CN108062593A - Water use forecast method and prediction meanss - Google Patents

Water use forecast method and prediction meanss Download PDF

Info

Publication number
CN108062593A
CN108062593A CN201711055174.8A CN201711055174A CN108062593A CN 108062593 A CN108062593 A CN 108062593A CN 201711055174 A CN201711055174 A CN 201711055174A CN 108062593 A CN108062593 A CN 108062593A
Authority
CN
China
Prior art keywords
water
target
time
matrix
prediction
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.)
Pending
Application number
CN201711055174.8A
Other languages
Chinese (zh)
Inventor
张树荣
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shenzhen H&T Intelligent Control Co Ltd
Original Assignee
Shenzhen H&T Intelligent Control Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Shenzhen H&T Intelligent Control Co Ltd filed Critical Shenzhen H&T Intelligent Control Co Ltd
Priority to CN201711055174.8A priority Critical patent/CN108062593A/en
Publication of CN108062593A publication Critical patent/CN108062593A/en
Pending legal-status Critical Current

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation, e.g. linear programming, "travelling salesman problem" or "cutting stock problem"
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06Electricity, gas or water supply

Abstract

The embodiment of the invention discloses a kind of water use forecast method and prediction meanss.Wherein, which may include:Structure objective matrix is recorded according to target water, the history that the target is recorded as with water in first time period is recorded with water;It will be predicted in the objective matrix input linear regressive prediction model, obtain prediction matrix, the prediction matrix is the corresponding matrix of target water information, and the target is to use water information in the target time section predicted with water information;The target water information is determined according to the prediction matrix.Using the embodiment of the present invention, the accuracy that prediction meanss predict user's water can not only be improved, additionally it is possible to intelligently realize prediction, improve the efficiency of prediction meanss prediction, improve the satisfaction of user.

Description

Water use forecast method and prediction meanss
Technical field
The present invention relates to Smart Home technical field more particularly to a kind of water use forecast methods and prediction meanss.
Background technology
At present in China's most area, the use of storage-type electric water heater in the family is than wide.In real life In, user can carry out in the following way when using storage-type electric water heater, and one kind is manually opened or closing electric hot water Device;Another kind is turned on and off by terminal control electric heater.The first needs user manually opened or closes, second Kind user needs to pre-set heating time, so that terminal is heated according to pre-set heating time.
Using above-mentioned technical proposal, the rule of life of user, inefficiency can not be intelligently adapted to.
The content of the invention
The embodiment of the present invention provides a kind of water use forecast method and prediction meanss, can not only improve prediction meanss prediction and use The accuracy of family water, additionally it is possible to intelligently realize prediction, improve the efficiency of prediction meanss prediction, improve the satisfaction of user.
In a first aspect, an embodiment of the present invention provides a kind of water use forecast method, including:
Structure objective matrix is recorded according to target water, the target is recorded as the history water in first time period with water Record;
It will be predicted in the objective matrix input linear regressive prediction model, obtain prediction matrix, the prediction square Battle array is the corresponding matrix of target water information, and the target is to use water information in the target time section predicted with water information;
The target water information is determined according to the prediction matrix.
It is described to be recorded according to target water before building objective matrix in an optional realization method, the method It further includes:
The target is obtained to be recorded with water;
Determine that the target records the use water information of corresponding N number of time cycle with water;
N number of time cycle with water information is encoded, obtains target code, the target code is the N A time cycle uses the corresponding N number of coding of water information;
It is described to include according to target water record structure objective matrix:
The objective matrix is built according to the target code.
It is described to be recorded according to target water before building objective matrix in an optional realization method, the method It further includes:
Structure training matrix is recorded according to the water of the history in second time period;
Linear regression model (LRM) is trained using the training matrix, obtains the Linear Regression Forecasting Model.
In an optional realization method, N number of time cycle includes cycle first time and second time period, Respectively comprising M time, the M time is by the first time for cycle first time and the second time period Obtained time after cycle or the second time period are divided according to default resolution ratio, the target code include institute It states the M time corresponding coding with water information in cycle first time and the M time of the second time period corresponds to The coding with water information.
In an optional realization method, the Linear Regression Forecasting Model is as follows:
x(n)=X(N)*W
Wherein, x(n)For with the prediction matrix, X(N)The corresponding matrix of water information is used for N number of time cycle, W is Regression coefficient.
In an optional realization method, it is described determine the target water information according to the prediction matrix after, The method further includes:
After generating control instruction with water information according to the target, the control instruction is exported.
Second aspect, an embodiment of the present invention provides a kind of water use forecast device, including:
Construction unit, for recording structure objective matrix according to target water, the objective matrix is in first time period History recorded with water;
Predicting unit for will be predicted in the objective matrix input linear regressive prediction model, obtains prediction square Battle array, the prediction matrix are the corresponding matrix of target water information, and the target is in the target time section predicted with water information Use water information;
First determination unit, for determining the target water information according to the prediction matrix.
In an optional realization method, described device further includes:
Acquiring unit is recorded for obtaining the target with water;
Second determination unit records the use water information of corresponding N number of time cycle for determining the target with water;
Coding unit for being encoded to N number of time cycle with water information, obtains target code, the mesh Mark is encoded to the corresponding N number of coding of use water information of N number of time cycle;
The construction unit, specifically for building the objective matrix according to the target code.
In an optional realization method, the construction unit is additionally operable to according to the history water in second time period Record structure training matrix;
Described device further includes:
Training unit for being trained using the training matrix to linear regression model (LRM), obtains the linear regression Prediction model.
In an optional realization method, N number of time cycle includes cycle first time and second time period, Respectively comprising M time, the M time is by the first time for cycle first time and the second time period Obtained time after cycle or the second time period are divided according to default resolution ratio, the target code include institute It states the M time corresponding coding with water information in cycle first time and the M time of the second time period corresponds to The coding with water information.
In an optional realization method, the Linear Regression Forecasting Model is as follows:
x(n)=X(N)*W
Wherein, x(n)For the prediction matrix, X(N)The corresponding target square of water information is used for N number of time cycle Battle array, W are the corresponding matrix of regression coefficient of N number of time cycle.
In an optional realization method, described device further includes:
Generation unit, for generating control instruction according to target water information;
Output unit, for exporting the control instruction.
The third aspect, an embodiment of the present invention provides a kind of prediction meanss, including processor, transceiver and memory, In:The processor, the transceiver and the memory are connected with each other, and the memory is for storing computer program, institute Stating computer program includes program instruction, and the processor is arranged to call described program instruction, performs such as first aspect Or any one possible described method of realization method of first aspect.
Fourth aspect, the embodiment of the present invention provide a kind of computer readable storage medium, the computer-readable storage medium Matter is stored with computer program, and the computer program includes program instruction, and described program instruction is when by sound of snoring detection device When processor performs, the processor is made to perform any one possible realization method institute of above-mentioned first aspect or first aspect The method of description.
5th aspect, the embodiment of the present invention provides a kind of computer program product for including program instruction, when it is being calculated When being run on machine, the computer is made to perform above-mentioned first aspect or any one possible described method of realization method.
Implement the embodiment of the present invention, after recording structure objective matrix with water by target, by above-mentioned objective matrix input line It is predicted in property regressive prediction model, obtains prediction matrix;Target water information, above-mentioned mesh are exported according to above-mentioned prediction matrix Mark with water information be prediction target time section in use water information, controlled by the control instruction corresponding equipment to water into Row processing, and then facilitate user's water, and the rule of life of user is more intelligently adapted to, improve the water-use efficiency of user.
Description of the drawings
Technical solution in order to illustrate the embodiments of the present invention more clearly or in background technology below will be implemented the present invention Attached drawing illustrates needed in example or background technology.
Fig. 1 is a kind of flow diagram of water use forecast method provided in an embodiment of the present invention;
Fig. 2 is the flow diagram of another water use forecast method provided in an embodiment of the present invention;
Fig. 3 is a kind of structure diagram of objective matrix provided in an embodiment of the present invention;
Fig. 4 is a kind of schematic diagram of a scenario of water use forecast method provided in an embodiment of the present invention;
Fig. 5 is a kind of structure diagram of prediction meanss provided in an embodiment of the present invention;
Fig. 6 is the structure diagram of another prediction meanss provided in an embodiment of the present invention;
Fig. 7 is the structure diagram of another prediction means provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical solutions and advantages of the present invention clearer, the present invention is made below in conjunction with attached drawing into One step it is described in detail.
It should be noted that the detailed description illustrated with reference to attached drawing is intended as the description to various configurations, without purport Unique configuration of concepts described herein can be wherein put into practice in expression.Recorded device embodiment and method are real herein Applying example will be described in the following detailed description, and pass through various frames, module, unit, component, circuit, step in the accompanying drawings Suddenly, process, algorithm etc. (being referred to as " element ") are shown.These elements can use electronic hardware, computer software Or it is combined to realize.Hardware or software are implemented as these elements, depending on specific application and is applied to Design constraint on total system.If the term in description and claims of this specification and Figure of description uses The descriptions such as " first ", " second ", this kind description are for distinguishing different objects rather than for describing particular order.
It should be appreciated that ought use in this specification and in the appended claims, term " comprising " and "comprising" instruction Described feature, entirety, step, operation, the presence of element and/or component, but it is not precluded from one or more of the other feature, whole Body, step, operation, element, component and/or its presence or addition gathered.It is also understood that the institute in this description of the invention The term used is not intended to limit the present invention merely for the sake of the purpose of description specific embodiment.Such as in description of the invention With it is used in the attached claims like that, other situations unless the context is clearly specified, otherwise singulative " one ", "one" and "the" are intended to include plural form.It it will be further appreciated that will in description of the invention and appended right The term used in book "and/or" is asked to refer to any combinations of one or more of the associated item listed and be possible to Combination, and including these combinations.
It should be noted that in the case of the special instruction do not expressed, every technology in various embodiments of the present invention As long as feature, which can be considered to be mutually combined, is either combined this kind combination or with reference to nothing the reason for being not as technology Method is implemented.In order to more sufficiently illustrate the present invention, some are illustrative, and optionally or preferred feature is of the invention each It is combined together and is described with other technical characteristics in embodiment, but this combination is not essential, and should be appreciated that this and show Example property, optionally either preferred feature and other technical characteristics are all separable or independent each other, as long as should It plants the reason for separating or being independently not as technology and can not implement.Some functions of technical characteristic in embodiment of the method Property description can be understood as performing the function, method or step, some functionality of the technical characteristic in device embodiment are retouched It states and can be understood as performing the function, method or step using this kind of device.
Fig. 1 is referred to, Fig. 1 is a kind of flow diagram of water use forecast method provided in an embodiment of the present invention, this uses water Forecasting Methodology can be applied to prediction meanss, which can be that water heater, kettle, water purifier, water filter etc. carry out water The device of processing, specifically, the prediction meanss can be device for carrying out water heating etc..It is understood that the prediction fills It puts or server, the i.e. processor of the server runs the water use forecast by calling the program instruction in memory Method, the prediction meanss or Cloud Server or for other kinds of server etc..As shown in Figure 1, this uses water Forecasting Methodology may include:
101st, structure objective matrix is recorded according to target water, above-mentioned target is recorded as the history in first time period with water It is recorded with water;
In the embodiment of the present invention, target can be that prediction meanss carry out being recorded with water for water process with water record, specifically, Target water record can be that can such as be included in first time period with the relevant record of user's water behavior in first time period It is each at the beginning of water behavior, with information such as water duration, end times.For example, the form of target water record can To be " with the water time started:XX divides XX seconds during XXX XX month XX day XX;With the water end time:XX during XXX XX month XX day XX Divide XX seconds ".Refer to the behavior with water with water behavior, such as discharged water by tap, hot water is for another example used by water heater.It is above-mentioned Prediction meanss can include data acquisition module and logging modle, and above-mentioned data acquisition module can detect the beginning with water behavior Time, end time and other can also be gathered with water information such as water temperature etc.;Above-mentioned logging modle can record above-mentioned use At the beginning of water behavior, end time and above-mentioned acquisition module collect other with water information etc..
It is water heater when carrying out the device of water process in prediction meanss, it is pre- that above-mentioned history water record can be stored in this It surveys in the memory of device, be stored in the server being connected with the prediction meanss etc., the embodiment of the present invention is not made It limits.
102nd, it will be predicted in above-mentioned objective matrix input linear regressive prediction model, obtain prediction matrix, it is above-mentioned pre- Survey matrix is the corresponding matrix of target water information, and above-mentioned target is being believed with water in the target time section predicted with water information Breath;
In the embodiment of the present invention, Linear Regression Forecasting Model can be opposite with the watering law of user after training The regression coefficient of the model answered, i.e. Linear Regression Forecasting Model has had determined or the Linear Regression Forecasting Model Regression coefficient can be continuously increased and constantly update etc. with what history water recorded, and objective matrix is input to this linear returns After returning prediction model, it is possible to obtain prediction matrix.
103rd, above-mentioned target water information is determined according to above-mentioned prediction matrix.
After determining target water information according to prediction matrix, target user's information can also be exported, first will such as can be Above-mentioned prediction matrix is converted into above-mentioned target water information, then exports above-mentioned target water information.Above-mentioned prediction matrix can wrap Containing at least one column vector, water information is used in each one prediction of column vector expression.
Optionally, the embodiment of the present invention additionally provides a kind of realization method for exporting target user's information, above-mentioned according to upper It states after prediction matrix determines above-mentioned target water information, the above method can also include:
