according to the number of target acquisition bodies and the aggregation degree of the target acquisition bodies in the refrigeration subarea, effective physiological signal data for determining the target acquisition bodies are obtained through noise reduction according to the physiological data of a human body of the single target acquisition body in the refrigeration subarea, and whether the target acquisition bodies are thermal comfort reference targets or not is determined according to the leaving frequency of the target acquisition bodies in the refrigeration subarea, so that the thermal comfort reference values are more in accordance, the judgment precision of the method is further improved, and the secondary regulation rate of an air conditioner is reduced.
Further, the data analysis unit determines whether the refrigeration subarea is cyclically started and stopped for refrigeration according to the number of target acquisition bodies in the refrigeration subarea under the first data analysis condition, and determines the initial temperature of the refrigeration subarea according to the Euclidean distance between the target acquisition bodies in the refrigeration subarea, so that the judgment of the initial temperature is more in line with the actual scene requirement.
Furthermore, the thermal comfort model unit performs noise reduction processing on the physiological signal data to be noise reduced of the user, judges whether the physiological signal data to be noise reduced is effective physiological signal data according to the comparison result of the peak detection parameter and the preset peak detection parameter, and improves the signal to noise ratio of the signal by removing the noise in the signal, so that the signal analysis is more accurate and reliable, and the judgment accuracy of the thermal comfort model unit is improved.
Further, the thermal comfort model unit determines whether the target acquisition body is a thermal comfort reference target body according to the departure frequency of the target acquisition body in the refrigeration subarea, wherein the departure frequency is the times of the target acquisition body entering and leaving the refrigeration subarea in unit time, so that the data processing capacity is reduced, and meanwhile, the temperature regulation is more in accordance with the requirements of users in the refrigeration subarea.
Further, in the present inventionThe thermal comfort model unit judges that the single thermal comfort value is an effective single thermal comfort value, so that misjudgment of the single thermal comfort value due to individual specification or acquisition problem of the device is avoided, and the judgment precision of the single thermal comfort value is further improved.
Drawings
FIG. 1 is a unit connection diagram of an intelligent air conditioning control system based on a wearable device and a convolutional neural network of the present invention;
FIG. 2 is a schematic diagram of a CNN model according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a refrigeration end according to an embodiment of the present invention;
in the figure: 1, manufacturing a cold end; 2, refrigerating subareas; and 3, wearing a target acquisition body of the wearable equipment.
Detailed Description
In order that the objects and advantages of the invention will become more apparent, the invention will be further described with reference to the following examples; it should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are merely for explaining the technical principles of the present invention, and are not intended to limit the scope of the present invention.
It should be noted that, in the description of the present invention, terms such as "upper," "lower," "left," "right," "inner," "outer," and the like indicate directions or positional relationships based on the directions or positional relationships shown in the drawings, which are merely for convenience of description, and do not indicate or imply that the apparatus or elements must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be either fixedly connected, detachably connected, or integrally connected, for example; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communication between two elements. The specific meaning of the above terms in the present invention can be understood by those skilled in the art according to the specific circumstances.
Referring to fig. 1, the present invention provides an intelligent air conditioner control system based on a wearable device and a convolutional neural network, comprising:
the data acquisition unit comprises a wearable device for acquiring human physiological data of a target acquisition body and a monitoring device for acquiring environment information of a refrigeration area;
the data analysis unit is connected with the data acquisition unit and is used for determining the refrigeration mode of the refrigeration subarea according to the number of target acquisition bodies in the refrigeration subarea and determining the initial temperature of the refrigeration subarea according to the Euclidean distance between the target acquisition bodies or the number of target acquisition bodies;
the thermal comfort model unit is connected with the data analysis unit and comprises a data noise reduction sub-module, a Z-score module and a model generation module, wherein the data noise reduction sub-module is used for carrying out noise reduction treatment on the physiological signal data which is required to be reduced by a user so as to determine effective physiological signal data, the Z-score module is used for processing the effective physiological signal data into a training set, and the model generation module is used for generating a thermal comfort index model according to the data of the training set and the corresponding environment temperature;
the data screening unit is connected with the data acquisition unit and the thermal comfort model unit, and is used for determining whether a target acquisition body is a thermal comfort reference target body according to the departure frequency of the target acquisition body in the refrigeration subarea, and determining the temperature regulation mode of the refrigeration subarea according to the effective physiological signal data of the thermal comfort reference target body, wherein the thermal comfort model unit is used for respectively generating a single thermal comfort value of each effective physiological signal data through a thermal comfort index model, and determining the effective single thermal comfort value according to a thermal comfort value mean value and a thermal comfort standard deviation;
the refrigeration control unit is connected with the thermal comfort model unit and is used for calculating a target thermal comfort value average value according to each effective monomer thermal comfort value and correspondingly adjusting the temperature of the refrigeration subarea;
wherein the human physiological data comprises the human skin temperature, the brain electrical signal and the electrocardiosignal of the target acquisition body; the physiological signal data to be noise-reduced are brain electrical signals and electrocardiosignals, and the skin temperature of the human body is effective physiological signal data; the refrigeration area environment information comprises the number of target acquisition bodies, the environment temperature and the positions of the target acquisition bodies;
the refrigeration end comprises a plurality of refrigeration subareas with refrigeration function.
