CN103093284A - Hourly water consumption forecasting method of island water supply system - Google Patents
Hourly water consumption forecasting method of island water supply system Download PDFInfo
- Publication number
- CN103093284A CN103093284A CN2013100181229A CN201310018122A CN103093284A CN 103093284 A CN103093284 A CN 103093284A CN 2013100181229 A CN2013100181229 A CN 2013100181229A CN 201310018122 A CN201310018122 A CN 201310018122A CN 103093284 A CN103093284 A CN 103093284A
- Authority
- CN
- China
- Prior art keywords
- water consumption
- data
- days
- pattern
- water
- 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
Links
Abstract
The invention discloses an hourly water consumption forecasting method of an island water supply system. Utilizing historical data of past N days, the hourly water consumption forecasting method of the island water supply system creates a historical trend pattern. On the basis, hourly water consumption is forecasted by comparing the water consumption data and the pattern data of the last hours of the previous day. In order to improve precision, the historical trend pattern is combined with a linear shift average method. Final water consumption forecasting data can be obtained by combining the forecasting water consumption obtained by the pattern and the forecasting water consumption obtained by the linear shift average method. The hourly water consumption forecasting method of the island water supply system is high in forecasting accuracy and has a good effect in a situation, wherein the situation is obvious in hour trend law and unobvious in day trend law.
Description
Technical field
The invention belongs to the water resource technical field, especially a kind of freshwater supply to sea island system hourly water consumption Forecasting Methodology.
Background technology
Water consumption is predicted at aspects such as water-supply systems scheduling, physical construction planning, urban industrial structure adjustment all important roles.The water consumption prediction divides by predetermined period can be divided into short-term forecasting and medium-and long-term forecasting.General short-term water consumption prediction can be divided into hourly water consumption prediction and daily water consumption prediction etc.; Medium-term and long-term water consumption prediction can be divided into water consumption per month prediction and year water consumption prediction etc.From prediction principle, can be divided into cause and effect type predicted method and trend extrapolation again.See also from processing observation data angle and can be divided into grey method, time series analysis method etc.
The prediction of short-term water consumption is mainly used in the water supply scheduling of public supply mains.Its accuracy is directly connected to reliability and the superiority of scheduling result.For seawater desalination system, its short-term water consumption prediction used is different on target with the urban water consumption prediction, water consumption predicts the outcome as the initial conditions of desalinization water processed, water supply unit Optimized Operation, in order to reasonably water-making machine group plan for start-up and shut-down, product pond water supply plan in the formulation cycle.
The area of adopting desalination technology to prepare fresh water is mainly the island, although this area's town water can be divided into town water and resident living water, but adopt identical transmission pipeline network, it is less that daily water consumption is subject to the impact of the aspects such as festivals or holidays, water consumption has comparatively significantly trend feature, show take 24 hours as one-period, present periodically variable rule.In one-period (24 hours), water consumption in the morning 9 o'clock to 10 o'clock and at 7 o'clock in evening to 8 o'clock two periods reach water use peak, touch the bottom in 3:00 AM to 5 water consumption, the water consumption of every day presents " M " type and changes.
Summary of the invention
The present invention realizes water processed and for water integrated, needs predict hourly water consumption in order to satisfy desalinization scheduling requirement, instructs seawater desalination system to dispatch, and reaches energy-saving and cost-reducing target and has proposed a kind of freshwater supply to sea island system hourly water consumption Forecasting Methodology.
The present invention adopts historical data to set up historical trend pattern, utilizes on this basis water usage data and the pattern data of last several hours of the previous day relatively to predict hourly water consumption.In order to improve precision, the historical trend pattern that proposes combines with the Linear-moving averaging method, and the prediction water consumption that the prediction water consumption that pattern is obtained and Linear-moving averaging method obtain merges, and obtains final water consumption predicted data.The method precision of prediction is high, for the time trend rule obviously and day unconspicuous occasion of trend has better effect.
The concrete steps of the inventive method are as follows:
Step (1). build initial historical trend pattern
Choose over
It each water usage data constantly, wherein L is the number between 10~15.Each water consumption observed reading constantly is
, here
Be constantly, unit is hour,
,
Be number of days, unit is the sky,
By finding the solution the least square objective function
Obtain
,
Expression the
Hour the water consumption predicted value.Obtain the water consumption predicted value in 24 moment
After, with 24
Couple together, form by these 24
p i The initial historical trend pattern that value consists of.
Step (2). the correction of historical trend pattern and renewal
After obtaining initial historical trend pattern, in order to ensure the real-time of pattern, need every day and upgrade pattern.For this reason, after the water consumption observed reading that obtains the same day, need to add the water consumption predicted value on the same day, then pick out old historical data, step (1) be participated in the water consumption observed reading of each day of calculating for this reason
With obtain
Sequence is asked its degree of correlation, obtains
Individual relevance degree sequence
, reject degree of correlation minimum value corresponding that day of historical data (namely deleting oscillation on large scale, to get rid of the water consumption observation data under large interference), then add the new historical data of a day, recomputate new historical trend pattern, obtain new
Sequence.The degree of correlation
Computing formula is as follows:
Step (3). obtain based on water consumption predicted value under historical trend pattern
After acquisition the historical water usage data and historical trend pattern of Q days, in order to predict each water consumption constantly in Q+1 days, with Q days
In the data of hourly consumption and pattern, corresponding data relatively, obtains the water consumption predicted value of each hour in Q+1 days.Wherein
Represent the 19th~24 hour.Circular is as follows:
If Q days
Hour the water consumption observed reading be
,
, order
,
,
Expression is last
The mean absolute deviation of hour data,
Expression is last
The mean relative deviation of hour data.Adopt the water consumption predicted value of each hour in Q+1 days that historical trend pattern obtains to be
Step (4). adopt linear averaging method prediction hourly water consumption
Q represents the current water consumption cycle;
Expression
The hourly water consumption in the Q+1 days moment;
Expression
The water consumption observed reading in the Q days moment;
nFor the moving window size, generally get 5~7.
Step (5). step (3) and step (4) are obtained
With
Carry out linear data and merge, obtain final water usage data predicted value.The method that adopts linear data to merge, i.e. final water usage data predicted value
,
,
Represent final
The water usage data predicted value in the Q+1 days moment, wherein
Value adopts following rule: if
More than or equal to 0.20,
If
Less than 0.20,
Adopt beneficial effect of the present invention to be: the precision of hourly water consumption prediction obviously improves, and has directive significance for industry and domestic water scheduling on the island.Be applicable to hourly water consumption and have obvious trend feature, relatively large, the regular not strong situation of amplitude of variation before and after its daily water consumption.The method has the advantages such as precision of prediction is high, applicability is strong.
Embodiment:
Step (1). build initial historical trend pattern
Choose over
It each water usage data constantly, wherein L is the number between 10~15.Each water consumption observed reading constantly is
, here
Be constantly, unit is hour,
,
Be number of days, unit is the sky,
In order to obtain best historical trend pattern, at first by finding the solution the least square objective function
Obtain
, here
Expression the
Hour the water consumption predicted value.Obtain the water consumption predicted value in 24 moment
After, with 24
Couple together, form by these 24
p i The initial historical trend pattern that value consists of.
Step (2). the correction of historical trend pattern and renewal
After obtaining initial historical trend pattern, in order to ensure the real-time of pattern, need every day and upgrade pattern.For this reason, after the water consumption observed reading that obtains the same day, need to add the water consumption predicted value on the same day, then pick out old historical data, step (1) be participated in the water consumption observed reading of each day of calculating for this reason
With obtain
Sequence is asked its degree of correlation, obtains
Individual relevance degree sequence
, reject degree of correlation minimum value corresponding that day of historical data (namely deleting oscillation on large scale, to get rid of the water consumption observation data under large interference), then add the new historical data of a day, recomputate new historical trend pattern, obtain new
Sequence.The degree of correlation
Computing formula is as follows:
Step (3). obtain based on water consumption predicted value under historical trend pattern
After acquisition the historical water usage data and historical trend pattern of Q days, in order to predict each water consumption constantly in Q+1 days, with Q days
In the data of hourly consumption and pattern, corresponding data relatively, obtains the water consumption predicted value of each hour in Q+1 days.Wherein
Represent the 19th~24 hour.Circular is as follows:
If Q days
Hour the water consumption observed reading be
,
, order
,
,
Expression is last
The mean absolute deviation of hour data,
Expression is last
The mean relative deviation of hour data.Adopt the water consumption predicted value of each hour in Q+1 days that historical trend pattern obtains to be
Step (4). adopt linear averaging method prediction hourly water consumption
Q represents the current water consumption cycle;
Expression
The hourly water consumption in the Q+1 days moment;
Expression
The water consumption observed reading in the Q days moment;
nFor the moving window size, generally get 5~7.
Step (5). step (3) and step (4) are obtained
With
Carry out linear data and merge, obtain final water usage data predicted value.The method that adopts linear data to merge, i.e. final water usage data predicted value
,
,
Represent final
The water usage data predicted value in the Q+1 days moment, wherein
Value adopts following rule: if
More than or equal to 0.20,
If
Less than 0.20,
Claims (1)
1. freshwater supply to sea island system hourly water consumption Forecasting Methodology is characterized in that the method comprises the following steps:
Step (1). build initial historical trend pattern
Choose over
It each water usage data constantly, wherein L is the number between 10~15; Each water consumption observed reading constantly is
,
Be constantly, unit is hour,
,
Be number of days, unit is the sky,
By finding the solution the least square objective function
Obtain
, here
Expression the
Hour the water consumption predicted value; Obtain the water consumption predicted value in 24 moment
After, with 24
Couple together, form by these 24
p i The initial historical trend pattern that value consists of;
Step (2). the correction of historical trend pattern and renewal
After obtaining initial historical trend pattern, in order to ensure the real-time of pattern, need every day and upgrade pattern; After the water consumption observed reading that obtains the same day, need to add the water consumption predicted value on the same day, then pick out old historical data, step (1) is participated in the water consumption observed reading of each day of calculating for this reason
With obtain
Sequence is asked its degree of correlation, obtains
Individual relevance degree sequence
, reject degree of correlation minimum value corresponding that day of historical data, then add the new historical data of a day, recomputate new historical trend pattern, obtain new
Sequence; The degree of correlation
Computing formula is as follows:
Step (3). obtain based on water consumption predicted value under historical trend pattern
After acquisition the historical water usage data and historical trend pattern of Q days, in order to predict each water consumption constantly in Q+1 days, with Q days
In the data of hourly consumption and pattern, corresponding data relatively, obtains the water consumption predicted value of each hour in Q+1 days; Wherein
Represent the 19th~24 hour; Circular is as follows:
If Q days
Hour the water consumption observed reading be
,
, order
,
,
Expression is last
The mean absolute deviation of hour data,
Expression is last
The mean relative deviation of hour data; Adopt the water consumption predicted value of each hour in Q+1 days that historical trend pattern obtains to be
Step (4). adopt linear averaging method prediction hourly water consumption
Q represents the current water consumption cycle;
Expression
The hourly water consumption in the Q+1 days moment;
Expression
The water consumption observed reading in the Q days moment;
nFor the moving window size, generally get 5~7;
Step (5). step (3) and step (4) are obtained
With
Carry out linear data and merge, obtain final water usage data predicted value; The method that adopts linear data to merge, i.e. final water usage data predicted value
,
,
The water usage data predicted value that represents the final Q+1 days moment of i, wherein
Value adopts following rule: if
More than or equal to 0.20,
If
Less than 0.20,
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013100181229A CN103093284A (en) | 2013-01-18 | 2013-01-18 | Hourly water consumption forecasting method of island water supply system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2013100181229A CN103093284A (en) | 2013-01-18 | 2013-01-18 | Hourly water consumption forecasting method of island water supply system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103093284A true CN103093284A (en) | 2013-05-08 |
Family
ID=48205830
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2013100181229A Pending CN103093284A (en) | 2013-01-18 | 2013-01-18 | Hourly water consumption forecasting method of island water supply system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103093284A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN106228460A (en) * | 2015-06-02 | 2016-12-14 | Ls产电株式会社 | Method of supplying water |
CN106839468A (en) * | 2017-04-14 | 2017-06-13 | 广州机智云物联网科技有限公司 | A kind of solar water heater runoff investigation method and system |
CN108302768A (en) * | 2017-12-29 | 2018-07-20 | 深圳和而泰数据资源与云技术有限公司 | Based on the real time checking method, device and storage medium with water behavior |
CN111259334A (en) * | 2020-01-14 | 2020-06-09 | 杭州电子科技大学 | Monitoring and early warning method for water use abnormity of large users of industrial enterprises |
CN113551296A (en) * | 2021-06-21 | 2021-10-26 | 顺德职业技术学院 | Daily water consumption adjusting method based on periodic variation |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002324112A (en) * | 2001-04-25 | 2002-11-08 | Osaka Gas Co Ltd | Prediction system and method for utility consumption |
CN102629106A (en) * | 2012-04-11 | 2012-08-08 | 广州东芝白云自动化系统有限公司 | Water supply control method and water supply control system |
-
2013
- 2013-01-18 CN CN2013100181229A patent/CN103093284A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002324112A (en) * | 2001-04-25 | 2002-11-08 | Osaka Gas Co Ltd | Prediction system and method for utility consumption |
CN102629106A (en) * | 2012-04-11 | 2012-08-08 | 广州东芝白云自动化系统有限公司 | Water supply control method and water supply control system |
Non-Patent Citations (2)
Title |
---|
王剑俊: "《城市供水系统短期负荷预测方法的研究》", 《万方学位论文》, 27 February 2006 (2006-02-27) * |
蒋镇军: "《海水淡化优化运行与调度系统的研究与开发》", 《万方学位论文》, 31 July 2012 (2012-07-31), pages 1 - 13 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
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 |
CN106228460A (en) * | 2015-06-02 | 2016-12-14 | Ls产电株式会社 | Method of supplying water |
CN106839468A (en) * | 2017-04-14 | 2017-06-13 | 广州机智云物联网科技有限公司 | A kind of solar water heater runoff investigation method and system |
CN108302768A (en) * | 2017-12-29 | 2018-07-20 | 深圳和而泰数据资源与云技术有限公司 | Based on the real time checking method, device and storage medium with water behavior |
CN111259334A (en) * | 2020-01-14 | 2020-06-09 | 杭州电子科技大学 | Monitoring and early warning method for water use abnormity of large users of industrial enterprises |
CN111259334B (en) * | 2020-01-14 | 2023-06-23 | 杭州电子科技大学 | Abnormal water consumption monitoring and early warning method for large users of industrial enterprises |
CN113551296A (en) * | 2021-06-21 | 2021-10-26 | 顺德职业技术学院 | Daily water consumption adjusting method based on periodic variation |
CN113551296B (en) * | 2021-06-21 | 2022-06-07 | 顺德职业技术学院 | Daily water consumption adjusting method based on periodic variation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103093284A (en) | Hourly water consumption forecasting method of island water supply system | |
Pinson et al. | Probabilistic forecasting of the wave energy flux | |
Santhosh et al. | Real-time economic dispatch for the supply side of the energy-water nexus | |
Saif et al. | Water security in the GCC countries: Challenges and opportunities | |
Soares et al. | Modeling and forecasting short-term electricity load: A comparison of methods with an application to Brazilian data | |
US20180340515A1 (en) | Adaptive power generation management | |
Alam | Coping with blackouts: Power outages and firm choices | |
CN105323321A (en) | Water networking system | |
CN103279804A (en) | Super short-period wind power prediction method | |
CN105447594A (en) | Electric power system grey load prediction method based on exponential smoothing | |
CN108520165A (en) | Rainfall Runoff Forecasting | |
CN104050515A (en) | Water resource allocation method based on annual sequential decision | |
CN103996071A (en) | Wind power plant wind speed prediction method based on Markov theory | |
Smith et al. | Reducing energy use for water supply to China’s high-rises | |
Lam et al. | Life-cycle energy impacts for adapting an urban water supply system to droughts | |
CN107330538B (en) | Method for compiling reservoir adaptive scheduling rules under climate change condition | |
CN111008727A (en) | Power distribution station load prediction method and device | |
CN103276704A (en) | Determination method of hydropower station water storing and releasing dispatch and operation scheme based on energy storage analysis | |
Spencer et al. | Selecting sensors when forecasting temperature in smart buildings | |
Fane et al. | Integrated water resource planning in the context of climate uncertainty | |
Smith et al. | Research and innovation in agricultural water management for a water‐secure world | |
JP6829574B2 (en) | Water operation control system, water operation control method, and water operation control device | |
Ogiołda et al. | The irregularity of water consumption in settlements with varying numbers of inhabitants | |
Arampatzis et al. | A water demand forecasting methodology for supporting day-to-day management of water distribution systems | |
Nguyen et al. | Renewable Powered Desalination in the Coastal Mekong Delta |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130508 |