After generating control instruction with water information according to the target, the control instruction is exported.
Specifically, in the case where the prediction meanss carry out the device of water process for water heater etc., by exporting the control Instruction can effectively realize the control with water information to user, and user's water is controlled with water information according to the target predicted At the beginning of, end time etc. so that the prediction meanss are more intelligent.
In the case where the prediction meanss is servers, exporting above-mentioned control instruction includes:Above-mentioned control instruction is sent The device of water process is carried out to water heater etc. so that the device of the progress water process such as the water heater can realize the processing to water, Since server handling ability is better than water heater processing capacity, above-mentioned water use forecast method is realized by server, it can be with The running memory loss that water heater etc. carries out the device of water process is efficiently reduced, extends the dress that the water heater etc. carries out water process The service life put improves the satisfaction of user.
Implement the embodiment of the present invention, after recording structure objective matrix with water by target, by above-mentioned objective matrix input line It is predicted in property regressive prediction model, obtains prediction matrix;Target water information, above-mentioned mesh are exported according to above-mentioned prediction matrix Mark with water information be prediction target time section in use water information, controlled by the control instruction corresponding equipment to water into Row processing, and then facilitate user's water, and the rule of life of user is more intelligently adapted to, improve the water-use efficiency of user.
Based on water use forecast method described by Fig. 1, Fig. 2 is referred to, Fig. 2 is another kind water provided in an embodiment of the present invention Forecasting Methodology, which is that further optimization obtains on the basis of Fig. 1, as shown in Fig. 2, the water use forecast method Including but not limited to following steps:
201st, structure training matrix is recorded according to the water of the history in second time period;
202nd, linear regression model (LRM) is trained using above-mentioned training matrix, obtains above-mentioned Linear Regression Forecasting Model;
In the embodiment of the present invention, second time period can be a week, one month, half a year, 1 year etc..Above-mentioned foundation The history that history water record structure training matrix in two periods can be to determine in above-mentioned second time period is recorded with water Corresponding coding records corresponding coding with water using the history in above-mentioned second time period and builds above-mentioned training matrix.It is above-mentioned The building mode of training matrix can objective matrix as described below building mode.
Linear regression model (LRM) is trained using training matrix, obtains the method for Linear Regression Forecasting Model, can be wrapped Include following two situations:
Situation one,
Training linear regression model (LRM) is recorded using the water of the history in second time period, after obtaining regression coefficient, utilizes this Regression coefficient predicts the user information of user.In this case, second time period and target time section can be nothings Temporal contact, that is, after determining regression coefficient, prediction meanss can use the corresponding linear regression mould of the regression coefficient Type is predicted, can effectively reduce the power consumption penalty of the prediction meanss, avoids increasing the burden of the prediction meanss.
Situation two,
Second time period is variation;Specifically, which can be corresponding with target time section, optionally, First time period can also be corresponding with target time section.In this case, which can be with second time period Variation, and then regression coefficient is constantly updated, the accuracy of prediction meanss prediction can be improved, is more in line with user when current Between section water use variation, improve the efficiency of prediction.
Second time period is corresponding with target time section and first time period and target time section it is corresponding, citing comes It says, target time section uses water information for September user's on the 27th, then first time period can be September 20 days to September user on the 26th With water information, and can be September 19 days use water information to second time period to September user on the 25th.It is understood that above-mentioned One period and the correspondence of second time period and target time section are only a kind of citing, be should not be construed as to of the invention real Example is applied with limiting meaning.
203rd, obtain target to be recorded with water, the history that above-mentioned target is recorded as with water in first time period is recorded with water;
204th, determine that above-mentioned target records the use water information of corresponding N number of time cycle with water;
In the embodiment of the present invention, the time cycle can be one week, one day, 12 hours, two hours etc..Determine that target is used What water recorded corresponding N number of time cycle can first determine that above-mentioned target water is remembered according to the sequencing of time with water information Record corresponding N number of time cycle;Then, it is determined that corresponding water information of each time cycle in above-mentioned N number of time cycle.Citing For, it is assumed that target was recorded with water record comprising 5 days with water, i.e., first time period when a length of 5 days, the time cycle one My god;Then determine that the target records the use water information of corresponding 5 time cycles with water.It is above-mentioned with water information can be from history use Some water data for filtering out in water record, such as with duration at the beginning of water behavior and with water behavior etc.. For example, target records corresponding cycle first time to the 4th time cycle with water, it may be determined that cycle first time to the Respectively at the beginning of water behavior and duration in four time cycles.Above-mentioned N is the integer more than or equal to 1.
205th, above-mentioned N number of time cycle with water information is encoded, obtains target code, above-mentioned target code is upper That states N number of time cycle uses the corresponding N number of coding of water information;
Specifically, above-mentioned N number of time cycle includes cycle first time and second time period, above-mentioned cycle first time With above-mentioned second time period respectively comprising M time, the M time for by cycle first time or second time period according to Default resolution ratio obtained time after being divided, above-mentioned target code include the M time pair in above-mentioned cycle first time Answer with the coding of water information and the coding of M time of above-mentioned second time period corresponding water information.It is of the invention real It applies in example, with one day for the time cycle, default resolution ratio can be minute, so as to be carried out to N number of time cycle with water information Coding, you can after each time cycle to be divided according to default resolution ratio, M time is obtained, to the M after division Time is encoded.Such as using day as the time cycle after, then using minute as resolution ratio, one day is divided into 1440 minutes, so as to one It 1440 minutes in a time cycle is encoded with water information.
For example, it is assumed that target was recorded with water record comprising 5 days with water, i.e., first time period when a length of 5 days, when Between the cycle be one day, it is minute to preset resolution ratio, then the target with water record comprising five time cycles, wherein, each time Be divided into cycle not Dui Ying 1440 with water information.In the embodiment of the present invention, it can represent to exist with 1 and use water behavior, be represented with 0 There is no with water behavior, so as to being encoded in N number of time cycle with water information.It is provided using the embodiment of the present invention Coding mode is encoded, and can effectively be indicated whether to exist and be used water behavior, record user exactly uses water behavior, uses this Kind coding mode form is simple, is more easy to realize.
206th, above-mentioned objective matrix is built according to above-mentioned target code;
Wherein, the row of objective matrix can represent to use water information in a time cycle, and one can be represented with the corresponding row of row M time corresponding water information in a time cycle, as shown in figure 3, Fig. 3 provides a kind of structural representation of objective matrix Figure, wherein, first row can represent to use water information in cycle first time, and the first row can represent and such as first minute M time Water information is used in corresponding N number of time cycle.
207th, it will be predicted in above-mentioned objective matrix input linear regressive prediction model, obtain prediction matrix, it is above-mentioned pre- Survey matrix is the corresponding matrix of target water information, and above-mentioned target is being believed with water in the target time section predicted with water information Breath;
Above-mentioned Linear Regression Forecasting Model is as follows:
x(n)=X(N)*W
Wherein, x(n)For above-mentioned prediction matrix, X(N)The corresponding above-mentioned target square of water information is used for above-mentioned N number of time cycle Battle array, W are the corresponding matrix of regression coefficient of N number of time cycle.
In the embodiment of the present invention, if with one day for the time cycle, x(n)Intraday for prediction uses water information, i.e., pre- The form for surveying matrix is 1440 × 1, X(N)Water information, i.e. target square are used for N number of 1400 minutes corresponding to N number of time cycle The form of battle array is 1440 × N, and W is N × 1.It it is understood that should not be by x of the embodiment of the present invention(n)In n be interpreted as to this hair The prediction model that bright embodiment is provided has limiting meaning.
Described in the situation two according to structure objective matrix, second time period can be 1 to N-1 time Water information is used in cycle, first time period can be to use water information in 2 to N number of time cycle, utilize week preceding N-1 time Linear regression model (LRM) is trained with water information in phase, obtains regression coefficient W (i.e. trained Linear Regression Forecasting Model).Then Using the use water information architecture objective matrix in 2 to N number of time cycle, as the input of Linear Regression Forecasting Model, so as to pre- Survey the use water information of the N+1 time cycle.
208th, above-mentioned target water information is determined according to above-mentioned prediction matrix.
Implement the embodiment of the present invention, disclosure satisfy that the requirement of total management system, effective user's future behaviour is provided Prediction result controls so as to fulfill the equipment of optimization, meets user demand.Meanwhile linear regression model (LRM) is used so that calculating Mode more lightweight adapts to the limited condition of computing resource.
In order to make it easy to understand, based on the described water use forecast methods of Fig. 2, Fig. 4 is referred to, Fig. 4 is the embodiment of the present invention The schematic diagram of a scenario of a kind of water use forecast method provided, as shown in figure 4, the scene specifically may include:
401st, the history in user 8 days is called to be recorded with water;
Wherein, it is minute to preset resolution ratio, then it is 1440*8 that the history recalled records corresponding matrix with water.
402nd, the history of former 7 days records corresponding matrix as independent variable by the use of water, is recorded as because of change by the use of water within the 8th day Amount calculates regression coefficient W by formula (1);
Wherein, N can be 8, and 1~N-1 can be understood as 1~7, that is, represent the history water record pair in the 1st day to the 7th day The matrix answered, specifically,It is construed as recording corresponding 1440*7 with water to the history in the 1st day to the 7th day Matrix required by inverse matrix.x(n)Represent the corresponding matrix of user record of the 8th day.
403rd, corresponding matrix (i.e. objective matrix) is recorded as input with water with the history in the 2nd day to the 8th day, utilizes step The regression coefficient calculated in rapid 402 calculates prediction matrix by formula (2);
x(n+1)=X(2~N)*W (2)
Wherein, X(2~N)It is construed as recording corresponding matrix with water to the history in the 2nd day to the 8th day, W is to pass through The regression coefficient that formula (1) is calculated, x(n+1)Represent the 9th day of prediction uses the corresponding matrix of water information.
404th, the use water information of the 9th day is determined according to prediction matrix.
It specifically, can be according to the x that formula (2) are calculated(n+1)Determine the use water information of the 9th day.
It is understood that the specific implementation of the water use forecast method provided the embodiment of the present invention, please also join See the specific implementation of other embodiment, no longer repeat one by one here.
Implement the embodiment of the present invention, can not only predict user exactly uses water information, but also calculation amount is small, reduces pre- Survey the workload of device.
The device of the embodiment of the present invention is provided below in the above-mentioned method for illustrating the embodiment of the present invention.
Refer to Fig. 5, Fig. 5 is a kind of structure diagram of prediction meanss provided in an embodiment of the present invention, the prediction meanss It can include:
Construction unit 501, for recording structure objective matrix according to target water, above-mentioned objective matrix is first time period Interior history is recorded with water;
Predicting unit 502 for will be predicted in above-mentioned objective matrix input linear regressive prediction model, is predicted Matrix, above-mentioned prediction matrix be the corresponding matrix of target water information, above-mentioned target with water information be prediction target time section Interior uses water information;
First determination unit 503, for determining above-mentioned target water information according to above-mentioned prediction matrix.
Implement the embodiment of the present invention, after recording structure target square with water by target, by above-mentioned objective matrix input linear It is predicted in regressive prediction model, obtains prediction matrix;Target water information, above-mentioned target are exported according to above-mentioned prediction matrix It is to use water information in the target time section of prediction with water information, corresponding equipment is controlled to carry out water by the control instruction Processing, and then facilitate user's water, and the rule of life of user is more intelligently adapted to, improve the water-use efficiency of user.
Optionally, as shown in fig. 6, above device further includes:
Acquiring unit 504 is recorded for obtaining above-mentioned target with water;
Second determination unit 505 records the use water information of corresponding N number of time cycle for determining above-mentioned target with water;
Coding unit 506 for being encoded to above-mentioned N number of time cycle with water information, obtains target code, above-mentioned Target code uses the corresponding N number of coding of water information for above-mentioned N number of time cycle;
Above-mentioned construction unit 501, specifically for building above-mentioned objective matrix according to above-mentioned target code.
Implement the embodiment of the present invention, being encoded with water information for corresponding N number of time cycle recorded with water according to target, Obtain target code;According to the target code build objective matrix, the objective matrix can with input linear regressive prediction model into Row prediction;Target can be rapidly converted into Linear Regression Forecasting Model acceptable input form with water record, realized Simply.
Optionally, above-mentioned construction unit 501 is additionally operable to record structure training square according to the water of the history in second time period Battle array;
As shown in fig. 6, above device further includes:
Training unit 507 for being trained using above-mentioned training matrix to linear regression model (LRM), is obtained above-mentioned linear time Return prediction model.
Implement the embodiment of the present invention, linear regression model (LRM) regression coefficient with the change of second time period and constantly more Under news, by constantly updating regression coefficient, the accuracy of prediction meanss prediction can be not only improved, but also can be more Add the water use variation for meeting user in current slot, improve the efficiency of prediction.
Specifically, above-mentioned N number of time cycle includes cycle first time and second time period, above-mentioned cycle first time With above-mentioned second time period respectively comprising M time, the above-mentioned M time for by above-mentioned cycle first time or it is above-mentioned second when Between the cycle divided according to default resolution ratio after the obtained time, above-mentioned target code includes above-mentioned cycle first time M time is corresponding with the coding of water information and the coding of M time of above-mentioned second time period corresponding water information.
Specifically, above-mentioned Linear Regression Forecasting Model is as follows:
x(n)=X(h)w
Wherein, x(n)For above-mentioned prediction matrix, X(h)The corresponding above-mentioned target square of water information is used for above-mentioned N number of time cycle Battle array, w are the corresponding matrix of regression coefficient of N number of time cycle.
Optionally, as shown in fig. 6, above device further includes:
Generation unit 508, for generating control instruction according to above-mentioned target water information;
Output unit 509, for exporting above-mentioned control instruction.
Implement the embodiment of the present invention, by exporting control instruction, so that water treatment facilities can be carried out at water in time Reason, realization is simple, user is not required to be operated.
It should be noted that the realization of unit can also correspond to the phase referring to figs. 1 to embodiment of the method shown in Fig. 4 It should describe.
Fig. 7 is referred to, Fig. 7 is another prediction means provided in an embodiment of the present invention, which includes processor 701st, memory 702 and transceiver 703, the processor 701, memory 702 and transceiver 703 are mutually interconnected by bus 704 It connects.
Memory 702 includes but not limited to be random access memory (English:Random Access Memory, referred to as: RAM), read-only memory (English:Read-Only Memory, referred to as:ROM), Erasable Programmable Read Only Memory EPROM (English: Erasable Programmable Read Only Memory, referred to as:EPROM) or portable read-only memory is (English: Compact Disc Read-Only Memory, referred to as:CD-ROM), which is used for dependent instruction and data.
Transceiver 703 is used to send and receive data, it may include a receiver and a transmitter, such as less radio-frequency Module.
Processor 701 can be one or more central processing units (English:Central Processing Unit, letter Claim:CPU), in the case where processor 701 is a CPU, which can be monokaryon CPU or multi-core CPU.
Processor 701 in the prediction meanss is used to read the program code stored in memory 702, performs following behaviour Make:
Structure objective matrix is recorded according to target water, above-mentioned target is recorded as the history water in first time period with water Record;It will be predicted in above-mentioned objective matrix input linear regressive prediction model, obtain prediction matrix, above-mentioned prediction matrix is The target corresponding matrix of water information, above-mentioned target are to use water information in the target time section predicted with water information;According to upper It states prediction matrix and determines above-mentioned target water information.
In an optional realization method, the processor 701 in the prediction meanss can be also used for reading memory 702 The program code of middle storage performs following operation:
Above-mentioned target is obtained to be recorded with water;
Determine that above-mentioned target records the use water information of corresponding N number of time cycle with water;
Above-mentioned N number of time cycle with water information is encoded, obtains target code, above-mentioned target code is above-mentioned N A time cycle uses the corresponding N number of coding of water information;
The specific implementation that above-mentioned processor 701 records structure objective matrix according to target water can include:
Above-mentioned objective matrix is built according to above-mentioned target code.
In an optional realization method, the processor 701 in the prediction meanss can be also used for reading memory 702 The program code of middle storage performs following operation:
Structure training matrix is recorded according to the water of the history in second time period;
Linear regression model (LRM) is trained using above-mentioned training matrix, obtains above-mentioned Linear Regression Forecasting Model.
Specifically, above-mentioned N number of time cycle includes cycle first time and second time period, above-mentioned cycle first time With above-mentioned second time period respectively comprising M time, the above-mentioned M time for by above-mentioned cycle first time or it is above-mentioned second when Between the cycle divided according to default resolution ratio after the obtained time, above-mentioned target code includes above-mentioned cycle first time M time is corresponding with the coding of water information and the coding of M time of above-mentioned second time period corresponding water information.
Specifically, above-mentioned Linear Regression Forecasting Model is as follows:
x(n)=X(h)w
Wherein, x(n)For above-mentioned prediction matrix, X(h)The corresponding above-mentioned target square of water information is used for above-mentioned N number of time cycle Battle array, w are the corresponding matrix of regression coefficient of N number of time cycle.
In an optional realization method, the processor 701 in the prediction meanss can be also used for reading memory 702 The program code of middle storage performs following operation:
After generating control instruction with water information according to above-mentioned target, above-mentioned control instruction is exported.
It should be noted that the realization of each operation can also correspond to the phase referring to figs. 1 to embodiment of the method shown in Fig. 4 It should describe.
The embodiment of the present invention also provides a kind of computer readable storage medium, and above computer readable storage medium storing program for executing is stored with Computer program, above computer program include program instruction, and above procedure instruction is realized when being executed by processor:
Structure objective matrix is recorded according to target water, above-mentioned target is recorded as the history water in first time period with water Record;It will be predicted in above-mentioned objective matrix input linear regressive prediction model, obtain prediction matrix, above-mentioned prediction matrix is The target corresponding matrix of water information, above-mentioned target are to use water information in the target time section predicted with water information;According to upper It states prediction matrix and determines above-mentioned target water information.
Above computer readable storage medium storing program for executing can be the internal storage unit of the prediction meanss of foregoing any embodiment, example Such as the hard disk or memory of prediction meanss.Above computer readable storage medium storing program for executing can also be that the external storage of above-mentioned prediction meanss is set Plug-in type hard disk that is standby, such as being equipped in above-mentioned prediction meanss, intelligent memory card (Smart Media Card, SMC), safe number Word (Secure Digital, SD) blocks, flash card (Flash Card) etc..Further, above computer readable storage medium storing program for executing The internal storage unit of above-mentioned prediction meanss can also both be included or including External memory equipment.Above computer readable storage medium Matter is used to store above computer program and other programs and data needed for above-mentioned prediction meanss.Above computer is readable to deposit Storage media can be also used for temporarily storing the data that has exported or will export.
One of ordinary skill in the art will appreciate that realize all or part of flow in above-described embodiment method, the flow Relevant hardware can be instructed to complete by computer program, which can be stored in computer read/write memory medium, should Program is upon execution, it may include such as the flow of above-mentioned each method embodiment.And foregoing storage medium includes:ROM is deposited at random Store up the medium of the various program storage codes such as memory body RAM, magnetic disc or CD.