Through dividing into a plurality of refrigeration subregions with refrigerating end to avoid refrigerating to the large tracts of land region when, the air conditioner temperature is difficult to satisfy the problem of actual working scene personnel demand, the setting of refrigeration subregion makes the control regulation of air conditioner accord with actual working condition more.
In implementation, the wearable device is not limited in setting mode, but should have the function of detecting electrocardiosignals, electroencephalograms and human skin temperature data of the target acquisition body.
Specifically, the data denoising sub-module is connected with the data acquisition unit and is used for denoising the physiological signal data to be denoised of the target acquisition body through a Butterworth filter so as to generate effective physiological signal data;
the Z-score module is connected with the data denoising submodule and is used for combining effective physiological signal data into an input set, normalizing the input set by using a Z-score model to obtain a normal input set, randomly distributing the data of the normal input set to a training set and a verification set, wherein in implementation, the training set generally occupies 70% of the data of the normal input set, and the verification set occupies 30% of the data of the normal input set;
the model generation module is used for training the CNN model by using the training set to generate a thermal comfort index model which meets the requirements of users;
the thermal comfort value is a difference between a current ambient temperature and a target ambient temperature corresponding to a human comfort state.
Specifically, the data analysis unit determines the refrigeration mode of the refrigeration subarea according to the target collection body quantity in the refrigeration subarea under a first data analysis condition;
if the number of the target acquisition bodies is smaller than the reference value of the preset signal number, the data analysis unit judges that the refrigeration subarea starts circulating refrigeration, and determines the initial temperature of the refrigeration subarea according to the Euclidean distance between the target acquisition bodies in the refrigeration subarea;
if the number of the target acquisition bodies is greater than or equal to the preset signal number reference value, the data analysis unit judges that the refrigeration subarea is started to continuously refrigerate, and determines the initial temperature of the refrigeration subarea according to the number difference delta N between the number N of the target acquisition bodies and the preset signal number reference value N0;
wherein Δn=n—n0, and the first data analysis condition is that the data acquisition unit detects that a target acquisition body exists in the refrigeration subarea.
Specifically, the Euclidean distance calculating method includes that the distance difference between two target collectors in the horizontal direction is calculated, the distance difference between two points in the vertical direction is calculated, and then the square sum and the open square of the two distance differences are calculated to obtain the Euclidean distance.
The number of the target collection bodies and the reference value of the preset signal number reflect the number of people in a single refrigeration subarea so as to avoid the problem of temperature rise in the refrigeration subarea caused by excessive number of people;
the circulating refrigeration is that the starting state and the standby state of the refrigeration device in the refrigeration subarea are periodically switched, and the starting time and the standby time are the same; the value of the preset signal quantity reference value is related to the area of the single refrigerating subarea, and the larger the area of the single refrigerating subarea is, the larger the preset signal quantity reference value is because the more personnel are in the fixed space, the higher the indoor temperature is.
Specifically, the data analysis unit calculates Euclidean distances between a single target acquisition body and other target acquisition bodies according to the plane coordinate positions of the target acquisition bodies under the second data analysis condition;
the data analysis unit is provided with a preset Euclidean distance threshold value;
if the Euclidean distance between the two target acquisition bodies is smaller than a preset Euclidean distance threshold, the data analysis unit judges that the two target acquisition bodies are in an aggregation relationship;
the data analysis unit determines the initial temperature of the refrigeration subarea according to the aggregation relation quantity of the single target acquisition body with the largest aggregation relation;
the second data analysis condition is that the number of target acquisition bodies of the single refrigeration subarea is smaller than a preset signal number reference value.
Wherein the Euclidean distance represents the aggregation degree of each target acquisition body in a single refrigeration subarea so as to avoid the problems of temperature rise and somatosensory temperature rise of the refrigeration subarea caused by the aggregation of the target acquisition bodies
The preset Euclidean distance threshold is related to the demand of a user for refrigeration, namely, the user can determine the preset Euclidean distance threshold according to the distance of a target acquisition body which influences the comfort level of the user in advance.
Specifically, the data denoising sub-module performs denoising processing on the physiological signal data to be denoised of a user, calculates a peak detection parameter F in the physiological signal data to be denoised, determines whether the physiological signal data to be denoised after denoise is effective physiological signal data according to the peak detection parameter F, and respectively corresponds to preset peak detection parameters with different values of the electroencephalogram signal and the electrocardiosignal signal;
if F is smaller than the preset peak detection parameter, the data denoising sub-module judges that the denoising physiological signal data to be denoised is effective physiological signal data;
wherein, the calculation formula of F is:
F＝FPR＋FNR
wherein FPR is false positive rate, FNR is false negative rate, wherein electrocardiosignals and electroencephalogram signals respectively correspond to different FPRs and FNRs,
the calculation formula of the FPR and FNR corresponding to the electroencephalogram signals is FPR= (FP/TN) multiplied by 100 percent, FNR= (FN/TP) multiplied by 100 percent,
the calculation formula of the FPR and FNR corresponding to the electrocardiosignal is FPR= [ FP/(FP+TN) ]. Times.100%, FNR= [ FN/(TP+FN) ]. Times.100%,
here, FP represents the number of times classified as intended but not actually intended, FN represents the number of times classified as not intended but not actually intended, TP represents the number of times classified as intended and actually intended, and TN represents the number of times classified as not intended and not actually intended.
FPR and FNR respectively represent the ratio of erroneously predicting negative samples as positive samples and the ratio of erroneously predicting positive samples as negative samples, and the problem of subsequent judgment errors caused by that noise reduction does not reach the noise reduction requirement of a user is avoided by comparing F with preset peak detection parameters.
The preset peak detection parameters can obtain a large amount of noise-reduction physiological signal data of the user after noise reduction through experiments, record F corresponding to the noise-reduction physiological signal data of the user meeting the user requirements, calculate an average value of the F as the preset peak detection parameters, but are worth noting that blurring of the noise-reduction signals should be avoided, the value of the preset peak detection parameters is required to keep high classification accuracy and recall, and the content is easy to understand by a person skilled in the art, and the value of the preset peak detection parameters F0, F0=0.2 corresponding to the electroencephalogram signals and F0=0.15 corresponding to the electrocardiosignals are provided.
Specifically, the thermal comfort model unit determines whether the target acquisition body is a thermal comfort reference target body according to the leaving frequency of the target acquisition body in the refrigeration subarea, and determines a temperature regulation mode of the refrigeration subarea according to the effective physiological signal data of the thermal comfort reference target body;
and if the departure frequency of the target acquisition body is in a first preset departure frequency state, the thermal comfort model unit judges that the target acquisition body is a thermal comfort reference target body.
The leaving frequency is the number of times of leaving the refrigerating subarea where the target collector is positioned in unit time, and the value of the first preset leaving frequency state is the average value of the leaving frequencies of the target collectors in the refrigerating subarea. Preferably, the present embodiment provides a value range of a first preset departure frequency state, where the first preset departure frequency state is that the departure frequency is less than 3 times/1 h.
Specifically, the thermal comfort model unit generates a plurality of single thermal comfort values Xi through the thermal comfort index model based on the effective physiological signal data of the thermal comfort reference target body of the refrigeration subarea under the first thermal comfort judging condition, wherein i=1, 2,3, … …, n, n is the total number of the thermal comfort reference target bodies;
wherein the first thermal comfort determination condition is that thermal comfort reference target determination is completed.
Generating a plurality of single thermal comfort values for each thermal comfort reference target body to reflect the thermal comfort level of the single target acquisition body, thereby determining the subsequent temperature adjustment mode of the refrigeration subarea.
Specifically, the thermal comfort model unit calculates a thermal comfort value average value from each of the individual thermal comfort values under a second thermal comfort determination conditionAnd a thermal comfort standard deviation delta, and the individual thermal comfort value and the thermal comfort value mean +.>And a comparison of the thermal comfort standard deviation delta, determining whether the individual thermal comfort value is a valid individual thermal comfort value;
if it isThe thermal comfort model unit determines that the single thermal comfort value isAn effective monomer thermal comfort value;
wherein, setting，/>The method comprises the steps of carrying out a first treatment on the surface of the u is a range compensation coefficient, u=3;
the second thermal comfort determination condition is that the generation of the individual thermal comfort values is completed.
By judging the effective monomer thermal comfort value, the problem of low control and regulation precision caused by the fact that the monomer thermal comfort value has larger specificity due to the data acquisition problem is avoided.
Specifically, the refrigeration control unit calculates a target thermal comfort value average value from each of the effective individual thermal comfort valuesAnd according to->Determining the environmental temperature to be reached by the air conditioner;
setting upXc is the c-th effective monomer thermal comfort value, and cmax is the total number of effective monomer thermal comfort values.
Specifically, the refrigeration control unit controls the refrigeration capacity of the air conditioner according to the target thermal comfort value average value so as to enable the ambient temperature to reach the target ambient temperature;
if the target thermal comfort value average value is a negative value, the refrigeration control unit regulates down the refrigeration capacity so as to improve the temperature of the refrigeration subarea;
if the target thermal comfort value average value is a positive value, the refrigeration control unit adjusts the refrigeration capacity to reduce the temperature of the refrigeration subarea;
the target thermal comfort value average value is the difference value of the current ambient temperature minus the target ambient temperature of the comfort state of the target acquisition body.
The refrigeration control unitCalculating a target thermal comfort value mean value according to each effective monomer thermal comfort valueSo that the temperature of the regulated refrigeration subarea can meet the requirements of each target collector in the subarea.
Examples: referring to fig. 2 to 3, in the present embodiment, the area of a refrigeration side 1 is 400 square meters, the area of a refrigeration sub-area 2 is 100 square meters, and a monitoring device detects that a plurality of target acquisition bodies 3 wearing wearable devices are present in the refrigeration sub-area 2;
the data denoising submodule uses a Butterworth band-pass filter to denoise physiological signal data to be denoised of a single target acquisition body, the order of the filter is 4, and the passband range of the filter is 150Hz;
the Z-score module combines the effective physiological signal data into an input set, performs normalization processing on the input set by using the Z-score model to obtain a normal input set, randomly distributes the normal input set into a training set and a verification set, wherein the training set accounts for 70% of the normal input set, and the verification set accounts for 30% of the normal input set;
converting an input set into a matrix, and training through a CNN model to generate a thermal comfort index model meeting the user requirement, wherein the CNN model comprises 12 layers, namely an input layer, a first convolution layer, a first batch normalization layer, a first ReLu layer, a first maximum pooling layer, a second convolution layer, a second batch normalization layer, a second ReLu layer, a second maximum pooling layer, a full connection layer, a softmax layer and an output layer;
after an input set is input into an input layer, data of the input layer are convolved in a first layer convolution layer, the first layer batch normalization layer is performed for normalization after the convolution layer is completed, then the first layer ReLu layer is entered for normalization, the first layer maximum pooling layer is entered after normalization is performed for normalization, the second layer convolution, batch normalization, reLu and maximum pooling are continued, finally pooled data are sequentially connected, one-dimensional data are output, finally classification is performed according to a softmax layer, all thermal comfort values are mapped into a (0, 1) interval, the output of all the thermal comfort values in the all the connection layers are compared, the process can be regarded as a probability, the probability of which output belongs to is judged to be higher, and the process is iterated 500 times to generate a thermal comfort index model.
It should be noted that the acquisition and processing of physiological signals requires that the accuracy and reliability of the data be guaranteed, while the training and testing of the model needs to be performed on a sufficient data set to ensure the generalization ability and predictive effect of the model.
Thus far, the technical solution of the present invention has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present invention is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present invention, and such modifications and substitutions will be within the scope of the present invention.
The foregoing description is only of the preferred embodiments of the invention and is not intended to limit the invention; various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. An intelligent air conditioner control system based on wearable equipment and convolutional neural network, which is characterized by comprising:
the data acquisition unit comprises a wearable device for acquiring human physiological data of a target acquisition body and a monitoring device for acquiring environment information of a refrigeration area;
the data analysis unit is connected with the data acquisition unit and is used for determining the refrigeration mode of the refrigeration subarea according to the number of target acquisition bodies in the refrigeration subarea and determining the initial temperature of the refrigeration subarea according to the Euclidean distance between the target acquisition bodies or the number of target acquisition bodies;
the thermal comfort model unit is connected with the data analysis unit and comprises a data noise reduction sub-module, a Z-score module and a model generation module, wherein the data noise reduction sub-module is used for carrying out noise reduction treatment on the physiological signal data which is required to be reduced by a user so as to determine effective physiological signal data, the Z-score module is used for processing the effective physiological signal data into a training set, and the model generation module is used for generating a thermal comfort index model according to the data of the training set and the corresponding environment temperature;
the data screening unit is connected with the data acquisition unit and the thermal comfort model unit, and is used for determining whether a target acquisition body is a thermal comfort reference target body according to the departure frequency of the target acquisition body in the refrigeration subarea, and determining the temperature regulation mode of the refrigeration subarea according to the effective physiological signal data of the thermal comfort reference target body, wherein the thermal comfort model unit is used for respectively generating a single thermal comfort value of each effective physiological signal data through a thermal comfort index model, and determining the effective single thermal comfort value according to a thermal comfort value mean value and a thermal comfort standard deviation;
the refrigeration control unit is connected with the thermal comfort model unit and is used for calculating a target thermal comfort value average value according to each effective monomer thermal comfort value and correspondingly adjusting the temperature of the refrigeration subarea;
wherein the human physiological data comprises the human skin temperature, the brain electrical signal and the electrocardiosignal of the target acquisition body; the physiological signal data to be noise-reduced are brain electrical signals and electrocardiosignals, and the skin temperature of the human body is effective physiological signal data; the refrigeration area environment information comprises the number of target acquisition bodies, the environment temperature and the positions of the target acquisition bodies;
the refrigerating end comprises a plurality of refrigerating subareas with refrigerating functions;
the data denoising sub-module is connected with the data acquisition unit and is used for denoising the physiological signal data to be denoised of the target acquisition body through a Butterworth filter so as to generate effective physiological signal data;
the Z-score module is connected with the data denoising submodule and is used for combining effective physiological signal data into an input set, normalizing the input set by using a Z-score model to obtain a normal input set, and randomly distributing the data of the normal input set to a training set and a verification set;
the model generation module is used for training the CNN model by using the training set to generate a thermal comfort index model which meets the requirements of users;
the thermal comfort value is the difference between the current ambient temperature and the target ambient temperature corresponding to the comfort state of the human body;
the data analysis unit determines the refrigeration mode of the refrigeration subarea according to the target collection body quantity in the refrigeration subarea under a first data analysis condition;
if the number of the target acquisition bodies is smaller than the reference value of the preset signal number, the data analysis unit judges that the refrigeration subarea starts circulating refrigeration, and determines the initial temperature of the refrigeration subarea according to the Euclidean distance between the target acquisition bodies in the refrigeration subarea;
if the number of the target acquisition bodies is greater than or equal to the preset signal number reference value, the data analysis unit judges that the refrigeration subarea is started to continuously refrigerate, and determines the initial temperature of the refrigeration subarea according to the number difference delta N between the number N of the target acquisition bodies and the preset signal number reference value N0;
wherein Δn=n—n0, and the first data analysis condition is that the data acquisition unit detects that a target acquisition body exists in the refrigeration subarea.
2. The intelligent air conditioner control system based on the wearable device and the convolutional neural network according to claim 1, wherein the data analysis unit calculates the Euclidean distance between the single target acquisition body and other target acquisition bodies according to the plane coordinate position of each target acquisition body under the second data analysis condition;
the data analysis unit is provided with a preset Euclidean distance threshold value;
if the Euclidean distance between the two target acquisition bodies is smaller than a preset Euclidean distance threshold, the data analysis unit judges that the two target acquisition bodies are in an aggregation relationship;
the data analysis unit determines the initial temperature of the refrigeration subarea according to the aggregation relation quantity of the single target acquisition body with the largest aggregation relation;
the second data analysis condition is that the number of target acquisition bodies of the single refrigeration subarea is smaller than a preset signal number reference value.
3. The intelligent air conditioner control system based on the wearable device and the convolutional neural network according to claim 2, wherein the data denoising sub-module performs denoising processing on the physiological signal data to be denoised of the user and calculates a peak detection parameter F in the physiological signal data to be denoised after denoising, and determines whether the physiological signal data to be denoised after denoising is effective physiological signal data according to the peak detection parameter F;
if F is smaller than the preset peak detection parameter, the data denoising sub-module judges that the denoising physiological signal data to be denoised is effective physiological signal data;
wherein, the calculation formula of F is:
F＝FPR＋FNR
wherein FPR is false positive rate, FNR is false negative rate, wherein electrocardiosignals and electroencephalogram signals respectively correspond to different FPRs and FNRs,
the calculation formula of the FPR and FNR corresponding to the electroencephalogram signals is FPR= (FP/TN) multiplied by 100 percent, FNR= (FN/TP) multiplied by 100 percent,
the calculation formula of the FPR and FNR corresponding to the electrocardiosignal is FPR= [ FP/(FP+TN) ]. Times.100%, FNR= [ FN/(TP+FN) ]. Times.100%,
here, FP represents the number of times classified as intended but not actually intended, FN represents the number of times classified as not intended but not actually intended, TP represents the number of times classified as intended and actually intended, and TN represents the number of times classified as not intended and not actually intended.
4. The intelligent air conditioner control system based on the wearable device and the convolutional neural network according to claim 3, wherein the thermal comfort model unit determines whether the target acquisition body is a thermal comfort reference target body according to the departure frequency of the target acquisition body in the refrigeration subarea, and determines a plurality of temperature adjustment modes of the refrigeration subarea according to the effective physiological signal data of the thermal comfort reference target body;
and if the departure frequency of the target acquisition body is in a first preset departure frequency state, the thermal comfort model unit judges that the target acquisition body is a thermal comfort reference target body.
5. The intelligent air conditioner control system based on a wearable device and a convolutional neural network according to claim 4, wherein the thermal comfort model unit generates a number of individual thermal comfort values Xi by the thermal comfort index model from the effective physiological signal data of the thermal comfort reference target of the refrigeration sub-region under a first thermal comfort decision condition, wherein i = 1,2,3, … …, n, n is the total number of thermal comfort reference targets;
wherein the first thermal comfort determination condition is that thermal comfort reference target determination is completed.
6. The wearable device and convolutional neural network-based intelligent air conditioner control system of claim 5, wherein the thermal comfort model unit calculates a thermal comfort value mean value from each of the individual thermal comfort values under a second thermal comfort determination conditionAnd a thermal comfort standard deviation delta, and the individual thermal comfort value and the thermal comfort value mean +.>And a comparison of the thermal comfort standard deviation delta, determining whether the individual thermal comfort value is a valid individual thermal comfort value;
if it isThe thermal comfort model unit determines the individual thermal comfort value as a valid individual thermal comfort value;
wherein, setting，/>The method comprises the steps of carrying out a first treatment on the surface of the u is a range compensation coefficient, and u is more than or equal to 1;
the second thermal comfort determination condition is that the generation of the individual thermal comfort values is completed.
7. The intelligent air conditioning control system based on a wearable device and a convolutional neural network of claim 6, wherein the refrigeration control unit calculates a target thermal comfort value mean value from each of the effective individual thermal comfort valuesAnd according to->Determining the environmental temperature to be reached by the air conditioner;
setting upXc is the c-th effective individual thermal comfort value, c=1, 2,3, … …, cmax, cmax being the total number of effective individual thermal comfort values.
8. The intelligent air conditioner control system based on the wearable device and the convolutional neural network according to claim 7, wherein the refrigeration control unit controls the refrigeration capacity of the air conditioner according to the target thermal comfort value average value so that the ambient temperature reaches the target ambient temperature;
if the target thermal comfort value average value is a negative value, the refrigeration control unit regulates down the refrigeration capacity so as to improve the temperature of the refrigeration subarea;
if the target thermal comfort value average value is a positive value, the refrigeration control unit adjusts the refrigeration capacity to reduce the temperature of the refrigeration subarea;
the target thermal comfort value average value is the difference value of the current ambient temperature minus the target ambient temperature of the comfort state of the target acquisition body.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311011802.8A CN116734389B (en) | 2023-08-11 | 2023-08-11 | Intelligent air conditioner control system based on wearable equipment and convolutional neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311011802.8A CN116734389B (en) | 2023-08-11 | 2023-08-11 | Intelligent air conditioner control system based on wearable equipment and convolutional neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116734389A CN116734389A (en) | 2023-09-12 |
| CN116734389B true CN116734389B (en) | 2023-10-20 |
Family
ID=87918974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311011802.8A Active CN116734389B (en) | 2023-08-11 | 2023-08-11 | Intelligent air conditioner control system based on wearable equipment and convolutional neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116734389B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0835702A (en) * | 1994-07-22 | 1996-02-06 | Kubota Corp | Air conditioning facility |
| CN104808723A (en) * | 2015-03-03 | 2015-07-29 | 南京邮电大学 | Semi-intelligent office air conditioner regulation system based on wireless sensor network |
| CN105398307A (en) * | 2015-11-23 | 2016-03-16 | 惠州学院 | Electric automobile temperature adjusting system and method |
| CN106200400A (en) * | 2016-08-18 | 2016-12-07 | 南昌大学 | A kind of house control system based on brain electricity APP |
| CN108019901A (en) * | 2017-11-16 | 2018-05-11 | 青岛安森克电子有限公司 | A kind of orientation adjustment air-conditioning |
| CN108613332A (en) * | 2018-04-12 | 2018-10-02 | 南京信息工程大学 | A kind of energy-saving building film micro area personnel interactive mode hot comfort adjusting method |
-
2023
- 2023-08-11 CN CN202311011802.8A patent/CN116734389B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0835702A (en) * | 1994-07-22 | 1996-02-06 | Kubota Corp | Air conditioning facility |
| CN104808723A (en) * | 2015-03-03 | 2015-07-29 | 南京邮电大学 | Semi-intelligent office air conditioner regulation system based on wireless sensor network |
| CN105398307A (en) * | 2015-11-23 | 2016-03-16 | 惠州学院 | Electric automobile temperature adjusting system and method |
| CN106200400A (en) * | 2016-08-18 | 2016-12-07 | 南昌大学 | A kind of house control system based on brain electricity APP |
| CN108019901A (en) * | 2017-11-16 | 2018-05-11 | 青岛安森克电子有限公司 | A kind of orientation adjustment air-conditioning |
| CN108613332A (en) * | 2018-04-12 | 2018-10-02 | 南京信息工程大学 | A kind of energy-saving building film micro area personnel interactive mode hot comfort adjusting method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116734389A (en) | 2023-09-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2017008321A1 (en) | Smart home energy management method based on smart wearable device behavior detection | |
| CN106352506A (en) | Air conditioner and automatic control method thereof | |
| CN110513840A (en) | Temprature control method and system based on smart home operating system | |
| CN103375869A (en) | Air conditioner control method, device and air conditioner | |
| CN111696575B (en) | Low ventilation and apnea detection and identification system based on hybrid neural network model | |
| CN112254287B (en) | Variable-weight multi-model comprehensive prediction central air conditioner tail end air supply control method | |
| CN106236013A (en) | A kind of sleep monitor method and device | |
| CN111870793B (en) | Temperature pre-adjusting method and device, electronic equipment and readable storage medium | |
| CN108954692B (en) | Environment temperature calculation method and air conditioner | |
| CN111897230A (en) | Sleep quality monitoring method and device, electrical equipment, storage medium and processor | |
| CN111609537B (en) | Air conditioner, control method and device thereof and air conditioning system | |
| CN109044272A (en) | Sleep state monitoring and sleep environment parameter regulation method, system and terminal device | |
| CN109812936B (en) | Air conditioner and air conditioner air outlet temperature adjusting method and device thereof | |
| CN106765970A (en) | Air-conditioner and its control method | |
| CN116734389B (en) | Intelligent air conditioner control system based on wearable equipment and convolutional neural network | |
| CN113858915A (en) | Control method, system, equipment and storage medium for vehicle cabin air conditioner | |
| CN205279345U (en) | Intelligent air conditioner | |
| CN112303850A (en) | Control method and device of air conditioner | |
| CN112097378A (en) | Air conditioner comfort level adjusting method based on feedforward neural network | |
| CN113531804B (en) | Automatic control method and device of air conditioner, electronic equipment and storage medium | |
| CN115185320A (en) | Intelligent vehicle-mounted sleep system and control method thereof | |
| CN109084437B (en) | Control method and control device for automatic temperature adjustment of air conditioner | |
| CN113776105A (en) | Intelligent variable air volume range hood based on behavior monitoring of cooking personnel | |
| CN111271810A (en) | Air conditioner control method, control system, air conditioner and computer storage medium | |
| CN117168558B (en) | High-end intelligent real-time monitoring method for fuel consumption of caravan |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |