CN102749664B - Method for predicting icing degree of power gird - Google Patents

Abstract

The invention discloses a method for predicting the icing degree of a power gird, comprising the following steps of: finding out seventeen meteorological impact factors with high relativity by using historical icing data, the geographic information meteorological data, computing the weight coefficient of each meteorological impact factor by an analytic hierarchy process, and synthesizing the seventeen meteorological impact factors according to the weight to predict the icing degree of the power gird. The method can predict the icing degree of the power gird one moth in advance, so that a corresponding emergency disposal plan can be made according to a predicting conclusion, a most economical and feasible power gird icing solution can be selected, the power gird deicing equipment, material assets and team can be reserved, the safe and stable running when the power gird is iced can be guaranteed, the early response to the icing of the power gird can be realized, and the loss caused by the icing of the power gird can be reduced.

Description

Electrical network icing degree Forecasting Methodology

Technical field

The invention belongs to power transmission and distribution technical field, be specifically related to a kind of electrical network icing degree Forecasting Methodology.

Background technology

The safe and stable operation of icing serious threat electrical network, southern ice damage at the beginning of 2008, national grid 220kV and above transmission line of electricity fall tower base more than 1300, more than 120 hundred million yuan of property losses, factory, hospital and residential block have a power failure, the wide electric railway in capital is stopped transport, and social stability and people's production have been formed to serious threat with life.Carry out in advance the long-term forecasting of electrical network icing, can predict the outcome and carry out in advance counter-measure according to electrical network icing, give warning in advance, reduce the loss that electrical network icing causes.

The transmission line of electricity microprocess that icing increases under specific meteorological condition is mainly considered in current powerline ice-covering forecasting research both domestic and external, as representative Imai model, Lenhard model, Goodwin model, Chaine model and Makkonen model, belong to the short-term forecasting of powerline ice-covering rate of rise, and forecast model is very complicated, be only suitable for the short-term icing prediction under supposition meteorological condition, do not explore the formation rule of powerline ice-covering weather.

Summary of the invention

The object of this invention is to provide a kind of electrical network icing degree Forecasting Methodology that can carry out long-term icing prediction.

Electrical network icing degree Forecasting Methodology provided by the invention, comprises the following steps:

(1) Historical Meteorological Information of 60 years and electrical network icing in collection, arrangement, statistical study provincial region

Data, counts average icing number of days, divides electrical network icing degree rank;

(2) according to icing correlation calculations formula, calculate the related coefficient of historical meteorologic parameter and average icing number of days, icing correlation calculations formula is:

$R_X=\frac{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_i{d}_i-\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_i\·\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{d}_i}{\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x_i}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{x}_i\right)}^2\·\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{d_i}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{d}_i\right)}^2}}}$

R in formula _xfor the related coefficient of meteorologic parameter with average icing number of days, x _ifor the value of meteorologic parameter, i is time sequence number, and n is historical time quantity, d _ifor the average icing number of days in a certain time, the icing number of days that average icing number of days is each icing monitoring station is cumulative divided by monitoring station number mutually;

(3) from the result of calculation of step (2), filter out 17 meteorological effect factors that correlativity is the strongest;

(4) utilize statistical study, find out the characteristic rule of 17 described meteorological effect Effects of Factors electrical network icing of step (3);

(5) utilize mathematical statistics, obtain the statistical probability of the various icing degree based on characteristic rule of 17 meteorological effect factors, set up the icing degree occurrence frequency statistical form of each factor characteristic rule;

(6) compile the weather data that needs the prediction electrical network icing degree time, the data in 17 factor prediction times that input is corresponding;

(7) utilize the data in prediction time, comprehensive characteristics rule, and inquire about icing degree occurrence frequency statistical form, show that the slight icing frequency in prediction time is s _i, moderate icing frequency is m _i, serious icing frequency is l _i, especially severe frequency is n _i, i=1,2 ..., 17;

(8) experience according to weather report, provides the weight coefficient of 17 factor pair electrical network icing numbers of days impacts, is respectively ɑ ₁, ɑ ₂... ɑ ₁₇;

(9) utilize weight system, by weighted mean, calculate total prediction probability of the various icing degree of prediction time under 17 Effects of Factors, computing formula is:

Slight icing probability of happening:

Moderate icing probability of happening: $M=\underset{i=1}{\overset{17}{\mathrm{\Σ}}}{a}_i\×m_i$

Serious icing probability of happening:

Especially severe icing probability of happening:

The climate law that the present invention forms from icing is carried out the research of macroscopical aspect, utilize historical icing, geography information meteorological data, find out 17 meteorological effect factors that correlativity is strong, utilize analytical hierarchy process to calculate the weight coefficient of each meteorological effect factor, according to comprehensive 17 the meteorological effect factors of weight, form electrical network icing degree Forecasting Methodology, before annual icing arriving in winter, carry out the prediction of grid power transmission circuit icing degree.To provincial power network region, one month earlier to winter electrical network icing degree predict, solved a difficult problem for electrical network icing long-term forecasting.According to prediction conclusion, carry out corresponding emergency disposal prediction scheme, select the most economically viable electrical network icing counter-measure, deposit electrical network deicing equipment, goods and materials and troop, safe and stable operation while guaranteeing electrical network icing, realizes the reply in advance of electrical network icing, reduces the loss that electrical network icing causes.

Accompanying drawing explanation

Fig. 1 is the FB(flow block) of electrical network icing degree Forecasting Methodology of the present invention.

Embodiment

Below an embodiment to the prediction of electric network in Hunan province icing degree by the inventive method: from 2008-2009, carried out the Long-term forecasting of Hunan Electric Grid icing annual November, and up to now, Long-term forecasting is all accurate.Its concrete grammar is:

(1), first collect, Historical Meteorological Information and the electrical network icing data of 60 years in the provincial level of arrangement, statistical study Hunan Province region, count average icing number of days, division electrical network icing degree rank.Dividing electrical network icing degree rank is:

Slight icing: the 0 day average icing of < number of days≤3 day;

Moderate icing: the 3 days average icing of < number of days≤5 day;

Serious icing: the average icing number of days of < < was 11 days in 5 days;

Especially severe icing: average icing number of days >=11 day.

Average icing number of days is: the icing number of days of each icing monitoring station is cumulative divided by monitoring station number mutually.

(2), utilize icing correlation calculations formula

$R_X=\frac{n\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{x}_i{d}_i-\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{x}_i\&CenterDot;\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{d}_i}{\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{x_i}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{x}_i\right)}^2\&CenterDot;\sqrt{n\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{d_i}^2-{\left(\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{d}_i\right)}^2}}}$

In formula, R _xfor the related coefficient of meteorologic parameter X with average icing number of days, x _ifor the value of meteorologic parameter, i is time sequence number, and n is historical time quantity, d _iaverage icing number of days for a certain time.

Calculate the related coefficient of the historical meteorological effect factor and average icing number of days, from result of calculation, filter out 17 meteorological effect factors of related coefficient maximum, these 17 meteorological effect factors are respectively: sunspot, sea temperature abnormality, general circulation, subtropical high, Strength of Subtropical High conversion in season, subtropical high area and crestal line, subtropical high and analog year comparison, East Asian Circulation, drought and waterlogging, February the highest temperature, mid-September temperature on average, the second half of the month in September high temperature number of days, October the highest temperature, four May temperature depature, September quantity of precipitation, Arctic Oscillation, after serious and especially severe icing year.According to this method, collect 17 the strongest meteorological effect factor data of correlativity in prediction time.Utilize statistical study, find out the characteristic rule of 17 meteorological effect Effects of Factors electrical network icing, in this test, the characteristic rule of 17 meteorological effect Effects of Factors electrical network icing is:

Sunspot: when sunspot number is in the paddy time, electrical network especially severe icing easily occurs;

Sea temperature abnormality: occur that EI Nino or La Nina Phenomenon easily cause the serious and especially severe icing of electrical network in winter;

General circulation: area index is less than normal when the Northern Hemisphere utmost point in August whirlpool, and winter, cold air was on the weak side, is difficult for icing occurs, and when the Northern Hemisphere utmost point in August whirlpool, area index is bigger than normal, and winter, cold air was partially strong, and cold wave easily occurs, and icing easily occurs;

Subtropical high: Yuexi, subtropical ridge point in July position, winter, cold air was stronger, more easily occurred especially severe icing;

Strength of Subtropical High conversion in season: spring and summer and the conversion of Xia Qiu Strength of Subtropical High are larger, more easily occur then icing winter;

Subtropical high crestal line: the beginning of the year (February) crestal line by north and year in when (June), crestal line was by north, easily there is icing winter;

Subtropical high and analog year comparison: similar when the Strength Changes of subtropical high, icing situation is also similar;

East Asian Circulation: when April, zonal circulation and August, meridional circulation was less than normal, icing easily occurs then winter;

Drought and waterlogging: the easy icing in winter in the year of drought, is not easy icing the winter in waterlogging year;

February the highest temperature: when February, the highest temperature was higher than mean value, easily there is then icing winter;

Mid-September temperature on average: during temperature on average, easily there is serious icing higher than historical same period winter in temperature on average mid-September;

The second half of the month in September high temperature number of days: in the second half of the month in September in many times of high temperature number of days, winter temperature is higher, is difficult for occurring serious icing.

October the highest temperature: when October, the highest temperature was higher than 29 ℃, easily there is icing winter;

Four May temperature depature: when Changsha temperature depature value in four Mays is all large, easily there is icing;

September quantity of precipitation: Changsha quantity of precipitation in September is less than mean value, and icing easily occurs winter;

Arctic Oscillation: Arctic Oscillation, when positive phase, icing easily occurs;

After serious and especially severe icing year: seriously and after especially severe icing year be not easy to occur serious and above degree icing.(3), utilize mathematical statistics, count the frequency (occurrence frequency) of the various icing degree based on characteristic rule of 17 factors of influence, set up the icing degree occurrence frequency statistical form of each factor characteristic rule, as shown in table 1; Collect the characteristic rule of 17 predictor in prediction time, as shown in table 2; Utilize the data in prediction time, comprehensive characteristics rule, and inquire about icing degree occurrence frequency statistical form, show that the slight icing frequency in prediction time is s _i, medium icing frequency is m _i, serious icing frequency is l _i, especially severe frequency is n _i, i=1,2 ..., 17(is if sunspot 2008 is in valley near year, and by question blank 1 known (blue mark), the occurrence frequency of four kinds of degree icing in valley year is exactly the occurrence frequency s of four kinds of degree icing of 2008 ₁=40%, m ₁=40%, l ₁=20%, n ₁=0); According to forecast experience for many years, by subtropical high area in 17 factors and ridge point, February the highest temperature, four May temperature depature, serious and especially severe icing year rear four factors weight be made as 2, remaining is made as 1, by normalization, can draw each Factors Weighting average system ɑ ₁, ɑ ₂... ɑ ₁₇, as shown in table 2;

Utilize weight system, by weighted mean, calculate total prediction probability of the various icing degree of prediction time under 17 Effects of Factors, computing formula is:

Slight icing probability of happening:

Moderate icing probability of happening: $M=\underset{i=1}{\overset{17}{\mathrm{\&Sigma;}}}{a}_i\&times;m_i$

Serious icing probability of happening:

Especially severe icing probability of happening:

According to the weighting coefficient of step analysis, can draw total icing probability after weighted mean, as shown in table 2.

Table 1 icing degree occurrence frequency statistical form (sunspot)

Table 2 multiple-factor coupling 2008-2009 forecast result

2008-2009 icing in winter prediction conclusion is slight icing, and concrete probability of happening is: slight 60%, medium 29%, serious 8%, especially severe 3%, and actual icing number of days 1.7 days, belongs to slight icing, and prediction is accurately.

Claims (2)

1. an electrical network icing degree Forecasting Methodology, is characterized in that comprising the following steps:

(1) in collection, arrangement, statistical study provincial region, the Historical Meteorological Information of 60 years and electrical network icing data, count average icing number of days, divides electrical network icing degree rank;

(2) according to icing correlation calculations formula, calculate the related coefficient of historical meteorologic parameter and average icing number of days, icing correlation calculations formula is:

R in formula _xfor the related coefficient of meteorologic parameter with average icing number of days, x _ifor the value of meteorologic parameter, i is time sequence number, and n is historical time quantity, d _ifor the average icing number of days in a certain time, the icing number of days that average icing number of days is each icing monitoring station is cumulative divided by monitoring station number mutually;

(3) from the result of calculation of step (2), filter out 17 meteorological effect factors that correlativity is the strongest;

(4) utilize statistical study, find out the characteristic rule of 17 described meteorological effect Effects of Factors electrical network icing of step (3);

(5) utilize mathematical statistics, obtain the statistical probability of the various icing degree based on characteristic rule of 17 meteorological effect factors, set up the icing degree occurrence frequency statistical form of each factor characteristic rule;

(6) compile the weather data that needs the prediction electrical network icing degree time, the data in 17 factor prediction times that input is corresponding;

(7) utilize the data in prediction time, according to characteristic rule, inquiry icing degree occurrence frequency statistical form, show that the slight icing frequency in prediction time is s _i, moderate icing frequency is m _i, serious icing frequency is l _i, especially severe frequency is n _i, i=1,2 ..., 17;

(8) experience according to weather report, provides the weight coefficient of 17 factor pair electrical network icing numbers of days impacts, is respectively ɑ ₁, ɑ ₂... ɑ ₁₇;

(9) utilize weight system, by weighted mean, calculate total prediction probability of the various icing degree of prediction time under 17 Effects of Factors, computing formula is:

Slight icing probability of happening:

Moderate icing probability of happening:

Serious icing probability of happening:

Especially severe icing probability of happening:

Slight icing: the 0 day average icing of < number of days≤3 day;

Moderate icing: the 3 days average icing of < number of days≤5 day;

Serious icing: the average icing number of days of < < was 11 days in 5 days;

Especially severe icing: average icing number of days >=11 day;

Average icing number of days is: the icing number of days of each icing monitoring station is cumulative divided by monitoring station number mutually.

2. electrical network icing degree Forecasting Methodology according to claim 1, it is characterized in that: 17 meteorological effect factors that described correlativity is the strongest are respectively for Hunan Province: sunspot, sea temperature abnormality, general circulation, subtropical high, Strength of Subtropical High conversion in season, subtropical high area and crestal line, subtropical high and analog year comparison, East Asian Circulation, drought and waterlogging, February the highest temperature, mid-September temperature on average, the second half of the month in September high temperature number of days, October the highest temperature, four May temperature depature, September quantity of precipitation, Arctic Oscillation, after serious and especially severe icing year, the feature of described 17 meteorological effect Effects of Factors electrical network icing is respectively:

Sunspot: when sunspot number is in the paddy time, electrical network especially severe icing easily occurs;

Sea temperature abnormality: then or occur that EI Nino or La Nina Phenomenon easily cause the serious and especially severe icing of electrical network in winter the previous year;

General circulation: area index is less than normal when the Northern Hemisphere utmost point in August whirlpool, and winter, cold air was on the weak side, is difficult for icing occurs, and when the Northern Hemisphere utmost point in August whirlpool, area index is bigger than normal, and winter, cold air was partially strong, and cold wave easily occurs, and icing easily occurs;

Subtropical high: Yuexi, subtropical ridge point in July position, winter, cold air was stronger, more easily occurred especially severe icing;

Strength of Subtropical High conversion in season: spring and summer and the conversion of Xia Qiu Strength of Subtropical High are larger, more easily occur then icing winter;

Subtropical high crestal line: the beginning of the year crestal line by north and year in crestal line when by north, easily there is icing winter;

Subtropical high and analog year comparison: similar when the Strength Changes of subtropical high, icing situation is also similar;

East Asian Circulation: when April, zonal circulation and August, meridional circulation was less than normal, icing easily occurs then winter;

Drought and waterlogging: the easy icing in winter in the year of drought, is not easy icing the winter in waterlogging year;

February the highest temperature: when February, the highest temperature was higher than mean value, easily there is then icing winter;

Mid-September temperature on average: during temperature on average, easily there is serious icing higher than historical same period winter in temperature on average mid-September;

The second half of the month in September high temperature number of days: in the second half of the month in September in many times of high temperature number of days, winter temperature is higher, is difficult for occurring serious icing;

October the highest temperature: when October, the highest temperature was lower than 29 ℃, easily there is icing winter;

Four May temperature depature: when four Mays, temperature depature value was all large, easily there is icing;

September quantity of precipitation: September, quantity of precipitation was less than mean value, and icing easily occurs winter;

Arctic Oscillation: Arctic Oscillation, when positive phase, icing easily occurs;

After serious and especially severe icing year: seriously and after especially severe icing year be not easy to occur serious and above degree icing.