CN104866926A - Fault amount predicting method of power distribution network based on meteorological factors and time sequence analysis - Google Patents
Fault amount predicting method of power distribution network based on meteorological factors and time sequence analysis Download PDFInfo
- Publication number
- CN104866926A CN104866926A CN201510289912.XA CN201510289912A CN104866926A CN 104866926 A CN104866926 A CN 104866926A CN 201510289912 A CN201510289912 A CN 201510289912A CN 104866926 A CN104866926 A CN 104866926A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- network failure
- model
- time series
- fault amount
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a fault amount predicting method of the power distribution network based on meteorological factors and time sequence analysis. The method comprises the steps that 1) external meteorological data in the prediction period is obtained to determine the season of the prediction period, a prediction regression model of the fault amount of the power distribution network is established correspondingly according to the season, and the fault amount included by the external meteorological factors is obtained; 2) an ARIMA model of the fault amount of the power distribution network is established, and the fault amount influenced by factors except the external meteorological factors is obtained; and 3) the fault amount in the step 1) is added to the fault amount in the step 2) to obtain a final predication value of the fault amount of the power distribution network. Compared with the prior art, the prediction precision of the fault amount is high.
Description
Technical field
The present invention relates to distribution network failure quantitative forecast technical field, especially relate to a kind of distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis.
Background technology
Along with the development of China's economic society, the requirement of client to power supply reliability is more and more higher.Power distribution network is as the end Network of whole electric system, complex structure, in large scale, and electric company will configure ample resources every day and solve distribution network failure.For reaching higher electric service level, shorten the breakdown repair time as far as possible, electric company needs the look-ahead number of faults of follow-up some days, so that the resource of config failure repairing in advance.Therefore, realize the short-term forecasting comparatively accurately of distribution network failure quantity, to raising electric service level, promote the repairing level of resources utilization, significant.
At present, both at home and abroad mainly concentrate on localization of fault and diagnosis for the research of distribution network failure aspect, rush to repair task matching, breakdown repair strategy and rush to repair the aspect such as optimization in flow process and path, the research for distribution network failure quantitative forecast problem is relatively less.
Summary of the invention
Object of the present invention is exactly provide a kind of distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis to overcome defect that above-mentioned prior art exists.
Object of the present invention can be achieved through the following technical solutions:
Based on a distribution network failure quantitative forecasting technique for meteorologic factor and time series analysis, comprise the following steps:
1) obtain predictive period outside weather data and judge season residing for predictive period, set up corresponding distribution network failure quantitative forecast regression model according to residing season, obtain the number of faults under being affected by outside weather factor;
2) set up distribution network failure quantitative forecast ARIMA model, obtain the number of faults under the other factors impact beyond by outside weather factor;
3) by step 1) and step 2) number of faults that obtains sums up, and obtains final Distribution Network Failure quantitative forecast value.
Step 1) in, described season comprises winter, summer and spring and autumn, judges that residing for predictive period, season is specially according to predictive period outside weather data:
If lowest temperature mean value was lower than 10 DEG C in 7 days, be then judged to be winter;
If highest temperature mean value was higher than 26 DEG C in 7 days, be then judged to be winter;
Other is then spring and autumn.
Step 1) in, described distribution network failure quantitative forecast regression model comprise forecast model in winter, summer forecast model and spring and autumn forecast model.
In described winter forecast model, with Distribution Network Failure quantity for dependent variable, take daily minimal tcmperature as independent variable;
In described spring and autumn forecast model, with Distribution Network Failure quantity for dependent variable, take daily mean temperature as independent variable;
In described summer forecast model, with Distribution Network Failure quantity for dependent variable, with daily maximum temperature and thunderstorm weather for independent variable.
Described thunderstorm weather enters in forecast model in summer with dummy variable form, described dummy variable D
thunderstormbe specially:
Described step 2) in, set up distribution network failure quantitative forecast ARIMA model and be specially:
201) the sample fault amount time series in setting-up time section and corresponding outside weather data are obtained, described outside weather data are substituted in distribution network failure quantitative forecast regression model, obtain the sample fault amount time series under being affected by outside weather factor;
202) reject the sample fault amount time series under being affected by outside weather factor in described fault sample data, obtain the sample fault amount time series under the other factors impact beyond by outside weather factor;
203) whether the sample fault amount time series under the impact of described other factors is steady to adopt unit root test method to judge, if, then perform step 204), if not, then step 204 is performed after stationarity conversion being carried out on the sample fault amount time series under described other factors impact);
204) according to step 203) the stable sample fault amount time series that obtains chooses required optimum prediction model.
Described forecast model comprises AR model, MA model or ARIMA model.
Compared with prior art, the present invention has the following advantages:
(1) the present invention adopts innovative criterion foundation in season for the distribution network failure quantitative forecast regression model of Various Seasonal, and the number of faults precision of prediction is high;
(2) the present invention is directed to the residue fault amount rejecting meteorological factor influence, build ARIMA time series predicting model, catch the time series variation trend of fault amount, effectively predict the number of faults under being affected by the other factors beyond outside weather factor;
(3) the present invention can realize the short-term forecasting of distribution network failure quantity degree of precision accurately.
Accompanying drawing explanation
Fig. 1 is schematic flow sheet of the present invention;
Fig. 2 is Distribution Network Failure volume trends figure;
Fig. 3 is the correlationship schematic diagram of weather and fault amount;
Fig. 4 is the correlationship schematic diagram of wind-force and fault amount;
Fig. 5 is Regression Model Simulator result schematic diagram in point season;
Fig. 6 is the fault amount sequence auto-correlation and the partial autocorrelation figure that reject meteorological factor influence;
Fig. 7 is ARIMA (3,0,4) model autocorrelation of residuals and partial autocorrelation figure;
Fig. 8 is total breakdown quantity approximating and forecasting situation schematic diagram.
Embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is described in detail.The present embodiment is implemented premised on technical solution of the present invention, give detailed embodiment and concrete operating process, but protection scope of the present invention is not limited to following embodiment.
The present embodiment provides a kind of distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis, comprehensive employing multiple regression and time series analysis means, build the meteorological effect fault amount forecast model in point season, determine the quantitative relationship of the meteorologic factor such as temperature, weather and fault amount, and for rejecting the residue fault amount of meteorological factor influence, build ARIMA time series predicting model, catch the time series variation trend of fault amount.By the integrated application of above-mentioned model, achieve the short-term forecasting of distribution network failure quantity degree of precision.
Cause the factor of Distribution Network Failure a lot, two classes can be divided into substantially: a class is external environmental factor, the weather conditions such as such as temperature, wind-force, sleet; Another kind of is the factor of equipment itself, such as device type, manufacturer, performance parameter, the operation time limit, maintaining etc.
External environmental factor relatively easily quantizes, and the influence factor of equipment itself is because device type is various, wide material sources, performance parameter differ, data accumulation is deficient, is difficult to carry out refinement and quantification, and then cannot enters forecast model.Therefore, we select to adopt compromise proposal, namely first by regression model determination outside weather factor on the impact of Distribution Network Failure quantity, reject the number of faults that outside weather factor is explained afterwards, again time series modeling is carried out to the unaccounted number of faults of residue, adopt ARIMA method to predict, finally adopt reverse operating, the number of faults explain meteorologic factor and the number of faults of time series forecasting sum up, and namely obtain final Distribution Network Failure quantitative forecast value.
The present embodiment reports data for repairment based on East China city distribution network failure, carries out distribution network failure quantity short-term forecasting research.The present embodiment adopt the sample interval of data to be on Dec 31,1 day to 2014 January in 2014, data acquisition is in units of sky.Fault data reports record (rejecting client's wrong report and client's internal fault) for repairment from the Distribution Network Failure in East China city electric power company's T CM system, and weather data is from China Meteorological Administration.Use analysis software is Excel2007 and EViews8.
As shown in Figure 1, the method comprises the following steps:
Step S1, obtains predictive period outside weather data and judges season residing for predictive period, set up corresponding distribution network failure quantitative forecast regression model according to residing season, obtain the number of faults under being affected by outside weather factor;
Step S2, sets up distribution network failure quantitative forecast ARIMA model, obtains the number of faults under the other factors impact beyond by outside weather factor;
Step S3, sums up the number of faults that step S1 and step S2 obtains, obtains final Distribution Network Failure quantitative forecast value.
1, distribution network failure quantitative forecast regression model
Distribution Network Failure quantity has trend in obvious season, as shown in Figure 2, and under Various Seasonal temperature on number of faults to affect form different from direction, therefore need study the relation of temperature factor and fault amount point season.Described season comprises winter, summer and spring and autumn, judges that season residing for predictive period is specifically as shown in table 1 according to predictive period outside weather data:
Table 1 judgment basis in season
(1) distribution network failure quantitative forecast regression model independent variable is chosen
According to origin peculiarity and the practical experience of Distribution Network Failure, preliminary selected temperature, weather (without rain, light rain, moderate rain, heavy rain, heavy rain, thunder shower), wind-force are as the meteorologic factor affecting number of faults, analyze the correlationship of they and fault amount, select have the meteorologic factor of significant correlation to enter forecast model with fault measurer.
1) correlation analysis of temperature factor and fault amount
Under Various Seasonal, the correlationship of temperature and fault amount is as shown in table 2.The absolute value of all related coefficients, all more than 0.5, shows that temperature and fault amount have stronger correlativity, positive and negative as can be seen from related coefficient, winter and the lower fault of spring and autumn temperature more, the higher fault of summer temp is more.
The related coefficient of table 2 temperature and fault amount
2) weather conditions variance analysis that fault amount is affected
Weather conditions are qualitative variable, with the correlationship of fault amount as shown in Figure 3.We adopt variance analysis method to judge, and whether this factor and fault amount have significant correlation.First to the one-way analysis of variance carried out without the fault amount under rain, light rain, moderate rain, heavy rain (rainstorm weather only occurred one day in 2014, and its data are incorporated to heavy rain category) weather under 0.05 level of significance, result is as shown in table 3.The P-value=0.5857 > 0.05 of group difference, shows without weather conditions such as rain, light rain, moderate rain, heavy rain not remarkable on the impact of fault amount, does not consider in the prediction of consequent malfunction amount.
Table 3 weather conditions (not containing thunderstorm weather) variance analysis
But in above-mentioned weather conditions, add thunderstorm weather, carry out variance analysis again, as shown in table 4, it is 0.0004 that P-value declines to a great extent, much smaller than the level of significance of 0.05, show that thunderstorm weather has appreciable impact to fault measurer, and during sample, thunderstorm weather only appears at summer, therefore considers thunderstorm factor in summer in fault amount forecast model.
Table 4 weather conditions (containing thunderstorm weather) variance analysis
3) wind-force factor variance analysis that fault amount is affected
Wind-force factor is similarly qualitative variable, with the correlationship of fault amount as shown in Figure 4.We to 3 grades and following, 4 grades, 5 grades, 6-7 level wind-force time fault amount carry out under 0.05 level of significance single factor test methods analyst, result is as shown in table 5.P-value=0.2098 > 0.05, shows that wind-force is not remarkable on the impact of fault amount, does not consider in the prediction of consequent malfunction amount.
Table 5 wind-force analysis of variance
(2) foundation of distribution network failure quantitative forecast regression model
Distribution network failure quantitative forecast regression model comprise forecast model in winter, summer forecast model and spring and autumn forecast model, wherein,
In winter forecast model, with Distribution Network Failure quantity for dependent variable, take daily minimal tcmperature as independent variable;
In spring and autumn forecast model, with Distribution Network Failure quantity for dependent variable, take daily mean temperature as independent variable;
In summer forecast model, with Distribution Network Failure quantity for dependent variable, with daily maximum temperature and thunderstorm weather for independent variable.
Thunderstorm weather enters in forecast model in summer with dummy variable form, described dummy variable D
thunderstormbe specially:
The regression result of distribution network failure quantitative forecast regression model is as shown in table 6.
The distribution network failure quantitative forecast regression model in table 6 point season
Known based on regression result, the Significance F value of three seaconal models, all much smaller than 0.05, shows that regression equation is remarkable; The P-value of each regression parameter is all less than 0.05, shows that each independent variable is all very remarkable on the impact of fault amount; The modified R 2 of three seaconal models is all lower, shows that each independent variable is inadequate to the explanation degree of fault amount, and reason is that equipment itself waits other influences factor not enter model in addition.Therefore, we carry out time series forecasting, to improve the precision of prediction of unified model to the unaccounted fault amount of regression model.
2, distribution network failure quantitative forecast ARIMA model
ARIMA model full name autoregression difference moving average model(MA model) is a kind of famous Time Series Forecasting Methods.Its basic thought is, forecasting object to be passed in time and the data sequence formed is considered as a random series, carry out this sequence of approximate description by certain mathematical model, this model is once just can from seasonal effect in time series past value and present value to predict future value after being identified; Its general literary style is ARIMA (p, d, q), and wherein p represents autoregressive process exponent number, and d represents difference order, and q represents moving average process exponent number; Its general type is X
t=(α
1x
t-1+ α
2x
t-3+ ... + α
px
t-p)+(β
1ε
t-1+ β
2ε
t-2+ ... + β
qε
t-q); Its modeling process generally comprises the steps such as sequence stationary process, Model Identification, model testing, models fitting and prediction.
In step S2, set up distribution network failure quantitative forecast ARIMA model and be specially:
201) the sample fault amount time series in setting-up time section and corresponding outside weather data are obtained, described outside weather data are substituted in distribution network failure quantitative forecast regression model, obtain the sample fault amount time series under being affected by outside weather factor.
202) reject the sample fault amount time series under being affected by outside weather factor in described fault sample data, obtain the sample fault amount time series under the other factors impact beyond by outside weather factor, as shown in Figure 5.
203) whether the sample fault amount time series under the impact of described other factors is steady to adopt unit root test method to judge, if, then perform step 204), if not, then step 204 is performed after stationarity conversion being carried out on the sample fault amount time series under described other factors impact).
We judge the stationarity of the fault amount sequence of rejecting meteorologic factor as shown in Figure 5 by unit root (ADF) inspection.Unit root test result is as shown in table 7, and t statistic is-4.9371, is less than the critical value under 1%, 5%, 10% level of significance, by unit root test, shows that this sequence is a stable time series, without the need to carrying out the conversion such as difference, i.e. d=0.
The fault amount sequence unit root inspection of meteorological factor influence rejected by table 7
204) according to step 203) the stable sample fault amount time series that obtains chooses required optimum prediction model.
When choosing required optimum prediction model, by partial correlation coefficient and the coefficient of autocorrelation judgement of stationary time series.If the partial correlation coefficient of stationary time series is truncation, and coefficient of autocorrelation is hangover, then can conclude that this sequence is applicable to AR model; If the partial correlation coefficient of stationary time series is hangover, and coefficient of autocorrelation is truncation, then can conclude that this sequence is applicable to MA model; If the partial correlation coefficient of stationary time series and coefficient of autocorrelation are all hangovers, then this sequence is applicable to arma modeling.
The auto-correlation of the fault amount sequence of above-mentioned rejecting meteorological factor influence and partial autocorrelation figure are as shown in Figure 6, can find out, coefficient of autocorrelation obviously trails, PARCOR coefficients feature is not obvious, can think 1 rank truncation, also can think hangover, therefore we attempt the various combination of p, q, adopt optimum criterion function method of fixing price, namely AIC criterion selects optimization model.Known by table 8, when p=3, q=4, AIC obtains minimum value, still final selected ARIMA (3,0,4) model.
Table 8 ARIMA Model Selection
Utilize Eviews software to carry out ARIMA (3,0,4) model to calculate, result is as shown in table 9, except the P-value=0.0563 of MA (2), outside 0.05, the P-value of other parameters, all much smaller than 0.05, shows that each estimates of parameters has good conspicuousness.
Table 9 ARIMA (3,0,4) model estimated result
Set up distribution network failure quantitative forecast ARIMA model to need to verify model, judge whether the information of former sequence is extracted fully, being embodied in model residual error item is white-noise process.If model by inspection, then can carry out subsequent prediction.Above-mentioned ARIMA (3,0,4) model residual error is tested, obtains autocorrelation of residuals figure and partial autocorrelation figure (Fig. 7) and unit root test result (table 10).Therefrom known, autocorrelation of residuals and PARCOR coefficients are all in fiducial interval, and the t statistic of residual error unit root test is much smaller than the critical value under each level of significance, therefore, residual error passes through white noise verification, ARIMA (3,0,4) model is effective, and its final expression formula is:
Fault amount
t=-0.3181 fault amount
t-1+ 0.2586 fault amount
t-2+ 0.8033 fault amount
t-3+ 0.8882 ε
t-1+ 0.2047 ε
t-2-0.6931 ε
t-3-0.1846 ε
t-4
The unit root test of table 10 ARIMA (3,0,4) model residual error
3, predict the outcome
Predicted by the sample data of said method to the present embodiment, the approximating and forecasting result of forecast model as shown in Figure 8, can be found out, model captures the basic trend of fault amount change substantially, and overall fit situation is better.The predicted value of model to first week distribution network failure amount in 2015 is respectively: 1843,1925,1767,1533,1141,1509,1599.Forecasting Methodology of the present invention, to raising electric service level, promotes the distribution repairing level of resources utilization, tool significance.
Claims (7)
1., based on a distribution network failure quantitative forecasting technique for meteorologic factor and time series analysis, it is characterized in that, comprise the following steps:
1) obtain predictive period outside weather data and judge season residing for predictive period, set up corresponding distribution network failure quantitative forecast regression model according to residing season, obtain the number of faults under being affected by outside weather factor;
2) set up distribution network failure quantitative forecast ARIMA model, obtain the number of faults under the other factors impact beyond by outside weather factor;
3) by step 1) and step 2) number of faults that obtains sums up, and obtains final Distribution Network Failure quantitative forecast value.
2. the distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis according to claim 1, it is characterized in that, step 1) in, described season comprises winter, summer and spring and autumn, judges that residing for predictive period, season is specially according to predictive period outside weather data:
If lowest temperature mean value was lower than 10 DEG C in 7 days, be then judged to be winter;
If highest temperature mean value was higher than 26 DEG C in 7 days, be then judged to be winter;
Other is then spring and autumn.
3. the distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis according to claim 1, it is characterized in that, step 1) in, described distribution network failure quantitative forecast regression model comprise forecast model in winter, summer forecast model and spring and autumn forecast model.
4. the distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis according to claim 3, is characterized in that, in described winter forecast model, with Distribution Network Failure quantity for dependent variable, take daily minimal tcmperature as independent variable;
In described spring and autumn forecast model, with Distribution Network Failure quantity for dependent variable, take daily mean temperature as independent variable;
In described summer forecast model, with Distribution Network Failure quantity for dependent variable, with daily maximum temperature and thunderstorm weather for independent variable.
5. the distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis according to claim 4, it is characterized in that, described thunderstorm weather enters in forecast model in summer with dummy variable form, described dummy variable D
thunderstormbe specially:
。
6. the distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis according to claim 1, is characterized in that, described step 2) in, set up distribution network failure quantitative forecast ARIMA model and be specially:
201) the sample fault amount time series in setting-up time section and corresponding outside weather data are obtained, described outside weather data are substituted in distribution network failure quantitative forecast regression model, obtain the sample fault amount time series under being affected by outside weather factor;
202) reject the sample fault amount time series under being affected by outside weather factor in described fault sample data, obtain the sample fault amount time series under the other factors impact beyond by outside weather factor;
203) whether the sample fault amount time series under the impact of described other factors is steady to adopt unit root test method to judge, if, then perform step 204), if not, then step 204 is performed after stationarity conversion being carried out on the sample fault amount time series under described other factors impact);
204) according to step 203) the stable sample fault amount time series that obtains chooses required optimum prediction model.
7. the distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis according to claim 6, it is characterized in that, described forecast model comprises AR model, MA model or ARIMA model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510289912.XA CN104866926B (en) | 2015-05-30 | 2015-05-30 | Distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510289912.XA CN104866926B (en) | 2015-05-30 | 2015-05-30 | Distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104866926A true CN104866926A (en) | 2015-08-26 |
| CN104866926B CN104866926B (en) | 2018-09-18 |
Family
ID=53912746
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510289912.XA Active CN104866926B (en) | 2015-05-30 | 2015-05-30 | Distribution network failure quantitative forecasting technique based on meteorologic factor and time series analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104866926B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243449A (en) * | 2015-10-13 | 2016-01-13 | 北京中电普华信息技术有限公司 | Method and device for correcting prediction result of electricity selling amount |
| CN105426991A (en) * | 2015-11-06 | 2016-03-23 | 深圳供电局有限公司 | Method and system for predicting defect rate of transformer |
| CN107993033A (en) * | 2017-11-14 | 2018-05-04 | 广东电网有限责任公司物流服务中心 | A kind of Power Material Forecasting Methodology |
| CN109063924A (en) * | 2018-08-22 | 2018-12-21 | 国网湖南省电力有限公司 | Power distribution network based on meteorological data repairs work order quantitative forecasting technique |
| CN109469896A (en) * | 2018-12-28 | 2019-03-15 | 佛山科学技术学院 | A kind of diagnostic method and system based on time series analysis Industrial Boiler failure |
| CN109657835A (en) * | 2018-10-31 | 2019-04-19 | 中国电力科学研究院有限公司 | A kind of power distribution network area fault number prediction technique and system |
| CN109800995A (en) * | 2019-01-30 | 2019-05-24 | 北京数介科技有限公司 | A kind of grid equipment fault recognition method and system |
| CN110570322A (en) * | 2019-09-17 | 2019-12-13 | 西北农林科技大学 | Agricultural meteorological index insurance rate determination method based on time series simulation |
| CN112015169A (en) * | 2020-10-19 | 2020-12-01 | 金税信息技术服务股份有限公司 | Method, device and equipment for monitoring and maintaining equipment running state of intelligent equipment box |
| CN112187554A (en) * | 2020-12-01 | 2021-01-05 | 北京蒙帕信创科技有限公司 | Operation and maintenance system fault positioning method and system based on Monte Carlo tree search |
| CN112613638A (en) * | 2020-11-30 | 2021-04-06 | 国网北京市电力公司 | Distribution network fault quantity overall trend prediction method, prediction device and processor |
-
2015
- 2015-05-30 CN CN201510289912.XA patent/CN104866926B/en active Active
Non-Patent Citations (2)
| Title |
|---|
| 崔和瑞等: "基于季节ARIMA模型的华北电网售电量预测研究", 《华东电力》 * |
| 段大鹏等: "气象因素对北京电网设备影响的统计规律及时空分布特征", 《高压电器》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105243449A (en) * | 2015-10-13 | 2016-01-13 | 北京中电普华信息技术有限公司 | Method and device for correcting prediction result of electricity selling amount |
| CN105426991A (en) * | 2015-11-06 | 2016-03-23 | 深圳供电局有限公司 | Method and system for predicting defect rate of transformer |
| CN107993033A (en) * | 2017-11-14 | 2018-05-04 | 广东电网有限责任公司物流服务中心 | A kind of Power Material Forecasting Methodology |
| CN109063924A (en) * | 2018-08-22 | 2018-12-21 | 国网湖南省电力有限公司 | Power distribution network based on meteorological data repairs work order quantitative forecasting technique |
| CN109657835A (en) * | 2018-10-31 | 2019-04-19 | 中国电力科学研究院有限公司 | A kind of power distribution network area fault number prediction technique and system |
| CN109469896A (en) * | 2018-12-28 | 2019-03-15 | 佛山科学技术学院 | A kind of diagnostic method and system based on time series analysis Industrial Boiler failure |
| CN109800995A (en) * | 2019-01-30 | 2019-05-24 | 北京数介科技有限公司 | A kind of grid equipment fault recognition method and system |
| CN110570322A (en) * | 2019-09-17 | 2019-12-13 | 西北农林科技大学 | Agricultural meteorological index insurance rate determination method based on time series simulation |
| CN110570322B (en) * | 2019-09-17 | 2023-09-12 | 西北农林科技大学 | Agricultural meteorological index insurance rate calibrating method based on time sequence simulation |
| CN112015169A (en) * | 2020-10-19 | 2020-12-01 | 金税信息技术服务股份有限公司 | Method, device and equipment for monitoring and maintaining equipment running state of intelligent equipment box |
| CN112015169B (en) * | 2020-10-19 | 2021-02-09 | 金税信息技术服务股份有限公司 | Method, device and equipment for monitoring and maintaining equipment running state of intelligent equipment box |
| CN112613638A (en) * | 2020-11-30 | 2021-04-06 | 国网北京市电力公司 | Distribution network fault quantity overall trend prediction method, prediction device and processor |
| CN112187554A (en) * | 2020-12-01 | 2021-01-05 | 北京蒙帕信创科技有限公司 | Operation and maintenance system fault positioning method and system based on Monte Carlo tree search |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104866926B (en) | 2018-09-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104866926A (en) | Fault amount predicting method of power distribution network based on meteorological factors and time sequence analysis | |
| Jamil | Hydroelectricity consumption forecast for Pakistan using ARIMA modeling and supply-demand analysis for the year 2030 | |
| US10962576B2 (en) | Estimation of shading losses for photovoltaic systems from measured and modeled inputs | |
| Mahsin et al. | Modeling rainfall in Dhaka division of Bangladesh using time series analysis | |
| CN105740991B (en) | Climate change prediction method and system based on improved BP neural network fitting of multiple climate modes | |
| US9460478B2 (en) | System and method for wind generation forecasting | |
| Dirnberger et al. | PV module energy rating: opportunities and limitations | |
| CN105740975B (en) | A kind of equipment deficiency assessment and prediction technique based on data correlation relation | |
| US8532808B2 (en) | Systems and methods for measuring and verifying energy savings in buildings | |
| Kang et al. | Big data analytics in China's electric power industry: modern information, communication technologies, and millions of smart meters | |
| Dong et al. | Volatility of electricity price in Denmark and Sweden | |
| CN104252649A (en) | Regional wind power output prediction method based on correlation between multiple wind power plants | |
| US20150142347A1 (en) | Solar Energy Disaggregation Techniques for Whole-House Energy Consumption Data | |
| Nguyen et al. | Multivariate cointegration and causality between electricity consumption, economic growth, foreign direct investment and exports: Recent evidence from Vietnam | |
| CN105488592A (en) | Method for predicting generated energy of photovoltaic power station | |
| CN105023066A (en) | Business expansion analytical prediction system and method based on seasonal adjustment | |
| CN110555556A (en) | distribution network line variable relation identification method based on linear integer programming | |
| Otache et al. | ARMA modelling of Benue River flow dynamics: comparative study of PAR model | |
| Livera et al. | Intelligent cloud-based monitoring and control digital twin for photovoltaic power plants | |
| Jalori et al. | A Unified Inverse Modeling Framework for Whole-Building Energy Interval Data: Daily and Hourly Baseline Modeling and Short-Term Load Forecasting. | |
| Kong et al. | Long term forecast of local electrical demand and evaluation of future impacts on the electricity distribution network | |
| Yang et al. | Short-term wind power prediction based on multi-parameters similarity wind process matching and weighed-voting-based deep learning model selection | |
| CN114399076A (en) | Method for screening electricity stealing suspicion users based on big data analysis | |
| Maalej et al. | Sensor data augmentation strategy for load forecasting in smart grid context | |
| CN104933476A (en) | Power generation asset generating capacity prediction system and method and value estimation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |