CN110411608A - A kind of modification method and system of list firewire switch temperature measurement error - Google Patents
A kind of modification method and system of list firewire switch temperature measurement error Download PDFInfo
- Publication number
- CN110411608A CN110411608A CN201910668601.2A CN201910668601A CN110411608A CN 110411608 A CN110411608 A CN 110411608A CN 201910668601 A CN201910668601 A CN 201910668601A CN 110411608 A CN110411608 A CN 110411608A
- Authority
- CN
- China
- Prior art keywords
- straight line
- under
- time point
- section
- fitting
- 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
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K15/00—Testing or calibrating of thermometers
- G01K15/005—Calibration
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
- Fire-Detection Mechanisms (AREA)
Abstract
The invention discloses the modification methods and system of a kind of single firewire switch temperature measurement error, belong to smart home person's electro-technical field, comprising: in the case where room temperature is kept constant, obtain history data set；The data under sectional straight line fitting and off-position are carried out to the data under energized state using history data set and carry out piecewise fitting, determine heating split time point and cooling split time point；In active user's first time on/off electric process, according to heating split time point and cooling split time point, segmentation calculates the approximate regression coefficient that the approximate regression coefficient of each straight line and segmentation under energized state calculate each straight line under off-position；According to heating split time point, cooling split time point, under energized state under the approximate regression coefficient of each section of straight line and off-position each section of straight line approximate regression coefficient, correct single firewire and switch temperature measurement error.The present invention substantive can solve the problems, such as measurement temperature error caused by single firewire switch integrated circuit plate fever.
Description
Technical field
The present invention relates to smart home person's electro-technical field, in particular to a kind of amendment side of single firewire switch temperature measurement error
Method and system.
Background technique
The integrated circuit board of single firewire switch can generate heat in energization, since switch is directed to different use
Family family, size of current is different, and the heat of generation is also just different.For different user, in galvanization, equipment is to reality
The temperature error of measurement is different.
Currently, single firewire thermometric panel separates temperature sensor and integrated circuit board separation, to reduce integrated circuit board
Fever is influenced caused by temperature sensor, has the disadvantage that this single firewire thermometric panel only in structure
It improves, does not account for the influence that integrated circuit board electrical current component itself generates heat to thermometric, solve to rise there is no substantive
The influence of temperature.
Summary of the invention
It is an object of the invention to solve the problems, such as that above-mentioned background technique exists, to be automatically corrected to temperature error.
In order to achieve the above object, a kind of modification method of the present invention using single firewire switch temperature measurement error, including walk as follows
It is rapid:
In the case where room temperature is kept constant, obtain history data set, the history data set include certain amount user open/
Any point-in-time and its corresponding temperature difference during turning off the light；
Using history data set to the data under energized state carry out the data under sectional straight line fitting and off-position into
Row sectional straight line fitting determines heating split time point and cooling split time point；
In active user's first time on/off electric process, according to heating split time point and cooling split time point, segmentation
Calculate the approximate regression coefficient that the approximate regression coefficient of each straight line and segmentation under energized state calculate each straight line under off-position；
According to the approximate regression coefficient of each straight line under the heating split time point, cooling split time point, energized state
With the approximate regression coefficient of straight line each under off-position, corrects single firewire and switch temperature measurement error.
Further, described that at least two sections of straight line fittings and right are carried out to the data under energized state using history data set
Data under off-position carry out at least two sections of straight line fittings.
Further, described that sectional straight line fitting and power-off shape are carried out to the data under energized state using history data set
Data under state carry out sectional straight line fitting, determine heating split time point and cooling split time point, comprising:
It concentrates the data under energized state to be at least divided into two continuous time periods the historical data, utilizes each time
Data in section obtain the regression coefficient under every section of fitting a straight line to sectional straight line fitting under energized state, then calculate segmentation
Interior all data in addition to every section of starting point to every section of starting point slope；In every section of fitting a straight line, by every section of fitting a straight line
Under regression coefficient and the immediate coordinate points of this section of straight slope corresponding at the time of as this section of straight segments calculate when
Between point, the split time point of every section of fitting a straight line function constitutes the heating split time point；
Data under the history data set power cutoff state are at least divided into two continuous time periods, utilize each time
Data in section obtain the regression coefficient under every section of fitting a straight line to sectional straight line fitting under off-position, then calculate segmentation
Interior all data in addition to every section of starting point to every section of starting point slope；In every section of fitting a straight line, every section is fitted
At the time of corresponding to regression coefficient under straight line and the immediate coordinate points of straight slope as this section of straight segments calculate when
Between point, the split time point of every section of fitting a straight line function constitutes the heating split time point.
Further, described in active user's first time on/off electric process, according to heating split time point and cooling minute
Section time point, segmentation calculate energized state under each straight line approximate regression coefficient and segmentation calculate off-position under each straight line it is close
Like regression coefficient, comprising:
In active user's first time on/off electric process, at the time of obtaining identical as the heating split time point
The moment is demarcated as heating, and records the heating calibration moment corresponding temperature difference；Moment and its corresponding is demarcated using the heating
The temperature difference calculates each straight line coefficient, the approximate recurrence as each section of fitting a straight line under energized state to segmentation under the energized state
Coefficient；
In active user's first time on/off electric process, at the time of obtaining identical as the cooling split time point
The moment is demarcated as cooling, and records the cooling calibration moment corresponding temperature difference；Moment and its corresponding is demarcated using the cooling
The temperature difference calculates each straight line coefficient, the approximate recurrence as each section of fitting a straight line under off-position to segmentation under the off-position
Coefficient.
Further, described according to each straight line under the heating split time point, cooling split time point, energized state
The approximate regression coefficient of each straight line under approximate regression coefficient and off-position corrects single firewire and switchs temperature measurement error, comprising:
Single firewire switchs revised temperature △ Y=Y-b1*t1-b2*t2-b3*t3+b4*t1 '+b5*t2 '+b6*t3 ',
Wherein Y indicates the moment temperature sensor value that is powered, and t1, t2, t3 are respectively the duration of each period under energized state, t1 ',
T2 ', t3 ' are respectively the duration of each period under off-position, and b1, b2, b3 are respectively the segmented linear under energized state
Approximate regression coefficient, b4, b5, b6 are respectively the approximate regression coefficient of the segmented linear under off-position, and * indicates product.
On the other hand, using a kind of update the system of single firewire switch temperature measurement error, including history data set obtain module,
Split time point determining module, regression coefficient computing module and temperature measurement error correction module；
History data set obtains module and is used in the case where room temperature is kept constant, and obtains history data set, the history number
It include any point-in-time and its corresponding temperature difference during certain amount user ON/OFF lamp according to collection；
Split time point determining module is used to carry out segmented linear to the data under energized state using history data set quasi-
It closes and carries out sectional straight line fitting with the data under off-position, determine heating split time point and the split time point that cools down；
Regression coefficient computing module is used in active user's first time on/off electric process, according to heating split time point
With cooling split time point, it is segmented the approximate regression coefficient of each straight line and segmentation under calculating energized state and calculates under off-position respectively
The approximate regression coefficient of straight line；
Temperature measurement error correction module is used for according under the heating split time point, cooling split time point, energized state
The approximate regression coefficient of each straight line under the approximate regression coefficient and off-position of each straight line corrects single firewire and switchs temperature measurement error.
Further, described that at least two sections of straight line fittings and right are carried out to the data under energized state using history data set
Data under off-position carry out at least two sections of straight line fittings.
Further, when the split time point determining module includes heating split time point determination unit and cooling segmentation
Between put determination unit；
Heating split time point determination unit is used to concentrate the data under energized state to be at least divided into the historical data
Two continuous time periods, it is quasi- to segmented linear under energized state using all data in each period in addition to every section of starting point
It closes, obtains the regression coefficient under every section of fitting a straight line, then calculate all data of segmentation to the slope of every section of starting point；Every
In section fitting a straight line, when by corresponding to the regression coefficient and the immediate coordinate points of this section of straight slope under every section of fitting a straight line
It carves the time point calculated as this section of straight segments, when the split time point of every section of fitting a straight line function constitutes heating segmentation
Between point；
Data under the history data set power cutoff state for being at least divided by cooling split time point determination unit
It is straight to obtain every section of fitting using the data in each period to sectional straight line fitting under off-position for two continuous time periods
Then regression coefficient under line calculates the slope of all data in segmentation in addition to every section of starting point to every section of starting point；
In every section of fitting a straight line, when by corresponding to the regression coefficient and the immediate coordinate points of straight slope under every section of fitting a straight line
It carves the time point calculated as this section of straight segments, when the split time point of every section of fitting a straight line function constitutes heating segmentation
Between point.
Further, the regression coefficient computing module includes the first regression coefficient computing unit and the second regression coefficient meter
Calculate unit；
First regression coefficient computing unit is used in active user's first time on/off electric process, is obtained and the heating
As the heating calibration moment at the time of split time point is identical, and record the heating calibration moment corresponding temperature difference；Utilize the liter
Temperature scale timing is carved and its corresponding temperature difference calculates each straight line coefficient to segmentation under the energized state, and approximation is as under energized state
Each section of fitting a straight line regression coefficient；
Second regression coefficient computing unit is used in active user's first time on/off electric process, is obtained and the cooling
As the cooling calibration moment at the time of split time point is identical, and record the cooling calibration moment corresponding temperature difference；Utilize the drop
Temperature scale timing is carved and its corresponding temperature difference calculates each straight line coefficient to segmentation under the off-position, and approximation is as under off-position
Each section of fitting a straight line regression coefficient.
Further, the temperature measurement error correction module is specifically used for:
Single firewire switchs revised temperature △ Y=Y-b1*t1-b2*t2-b3*t3+b4*t1 '+b5*t2 '+b6*t3 ',
Wherein Y indicates the moment temperature sensor value that is powered, and t1, t2, t3 are respectively the duration of each period under energized state, t1 ',
T2 ', t3 ' are respectively the duration of each period under off-position, and b1, b2, b3 are respectively that the segmentation under energized state calculates directly
The approximate regression coefficient of line, b4, b5, b6 are respectively the approximate regression coefficient of the segmentation calculating straight line under off-position, and * expression multiplies
Product.
Compared with prior art, there are following technical effects by the present invention: when single firewire thermometric panel is installed and used, installation
Environment and electrical appliance power be substantially it is constant, i.e., user is switched using single firewire every time, the temperature rise actually caused be it is fixed,
When user uses single firewire panel for the first time, using Principle of Statistics, single firewire of counting user switchs heating, cooling data,
Single firewire thermometric panel is applied to using these data approximate calculation regression coefficients, and by approximate regression coefficient.It is secondary in user
When using single firewire thermometric panel, the temperature of heating, the i.e. actual environment of user are subtracted according to the practical temperature obtained of sensor
The value complement actually to cool down when user turns off the light, according to law of drop temperature, is repaid sensor by temperature, is corrected single firewire switch and is closed benefit
Temperature is repaid, temperature rise caused by integrated circuit board is generated heat eliminates, and reaches substantive and solves single firewire switch integrated circuit plate
Temperature error problem is measured caused by fever.
Detailed description of the invention
With reference to the accompanying drawing, specific embodiments of the present invention will be described in detail:
Fig. 1 is a kind of flow diagram of the modification method of single firewire switch temperature measurement error；
Fig. 2 is the data fitting schematic diagram for not being segmented thermometric correction model；
Fig. 3 is the data fitting schematic diagram for being segmented thermometric correction model；
Fig. 4 is a kind of structural schematic diagram of the update the system of single firewire switch temperature measurement error.
Specific embodiment
In order to further explain feature of the invention, reference should be made to the following detailed description and accompanying drawings of the present invention.Institute
Attached drawing is only for reference and purposes of discussion, is not used to limit protection scope of the present invention.
As shown in Figure 1, including the following steps present embodiment discloses a kind of modification method of single firewire switch temperature measurement error
S1-S4:
S1, in the case where room temperature is kept constant, obtain history data set, which includes certain amount user
Any point-in-time and its corresponding temperature difference during ON/OFF lamp；
S2, using history data set the data under energized state are carried out with data under sectional straight line fitting and off-position
Sectional straight line fitting is carried out, determines heating split time point and cooling split time point；
S3, in active user's first time on/off electric process, according to heating split time point and cooling split time point,
Segmentation calculates the approximate regression system that the approximate regression coefficient of each straight line and segmentation under energized state calculate each straight line under off-position
Number；
S4, according to the heating split time point, cooling split time point, under energized state each straight line approximate regression system
The approximate regression coefficient of each straight line under several and off-position corrects single firewire and switchs temperature measurement error.
It should be noted that since single firewire switch is mainly used for ordinary family user, under experimental conditions, electrical appliance pair
The temperature rise that equipment generates, reached heat dissipation equilibrium state at n minutes, it is assumed that in the case that room temperature is kept constant, in user's energization feelings
Under condition, pass through time data, that is, any point-in-time x of backstage counting useriWith temperature change value yi, constitute history data set.Benefit
Sectional straight line fitting function is fitted to obtain split time point with the data that historical data is concentrated, a certain user is acquired and divides
At the time of section time point corresponds to and its temperature difference is answered to learn approximate regression coefficient using split time point and approximate regression coefficient
Use in single firewire thermometric panel, can warming temperature using being powered as power blackout situation under the compensation temperature that cools down
Degree, thoroughly solve integrated circuit board fever influences caused by temperature sensor.
Further, sectional straight line fitting above-mentioned steps S2: is carried out to the data under energized state using history data set
Sectional straight line fitting is carried out with the data under off-position, determines heating split time point and cooling split time point, including such as
Lower subdivided step S21-S22:
S21, the data under energized state are concentrated at least to be divided into two continuous time periods the historical data, using each
Data in period obtain the regression coefficient under every section of fitting a straight line, then calculate to sectional straight line fitting under energized state
It is segmented the slope of interior all data in addition to every section of starting point to every section of starting point；In every section of fitting a straight line, every section is fitted
It is calculated at the time of corresponding to regression coefficient and the immediate coordinate points of this section of straight slope under straight line as this section of straight segments
Time point, the split time point of every section of fitting a straight line function constitutes the heating split time point；
S22, the data under the history data set power cutoff state are at least divided into two continuous time periods, using each
Data in period obtain the regression coefficient under every section of fitting a straight line, then calculate to sectional straight line fitting under off-position
It is segmented the slope of interior all data in addition to every section of starting point to every section of starting point；It is quasi- by every section in every section of fitting a straight line
It is calculated at the time of conjunction corresponding to regression coefficient and the immediate coordinate points of straight slope under straight line as this section of straight segments
The split time point at time point, every section of fitting a straight line function constitutes the heating split time point.
It should be noted that as follows using the process that linear least square solves the regression coefficient of fitting a straight line:
(1) fitting a straight line y=a+b*x
(2) arbitrary coordinate point (xi, yi)//xiRepresent time point, yiRepresent the temperature difference
(3) error di=yi-(a+b*xi)
(4) whenWhen taking most minimum value, straight line fitting degree highest
(5)First-order partial derivative is asked to a, b respectively:
It enables:A, b are solved, is enabled
Then:
It is above exactly the process that linear least square solves regression coefficient, xiIndicate time point, yiIndicate temperature change
Value,X, the mean value of y are represented, * indicates product, and (q) * * 2 indicates square of q.
It should be noted that being converted into rectangular in form in Computer Program Implementation Process and being calculated:Y=[y1,y2,yj,…,ym], yjFor temperature change value, X=[[1, x1],[1,x2],[1,xi]…
[1,xn]], xiAny point-in-time, β=[[a], [b]], a, b are fitting one-variable linear regression coefficient, (XTX)-1Indicate XTX's is quasi-
Matrix, XTIndicate the transposition of X.
Further, the sectional straight line fitting under the sectional straight line fitting and the off-position under the energized state is equal
It is at least two sections of fitting a straight lines.It should be noted that the fitting effect ratio by using sectional straight line fitting data uses one section
The fitting effect of fitting a straight line will be got well, by taking three sections of fitting a straight line functions as an example:
As shown in Fig. 2, history data set is loaded if original data set fitting a straight line function is not segmented into y=a+b*x, benefit
It is derived with above-mentioned least square method and calculates regression coefficient a, b, the then loss function calculated
As shown in figure 3, original data set point 3 sections of progress data fittings, using the above method carry out calculating regression coefficient (a1,
B1), (a2, b2) (a3, b3), then the loss function calculated:
Comparison uses segment data fitting effect more preferably it is found that same data set Loss1 > > loss2.
Specifically, the present embodiment is fitted with point three sections of fitting a straight lines, the determination process of every section of split time point are as follows:
(1) be segmented first segment, calculate separately the slope (k1, k2, k3 ...) of each coordinate points and coordinate origin, find out and
The immediate k of b1 records current segmentation at the time of point n1 at the time of corresponding to k value.
(2) second segment being segmented, calculates separately the last one coordinate points of the coordinate points of second segment and the first segment of segmentation
Slope (w1, w2, w3 ...), find out with the immediate w of b2, record the time point n2 that w value is segmented at the time of correspond to as second.
(3) the third section being segmented, calculates separately the last one coordinate points of the coordinate points of third section and the second segment of segmentation
Slope (u1, u2, u3 ...), find out with the immediate u of b3, record the time point n3 that u value is segmented at the time of correspond to as third.
It should be understood that power process as galvanization, determines split time point n1, n2 of galvanization respectively
And split time point n4, n5 and n6 of n3 and power process.To calculate energized state according to determining split time point
Under segmented linear approximate regression coefficient and the segmented linear under off-position approximate regression coefficient.
Further, above-mentioned steps S3: in active user's first time on/off electric process, according to heating split time point
With cooling split time point, it is segmented the approximate regression coefficient of each straight line and segmentation under calculating energized state and calculates under off-position respectively
The approximate regression coefficient of straight line, including following subdivided step S31-S32:
S31, in active user's first time on/off electric process, obtain it is identical with the heating split time point
Moment records the heating calibration moment corresponding temperature difference as the heating calibration moment；Moment and its right is demarcated using the heating
The temperature difference answered calculates each straight line coefficient to segmentation under the energized state, and approximation is as each section of fitting a straight line under energized state
Regression coefficient.Specifically:
Obtain the heating calibration moment corresponding temperature difference:
(1) turn on light moment X, temperature Y are counted；
(2) statistical sectional time point n1, Xn1, Yn1 at the time of corresponding；
(3) statistical sectional time point n2, Xn2, Yn2 at the time of corresponding；
(4) statistical sectional time point n3, Xn3, Yn3 at the time of corresponding.
Calculate the regression coefficient of the sectional straight line fitting under energized state:
(1) turn on light first stage regression coefficient b1=(Yn1-Y)/(Xn1-X)；
(2) second stage of turning on light regression coefficient b2=(Yn2-Yn1)/(Xn2-Xn1)；
(3) turn on light phase III regression coefficient b3=(Yn3-Yn2)/(Xn3-Xn2).
S32, in active user's first time on/off electric process, obtain it is identical with the cooling split time point
Moment records the cooling calibration moment corresponding temperature difference as the cooling calibration moment；Moment and its right is demarcated using the cooling
The temperature difference answered calculates each straight line coefficient to segmentation under the off-position, and approximation is as each section of fitting a straight line under off-position
Regression coefficient.
Obtain the cooling calibration moment corresponding temperature difference:
(1) the time X ', temperature Y ' of close moment are counted；
(2) statistical sectional time point n4, X ' n4, Y ' n4 at the time of corresponding；
(3) statistical sectional time point n5, X ' n5, Y ' n5 at the time of corresponding；
(4) statistical sectional time point n6, X ' n6, Y ' n6 at the time of corresponding.
Calculate the regression coefficient of the sectional straight line fitting under off-position:
(1) Xo temperature Yo at the time of statistics lighting time is more than after 60 minutes
(2) turn off the light first stage regression coefficient b4=(Y ' n4-Yo)/(X ' n4-Xo)
(3) second stage of turning off the light regression coefficient b5=(Y ' n5-Y ' n4)/(X ' n5-X ' n4)
(4) turn off the light phase III regression coefficient b6=(Y ' n6-Y ' n5)/(X ' n6-X ' n5).
Further, above-mentioned steps S4: according under the heating split time point, cooling split time point, energized state
The approximate regression coefficient of each straight line under the approximate regression coefficient and off-position of each straight line corrects single firewire and switchs temperature measurement error,
It specifically includes:
Single firewire switchs revised temperature △ Y=Y-b1*t1-b2*t2-b3*t3+b4*t1 '+b5*t2 '+b6*t3 ',
Wherein Y indicates the moment temperature sensor value that is powered, and t1, t2, t3 are respectively the duration of each period under energized state, t1 ',
T2 ', t3 ' are respectively the duration of each period under off-position, and b1, b2, b3 are respectively that the segmentation under energized state calculates directly
The approximate regression coefficient of line, b4, b5, b6 are respectively the approximate regression coefficient of the segmentation calculating straight line under off-position, and * expression multiplies
Product.
It should be noted that during single firewire thermometric panel is installed and used, installation environment and electrical appliance power base
Originally be it is constant, i.e., user is switched using single firewire every time, the temperature rise actually caused be it is fixed, user is for the first time using single fire
Line panel learns the warming law of user, carrys out preservation model actual parameter, and user turns off the light for the first time, after study user turns off the light
Law of drop temperature.The equal secondary use of users, user open single firewire panel, subtract liter according to the practical temperature obtained of temperature sensor
The temperature of temperature, is the actual ambient temperature of user；When user turns off the light, according to law of drop temperature, the value complement actually to cool down is repaid
To sensor, it can thus accomplish that single firewire opens the purpose that compensation temperature is closed in amendment, integrated circuit board can thus be sent out
Temperature rise caused by heat is eliminated.
Particularly, if the installation environment of user and electrical appliance power change, whether just single-chip microcontroller judges subscriber's line circuit
Often, as powered off, it may be possible to installation environment changes or electrical appliance changes, the actual parameter value that at this moment original model is calculated,
It can be emptied, relearn heating and cooling rule, be truly realized hommization, intelligence.
As shown in figure 4, present embodiment discloses a kind of update the systems of single firewire switch temperature measurement error, comprising: history number
Module 10, split time point determining module 20, regression coefficient computing module 30 and temperature measurement error correction module 40 are obtained according to collection；
History data set obtains module 10 and is used in the case where room temperature is kept constant, and obtains history data set, the history
Data set includes any point-in-time and its corresponding temperature difference during certain amount user's ON/OFF lamp；
Split time point determining module 20 is used to carry out segmented linear to the data under energized state using history data set
Data under fitting and off-position carry out sectional straight line fitting, determine heating split time point and cooling split time point；
Regression coefficient computing module 30 is used in active user's first time on/off electric process, according to heating split time
Point and cooling split time point, segmentation calculate the approximate regression coefficient of each straight line and segmentation under energized state and calculate under off-position
The approximate regression coefficient of each straight line；
Temperature measurement error correction module 40 is used for according to the heating split time point, cooling split time point, energized state
Under the approximate regression coefficient of each straight line and the approximate regression coefficient of each straight line under off-position, correct single firewire switch thermometric and miss
Difference.
Further, the sectional straight line fitting under the sectional straight line fitting and the off-position under the energized state is equal
It is at least two sections of fitting a straight lines.It should be noted that theoretically number of fragments is higher, fitting effect is better, but be easy by
To the influence of the abnormal data of certain point, and the calculation amount of single-chip microcontroller and the parameter of preservation are very more, do not have practicability, this
Three sections of progress data fittings of embodiment point.
Further, the split time point determining module 20 includes heating split time point determination unit and cooling segmentation
Time point determination unit；
Heating split time point determination unit is used to concentrate the data under energized state to be at least divided into the historical data
It is straight to obtain every section of fitting using the data in each period to sectional straight line fitting under energized state for two continuous time periods
Then regression coefficient under line calculates the slope of all data in segmentation in addition to every section of starting point to every section of starting point；
It, will be corresponding to the regression coefficient and the immediate coordinate points of this section of straight slope under every section of fitting a straight line in every section of fitting a straight line
At the time of time point for being calculated as this section of straight segments, the split time point of every section of fitting a straight line function constitutes the heating minute
Section time point；
Data under the history data set power cutoff state for being at least divided by cooling split time point determination unit
It is straight to obtain every section of fitting using the data in each period to sectional straight line fitting under off-position for two continuous time periods
Then regression coefficient under line calculates the slope of all data in segmentation in addition to every section of starting point to every section of starting point；
In every section of fitting a straight line, when by corresponding to the regression coefficient and the immediate coordinate points of straight slope under every section of fitting a straight line
It carves the time point calculated as this section of straight segments, when the split time point of every section of fitting a straight line function constitutes heating segmentation
Between point.
Further, the regression coefficient computing module 30 includes the first regression coefficient computing unit and the second regression coefficient
Computing unit；
First regression coefficient computing unit is used in active user's first time on/off electric process, is obtained and the heating
As the heating calibration moment at the time of split time point is identical, and record the heating calibration moment corresponding temperature difference；Utilize the liter
Temperature scale timing is carved and its corresponding temperature difference calculates each straight line coefficient to segmentation under the energized state, and approximation is as under energized state
Each section of fitting a straight line regression coefficient；
Second regression coefficient computing unit is used in active user's first time on/off electric process, is obtained and the cooling
As the cooling calibration moment at the time of split time point is identical, and record the cooling calibration moment corresponding temperature difference；Utilize the drop
Temperature scale timing is carved and its corresponding temperature difference calculates each straight line coefficient to segmentation under the off-position, and approximation is as under off-position
Each section of fitting a straight line regression coefficient.
Further, the temperature measurement error correction module 40 is specifically used for:
Single firewire switchs revised temperature △ Y=Y-b1*t1-b2*t2-b3*t3+b4*t1 '+b5*t2 '+b6*t3 ',
Wherein Y indicates the moment temperature sensor value that is powered, and t1, t2, t3 are respectively the duration of each period under energized state, t1 ',
T2 ', t3 ' are respectively the duration of each period under off-position, and b1, b2, b3 are respectively that the segmentation under energized state calculates directly
The approximate regression coefficient of line, b4, b5, b6 are respectively the approximate regression coefficient of the segmentation calculating straight line under off-position, and * expression multiplies
Product.
It should be noted that this programme difference only improves in structure with the prior art, to reduce integrated circuit board
Fever is influenced caused by temperature sensor, but utilizes Principle of Statistics, is counted the data of a large number of users, is passed through user's history
Data carry out training pattern, and model parameter is finally applied to single firewire thermometric panel.Point to be protected: pass through a large number of users history number
Model parameter merging single-chip microcontroller is obtained according to training, watt level is gone in the electricity consumption used for different user installation environment and user
Difference, the specific model parameter numerical value of intelligent learning, then the purpose of the temperature difference is corrected in automation.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of modification method of list firewire switch temperature measurement error characterized by comprising
In the case where room temperature is kept constant, history data set is obtained, which includes certain amount user's ON/OFF lamp
Any point-in-time and its corresponding temperature difference in the process；
The data under sectional straight line fitting and off-position are carried out to the data under energized state using history data set to divide
Section straight line fitting determines heating split time point and cooling split time point；
In active user's first time on/off electric process, according to heating split time point and cooling split time point, segmentation is calculated
The approximate regression coefficient of each straight line and segmentation calculate the approximate regression coefficient of each straight line under off-position under energized state；
According to the approximate regression coefficient of each straight line under the heating split time point, cooling split time point, energized state and break
The approximate regression coefficient of each straight line under electricity condition corrects single firewire and switchs temperature measurement error.
2. the modification method of list firewire switch temperature measurement error as described in claim 1, which is characterized in that described to utilize history number
At least two sections of straight line fittings are carried out to the data under energized state according to collection and at least two sections are carried out directly to the data under off-position
Line fitting.
3. the modification method of list firewire switch temperature measurement error as claimed in claim 2, which is characterized in that described to utilize history number
The data under sectional straight line fitting and off-position are carried out to the data under energized state according to collection and carry out sectional straight line fitting, are determined
Heat up split time point and cooling split time point, comprising:
The data under energized state are concentrated at least to be divided into two continuous time periods the historical data, using in each period
Data to sectional straight line fitting under energized state, obtain the regression coefficient under every section of fitting a straight line, then calculate segmentation in remove
All data outside every section of starting point to every section of starting point slope；It, will be under every section of fitting a straight line in every section of fitting a straight line
The time point calculated at the time of corresponding to regression coefficient and the immediate coordinate points of this section of straight slope as this section of straight segments,
The split time point of every section of fitting a straight line function constitutes the heating split time point；
Data under the history data set power cutoff state are at least divided into two continuous time periods, using in each period
Data to sectional straight line fitting under off-position, obtain the regression coefficient under every section of fitting a straight line, then calculate segmentation in remove
All data except every section of starting point to every section of starting point slope；In every section of fitting a straight line, by every section of fitting a straight line
Under regression coefficient and the immediate coordinate points of straight slope corresponding at the time of as this section of straight segments calculate time point,
The split time point of every section of fitting a straight line function constitutes the heating split time point.
4. the modification method of list firewire switch temperature measurement error as claimed in claim 3, which is characterized in that described in active user
In first time on/off electric process, according to heating split time point and cooling split time point, segmentation calculates each straight under energized state
The approximate regression coefficient of line and segmentation calculate the approximate regression coefficient of each straight line under off-position, comprising:
In active user's first time on/off electric process, conduct at the time of obtaining identical as the heating split time point
The heating calibration moment, and record the heating calibration moment corresponding temperature difference；Moment and its corresponding temperature difference are demarcated using the heating
Each straight line coefficient, the approximate recurrence system as each section of fitting a straight line under energized state are calculated to segmentation under the energized state
Number；
In active user's first time on/off electric process, conduct at the time of obtaining identical as the cooling split time point
The cooling calibration moment, and record the cooling calibration moment corresponding temperature difference；Moment and its corresponding temperature difference are demarcated using the cooling
Each straight line coefficient, the approximate recurrence system as each section of fitting a straight line under off-position are calculated to segmentation under the off-position
Number.
5. the modification method of list firewire switch temperature measurement error as claimed in claim 4, which is characterized in that described according to the liter
Warm split time point, cooling split time point, each straight line under the approximate regression coefficient of each straight line and off-position under energized state
Approximate regression coefficient, correct single firewire and switch temperature measurement error, comprising:
Single firewire switchs revised temperature △ Y=Y-b1*t1-b2*t2-b3*t3+b4*t1 '+b5*t2 '+b6*t3 ', wherein Y
Indicate the moment temperature sensor value that is powered, t1, t2, t3 are respectively the duration of each period under energized state, t1 ', t2 ', t3 '
The duration of each period respectively under off-position, b1, b2, b3 are respectively the approximation of the segmentation calculating straight line under energized state
Regression coefficient, b4, b5, b6 are respectively the approximate regression coefficient of the segmentation calculating straight line under off-position, and * indicates product.
6. a kind of update the system of list firewire switch temperature measurement error characterized by comprising history data set obtains module, divides
Section time point determining module, regression coefficient computing module and temperature measurement error correction module；
History data set obtains module and is used in the case where room temperature is kept constant, and obtains history data set, the history data set
Including any point-in-time and its corresponding temperature difference during certain amount user's ON/OFF lamp；
Split time point determining module be used for using history data set under energized state data carry out sectional straight line fitting with
Data under off-position carry out sectional straight line fitting, determine heating split time point and cooling split time point；
Regression coefficient computing module is used in active user's first time on/off electric process, according to heating split time point and drop
Warm split time point, segmentation calculate the approximate regression coefficient of each straight line and segmentation under energized state and calculate each straight line under off-position
Approximate regression coefficient；
Temperature measurement error correction module is used for according to each straight under the heating split time point, cooling split time point, energized state
The approximate regression coefficient of each straight line under the approximate regression coefficient and off-position of line corrects single firewire and switchs temperature measurement error.
7. the update the system of list firewire switch temperature measurement error as claimed in claim 6, which is characterized in that described to utilize history number
At least two sections of straight line fittings are carried out to the data under energized state according to collection and at least two sections are carried out directly to the data under off-position
Line fitting.
8. the update the system of list firewire switch temperature measurement error as claimed in claim 7, which is characterized in that the split time point
Determining module includes heating split time point determination unit and cooling split time point determination unit；
Heating split time point determination unit is used to concentrate the data under energized state to be at least divided into two the historical data
Continuous time period is obtained under every section of fitting a straight line using the data in each period to sectional straight line fitting under energized state
Regression coefficient, then calculate in segmentation all data in addition to every section of starting point to the slope of every section of starting point；It is quasi- at every section
It closes in straight line, makees at the time of by the regression coefficient under every section of fitting a straight line with corresponding to the immediate coordinate points of this section of straight slope
Split time point for the time point that this section of straight segments calculate, every section of fitting a straight line function constitutes the heating split time
Point；
Cooling split time point determination unit is used to the data under the history data set power cutoff state being at least divided into two
Continuous time period is obtained under every section of fitting a straight line using the data in each period to sectional straight line fitting under off-position
Regression coefficient, then calculate in segmentation all data in addition to every section of starting point to the slope of every section of starting point；Every
In section fitting a straight line, make at the time of by the regression coefficient under every section of fitting a straight line with corresponding to the immediate coordinate points of straight slope
Split time point for the time point that this section of straight segments calculate, every section of fitting a straight line function constitutes the heating split time
Point.
9. the update the system of list firewire switch temperature measurement error as claimed in claim 8, which is characterized in that the regression coefficient meter
Calculating module includes the first regression coefficient computing unit and the second regression coefficient computing unit；
First regression coefficient computing unit is used in active user's first time on/off electric process, is obtained and is segmented with the heating
As the heating calibration moment at the time of time point is identical, and record the heating calibration moment corresponding temperature difference；It is marked using the heating
Timing is carved and its corresponding temperature difference calculates each straight line coefficient to segmentation under the energized state, and approximation is as each under energized state
The regression coefficient of section fitting a straight line；
Second regression coefficient computing unit is used in active user's first time on/off electric process, is obtained and is segmented with the cooling
As the cooling calibration moment at the time of time point is identical, and record the cooling calibration moment corresponding temperature difference；It is marked using the cooling
Timing is carved and its corresponding temperature difference calculates each straight line coefficient to segmentation under the off-position, and approximation is as each under off-position
The regression coefficient of section fitting a straight line.
10. the update the system of list firewire switch temperature measurement error as claimed in claim 9, which is characterized in that the temperature measurement error
Correction module is specifically used for:
Single firewire switchs revised temperature △ Y=Y-b1*t1-b2*t2-b3*t3+b4*t1 '+b5*t2 '+b6*t3 ', wherein Y
Indicate the moment temperature sensor value that is powered, t1, t2, t3 are respectively the duration of each period under energized state, t1 ', t2 ', t3 '
The duration of each period respectively under off-position, b1, b2, b3 are respectively the recurrence of the sectional straight line fitting under energized state
Coefficient, b4, b5, b6 are respectively the regression coefficient of the sectional straight line fitting under off-position, and * indicates product.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910668601.2A CN110411608B (en) | 2019-07-23 | 2019-07-23 | Correction method and system for temperature measurement error of single-live-wire switch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910668601.2A CN110411608B (en) | 2019-07-23 | 2019-07-23 | Correction method and system for temperature measurement error of single-live-wire switch |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110411608A true CN110411608A (en) | 2019-11-05 |
| CN110411608B CN110411608B (en) | 2021-01-19 |
Family
ID=68362577
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910668601.2A Active CN110411608B (en) | 2019-07-23 | 2019-07-23 | Correction method and system for temperature measurement error of single-live-wire switch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110411608B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011068293A1 (en) * | 2009-12-02 | 2011-06-09 | 한국전력공사 | Apparatus for measuring thermal diffusivity in a nanofluid |
| CN103604525A (en) * | 2013-12-09 | 2014-02-26 | 张金木 | Thermal resistor temperature measuring instrument based on verification data |
| CN104465370A (en) * | 2014-10-31 | 2015-03-25 | 苏州瑞晟纳米科技有限公司 | Temperature rise and fall control method in rapid temperature rise treatment process |
| CN105352630A (en) * | 2015-12-24 | 2016-02-24 | 深圳市博巨兴实业发展有限公司 | Segmental linear calibration system and method for temperature sensor chip |
| CN106679842A (en) * | 2016-11-28 | 2017-05-17 | 北京航天自动控制研究所 | Temperature measuring method and circuit adopting reference voltage compensation technology |
| CN107764435A (en) * | 2017-10-12 | 2018-03-06 | 中国计量大学 | A kind of dynamic characteristic test method of temperature switch |
-
2019
- 2019-07-23 CN CN201910668601.2A patent/CN110411608B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011068293A1 (en) * | 2009-12-02 | 2011-06-09 | 한국전력공사 | Apparatus for measuring thermal diffusivity in a nanofluid |
| CN103604525A (en) * | 2013-12-09 | 2014-02-26 | 张金木 | Thermal resistor temperature measuring instrument based on verification data |
| CN104465370A (en) * | 2014-10-31 | 2015-03-25 | 苏州瑞晟纳米科技有限公司 | Temperature rise and fall control method in rapid temperature rise treatment process |
| CN105352630A (en) * | 2015-12-24 | 2016-02-24 | 深圳市博巨兴实业发展有限公司 | Segmental linear calibration system and method for temperature sensor chip |
| CN106679842A (en) * | 2016-11-28 | 2017-05-17 | 北京航天自动控制研究所 | Temperature measuring method and circuit adopting reference voltage compensation technology |
| CN107764435A (en) * | 2017-10-12 | 2018-03-06 | 中国计量大学 | A kind of dynamic characteristic test method of temperature switch |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110411608B (en) | 2021-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10557640B2 (en) | Smart-home device telemetry circuits for detecting electrical anomalies in a population of devices | |
| US6816078B2 (en) | System and method for estimating power consumption of electric apparatus, and abnormality alarm system utilizing the same | |
| US20170059190A1 (en) | Smart thermostat robust against adverse effects from internal heat-generating components | |
| US10248143B2 (en) | Architecture for thermostat control during time-of-use intervals | |
| US20160154038A1 (en) | Monitoring apparatus, monitoring method, and storage medium | |
| KR102067095B1 (en) | Digital twin device | |
| Caicedo et al. | Sensor-driven lighting control with illumination and dimming constraints | |
| US10248092B2 (en) | Architecture for thermostat control during peak intervals | |
| WO2015125402A1 (en) | Appliance inference device and program | |
| US20140149056A1 (en) | Multi-modal data improvement for power disaggregation systems | |
| US9612586B2 (en) | Energy management apparatus and energy management system | |
| WO2015087470A1 (en) | Demand prediction device and program | |
| Xu et al. | Supply demand coordination for building energy saving: Explore the soft comfort | |
| Milanović et al. | Methodology for estimation of dynamic response of demand using limited data | |
| Panna et al. | Development of energy saving smart home prototype | |
| JP2011150650A (en) | Energy saving diagnostic system | |
| Mraz | The design of intelligent control of a kitchen refrigerator | |
| Ain et al. | Improving quality of experience using fuzzy controller for smart homes | |
| CN110411608A (en) | A kind of modification method and system of list firewire switch temperature measurement error | |
| KR102081218B1 (en) | Digital twin device | |
| KR20200066182A (en) | Energy consumption prediction apparatus and method thereof | |
| CN104363144A (en) | Detection system for network access equipment | |
| JP6292515B2 (en) | Demand forecasting device, program | |
| Aksanli et al. | Distributed control of a swarm of buildings connected to a smart grid: Demo abstract | |
| Ledva et al. | Benchmarking of aggregate residential load models used for demand response |
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 |