CN110017567A - A kind of condensation prevention control method and device - Google Patents
A kind of condensation prevention control method and device Download PDFInfo
- Publication number
- CN110017567A CN110017567A CN201910094295.6A CN201910094295A CN110017567A CN 110017567 A CN110017567 A CN 110017567A CN 201910094295 A CN201910094295 A CN 201910094295A CN 110017567 A CN110017567 A CN 110017567A
- Authority
- CN
- China
- Prior art keywords
- behavioural characteristic
- behavior
- data
- slope
- data collection
- 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.)
- Granted
Links
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
- F24F11/41—Defrosting; Preventing freezing
-
- 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
- F24F11/89—Arrangement or mounting of control or safety devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
- F24F2110/20—Humidity
Abstract
The present invention provides a kind of condensation prevention control methods, which comprises obtains real-time humidity data collection；The processing of slope variance, which is executed, for the real-time humidity data collection extracts goal behavior feature；The goal behavior feature is inputted into default behavioural characteristic library, calculates the class probability that the goal behavior feature belongs to default behavioural characteristic；When the goal behavior feature is user behavior characteristics to be controlled, radiation cold source is closed in control, wherein the user behavior characteristics to be controlled are the behavioural characteristic that the condensation risk that represents is high in the default behavioural characteristic.Anti-condensation measure is triggered the present invention is based on user behavior, avoids traditional control method and the poor problem of effect is weakened to dew condensation phenomenon.
Description
Technical field
The present invention relates to filed of air conditioning more particularly to a kind of condensation prevention control methods and device.
Background technique
As the improvement of people's living standards, air-conditioning equipment is gradually widely used in a variety of applications on various occasions.Air-conditioning
Dew condensation phenomenon in operational process also results in the concern of each side such as manufacturer and user.Especially in cooling in summer, work as radiation
When the temperature on surface is reduced to dew-point temperature or less, the surface dew condensation phenomenon of wall or other equipment facility is more prominent significant.
Currently, usually being monitored using indoor thermostat to mitigate damage of the dew condensation phenomenon to electrical equipment and other facilities
Humidity, radiation temperature is constant indoors and atmospheric pressure it is constant under conditions of, will lead to the liter of dew-point temperature T1 when humidity increases
Height then cuts off radiating surface cold source (for example, closing spoke once learning that dew-point temperature T1 is more than or equal to radiating surface temperature T2 in t moment
Penetrate face water route), pause refrigeration, to keep radiating surface temperature mobile to the direction far from dew-point temperature, by compeling after time Δ t
So that indoor radiating surface temperature is raised to above dew-point temperature, reduces condensation risk and its loss.
However, the existing measure for alleviating condensation, due to being the triggering cutting radiation when temperature condition reaches condensation condition
Face cold source, so, even if having cut off cold source, since the response of system and the raising of indoor radiating surface temperature are required to centainly
Time, and can not effectively weaken condensation risk, still be easier to generate dew condensation phenomenon.
Summary of the invention
The present invention provides a kind of condensation prevention control method and device, to solve to weaken effect to dew condensation phenomenon in the prior art
Poor problem.
In order to solve the above-mentioned technical problem, the present invention is implemented as follows:
A kind of condensation prevention control method, which comprises
Obtain real-time humidity data collection；
The processing of slope variance, which is executed, for the real-time humidity data collection extracts goal behavior feature；
The goal behavior feature is inputted into default behavioural characteristic library, the goal behavior feature is calculated and belongs to default behavior
The class probability of feature；
When the goal behavior feature be user behavior characteristics to be controlled when, control close radiation cold source, wherein it is described to
The high behavioural characteristic of the condensation risk that user behavior characteristics are controlled to represent in the default behavioural characteristic.
Optionally, described to execute slope variance processing extraction goal behavior feature, packet for the real-time humidity data collection
It includes:
The relation curve that humidity changes over time is drawn according to the real-time humidity data collection；
Calculate the slope of each data point of real-time humidity data collection described in the relation curve；
The slope square of each data point is calculated according to the slope；
To the slope square slide default item number it is cumulative obtain slope square it is cumulative and；
It screens the cumulative and data greater than preset threshold of the slope square and forms goal behavior data set, wherein is described
Goal behavior data set characterizes the goal behavior feature.
Optionally, before the acquisition real-time humidity data collection further include:
Obtain history humidity data collection；
The history humidity data collection is trained by behavioural characteristic analysis, establishes default behavioural characteristic library, wherein
It include a variety of different user behaviors in the default behavioural characteristic library, the user behavior is the row for causing indoor humidity to change
For.
Optionally, described that the history humidity data collection is trained by behavioural characteristic analysis, establish default behavior
Feature database includes:
The slope variance processing is executed to the history humidity data collection and generates historical behavior data set；
Clustering is executed to the historical behavior data set and generates each aggregate of data, each aggregate of data represents difference
User behavior；
Behavior label, which is added, for each aggregate of data establishes default behavioural characteristic library.
