CN110070203A - A kind of forecasting providing-water method, system, device and storage medium - Google Patents

A kind of forecasting providing-water method, system, device and storage medium Download PDF

Info

Publication number
CN110070203A
CN110070203A CN201910139355.1A CN201910139355A CN110070203A CN 110070203 A CN110070203 A CN 110070203A CN 201910139355 A CN201910139355 A CN 201910139355A CN 110070203 A CN110070203 A CN 110070203A
Authority
CN
China
Prior art keywords
interval
section
temperature
regression
algorithm model
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
Application number
CN201910139355.1A
Other languages
Chinese (zh)
Other versions
CN110070203B (en
Inventor
陈冬雷
周毓
饶明明
罗宇鹏
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.)
GUANGZHOU RUNNING WATER CORP
Original Assignee
GUANGZHOU RUNNING WATER CORP
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 GUANGZHOU RUNNING WATER CORP filed Critical GUANGZHOU RUNNING WATER CORP
Priority to CN201910139355.1A priority Critical patent/CN110070203B/en
Publication of CN110070203A publication Critical patent/CN110070203A/en
Application granted granted Critical
Publication of CN110070203B publication Critical patent/CN110070203B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00Administration; Management
    • G06Q10/04Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06Energy or water supply

Landscapes

  • Business, Economics & Management (AREA)
  • Engineering & Computer Science (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Strategic Management (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Marketing (AREA)
  • General Physics & Mathematics (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Quality & Reliability (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Development Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Public Health (AREA)
  • Water Supply & Treatment (AREA)
  • General Health & Medical Sciences (AREA)
  • Primary Health Care (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a kind of forecasting providing-water method, system, device and storage mediums, this method comprises: carrying out business feature matching treatment according to business feature, obtain the corresponding first interval of the business feature;Temperature Matching processing is carried out according to temperature, to obtain the corresponding second interval of the temperature from the first interval obtained;The corresponding interval algorithm model of second interval of the acquisition is obtained, and in the interval algorithm model that temperature and/or meteorological factor input are obtained, to calculate prediction water supply.By using the present invention, more scientific and reasonable algorithm model can be obtained according to business feature, so that prediction water supply is easily calculated, so that prediction water supply is more accurate.The present invention can be widely applied in computer assistant applications field as a kind of forecasting providing-water method, system, device and storage medium.

Description

A kind of forecasting providing-water method, system, device and storage medium
Technical field
The present invention relates to computer assistant applications field more particularly to a kind of forecasting providing-water method, system, device and Storage medium.
Background technique
In general, weather is hotter more with water, more cold then fewer with water.The meter that supplies water is making in the production division of water undertaking When drawing, daily scheduling scheme needs to predict water according to temperature forecast, this is practice.In addition to this, meteorological Middle other factors may also have an impact to the way of water.Other than the prediction to scheduling scheme entirety water, in zone metering In DMA, if calculate water and actual amount of water different cause, illustrate that area's water is abnormal.
However the relationship of temperature and water is not fixed and invariable, by the data verification of extreme case, the two is one As in the case of there is no too big correlation, be only more than it is certain it is extreme in the case where have dependence association, therefore it is bright between them True dependence should have a section.In addition, each department, each subregion are due to population distribution, the difference of city layout, it is such Dependence is also different.Problem is to be transported with computer when doing scheduling scheme or assessment DMA water daily It calculates, therefore is only capable of providing dependence to extreme case, operation is just artificially incorporated into when judging extreme case, this is to automatic It realizes to calculate and not help.It it would therefore be desirable to have a kind of mode, can not only include the case where handmarking's technical dates, but also can include The specific data (including: temperature, meteorological factor, water supply etc.) on the same day, then transport regression model intrinsic in this way Remove prediction water supply.Long holidays, the special large-scale activity having every year, water supply in even special circumstances, such as the Spring Festival, year It is possible that there is certain association with meteorological factor in these cases.
Regression analysis is to analyze the most popular method of meteorological factor and water consumption relationship.But, with which kind of regression analysis Preferably, this there is no final conclusion, because there is too many complex situations here.Therefore common method is typically all in some typical meanings Local section in justice is analyzed, such as situation of the Shanghai City for 20 degree of mean temperature or more, Xi'an exclusion winter Then the case where heating is opened in its interior selectes a kind of regression model again and goes to analyze.
Existing regression analysis model specifically includes that when dependent variable one or the other (for example water add drop is as dependent variable), It is analyzed using logistic regression algorithm;When linear relationship is presented between variable, analyzed using linear regression algorithm;If between variable not With linear relationship, but there is high-order characteristic (general in water relationship only to calculate to second order), using polynomial regression algorithm Analysis;It is coefficient for multiple factors, it is analyzed using the Stepwise Regression Algorithm.But existing means do not account for water supply industry Influence of the characteristic of being engaged in water supply, therefore have the shortcomings that predict that precision is lower.
Summary of the invention
In order to solve the above-mentioned technical problem, the object of the present invention is to provide a kind of forecasting providing-water method, system, device with And storage medium.
On the one hand, the present invention provides a kind of forecasting providing-water methods, comprising the following steps:
Business feature matching treatment is carried out according to business feature, obtains the corresponding first interval of the business feature;
Temperature Matching processing is carried out according to temperature, to obtain corresponding secondth area of the temperature from the first interval obtained Between;
The corresponding interval algorithm model of second interval of the acquisition is obtained, and temperature and/or meteorological factor input are obtained In the interval algorithm model obtained, to calculate prediction water supply.
Further, described that business feature matching treatment is carried out according to business feature, obtain the business feature corresponding the The step for one section, specifically:
According to business feature, matches and filter out and the business feature phase from several in advance ready-portioned first interval Corresponding first interval;
Several described preparatory ready-portioned first intervals are in the time where the business feature of history water supply data Several sections marked off in section, the first interval and business feature correspond.
Further, described that Temperature Matching processing is carried out according to temperature, to obtain the temperature from the first interval obtained The step for corresponding second interval, specifically:
Several ready-portioned second intervals in advance, several described second interval packets are obtained from the first interval obtained Contained in first interval;
According to temperature, matching filters out the temperature and is fallen into from several obtained preparatory ready-portioned second intervals Second interval;
Several described preparatory ready-portioned second intervals are the passes of temperature and water supply according to history water supply data System or temperature and the relationship of meteorological factor and water supply carry out several sections obtained after the screening division processing of section;
The step of section screening divides processing specifically includes: obtaining the corresponding humidity province of each first interval respectively Between, when the corresponding temperature range of the first interval meets logistic regression condition, by the corresponding humidity province of the first interval Between be used as logistic regression section, and calculate and obtain the corresponding regression algorithm model in the logistic regression section, the logistic regression Section is contained in second interval.
Further, the step of section screening divides processing further include:
The temperature range for meeting linear regression condition is filtered out from the first temperature range as linear regression section, and is counted It calculates and obtains the corresponding regression algorithm model in the linear regression section, first temperature range is the corresponding temperature of first interval Section be screened after remaining section, the linear regression section is contained in second interval.
Further, the step of section screening divides processing further include:
The temperature range for meeting quadratic polynomial recurrence condition is filtered out from the first temperature range as quadratic polynomial Section is returned, and calculates and obtains the corresponding regression algorithm model in quadratic polynomial recurrence section, first temperature range It is the remaining section after the corresponding temperature range of first interval is screened, the quadratic polynomial returns section and is contained in the secondth area Between.