Claims (10)

  1. A kind of 1. water use forecast method, which is characterized in that including:
    Structure objective matrix is recorded according to target water, the history that the target is recorded as with water in first time period is remembered with water Record;
    It will be predicted in the objective matrix input linear regressive prediction model, obtain prediction matrix, the prediction matrix is The target corresponding matrix of water information, the target are to use water information in the target time section predicted with water information;
    The target water information is determined according to the prediction matrix.
  2. 2. according to the method described in claim 1, it is characterized in that, it is described according to target with water record structure objective matrix it Before, the method further includes:
    The target is obtained to be recorded with water;
    Determine that the target records the use water information of corresponding N number of time cycle with water;
    N number of time cycle with water information is encoded, target code is obtained, when the target code is described N number of Between the cycle with the corresponding N number of coding of water information;
    It is described to include according to target water record structure objective matrix:
    The objective matrix is built according to the target code.
  3. 3. according to the method described in claim 2, it is characterized in that, it is described according to target with water record structure objective matrix it Before, the method further includes:
    Structure training matrix is recorded according to the water of the history in second time period;
    Linear regression model (LRM) is trained using the training matrix, obtains the Linear Regression Forecasting Model.
  4. 4. according to the method in claim 2 or 3, which is characterized in that N number of time cycle include cycle first time and Respectively comprising M time, the M time is for second time period, cycle first time and the second time period It is the obtained time after cycle first time or the second time period are divided according to default resolution ratio, described Target code includes the M time corresponding coding with water information in cycle first time and the second time period The M time corresponding coding with water information.
  5. 5. according to the method described in claim 2, it is characterized in that, the Linear Regression Forecasting Model is as follows:
    x(n)=X(N)*W
    Wherein, x(n)For the prediction matrix, X(N)The corresponding objective matrix of water information, W are used for N number of time cycle For the corresponding matrix of regression coefficient of N number of time cycle.
  6. 6. according to the method described in claim 1, it is characterized in that, described determine the target water according to the prediction matrix After information, the method further includes:
    After generating control instruction with water information according to the target, the control instruction is exported.
  7. 7. a kind of water use forecast device, which is characterized in that including:
    Construction unit, for recording structure objective matrix according to target water, the objective matrix is going through in first time period History is recorded with water;
    Predicting unit for will be predicted in the objective matrix input linear regressive prediction model, obtains prediction matrix, institute Prediction matrix is stated as the corresponding matrix of target water information, the target is to use water in the target time section predicted with water information Information;
    First determination unit, for determining the target water information according to the prediction matrix.
  8. 8. device according to claim 7, which is characterized in that described device further includes:
    Acquiring unit is recorded for obtaining the target with water;
    Second determination unit records the use water information of corresponding N number of time cycle for determining the target with water;
    Coding unit for being encoded to N number of time cycle with water information, obtains target code, and the target is compiled Code uses the corresponding N number of coding of water information for N number of time cycle;
    The construction unit, specifically for building the objective matrix according to the target code.
  9. 9. a kind of water use forecast device, which is characterized in that including:Processor, transceiver and memory, wherein:
    The processor, the transceiver and the memory are connected with each other, and the memory is for storing computer program, institute Stating computer program includes program instruction, and the processor is arranged to call described program instruction, performs such as claim 1 To the method described in 6 any one.
  10. 10. a kind of computer readable storage medium, which is characterized in that the computer-readable recording medium storage has computer journey Sequence, the computer program include program instruction, and described program instructs when being executed by a processor, and the processor is made to perform such as Method described in claim 1 to 6 any one.
CN201711055174.8A 2017-10-31 2017-10-31 Water use forecast method and prediction meanss Pending CN108062593A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201711055174.8A CN108062593A (en) 2017-10-31 2017-10-31 Water use forecast method and prediction meanss

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201711055174.8A CN108062593A (en) 2017-10-31 2017-10-31 Water use forecast method and prediction meanss

Publications (1)

Publication Number Publication Date
CN108062593A true CN108062593A (en) 2018-05-22

Family

ID=62135583

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201711055174.8A Pending CN108062593A (en) 2017-10-31 2017-10-31 Water use forecast method and prediction meanss

Country Status (1)

Country Link
CN (1) CN108062593A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108961129A (en) * 2018-06-11 2018-12-07 福建工程学院 Animation detection method and storage medium based on water meter water
CN110367897A (en) * 2019-07-25 2019-10-25 宁波方太厨具有限公司 Control method, system, equipment and the storage medium of the automatic warm dish of smart machine

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013141094A1 (en) * 2012-03-22 2013-09-26 Kabushiki Kaisha Toshiba Behavioral model generating device and method therefor
CN104134182A (en) * 2014-08-01 2014-11-05 武汉理工大学 Hospital industry dynamic water intake rationing method
CN104715292A (en) * 2015-03-27 2015-06-17 上海交通大学 City short-term water consumption prediction method based on least square support vector machine model
CN105725791A (en) * 2014-12-12 2016-07-06 佛山市顺德区美的饮水机制造有限公司 Control method and system of water dispenser, server and water dispenser

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2013141094A1 (en) * 2012-03-22 2013-09-26 Kabushiki Kaisha Toshiba Behavioral model generating device and method therefor
CN104134182A (en) * 2014-08-01 2014-11-05 武汉理工大学 Hospital industry dynamic water intake rationing method
CN105725791A (en) * 2014-12-12 2016-07-06 佛山市顺德区美的饮水机制造有限公司 Control method and system of water dispenser, server and water dispenser
CN104715292A (en) * 2015-03-27 2015-06-17 上海交通大学 City short-term water consumption prediction method based on least square support vector machine model

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108961129A (en) * 2018-06-11 2018-12-07 福建工程学院 Animation detection method and storage medium based on water meter water
CN110367897A (en) * 2019-07-25 2019-10-25 宁波方太厨具有限公司 Control method, system, equipment and the storage medium of the automatic warm dish of smart machine

Similar Documents

Publication Publication Date Title
CN108062593A (en) Water use forecast method and prediction meanss
CN108197705A (en) Convolutional neural networks hardware accelerator and convolutional calculation method and storage medium
CN107655212A (en) Heating treatment method and device
CN108091166B (en) Method, device, equipment and storage medium for predicting number change of available parking spaces
CN107728874A (en) The method, apparatus and equipment of user prompt operation are provided
Çunkaş et al. Long term electricity demand forecasting in Turkey using artificial neural networks
CN104731607B (en) Terminal Lifelong Learning processing method, device and system
CN107726632B (en) Heating control method and server
EP3822880A1 (en) Load prediction method and apparatus based on neural network
CN103455555A (en) Recommendation method and device based on mobile terminal similarity
CN109472350A (en) A kind of neural network acceleration system based on block circulation sparse matrix
CN109188924A (en) The power consumption control method and device of smart home system
CN110442689A (en) A kind of question and answer relationship sort method, device, computer equipment and storage medium
CN106374933A (en) Data compression device and method
CN105432038A (en) Application ranking calculating apparatus and usage information collecting apparatus
CN106548252A (en) Based on the distribution network structure constructing system for improving K means clusters
CN107818380B (en) Information processing method and server
CN111539841A (en) Electric quantity prediction method, device, equipment and readable storage medium
Konecny et al. Simulation of a daytime-based Q-learning control strategy for environmental Harvesting WSN nodes
CN110084407A (en) Load forecasting method and device based on Recognition with Recurrent Neural Network and meta learning strategy
CN108332423A (en) Information control method, server and computer-readable medium
CN109063900A (en) The training of user's conversion ratio prediction model and user's conversion ratio prediction technique and device
CN109190039A (en) Determine the method, apparatus and computer readable storage medium of analogical object
CN109116967A (en) Based on time-multiplexed neural network prediction device and its applicable electronic equipment
CN109062963A (en) From Media Recommendation Method, device and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20180522

RJ01 Rejection of invention patent application after publication