The present invention also provides a kind of anti-condensation control device, described device includes:
Real time data acquisition module, for obtaining real-time humidity data collection；
Extraction module extracts goal behavior feature for executing the processing of slope variance for the real-time humidity data collection；
Computing module calculates the goal behavior feature category for the behavioural characteristic to be inputted default behavioural characteristic library
In the class probability of default behavioural characteristic；
Control module, for when the goal behavior feature is user behavior characteristics to be controlled, it is cold that radiation is closed in control
Source, wherein the user behavior characteristics to be controlled are the behavioural characteristic that the condensation risk that represents is high in the default behavioural characteristic.
Optionally, the extraction module includes:
Rendering submodule, for drawing the relation curve that humidity changes over time according to the real-time humidity data collection；
Slope computational submodule, for calculate real-time humidity data collection described in the relation curve each data point it is oblique
Rate；
Slope square computational submodule, for calculating the slope square of each data point according to the slope；
Cumulative submodule, for the slope square slide default item number it is cumulative obtain slope square it is cumulative and；
Submodule is screened, forms goal behavior number for screening the cumulative and data greater than preset threshold of the slope square
According to collection, wherein the goal behavior data set characterizes the goal behavior feature.
Optionally, described device further include:
Historical data obtains module, for obtaining history humidity data collection；
Training module establishes default row for being trained by behavioural characteristic analysis to the history humidity data collection
It is characterized library, wherein include a variety of different user behaviors in the default behavioural characteristic library, the user behavior is to cause room
The behavior of interior humidity variation.
Optionally, the training module includes:
Data set generation submodule generates history for executing the slope variance processing to the history humidity data collection
Behavioral data collection；
Aggregate of data generates submodule, generates each aggregate of data for executing clustering to the historical behavior data set,
Each aggregate of data represents different user behaviors；
Setting up submodule establishes default behavioural characteristic library for adding behavior label for each aggregate of data.
Optionally, described device further include:
Compared with prior art, the present invention has at least the following advantages:
A kind of condensation prevention control method and device provided by the invention handle real-time humidity data collection by slope variance
Goal behavior feature therein is extracted, goal behavior feature is inputted in default behavioural characteristic library, goal behavior feature can be obtained
Belong to the class probability of default behavioural characteristic, to learn user behavior representated by real-time humidity data collection, works as goal behavior
When feature is user behavior characteristics to be controlled, it is believed that the risk that user behavior generates condensation is larger, and control can be instructed to close spoke
Cold source is penetrated, to trigger anti-condensation measure based on user behavior, traditional control method is avoided and effect is weakened to dew condensation phenomenon
Poor problem.
Detailed description of the invention
Fig. 1 is a kind of flow chart of condensation prevention control method provided in an embodiment of the present invention；
Fig. 2 is the flow chart of another condensation prevention control method provided in an embodiment of the present invention；
Fig. 3 is the curve synoptic diagram that humidity changes over time in the embodiment of the present invention；
Fig. 4 is the schematic diagram of result of the embodiment of the present invention by the classification of k-means clustering algorithm；
Fig. 5 is a kind of structural block diagram of anti-condensation control device provided in an embodiment of the present invention；
Fig. 6 is the structural block diagram of another anti-condensation control device provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are some of the embodiments of the present invention, instead of all the embodiments.Based on this hair
Embodiment in bright, every other implementation obtained by those of ordinary skill in the art without making creative efforts
Example, shall fall within the protection scope of the present invention.
A kind of air-conditioner control system provided by the invention is discussed in detail below by several specific embodiments are enumerated.
Embodiment one
Referring to Fig.1, the present invention provides a kind of condensation prevention control methods, which comprises
Step 101, real-time humidity data collection is obtained.
Specifically, air-conditioning equipment runs heating or refrigeration, therefore, spoke with a certain steady temperature due under normal conditions
It is basicly stable constant to penetrate temperature, and room atmosphere pressure will not change, will lead to the raising of dew-point temperature when humidity increases,
In contrast, i.e., the gap between radiation temperature and dew-point temperature reduces, and the two is further close, and condensation risk will increase, thus
Indoor humidity data is acquired as real-time humidity data collection by the temperature controller in equipment in real time in the method, for example, with 8 points
To 10 points at interval of 1 minute humidity data as real-time humidity data collection, based on humidity data for predicting user behavior,
Carry out anti-condensation control.
Step 102, the processing of slope variance is executed for the real-time humidity data collection extract goal behavior feature.
Specifically, since the real-time humidity data got represents the humidity data of certain a period of time, this time
It is interior to occur influencing the user behavior of humidity variation, such as: between 8 points to 10 points of the morning at weekend, 8 points to 9 points,
User is in resting state, changes without influence on indoor humidity, 9 points to 10 points, cleaning after user gets up, for a long time
The behavior of mopping floor increase indoor humidity may.The processing of slope variance thus is executed to real-time humidity data collection, it can be to data
Screening and filtering is carried out, the goal behavior feature that data imply is extracted, which is to characterize the behavioral implications of user
Indoor humidity variation has been arrived, there may be the risk for generating condensation.
Step 103, the goal behavior feature is inputted into default behavioural characteristic library, calculates the goal behavior feature and belongs to
The class probability of default behavioural characteristic.
Specifically, based on the goal behavior feature extracted is concentrated from aforementioned real-time humidity data, it can be by the target line
Be characterized and be input in the default behavioural characteristic library pre-established, by goal behavior feature with it is various in default behavioural characteristic library
Behavior is compared calculating, determines that goal behavior feature belongs to the class probability of default behavioural characteristic.Such as: default behavior is special
Levying the behavioural characteristic in library is A, B three classes, and goal behavior feature M is that M ownership can be obtained by calculation to specific behavioural characteristic
In the class probability P of AA, M belongs to the class probability P of BB.Pass through PA、PBSize relation, it is believed that class probability numerical value
The classification for the behavioural characteristic that biggish classification, as goal behavior feature M are belonged to.
Step 104, when the goal behavior feature is user behavior characteristics to be controlled, radiation cold source is closed in control,
In, the user behavior characteristics to be controlled are the behavioural characteristic that the condensation risk that represents is high in the default behavioural characteristic.
Specifically, according to aforementioned step can determine real-time humidity data collection representated by belonging to goal behavior feature
Specific category, however, be not that each type requires to carry out anti-condensation control, such as: A class behavior feature is represented to room
Interior humidity influences more faint user behavior, for example, the windowing behavior etc. of humid day moment, according to experimental monitoring, this kind of row
The condensation risk for being characterized the user behavior of representative is low, is not necessarily to anti-condensation control；B class behavior feature, which represents, influences indoor humidity
Heavier user behavior, for example, the windowing behavior of humid day, heating up water, blocked shot, mopping floor etc. according to experimental monitoring, this class behavior
The condensation risk for the user behavior that feature represents is higher than A class；Obviously, when the behavior for analyzing user according to goal behavior feature M
It is when belonging to A class behavior feature, condensation risk is substantially not present, and carries out anti-condensation control without sacrificing user's comfort.Example
Such as, existing a kind of it may be the case that M belongs to the class probability P of A in practiceA=60%, M belong to the class probability P of BB=
40% it is ensured that goal behavior feature M belongs to A class behavior feature, the user behavior low for this kind of condensation risk,
Without anti-condensation control.Existing another kind is it may be the case that M belongs to the class probability P of A in practiceA=45%, M are belonged to
The class probability P of BB=55%, it was determined that goal behavior feature M belongs to B class behavior feature, i.e. current target row
It is characterized user behavior representated by M, judges that it belongs to the high B class of condensation risk from class probability, it is high to represent condensation risk
User behavior need to control closing radiation cold source and room temperature made to increase, increase it is then desired to carry out anti-condensation control
The gap of radiation temperature and dew-point temperature weakens dew condensation phenomenon.Certainly, in practice it is also possible that a kind of situation be, M ownership
In the class probability P of AA=50%, M belong to the class probability P of BB=50%, in the present solution, purpose is anti-condensation control,
In the case of identical for this probability, the behavior that goal behavior feature M default is classified as this high risk of B class behavior feature is special
Sign, the anti-condensation control of triggering following, to avoid dew condensation phenomenon as far as possible.Therefore, P in the examples described aboveA> PBWhen, without anti-
Condensation control, works as PA≤PBWhen, it can determine whether that goal behavior feature M is B class behavior feature user i.e. to be controlled according to class probability
Behavioural characteristic.
In conclusion a kind of condensation prevention control method provided by the invention, passes through slope variance to real-time humidity data collection
Goal behavior feature is inputted in default behavioural characteristic library, target line can be obtained by goal behavior feature therein in processing extraction
It is characterized the class probability for belonging to default behavioural characteristic, to learn user behavior representated by real-time humidity data collection, works as mesh
When mark behavioural characteristic is user behavior characteristics to be controlled, it is believed that the risk that user behavior generates condensation is larger, can instruct to control
Radiation cold source is closed to avoid traditional control method to trigger anti-condensation measure based on user behavior and cut dew condensation phenomenon
The poor problem of weak effect.
Embodiment two
Referring to Fig. 2, the present invention provides another condensation prevention control methods, which comprises
Step 201, history humidity data collection is obtained.
Specifically, can be before acquisition humidity data in real time, acquisition is collected in a manner of experiment has adequate representation
A large amount of history humidity data history of forming humidity data collection, described adequate representation i.e. refer to history humidity data collection this
A little data volumes are The more the better, and covering time range, the longer the better, including user behavior type it is more abundanter better, so as to going through
History data set carries out analytic learning to construct default behavioural characteristic library.
Step 202, the history humidity data collection is trained by behavioural characteristic analysis, establishes default behavioural characteristic
Library, wherein include a variety of different user behaviors in the default behavioural characteristic library, the user behavior is to cause indoor humidity
The behavior of variation.
Specifically, since the history humidity data got represents over the humidity number of experimentation a period of time
According to, might not occur in this time influence humidity variation user behavior, such as: 8 points of the morning at weekend to 10 points it
Between, 8 points to 9 points, user is in resting state, changes without influence on indoor humidity, 9 points to 10 points, user sweeps after getting up
Health, the behavior of mopping floor increase indoor humidity may for a long time.Behavioural characteristic point thus is passed through to history humidity data collection
Analysis, can be trained data, find out the data with similar characteristics in most evidences of comforming, and to data classification, establish default
Behavioural characteristic library, wherein include a variety of different user behaviors in default behavioural characteristic library, user behavior is to cause indoor humidity
The behavior of variation, i.e. every kind of behavioural characteristic represent different classes of user behavior.
Optionally, in another embodiment of the invention, above-mentioned steps 202 may include sub-step 2021 to sub-step
2023。
Sub-step 2021 executes the slope variance processing to the history humidity data collection and generates historical behavior data
Collection.
Specifically, history humidity data collection is the collected initial data of temperature controller, in order to guarantee the representative of sample
Property, sample data quantity is more, is formed by the curve that humidity changes over time, and the behavior of user is not significant enough, such as Fig. 3 institute
To show, abscissa meaning is the time of data acquisition, and ordinate meaning is humidity value, to be converted to humidity value convenient for diagram,
Ordinate institute's drawing data is, multiplied by 1000, curve a is history humidity data collection institute after relative humidity %RH one decimal of reservation
The curve that the humidity of formation changes over time.The data point with behavioural characteristic can be amplified by being handled by slope variance, mesh
Be by behavioural characteristic from history humidity data concentration extract to generate historical behavior data set, i.e. historical behavior data
Integrate the secondary data collection as history humidity data collection after feature extraction, as shown in figure 3, curve b is historical behavior data set
It is formed by the curve that humidity changes over time.It can refer to following mistakes by the process that slope variance handles progress feature extraction
Journey:
Assuming that original history humidity data integrates as experimentation continuous 24 hours at interval of 1 minute collected humidity number
According to.
Firstly, calculate the slope of each data point, since interval time difference is 1, then slope x (n)=y (n)-y (n-1),
Wherein, n=2,3,4 ... ... 1440, at the time of x (1)=0, n is that data acquire, n=2,3,4 ... ... i.e. the 2nd minute, the 3rd point
Clock, the 4th minute ..., y (n) then indicated that humidity value corresponding with the moment, i.e. y (2) indicate the 2nd minute humidity value, y (3) table
Show the 3rd minute humidity ... ....
Secondly, being squared to slope, slope square p (n)=x is obtained2(n)=[y (n)-y (n-1)]2, pass through a square meter
It calculates, increases gap data with can dramatically, so that behavioural characteristic displays.
Then, to each slope square data, it is slided four times to historical time dimension direction, tiltedly by obtain five
Rate square data accumulation obtains slope square and slides five cumulative and a (n)=p (n)+p (n-1)+p (n-2)+p (n-3)+p (n-
4), wherein a (1)=0, a (2)=0, a (3)=0, a (4)=0.It is carried out by the cumulative of five data and with a preset threshold
Compare, will be more than five cumulative data concentrated with data as secondary data of slope square sliding of the threshold value, to realize
The extraction of behavioural characteristic.
It is understood that when above-mentioned slope variance processing carries out acquiring the interval of data in the reference process of feature extraction
Between, slope square sliding number be exemplary description, can be selected according to factors such as scale, the computational efficiencies of sample size
Select suitable parameter.
Sub-step 2022 executes clustering to the historical behavior data set and generates each aggregate of data, each number
Different user behaviors is represented according to cluster.
Specifically, the above-mentioned slope variance processing progress obtained historical behavior data set of feature extraction that passes through is generation
The data of table particular user behavior, however do not learn which data is similar one kind, it is common by artificial intelligence field
K-means clustering algorithm, clustering can be carried out to historical behavior data set, generate multiple aggregates of data, of aggregate of data
The number i.e. number of classification, that is, the classification number of behavioural characteristic, then each aggregate of data represents different user behaviors.For example,
Fig. 4 gives the schematic diagram of the result by the classification of k-means clustering algorithm, and abscissa indicates humidity duration, ordinate
Indicate humidity value, " " indicates cluster 1, i.e., Cluster 1 in legend in Fig. 4, indicates to influence indoor humidity more faint
User behavior, " ▲ " indicate cluster 2, i.e., Cluster 2 in legend, indicate to influence heavier user behavior to indoor humidity,
In, "×" indicates the central point of each cluster.
Sub-step 2023 adds behavior label for each aggregate of data and establishes default behavioural characteristic library.
Specifically, convenient for the pipe administrative analysis of data, can be sorted to be distinguished different aggregates of data
Aggregate of data subscripts label, is used to indicate behavioural characteristic, for example indicate cluster 1 with label C luster 1, representing influences indoor humidity
More faint user behavior indicates cluster 2 with label C luster 2, and representing influences heavier user behavior to indoor humidity
The corresponding relationship for presetting in behavioural characteristic library and describing aggregate of data and label, that is, determined aggregate of data, that is, may know that the data
Which kind of user behavior cluster represents.
Step 203, real-time humidity data collection is obtained.
Specifically, air-conditioning equipment runs heating or refrigeration, therefore, spoke with a certain steady temperature due under normal conditions
It is basicly stable constant to penetrate temperature, and room atmosphere pressure will not change, will lead to the raising of dew-point temperature when humidity increases,
In contrast, i.e., the gap between radiation temperature and dew-point temperature reduces, and the two is further close, and condensation risk will increase, thus
Indoor humidity data is acquired as real-time humidity data collection by the temperature controller in equipment in real time in the method, for example, with 8 points
To 10 points at interval of 1 minute humidity data as real-time humidity data collection, based on humidity data for predicting user behavior,
Carry out anti-condensation control.
Step 204, the relation curve that humidity changes over time is drawn according to the real-time humidity data collection.
Specifically, using the time as horizontal axis, using humidity value as the longitudinal axis, can lead to for the real-time humidity data collection got
It crosses the relation curve changed over time similar to humidity shown in Fig. 3 curve a and intuitively shows humidity substantially situation of change.
Step 205, the slope of each data point of real-time humidity data collection described in the relation curve is calculated.
Specifically, it is each to can refer to the calculating acquisition real-time humidity data collection of slope calculating process described in sub-step 2021
The slope of data point.
Step 206, the slope square of each data point is calculated according to the slope.
Real-time humidity data is obtained specifically, can refer to slope square calculating process described in sub-step 2021 and calculate
Collect the slope square of each data point.
Step 207, to the slope square slide default item number it is cumulative obtain slope square it is cumulative and.
Specifically, can refer to, slope square described in sub-step 2021 is cumulative and calculating process calculates and obtains real-time wet
Spend each data point of data set slope square it is cumulative and.
Step 208, the cumulative and data greater than preset threshold of the slope square are screened and form goal behavior data set,
In, the goal behavior data set characterizes the goal behavior feature.
It will be more than the threshold value specifically, can be cumulative by the multiple slopes square being calculated and compared with preset threshold
Slope square sliding it is cumulative and data are used as the data in goal behavior data set, goal behavior data set characterization goal behavior
Feature, thus the extraction to behavioural characteristic is realized.
Step 209, the goal behavior feature is inputted into default behavioural characteristic library, calculates the goal behavior feature and belongs to
The class probability of default behavioural characteristic.
Specifically, by presetting in behavioural characteristic library as based on a variety of different of the summarized classification of a large amount of historical datas
Goal behavior feature is inputted in default behavioural characteristic library, that is, is exactly by goal behavior data set and each data by user behavior
Cluster is compared, and belongs to the probability of which aggregate of data by specific data individual in calculating goal behavior data set to calculate mesh
Mark behavioural characteristic belongs to the class probability of default behavioural characteristic, so that it is determined that the classification of goal behavior feature.For example, can be by
Class probability is calculated in logistic recurrence, it is assumed that a certain data point of goal behavior data set is C, with cluster 1, cluster
The Euclidean distance of 2 central point is respectively X1、X2, then C belongs to cluster 1, the class probability of cluster 2 is respectively P1=X2/(X1+
X2)、P2=X1/(X1+X2).To pass through P1、P2Size relation, it is believed that it is larger that data point C belongs to class probability numerical value
That classification, when probability is identical, default be classified as the big one kind of condensation risk.
Step 210, when the goal behavior feature is user behavior characteristics to be controlled, radiation cold source is closed in control,
In, the user behavior characteristics to be controlled are the behavioural characteristic that the condensation risk that represents is high in the default behavioural characteristic.
Specifically, according to aforementioned step can determine real-time humidity data collection representated by belonging to goal behavior feature
Specific category, however, be not that each type requires to carry out anti-condensation control, such as: A class behavior feature is represented to room
Interior humidity influences more faint user behavior, for example, the windowing behavior etc. of humid day moment, according to experimental monitoring, this kind of row
The condensation risk for being characterized the user behavior of representative is low, is not necessarily to anti-condensation control；B class behavior feature, which represents, influences indoor humidity
Heavier user behavior, for example, the windowing behavior of humid day, heating up water, blocked shot, mopping floor, according to experimental monitoring, this class behavior
The condensation risk for the user behavior that feature represents is higher than A class；Obviously, when the behavior for analyzing user according to goal behavior feature M
It is when belonging to A class behavior feature, condensation risk is substantially not present, and carries out anti-condensation control without sacrificing user's comfort.Example
Such as, existing a kind of it may be the case that M belongs to the class probability P of A in practiceA=60%, M belong to the class probability P of BB=
40%, it was determined that goal behavior feature M belongs to A class behavior feature, the user row low for this kind of condensation risk
To be not necessarily to anti-condensation control.Existing another kind is it may be the case that M belongs to the class probability P of A in practiceA=45%, M return
Belong to the class probability P of BB=55%, it was determined that goal behavior feature M belongs to B class behavior feature, i.e., current mesh
User behavior representated by behavioural characteristic M is marked, judges that it belongs to the high B class of condensation risk from class probability, represents condensation wind
The high user behavior in danger needs to control closing radiation cold source and room temperature is made to increase it is then desired to carry out anti-condensation control,
The gap of Enhanced Radiation Reduced Blast temperature and dew-point temperature weakens dew condensation phenomenon.Certainly, in practice it is also possible that a kind of situation be M
Belong to the class probability P of AA=50%, M belong to the class probability P of BB=50%, in the present solution, purpose is anti-condensation control
Goal behavior feature M default in the case of identical for this probability, is classified as the behavior of this high risk of B class behavior feature by system
Feature, the anti-condensation control of triggering following, to avoid dew condensation phenomenon as far as possible.Therefore, P in the examples described aboveA> PBWhen, it is not necessarily to
Anti-condensation control, works as PA≤PBWhen, it can determine whether that goal behavior feature M is B class behavior feature use i.e. to be controlled according to class probability
Family behavioural characteristic.