Further, the step of section screening divides processing further include:
The temperature range for meeting meteorological factor successive Regression condition is filtered out from the first temperature range as successive Regression Section, and calculate and obtain the corresponding regression algorithm model in the successive Regression section, first temperature range is first interval Corresponding temperature range be screened after remaining section;
History water supply data in the successive Regression section meet successive Regression equation, the successive Regression section packet Contained in second interval, in the successive Regression equation: independent variable is temperature, and dependent variable is water supply, and the variable introduced one by one is Meteorological factor;
Remaining section after the corresponding temperature range of history water supply data is screened is described as unpredictable section Unpredictable section is contained in second interval.
Further, the corresponding interval algorithm model of second interval for obtaining the acquisition, and by temperature and/or meteorology In the interval algorithm model that factor input obtains, thus the step for calculating prediction water supply, specifically:
When acquired second interval is not belonging to unpredictable section, carry out returning judgement processing;
The recurrence judgement processing specifically: judge whether acquired second interval is successive Regression section, if so, The corresponding regression algorithm model in the successive Regression section is obtained, in conjunction with the relevant weather that is related to of regression algorithm model of acquisition The factor obtains in the regression algorithm model of the relevant weather factor input acquisition in temperature and meteorological factor in advance to calculate Survey water supply;Conversely, the corresponding regression algorithm model of the second interval is then obtained, the regression algorithm mould that temperature input is obtained In type, to calculate acquisition prediction water supply.
On the other hand, the present invention provides a kind of forecasting providing-water systems, comprising:
First interval obtains module, and for carrying out business feature matching treatment according to business feature, it is special to obtain the business The corresponding first interval of property;
Second interval obtains module, for carrying out Temperature Matching processing according to temperature, to obtain from the first interval obtained Take the corresponding second interval of the temperature;
Computing module, the corresponding interval algorithm model of second interval for obtaining the acquisition, and by temperature and/or gas In the interval algorithm model obtained as factor input, to calculate prediction water supply.
On the other hand, the present invention also provides a kind of forecasting providing-water devices, comprising:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized A kind of forecasting providing-water method.
On the other hand, the present invention also provides a kind of storage mediums, wherein it is stored with the executable instruction of processor, it is described The executable instruction of processor is used to execute when executed by the processor a kind of forecasting providing-water method.
The beneficial effects of the present invention are: the present invention matches first interval corresponding to business feature according to business feature, The corresponding second interval of temperature is obtained from first interval further according to temperature, so that the interval algorithm model of second interval is obtained, It will finally be calculated in temperature and/or meteorological factor input interval algorithm model, finally obtain prediction water supply, it can be according to business Characteristic carries out classification prediction to water supply, more has specific aim, has the advantages that accuracy is high.
Detailed description of the invention
Fig. 1 is the step flow chart of forecasting providing-water method provided in an embodiment of the present invention;
Fig. 2 is the structural block diagram of forecasting providing-water system provided in an embodiment of the present invention.
Specific embodiment
The present invention is described in further detail in the following with reference to the drawings and specific embodiments.In for the examples below Number of steps is arranged only for the purposes of illustrating explanation, does not do any restriction to the sequence between step, each in embodiment The execution sequence of step can be adaptively adjusted according to the understanding of those skilled in the art.
As shown in Figure 1, the embodiment of the invention provides a kind of forecasting providing-water methods, method includes the following steps:
S101, business feature matching treatment is carried out according to business feature, obtains the corresponding first interval of the business feature;
Specifically, it is found out according to business feature and belongs to the corresponding first interval of the business feature, the business feature includes: The date (such as discovery instrument fault data are wrong, discovery larger flow unauthorized with water etc.) of generation traffic issues, Day (such as hold big sport event, the whole city (or region) sanitary inspection etc.), conventional large-scale thing occur for unconventional large size event Part generation day (such as water factory's wash pool underproduction, large-scale network engineering, booster, or the valve being affected for water pattern is adjusted Deng), it is festivals or holidays, two-day weekend, daily, the first interval is the time interval as unit of day, i.e. section where the date, often A first interval has an one-to-one business feature label therewith.
S102, Temperature Matching processing is carried out according to temperature, it is corresponding to obtain the temperature from the first interval obtained Second interval;
Specifically, the second interval is temperature range, and the temperature is mean daily temperature, has per day temperature every day Degree and date, thus first interval may include multiple second intervals (for example, business feature be daily corresponding first interval just It may include 0-3 DEG C, 3-7 DEG C, 10-15 DEG C, 20-26 DEG C of multiple section, wherein each section is second interval, if to pre- Survey day business feature be it is daily, mean daily temperature is 15 DEG C, that matched second interval of institute is exactly 10-15 DEG C of this humidity province Between).
S103, the corresponding interval algorithm model of second interval for obtaining the acquisition, and temperature and/or meteorological factor is defeated Enter in the interval algorithm model of acquisition, to calculate prediction water supply;
Specifically, the second interval is temperature range, comprising: logistic regression section, linear regression section, secondary multinomial Formula returns types, the second intervals of each type such as section, successive Regression section and unpredictable section and also has difference, the Two sections are different, and corresponding interval algorithm is also may be different, therefore each second interval has corresponding interval algorithm mould Type, the interval algorithm model are regression function, and the regression function is function of the water supply about temperature, include: logic Regression algorithm model, linear regression algorithm model, quadratic polynomial regression algorithm model and meteorological factor the Stepwise Regression Algorithm mould (in meteorological factor the Stepwise Regression Algorithm model, independent variable is temperature to type, and dependent variable is water supply, and the variable introduced one by one is Meteorological factor) etc. types, input temp to logistic regression algorithm model, linear regression algorithm model and secondary multinomial can be passed through Prediction water supply is obtained in formula regression algorithm model, passes through input temp and meteorological factor to meteorological factor the Stepwise Regression Algorithm mould Prediction water supply is obtained in type, the meteorological factor includes: air pressure, humidity, weather, wind-force etc..
It is obtained by above-mentioned, by using the method for the embodiment of the present invention, water supply can be carried out according to business feature Classification prediction, in the identical situation of business feature, the relationship of temperature and water supply more have can regression, exclude simultaneously Influence of the different business characteristic to water supply, so that prediction more has specific aim, forecasting providing-water is more accurate.
It is further used as the preferred embodiment of this method, it is described that business feature matching treatment is carried out according to business feature, it obtains The business feature corresponding first interval the step for S101, specifically:
S1011, according to business feature, match and filter out and the business from several in advance ready-portioned first intervals The corresponding first interval of characteristic;
S1012, several described preparatory ready-portioned first intervals are where the business feature of history water supply data Time interval in several sections for marking off, the first interval and business feature correspond;
Specifically, each first interval is corresponding with business feature, such as: festivals or holidays, this service attribute was corresponding First interval includes 10 days -2 months on the 4th 2 months this time interval, and the corresponding first interval of this service attribute of two-day weekend includes 17 days-2 months on the 16th 2 months, 24 days-2 months on the 23rd 2 months and these time intervals on 2-March of March 3, if prediction day is March 3 Day, then first interval can be obtained according to the business feature of two-day weekend.Different business features cannot mix and do regression analysis, Come even if analyzing, rule is it is also apparent that be wrong.The sequential logic for dividing first interval may is that first dividing generation business asks The corresponding first interval of this business feature of the date of topic, when traffic issues occur, no matter whether subsequent various items Part, the water consumption on this kind of dates will receive very big sporadic influence, therefore this classification is primarily to exclude that The case where can returning a bit, because analysis is there will not be meaning, (for example the data on the day of some DMA (zone metering) subregion will not It brings into calculating);Then it divides unconventional large-scale event and the corresponding first interval of this business feature of day occurs, it is unconventional Large-scale event such case can greatly influence water, can quantization influence as one kind, it is possible to which there will be similar events Occur;Then it divides conventional large-scale event and the corresponding first interval of this business feature of day occurs, the large-scale event of conventional type is Relatively have reference significance because the events such as engineering and booster be it is recurrent, facilitate in these dates happened Predict the variation of water;Followed by festivals or holidays are divided, festivals or holidays usually all contain two-day weekend, therefore first divide festivals or holidays, have Help the continuous sexual intercourse for excluding to have comprising two-day weekend therein and festivals or holidays;Subsequently divide this business feature of two-day weekend Corresponding first interval, two-day weekend will have obvious difference with water on region, therefore can not be considered as daily The case where analyzed;Finally corresponding first interval of remaining date corresponds to this daily business feature.
It is obtained by above-mentioned, by using the method for the embodiment of the present invention, first interval and business feature can be carried out one by one Corresponding, so that prediction is more targeted, and prediction is more scientific and reasonable, has practical significance.
It is further used as the preferred embodiment of this method, it is described Temperature Matching processing is carried out according to temperature, with from obtaining The step for corresponding second interval of the temperature is obtained in first interval S102, specifically:
S1021, obtained from the first interval obtained several in advance ready-portioned second interval, it is described several second Section is contained in first interval;
S1022, according to temperature, from obtain several matching filters out the temperature in ready-portioned second intervals in advance The second interval fallen into;
Specifically, such as business feature is 3-10 DEG C, 10-15 DEG C and 15-18 DEG C three the in daily first interval Two sections, predict day business feature be it is daily, temperature is 12 DEG C, and 12 are greater than 10 and less than 15 known to judgement, then falls into for 12 DEG C Second interval be 10-15 DEG C, above-mentioned temperature is mean daily temperature.
S1023, several described preparatory ready-portioned second intervals are the temperature and water supply according to history water supply data The relationship or temperature and meteorological factor of amount and the relationship of water supply carry out several sections obtained after the screening division processing of section;
Specifically, logistic regression section, linear regression section and quadratic polynomial are divided and returns the division used when section Mode is: according to the relationship of the mean daily temperature of history water supply data and water supply, marking off different sections, divides gradually The division mode used when returning section is the mean daily temperature and water supply and meteorological factor according to history water supply data Relationship marks off the section that can do meteorological factor stepwise regression analysis.
The section screening divides the step S1023 of processing specifically:
S10231, the corresponding temperature range of each first interval is obtained respectively, when the corresponding humidity province of the first interval Between when meeting logistic regression condition, using the corresponding temperature range of the first interval as logistic regression section, and calculate acquisition The corresponding regression algorithm model in the logistic regression section, the logistic regression section is contained in second interval;
Specifically, logistic regression screening is done to first interval, judges whether the corresponding temperature range of first interval can patrol It collects and returns, specific judgement may is that a1, average to the water supply of the history water supply data in first interval;A2, it will count It is divided into two parts according to by water supply greater than average value and less than average value;A3, standard deviation and mean value in two parts are calculated separately Ratio (coefficient of standard deviation);If a4, ratio are less than a certain threshold value (such as: 5%, specific value can adjust according to the actual situation), Then the part can do logistic regression, otherwise the corresponding temperature range in the part is not suitable for logistic regression as logistic regression section.
It is further used as the preferred embodiment of this method, the section screening divides the S1023 of processing, further include:
S10232, the temperature range for meeting linear regression condition is filtered out from the first temperature range as linear regression area Between, and calculate and obtain the corresponding regression algorithm model in the linear regression section, first temperature range is first interval pair The temperature range answered be screened after remaining section, the linear regression section is contained in second interval;
Specifically, linear regression screening is done to the first interval after being screened, judges corresponding to first interval at this time Whether temperature range meets linear regression condition, and specific judgement may is that b1, sort according to temperature to first interval at this time, The window for being successively 5 with width continuously fetches to the first interval after sequence;The temperature and water supply of data in b2, calculation window Related coefficient;If b3, related coefficient absolute value are less than canonical correlation coefficient, (such as: 0.85), window moves back one, until calculating Related coefficient absolute value is greater than canonical correlation coefficient;If b4, can not find always, this first subregion is not suitable for linear regression; If b5, related coefficient absolute value are greater than canonical correlation coefficient, expand window 1, and the related coefficient of calculation window backward;b6, If related coefficient absolute value is greater than canonical correlation coefficient, continue to expand window, until related coefficient absolute value is less than standard Related coefficient bounces back window 1 at this time, then the range of window is the second interval as linear regression section;B7, from section Lower bound starts to create the window that width is 5, continues above-mentioned b2-b7 step, until first interval end (wherein, window width and The specific value of canonical correlation coefficient can adjust according to the actual situation), above-mentioned temperature is mean daily temperature.
It is obtained by above-mentioned, it, can be on the basis of first interval to drawing again by using the method for the embodiment of the present invention Divide second interval, so that interval division is more reasonable, and there is practical significance.
It is further used as the preferred embodiment of this method, the section screening divides the S1023 of processing, further include:
S10233, the temperature range for meeting quadratic polynomial recurrence condition is filtered out from the first temperature range as secondary Polynomial regression section, and calculate and obtain the corresponding regression algorithm model in quadratic polynomial recurrence section, first temperature Degree section is the remaining section after the corresponding temperature range of first interval is screened, and the quadratic polynomial returns section and is contained in Second interval;
Specifically, quadratic polynomial is done to the first interval after being screened and returns screening, judge first interval pair at this time Whether the temperature range answered, which meets quadratic polynomial, returns condition, specific judgement may is that c1, according to temperature to first at this time Section sequence;If the width of c2, first interval at this time carries out quadratic polynomial recurrence less than 5, to entire section, and counts Calculate coefficient of standard deviation;If c3, coefficient of standard deviation are less than coefficient of standard deviation threshold value, (such as: 10%), first interval at this time is corresponding Temperature range be as quadratic polynomial return section second interval;If the width of c4, first interval at this time is greater than 5, then the window that width is 5 is created, quadratic polynomial recurrence is carried out in window, and calculate coefficient of standard deviation;If c5, standard deviation Coefficient is greater than coefficient of standard deviation threshold value, and window is moved backward 1, is returned and calculated coefficient of standard deviation again, until mark Quasi- difference coefficient is less than coefficient of standard deviation threshold value;C6, when coefficient of standard deviation be less than coefficient of standard deviation threshold value when, window is expanded backward It 1, carries out quadratic polynomial recurrence and simultaneously calculates coefficient of standard deviation, until coefficient of standard deviation is greater than coefficient of standard deviation threshold value, then window Mouth retraction 1, then the range of window is the section for being suitble to quadratic polynomial to return;C7, since lower window edge, if subsequent zone Between width less than 5, then execute c1-c3 step, if it is greater than 5, then execute c4-c7 step, be repeated up to first interval end (wherein, The specific value of window width and coefficient of standard deviation threshold value can adjust according to the actual situation), above-mentioned temperature is mean daily temperature. Quadratic polynomial recurrence section, which is added, can make regression analysis specific in further detail, more be of practical significance.
It is further used as the preferred embodiment of this method, the section screening divides the S1023 of processing, further include:
S10234, filtered out from the first temperature range meet meteorological factor successive Regression condition temperature range be used as by Step returns section, and calculates and obtain the corresponding regression algorithm model in the successive Regression section, and first temperature range is the The corresponding temperature range in one section be screened after remaining section;
Specifically, the screening of meteorological factor successive Regression is done to the first interval after being screened, judges first interval at this time Whether corresponding temperature range meets meteorological factor successive Regression condition, specific judgement may is that d1, according to temperature at this time First interval sequence;If the width of d2, first interval at this time less than 5, carries out meteorological factor to entire section and gradually returns Return, and calculates coefficient of standard deviation;If d3, coefficient of standard deviation be less than coefficient of standard deviation threshold value (such as: 10%), the firstth area at this time Between corresponding temperature range be second interval as meteorological factor successive Regression section;If d4, first interval at this time Width is greater than 5, then creates the window that width is 5, meteorological factor successive Regression is carried out in window, and calculate coefficient of standard deviation; If d5, coefficient of standard deviation are greater than coefficient of standard deviation threshold value, window is moved backward 1, carries out meteorological factor successive Regression again And coefficient of standard deviation is calculated, until coefficient of standard deviation is less than coefficient of standard deviation threshold value;D6, when coefficient of standard deviation be less than standard deviation system When number threshold value, window is expanded to 1 backward, carry out meteorological factor successive Regression and calculate coefficient of standard deviation, until standard deviation system Number is greater than coefficient of standard deviation threshold value, then window bounces back 1, then the range of window is to be suitble to the section of meteorological factor successive Regression; D7, since lower window edge, if later span width less than 5, execute d1-d3 step, if it is greater than 5, then execute d4-d7 Step, being repeated up to first interval end, (wherein, the specific value of window width and coefficient of standard deviation threshold value can be according to the actual situation Adjustment), above-mentioned temperature is mean daily temperature.
Wherein, the history water supply data in the successive Regression section meet successive Regression equation, the successive Regression Section is contained in second interval, and in the successive Regression equation: independent variable is temperature, and dependent variable is water supply, is introduced one by one Variable is meteorological factor;
Specifically, meteorological factor herein is not fixed, can be selected according to regressive case.
S10235, the corresponding temperature range of history water supply data is screened after remaining section as unpredictable area Between, the unpredictable section is contained in second interval;
Specifically, will final remaining section as unpredictable section, the second interval belonging to the forecast date is not When predictable section, unpredictable information can return to, perhaps directly prediction terminates or returns 0 and (indicates that the date to be measured can not be into Row prediction).
Obtained by above-mentioned, bring of the embodiment of the present invention have the beneficial effect that meteorological factor successive Regression section is added can be with Alternatively, specific in further detail to the division in section so that regression analysis is more scientific and reasonable, with more practical Meaning
It is further used as the preferred embodiment of this method, the corresponding interval algorithm of second interval for obtaining the acquisition Model, and in the interval algorithm model that temperature and/or meteorological factor input are obtained, to calculate prediction this step of water supply Suddenly, specifically:
S1031, when acquired second interval is not belonging to unpredictable section, carry out return judgement processing;
The recurrence judgement processing specifically:
Whether the acquired second interval of judgement is successive Regression section, if so, obtaining the successive Regression section pair The regression algorithm model answered, in conjunction with the relevant weather factor that is related to of regression algorithm model of acquisition, by temperature and meteorological factor In the relevant weather factor input obtain regression algorithm model in, thus calculate acquisition prediction water supply;
Specifically, the relevant weather factor can be multiple, be also possible to one, after obtaining the Stepwise Regression Algorithm model Independent variable involved by the Stepwise Regression Algorithm model is obtained, is then inputted according to the corresponding parameter of forecast date into gradually In regression algorithm model, calculates and obtain prediction water supply.Such as: obtained the Stepwise Regression Algorithm model are as follows: prediction water supply is (single Position: ton)=20* temperature (unit: DEG C)+20* humidity+1.5* wind intensity+20, then can be according to the temperature (12 of forecast date DEG C), humidity (30%) and wind intensity (6), it is 275 tons that final prediction water supply, which is calculated, above-mentioned temperature be day it is flat Equal temperature.
Conversely, the corresponding regression algorithm model of the second interval is then obtained, the regression algorithm mould that temperature input is obtained In type, to calculate acquisition prediction water supply;
Specifically, different sections corresponds to different interval algorithm models, such as: predict day business feature be it is daily, Temperature be 15 DEG C, matched second interval be 10-16 DEG C, the corresponding algorithm model of second interval be prediction water supply (unit: Ton)=15* temperature (unit: DEG C)+7, then can calculate water supply is 232 tons;Another example is: the business feature of prediction day is Two-day weekend, temperature are 15 DEG C, and matched second interval is 12-18 DEG C, and the corresponding algorithm model of second interval is that prediction supplies at this time Water (unit: ton)=2* temperature (unit: DEG C) ^2-35, then can calculated water supply be 415 tons (citing, which is only done, to be joined Examine), above-mentioned temperature is mean daily temperature.
It is obtained by above-mentioned, bring of the embodiment of the present invention, which has the beneficial effect that, to use input not according to interval algorithm difference With value obtain prediction water supply, prediction more has practical significance.
As shown in Fig. 2, the embodiment of the invention also provides a kind of forecasting providing-water systems, comprising:
First interval obtains module 201, for carrying out business feature matching treatment according to business feature, obtains the business The corresponding first interval of characteristic.
Second interval obtains module 202, for carrying out Temperature Matching processing according to temperature, from the first interval obtained Obtain the corresponding second interval of the temperature.
Computing module 203, the corresponding interval algorithm model of second interval for obtaining the acquisition, and by temperature and/ Or in the interval algorithm model of meteorological factor input acquisition, to calculate prediction water supply.
Suitable for this system embodiment, this system embodiment is implemented content in above method embodiment Function is identical as above method embodiment, and the beneficial effect reached and above method embodiment beneficial effect achieved It is identical.
Based on forecasting providing-water method shown in FIG. 1, the embodiment of the invention also provides a kind of forecasting providing-water devices, should Device includes:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized A kind of above-mentioned forecasting providing-water method.
Suitable for present apparatus embodiment, present apparatus embodiment is implemented content in above method embodiment Function is identical as above method embodiment, and the beneficial effect reached and above method embodiment beneficial effect achieved It is identical.
In addition, the embodiment of the invention also provides a kind of storage mediums, wherein being stored with the executable instruction of processor, institute It states the executable instruction of processor and is used to execute any of the above-described kind of forecasting providing-water method when executed by the processor.
It is to be illustrated to preferable implementation of the invention, but the invention is not limited to the implementation above Example, those skilled in the art can also make various equivalent variations on the premise of without prejudice to spirit of the invention or replace It changes, these equivalent deformations or replacement are all included in the scope defined by the claims of the present application.

Claims (10)

1. a kind of forecasting providing-water method, it is characterised in that: the following steps are included:
Business feature matching treatment is carried out according to business feature, obtains the corresponding first interval of the business feature;
Temperature Matching processing is carried out according to temperature, to obtain the corresponding second interval of the temperature from the first interval obtained;
The corresponding interval algorithm model of second interval of the acquisition is obtained, and temperature and/or meteorological factor input are obtained In interval algorithm model, to calculate prediction water supply.
2. forecasting providing-water method according to claim 1, it is characterised in that: described to carry out business spy according to business feature Property matching treatment, the step for obtaining the business feature corresponding first interval, specifically:
According to business feature, matching is filtered out corresponding with the business feature from several in advance ready-portioned first interval First interval;
Several described preparatory ready-portioned first intervals are in the time interval where the business feature of history water supply data In several sections for marking off, the first interval and business feature correspond.
3. forecasting providing-water method according to claim 2, it is characterised in that: described to be carried out at Temperature Matching according to temperature Reason, the step for obtain the corresponding second interval of the temperature from the first interval obtained, specifically:
Several are obtained from the first interval obtained, and ready-portioned second interval, several described second intervals are contained in advance First interval;
According to temperature, from obtain several matching filters out that the temperature fallen into ready-portioned second intervals in advance the Two sections;
It is described several in advance ready-portioned second intervals be according to the temperature of history water supply data and the relationship of water supply or Temperature and the relationship of meteorological factor and water supply carry out several sections obtained after the screening division processing of section;
The step of section screening divides processing specifically includes: the corresponding temperature range of each first interval is obtained respectively, when When the corresponding temperature range of the first interval meets logistic regression condition, using the corresponding temperature range of the first interval as Logistic regression section, and calculate and obtain the corresponding regression algorithm model in the logistic regression section, the logistic regression section packet Contained in second interval.
4. forecasting providing-water method according to claim 3, it is characterised in that: the step of section screening divides processing Further include:
The temperature range for meeting linear regression condition is filtered out from the first temperature range as linear regression section, and is calculated and obtained The corresponding regression algorithm model in the linear regression section is obtained, first temperature range is the corresponding temperature range of first interval Remaining section after being screened, the linear regression section are contained in second interval.
5. forecasting providing-water method according to claim 3, it is characterised in that: the step of section screening divides processing Further include:
The temperature range for meeting quadratic polynomial recurrence condition is filtered out from the first temperature range as quadratic polynomial recurrence Section, and calculate and obtain the quadratic polynomial and return the corresponding regression algorithm model in section, first temperature range are the The corresponding temperature range in one section be screened after remaining section, the quadratic polynomial returns section and is contained in second interval.
6. forecasting providing-water method according to claim 3, it is characterised in that: the step of section screening divides processing Further include:
It is filtered out from the first temperature range and meets the temperature range of meteorological factor successive Regression condition as successive Regression section, And calculate and obtain the corresponding regression algorithm model in the successive Regression section, first temperature range is that first interval is corresponding Temperature range be screened after remaining section;
History water supply data in the successive Regression section meet successive Regression equation, and the successive Regression section is contained in Second interval, in the successive Regression equation: independent variable is temperature, and dependent variable is water supply, and the variable introduced one by one is meteorology The factor;
Remaining section after the corresponding temperature range of history water supply data is screened as unpredictable section, it is described can not Forecast interval is contained in second interval.
7. forecasting providing-water method according to claim 6, it is characterised in that: the second interval for obtaining the acquisition Corresponding interval algorithm model, and in the interval algorithm model that temperature and/or meteorological factor input are obtained, to calculate pre- The step for surveying water supply, specifically:
When acquired second interval is not belonging to unpredictable section, carry out returning judgement processing;
The recurrence judgement processing specifically: judge whether acquired second interval is successive Regression section, if so, obtaining The corresponding regression algorithm model in the successive Regression section, in conjunction with acquisition the relevant weather that is related to of regression algorithm model because Son is predicted in the regression algorithm model of the relevant weather factor input acquisition in temperature and meteorological factor to calculate Water supply;Conversely, the corresponding regression algorithm model of the second interval is then obtained, the regression algorithm model that temperature input is obtained In, to calculate acquisition prediction water supply.
8. a kind of forecasting providing-water system, it is characterised in that: include:
First interval obtains module, for carrying out business feature matching treatment according to business feature, obtains the business feature pair The first interval answered;
Second interval obtains module, for carrying out Temperature Matching processing according to temperature, to obtain institute from the first interval obtained State the corresponding second interval of temperature;
Computing module, the corresponding interval algorithm model of second interval for obtaining the acquisition, and by temperature and/or it is meteorological because In the interval algorithm model that son input obtains, to calculate prediction water supply.
9. a kind of forecasting providing-water device, it is characterised in that: include:
At least one processor;
At least one processor, for storing at least one program;
When at least one described program is executed by least one described processor, so that at least one described processor is realized as weighed Benefit requires a kind of any one of 1-7 forecasting providing-water method.
10. a kind of storage medium, wherein being stored with the executable instruction of processor, it is characterised in that: the processor is executable Instruction be used to execute a kind of forecasting providing-water method as described in claim any one of 1-7 when executed by the processor.
CN201910139355.1A 2019-02-25 2019-02-25 Water supply quantity prediction method, system, device and storage medium Active CN110070203B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910139355.1A CN110070203B (en) 2019-02-25 2019-02-25 Water supply quantity prediction method, system, device and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910139355.1A CN110070203B (en) 2019-02-25 2019-02-25 Water supply quantity prediction method, system, device and storage medium

Publications (2)

Publication Number Publication Date
CN110070203A true CN110070203A (en) 2019-07-30
CN110070203B CN110070203B (en) 2023-12-26

Family

ID=67366005

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910139355.1A Active CN110070203B (en) 2019-02-25 2019-02-25 Water supply quantity prediction method, system, device and storage medium

Country Status (1)

Country Link
CN (1) CN110070203B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115169771A (en) * 2022-01-18 2022-10-11 长安大学 Method for quickly evaluating efficiency damage of traffic network under influence of flood disasters

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105956690A (en) * 2016-04-25 2016-09-21 广州东芝白云自动化系统有限公司 Water supply prediction method and water supply prediction system
CN107273998A (en) * 2016-06-30 2017-10-20 国网江苏省电力公司南通供电公司 A kind of Temperature correction method predicted for platform area daily power consumption
CN107909195A (en) * 2017-11-08 2018-04-13 吴江华衍水务有限公司 A kind of design for commodities method
CN108133283A (en) * 2017-12-11 2018-06-08 中国水利水电科学研究院 Urban water system and the joint response regulation and control method of energy resource system reply climate change
CN109002937A (en) * 2018-09-07 2018-12-14 深圳供电局有限公司 Load Forecasting, device, computer equipment and storage medium

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105956690A (en) * 2016-04-25 2016-09-21 广州东芝白云自动化系统有限公司 Water supply prediction method and water supply prediction system
CN107273998A (en) * 2016-06-30 2017-10-20 国网江苏省电力公司南通供电公司 A kind of Temperature correction method predicted for platform area daily power consumption
CN107909195A (en) * 2017-11-08 2018-04-13 吴江华衍水务有限公司 A kind of design for commodities method
CN108133283A (en) * 2017-12-11 2018-06-08 中国水利水电科学研究院 Urban water system and the joint response regulation and control method of energy resource system reply climate change
CN109002937A (en) * 2018-09-07 2018-12-14 深圳供电局有限公司 Load Forecasting, device, computer equipment and storage medium

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115169771A (en) * 2022-01-18 2022-10-11 长安大学 Method for quickly evaluating efficiency damage of traffic network under influence of flood disasters

Also Published As

Publication number Publication date
CN110070203B (en) 2023-12-26

Similar Documents

Publication Publication Date Title
Xiao et al. An improved seasonal rolling grey forecasting model using a cycle truncation accumulated generating operation for traffic flow
Amral et al. Short term load forecasting using multiple linear regression
CN104899405B (en) Data predication method and system and alarm method and system
CN106779129A (en) A kind of Short-Term Load Forecasting Method for considering meteorologic factor
CN108230049A (en) The Forecasting Methodology and system of order
CN108647891A (en) Data exception classification, Reasons method and device
Krishnamurthy et al. Propagation of uncertainty in transportation land use models: Investigation of dram-empal and utpp predictions in austin, texas
CN111985701B (en) Power consumption prediction method based on power supply enterprise big data model base
CN103413443A (en) Short-term traffic flow forecasting method based on hidden Markov model
CN107644297B (en) Energy-saving calculation and verification method for motor system
CN105184388A (en) Non-linear regression method for urban power load short-period prediction
CN113537563B (en) Pollution emergency management and control effect evaluation method and device
CN115689396B (en) Pollutant emission control method, device, equipment and medium
Kacar et al. A comparison of multiple linear regression approaches for fitting clearing functions to empirical data
CN113988711A (en) Power consumption data-based monitoring method for stopping or limiting production of sewage disposal enterprises in control state
CN116843071B (en) Transportation network operation index prediction method and device for intelligent port
Zhang et al. Hybrid traffic speed modeling and prediction using real-world data
Truong et al. Forecasting time-series energy data in buildings using an additive artificial intelligence model for improving energy efficiency
CN113159374A (en) Data-driven urban traffic flow rate mode identification and real-time prediction early warning method
Zhang et al. An automatic real-time bus schedule redesign method based on bus arrival time prediction
CN107316096A (en) A kind of track traffic one-ticket pass passenger amount of entering the station Forecasting Methodology
CN110070203A (en) A kind of forecasting providing-water method, system, device and storage medium
JP2004023114A (en) Method and system for predicting communication traffic
Zhou et al. Forecasting daily gas load with OIHF-Elman neural network
CN108154259B (en) Load prediction method and device for heat pump, storage medium, and processor

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
CB02 Change of applicant information

Address after: 510660 No.12 Zhongshan 1st Road, Yuexiu District, Guangzhou, Guangdong Province

Applicant after: Guangzhou Water Supply Co.,Ltd.

Address before: 510600, No. 12, Zhongshan Road, Yuexiu District, Guangdong, Guangzhou

Applicant before: GUANGZHOU WATER SUPPLY CO.

CB02 Change of applicant information
GR01 Patent grant
GR01 Patent grant