In conclusion a kind of condensation prevention control method provided by the invention, based on the clustering algorithm of artificial intelligence to history
Humidity data carries out clustering, finds the similarity rules between data, and analysis forms different user behavior classifications and establishes in advance
If behavioural characteristic library, to real-time humidity data collection by goal behavior feature therein in the processing extraction of slope variance, by target
Behavioural characteristic inputs in default behavioural characteristic library, and the class probability that goal behavior feature belongs to default behavioural characteristic can be obtained, from
And learn user behavior representated by real-time humidity data collection, it, can when goal behavior feature is user behavior characteristics to be controlled
Think that the risk of user behavior generation condensation is larger, control can be instructed to close radiation cold source, to be triggered based on user behavior anti-
The measure of condensation avoids traditional control method and weakens the poor problem of effect to dew condensation phenomenon.
Embodiment three
Referring to Fig. 5, the present invention provides a kind of anti-condensation control device, described device includes:
Real time data acquisition module 501, for obtaining real-time humidity data collection.
Extraction module 502, it is special for executing slope variance processing extraction goal behavior for the real-time humidity data collection
Sign.
Computing module 503 calculates the goal behavior feature for the behavioural characteristic to be inputted default behavioural characteristic library
Belong to the class probability of default behavioural characteristic.
Control module 504, for when the goal behavior feature is user behavior characteristics to be controlled, radiation to be closed in control
Cold source, wherein the user behavior characteristics to be controlled are that the behavior that the condensation risk that represents is high in the default behavioural characteristic is special
Sign.
In conclusion a kind of condensation prevention control method provided by the invention, passes through slope variance to real-time humidity data collection
Goal behavior feature is inputted in default behavioural characteristic library, target line can be obtained by goal behavior feature therein in processing extraction
It is characterized the class probability for belonging to default behavioural characteristic, to learn user behavior representated by real-time humidity data collection, works as mesh
When mark behavioural characteristic is user behavior characteristics to be controlled, it is believed that the risk that user behavior generates condensation is larger, can instruct to control
Radiation cold source is closed to avoid traditional control method to trigger anti-condensation measure based on user behavior and cut dew condensation phenomenon
The poor problem of weak effect.
Example IV
Referring to Fig. 6, the present invention provides a kind of anti-condensation control device, described device includes:
Historical data obtains module 601, for obtaining history humidity data collection.
Training module 602 is established default for being trained by behavioural characteristic analysis to the history humidity data collection
Behavioural characteristic library, wherein include a variety of different user behaviors in the default behavioural characteristic library, the user behavior is to cause
The behavior of indoor humidity variation.
The training module 602 includes:
Data set generation submodule 6021 is generated for executing the slope variance processing to the history humidity data collection
Historical behavior data set.
Aggregate of data generates submodule 6022, generates each data for executing clustering to the historical behavior data set
Cluster, each aggregate of data represent different user behaviors.
Setting up submodule 6023 establishes default behavioural characteristic library for adding behavior label for each aggregate of data.
Real time data acquisition module 603, for obtaining real-time humidity data collection.
Extraction module 604, it is special for executing slope variance processing extraction goal behavior for the real-time humidity data collection
Sign.
The extraction module 604 includes:
Rendering submodule 6041, it is bent for drawing the relationship that humidity changes over time according to the real-time humidity data collection
Line；
Slope computational submodule 6042, for calculating each data point of real-time humidity data collection described in the relation curve
Slope；
Slope square computational submodule 6043, for calculating the slope square of each data point according to the slope；
Cumulative submodule 6044, for the slope square slide default item number it is cumulative obtain slope square it is cumulative and；
Submodule 6045 is screened, forms target line for screening the cumulative and data greater than preset threshold of the slope square
For data set, wherein the goal behavior data set characterizes the goal behavior feature.
Computing module 605 calculates the goal behavior for the goal behavior feature to be inputted default behavioural characteristic library
Feature belongs to the class probability of default behavioural characteristic；
Control module 606, for when the goal behavior feature is user behavior characteristics to be controlled, radiation to be closed in control
Cold source, wherein the user behavior characteristics to be controlled are that the behavior that the condensation risk that represents is high in the default behavioural characteristic is special
Sign.In conclusion a kind of condensation prevention control method provided by the invention, based on the clustering algorithm of artificial intelligence to history humidity number
According to clustering is carried out, the similarity rules between data are found, analysis forms different user behavior classifications and establishes default behavior
Feature database, to real-time humidity data collection by goal behavior feature therein in the processing extraction of slope variance, by goal behavior spy
Sign inputs in default behavioural characteristic library, and goal behavior feature can be obtained and belong to the class probability of default behavioural characteristic, to learn
User behavior representated by real-time humidity data collection, when goal behavior feature is user behavior characteristics to be controlled, it is believed that use
The risk that family behavior generates condensation is larger, control can be instructed to close radiation cold source, to trigger based on user behavior anti-condensation
Measure avoids traditional control method and weakens the poor problem of effect to dew condensation phenomenon.
It is apparent to those skilled in the art that for convenience and simplicity of description, foregoing description it is anti-condensation
Reveal control device specific work process, the corresponding process in the embodiment of aforementioned condensation prevention control method can be referred to, herein not
It repeats again.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, protection scope of the present invention should be subject to the protection scope in claims.
Claims (8)
1. a kind of condensation prevention control method, which is characterized in that the described method includes:
Obtain real-time humidity data collection；
The processing of slope variance, which is executed, for the real-time humidity data collection extracts goal behavior feature；
The goal behavior feature is inputted into default behavioural characteristic library, the goal behavior feature is calculated and belongs to default behavioural characteristic
Class probability；
When the goal behavior feature is user behavior characteristics to be controlled, radiation cold source is closed in control, wherein described to be controlled
User behavior characteristics are the behavioural characteristic that the condensation risk that represents is high in the default behavioural characteristic.
2. condensation prevention control method according to claim 1, which is characterized in that described to be directed to the real-time humidity data collection
It executes the processing of slope variance and extracts goal behavior feature, comprising:
The relation curve that humidity changes over time is drawn according to the real-time humidity data collection；
Calculate the slope of each data point of real-time humidity data collection described in the relation curve；
The slope square of each data point is calculated according to the slope；
To the slope square slide default item number it is cumulative obtain slope square it is cumulative and；
It screens the cumulative and data greater than preset threshold of the slope square and forms goal behavior data set, wherein the target
Behavioral data collection characterizes the goal behavior feature.
3. condensation prevention control method according to claim 1, which is characterized in that the acquisition real-time humidity data collection it
Before further include:
Obtain history humidity data collection；
The history humidity data collection is trained by behavioural characteristic analysis, establishes default behavioural characteristic library, wherein described
It include a variety of different user behaviors in default behavioural characteristic library, the user behavior is the behavior for causing indoor humidity to change.
4. condensation prevention control method according to claim 3, which is characterized in that described to be analyzed by behavioural characteristic to described
History humidity data collection is trained, and is established default behavioural characteristic library and is included:
The slope variance processing is executed to the history humidity data collection and generates historical behavior data set；
Clustering is executed to the historical behavior data set and generates each aggregate of data, each aggregate of data represents different use
Family behavior；
Behavior label, which is added, for each aggregate of data establishes default behavioural characteristic library.
5. a kind of anti-condensation control device, which is characterized in that described device includes:
Real time data acquisition module, for obtaining real-time humidity data collection；
Extraction module extracts goal behavior feature for executing the processing of slope variance for the real-time humidity data collection；
Computing module calculates the goal behavior feature category for the goal behavior feature to be inputted default behavioural characteristic library
In the class probability of default behavioural characteristic；
Control module, for controlling and closing radiation cold source when the goal behavior feature is user behavior characteristics to be controlled,
In, the user behavior characteristics to be controlled are the behavioural characteristic that the condensation risk that represents is high in the default behavioural characteristic.
6. anti-condensation control device according to claim 5, which is characterized in that the extraction module includes:
Rendering submodule, for drawing the relation curve that humidity changes over time according to the real-time humidity data collection；
Slope computational submodule, the slope of each data point for calculating real-time humidity data collection described in the relation curve；
Slope square computational submodule, for calculating the slope square of each data point according to the slope；
Cumulative submodule, for the slope square slide default item number it is cumulative obtain slope square it is cumulative and；
Submodule is screened, forms goal behavior data for screening the cumulative and data greater than preset threshold of the slope square
Collection, wherein the goal behavior data set characterizes the goal behavior feature.
7. anti-condensation control device according to claim 5, which is characterized in that described device further include:
Historical data obtains module, for obtaining history humidity data collection；
It is special to establish default behavior for being trained by behavioural characteristic analysis to the history humidity data collection for training module
Levy library, wherein include a variety of different user behaviors in the default behavioural characteristic library, the user behavior is to cause interior wet
Spend the behavior of variation.
8. anti-condensation control device according to claim 7, which is characterized in that the training module includes:
Data set generation submodule generates historical behavior for executing the slope variance processing to the history humidity data collection
Data set；
Aggregate of data generates submodule, generates each aggregate of data for executing clustering to the historical behavior data set, described
Each aggregate of data represents different user behaviors；
Setting up submodule establishes default behavioural characteristic library for adding behavior label for each aggregate of data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910094295.6A CN110017567B (en) | 2019-01-30 | 2019-01-30 | Anti-condensation control method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910094295.6A CN110017567B (en) | 2019-01-30 | 2019-01-30 | Anti-condensation control method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110017567A true CN110017567A (en) | 2019-07-16 |
| CN110017567B CN110017567B (en) | 2021-03-02 |
Family
ID=67188953
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910094295.6A Expired - Fee Related CN110017567B (en) | 2019-01-30 | 2019-01-30 | Anti-condensation control method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110017567B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102283283A (en) * | 2011-05-30 | 2011-12-21 | 广西大学 | Intelligentized superficial geothermal energy low-temperature grain storage system |
| CN102721117A (en) * | 2012-07-09 | 2012-10-10 | 广西电力工业勘察设计研究院 | Special air conditioner and dehumidifier for variable-air supply dew-point temperature-controlled hydropower station, and control method |
| JP2013047583A (en) * | 2011-08-29 | 2013-03-07 | Mitsubishi Electric Corp | Air conditioner |
| CN104076850A (en) * | 2014-07-21 | 2014-10-01 | 国家电网公司 | Method for achieving load switching based on single temperature and humidity sensor in control cubicle |
| CN106765744A (en) * | 2016-11-28 | 2017-05-31 | 天津商业大学 | It is determined that the method that radiation cooling system radiant panel surface temperature changes when lowering the temperature that supplies water |
| CN108917089A (en) * | 2018-04-18 | 2018-11-30 | 青岛海尔空调器有限总公司 | A kind of anti-condensation air conditioning control method and device |
-
2019
- 2019-01-30 CN CN201910094295.6A patent/CN110017567B/en not_active Expired - Fee Related
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102283283A (en) * | 2011-05-30 | 2011-12-21 | 广西大学 | Intelligentized superficial geothermal energy low-temperature grain storage system |
| JP2013047583A (en) * | 2011-08-29 | 2013-03-07 | Mitsubishi Electric Corp | Air conditioner |
| CN102721117A (en) * | 2012-07-09 | 2012-10-10 | 广西电力工业勘察设计研究院 | Special air conditioner and dehumidifier for variable-air supply dew-point temperature-controlled hydropower station, and control method |
| CN104076850A (en) * | 2014-07-21 | 2014-10-01 | 国家电网公司 | Method for achieving load switching based on single temperature and humidity sensor in control cubicle |
| CN106765744A (en) * | 2016-11-28 | 2017-05-31 | 天津商业大学 | It is determined that the method that radiation cooling system radiant panel surface temperature changes when lowering the temperature that supplies water |
| CN108917089A (en) * | 2018-04-18 | 2018-11-30 | 青岛海尔空调器有限总公司 | A kind of anti-condensation air conditioning control method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110017567B (en) | 2021-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11385611B2 (en) | Advanced identification and classification of sensors and other points in a building automation system | |
| US20090236432A1 (en) | Retrofitting a constant volume air handling unit with a variable frequency drive | |
| CN106196423A (en) | A kind of Indoor Environmental Quality based on model prediction controls optimization method | |
| CN107120794B (en) | Air conditioner operation condition adjusting method and air conditioner | |
| KR102170522B1 (en) | Energy management system for exhibition hall considering environment and occupancy changes | |
| CN113685962B (en) | Machine room temperature efficient control method and system based on correlation analysis | |
| Vanus et al. | New method for accurate prediction of CO2 in the Smart Home | |
| CN114997534A (en) | Similar rainfall forecasting method and equipment based on visual features | |
| CN111221249B (en) | Parameter optimization method for data center system group control | |
| CN112413820A (en) | Air conditioner defrosting method and device, air conditioning equipment and computer readable storage medium | |
| CN116089885A (en) | Unknown fault type diagnosis method based on random depth forest | |
| CN110017567A (en) | A kind of condensation prevention control method and device | |
| CN112329813B (en) | Energy consumption prediction feature extraction method and system | |
| CN113836808A (en) | PM2.5 deep learning prediction method based on heavy pollution feature constraint | |
| CN116109011B (en) | Energy consumption management method and terminal for intelligent park | |
| CN111242266A (en) | Operation data management system | |
| CN111896831A (en) | Non-invasive comprehensive energy load monitoring method and system | |
| CN117029236A (en) | Clean room environment control method and system | |
| CN103574856B (en) | Airconditioning control solution display unit and method | |
| Wolf et al. | CO2-based grey-box model to estimate airflow rate and room occupancy | |
| CN117010638B (en) | Intelligent management method and system for hotel equipment | |
| JP2007072991A (en) | Environment evaluation system and environment evaluation method | |
| CN114237072B (en) | Intelligent household door lock integrated control system and method applying cloud control technology | |
| CN110986318A (en) | Air conditioner control method and device based on temperature buffering and air conditioner | |
| CN116909339B (en) | Intelligent household safety early warning method and system based on artificial intelligence |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210302 Termination date: 20220130 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |