CN105678396A - Photovoltaic power station super-short-term power prediction device - Google Patents

Abstract

The invention discloses a photovoltaic power station super-short-term power prediction device which comprises a meteorological data receiving device used for receiving predicted meteorological data of a photovoltaic power station in the future, a power station monitoring device used for measuring the real-time meteorological data and real-time power of the photovoltaic power station, a historical power generation server used for recording and collecting historical generated power and corresponding real-time meteorological data of the photovoltaic power station, a measured meteorological change system used for analyzing change information of the real-time meteorological data, a power prediction system used for processing the data information of the meteorological data receiving device and the historical power generation server to get a predicted value of power, and a power prediction correction system used for correcting the predicted value of power of the photovoltaic power station. According to the invention, real-time change information of meteorological elements is introduced into the power prediction correction system, and the predicted power of the photovoltaic power station is corrected according to the laws of real-time meteorological change on the prediction day. Thus, the accuracy of photovoltaic power station super-short-term power prediction is improved.

Description

The super short time test prediction unit in photovoltaic electric station

Technical field

The present invention relates to the super short time test prediction unit in a kind of photovoltaic electric station.

Background technology

Photovoltaic electric station refer to be connected with electrical network and to electrical network carry electric power photovoltaic generating system, by the end of the year 2014, China's photovoltaic electric station adds up the super 28GW of installation, and rate of growth surpasses 60%, and photovoltaic power generation quantity about 25,000,000,000 KWh in 2014, increases by a year-on-year basis more than 200%.

Along with China's photovoltaic electric station scale constantly expands, it is to increase photovoltaic power generation power prediction precision, the safety and stablization being of value to electrical network are run. current predicting power of photovoltaic plant method mainly adopts direct forecast methods, direct forecast methods is exactly with history generated output, meteorological take off data, the information such as weather-forecast input as predictive model, export the power station power prediction value for following 24 hours to 72 hours, meteorological take off data is mainly with day irradiance average, temperature average, amass day cloud amount, the meteorological measuring parameter such as history cloud amount is input, statistical method conventional in predicting power of photovoltaic plant has multivariate linear regression algorithm, artificial neural network (ANN) algorithm, SVMs (SVM) algorithm and gray theory algorithm etc.

At present, by above-mentioned multivariate linear regression algorithm, artificial neural network (ANN) algorithm, the algorithms such as SVMs (SVM) algorithm and gray theory algorithm are set up predictive model and can photovoltaic power generation system output power be predicted, photovoltaic power generation power prediction can improve the solar energy power generating level of resources utilization, but these Forecasting Methodologies are just based on the history same cycle, the prediction obtained in type basis on the same day, history refers to such as one month same historical time cycle with the cycle, type refers to same weather condition such as fine day on the same day, the weak point of these Forecasting Methodologies is, prediction is just attributed to certain day type at day, such as fine day, think same history cycle, the meteorological change at the photovoltaic electric station of type has similarity with generated output on the same day, but the real-time change information processing of meteorological each key element of prediction day is inadequate, the cloudy weather that such as, happens suddenly when day type is fine day is difficult to realize super accurately predicting in short-term.

Summary of the invention

The present invention is directed to above-mentioned deficiency of the prior art, there is provided a kind of photovoltaic electric station super short time test prediction unit, this photovoltaic electric station super short time test prediction unit includes the real-time change information of each for meteorology key element in power prediction correcting system, real-time weather conversion rule according to prediction day corrects photovoltaic electric station predicted power, it is possible to improve the accuracy of the super short time test prediction in photovoltaic electric station.

In order to solve the problems of the technologies described above, the present invention adopts following technical scheme:

The super short time test prediction unit in photovoltaic electric station, comprises weather data receiving trap, power station monitoring device, history generating server, the meteorological change system of actual measurement, power prediction system and power prediction correcting system;

Described weather data receiving trap receives the prediction weather data in photovoltaic electric station future as weather station, and the method NO emissions reduction that the prediction weather data of the big scale received is combined with power by adding up, obtain the prediction weather data become more meticulous;

Described power station monitoring device is used for measuring in real time real time meteorological data and the photovoltaic electric station realtime power at photovoltaic electric station, and it is sent to history generating server, data comprise temperature, humidity, wind speed, intensity of illumination and the photovoltaic electric station realtime power that described power station monitoring device is monitored, the real time meteorological data at described power station monitoring device monitoring photovoltaic electric station and the resolving power of photovoltaic electric station realtime power are 15 minutes;

Described history generating server receives real time meteorological data and the photovoltaic electric station realtime power at the photovoltaic electric station that described power station monitoring device sends, history generating server is for the real time meteorological data of the history generated energy and the photovoltaic electric station corresponding with history generated energy that record and add up photovoltaic electric station, history generating server record and the data comprise photovoltaic electric station realtime power of statistics, temperature, humidity, wind speed and intensity of illumination, the resolving power of history generating server record and the above-mentioned data of statistics is 15 minutes; The prediction weather data that described weather data receiving trap receives is sent in history generating server;

Described history generating server to above-mentioned data according to 4 kinds day type classify, 4 kinds day type comprise A, B, C, D tetra-kinds, wherein A class day type is fine, fine with occasional clouds, cloudy with some sunny periods weather, B class day type is cloudy, the nether world is cloudy, cloudy, cloudy the moon, greasy weather gas, C class day type is rain with snow, light rain, sleet, drizzle or moderate rain, snow shower, slight snow, light to moderate snow shower, thundershower, thundershower is with hail weather, D class day type is extra torrential rain, moderate snow, heavy snow, sudden and violent snow, torrential rain is to extra torrential rain, moderate rain, heavy rain, heavy rain, torrential rain, moderate or heavy snow, heavy to torrential snow, sandstorm, moderate rain or heavy rain, heavy or torrential rain, thunderstorm is to Heavy Rain, on this basis, history generating server counts the temperature correction coefficient Kt of same history cycle one month, on the same day type_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_N; Its implementation is,

Step one, history generating server in, photovoltaic electric station historical data is monthly set up 4 kinds of day type set, 4 kinds day type as previously mentioned; Its specific implementation method is: taking the meteorological parameter of same cycle history and photovoltaic generation power as significant parameter foundation, monthly set up data acquisition, and monthly the meteorological change at the photovoltaic electric station of type and generated output have similarity on the same day;

Step 2, in history generating server, every data acquisition chooses the photovoltaic electric station historical data on 6 working dayss weekly, build together vertical monthly 24 groups working days data, the photovoltaic electric station historical data wherein often organizing working days at least comprises temperature T, humidity S, wind speed F, intensity of illumination G, photovoltaic electric station actual measurement power P_c, and the photovoltaic electric station predicted power P of power prediction system_y, the acquisition resolution of photovoltaic electric station historical data is 15 minutes;

Step 3, in history generating server, monthly calculates the temperature correction coefficient Kt of continuous two collection points of every data acquisition_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_N;

The photovoltaic electric station actual measurement power of N number of collection point and predicted power difference DELTA P_NFor:

${\mathrm{\ΔP}}_N={\mathrm{\ΔT}}_{N,M}\·{\mathrm{Kt}}_{N,M}+{\mathrm{\ΔS}}_{N_{,M}}\·{\mathrm{Ks}}_{N,M}+{\mathrm{\ΔF}}_{N,M}\·{\mathrm{Kf}}_{N,M}+{\mathrm{\ΔG}}_{N,M}\·{\mathrm{Kg}}_{N,M};$

Wherein, Δ T_NFor the temperature change value of N number of collection point on working days, Δ S_NFor the humidity changing value of N number of collection point on working days, Δ F_NFor the wind speed changing value of N number of collection point on working days, Δ G_NFor the intensity of illumination changing value of N number of collection point on working days; Δ P_N=P_c,N-P_y,N, wherein P_c,NFor the photovoltaic electric station actual measurement power of N number of collection point on working days, P_y,NFor the photovoltaic electric station predicted power of N number of collection point on working days;

For the history generating data at photovoltaic electric station, above-mentioned Δ P_NFormula there are 4 unknown numbers, i.e. Kt_N,M、Ks_N,M、Kf_N,M、Kg_N, it is necessary to 4 groups of data can obtain Kt_N,M、Ks_N,M、Kf_N,M、Kg_N, but in data acquisition correspondence often plant a day type exist 6 groups working days data, therefore can obtainPlant and separate, specifically, for temperature correction coefficient Kt_N, there is Kt_N,BIndividual solution, B is the number of 1 to 15, for humidity adjusted coefficient K s_N, there is Ks_N,BIndividual solution, B is the number of 1 to 15, for wind speed adjusted coefficient K f_N, there is Kf_N,BIndividual solution, B is the number of 1 to 15, for intensity of illumination adjusted coefficient K g_N, there is Kg_N,BIndividual solution, B is the number of 1 to 15; So,

The temperature correction coefficient Kt of the N collection point under one day type_NFor:

${\mathrm{Kt}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kt}}_{N,B}}{15};$

The humidity adjusted coefficient K s of the N collection point under one day type_NFor:

${\mathrm{Ks}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Ks}}_{N,B}}{15};$

The wind speed adjusted coefficient K f of the N collection point under one day type_NFor:

${\mathrm{Kf}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kf}}_{N,B}}{15};$

The intensity of illumination adjusted coefficient K g of the N collection point under one day type_NFor:

${\mathrm{Kg}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kg}}_{N,B}}{15};$

Said temperature adjusted coefficient K t_NFor rise in temperature 1 unit is to the influence value of photovoltaic electric station power, humidity adjusted coefficient K s_NFor humidity rises 1 unit to the influence value of photovoltaic electric station power, wind speed adjusted coefficient K f_NFor wind speed rises 1 unit to the influence value of photovoltaic electric station power, intensity of illumination adjusted coefficient K g_NFor intensity of illumination rises 1 unit to the influence value of photovoltaic electric station power, said temperature adjusted coefficient K t_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_NFor normalization coefficient;

The meteorological change system of described actual measurement receives the real time meteorological data at the photovoltaic electric station that described power station monitoring device sends, and then analyze the change information of real time meteorological data, the meteorological change system of actual measurement is the important evidence super short time test prediction in photovoltaic electric station corrected, the temperature T of the first two collection point that described power station monitoring device is collected by the meteorological change system of described actual measurement, humidity S, wind speed F, intensity of illumination G carry out data comparative analysis, calculate the temperature change value Δ T between the first two collection point_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, and intensity of illumination changing value Δ G_N; Its implementation is,

Step 4, in the meteorological change system of actual measurement, calculates the temperature change value Δ T between every workday continuous two collection points_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, intensity of illumination changing value Δ G_N;

Its implementation is:

ΔT_N=T_N-1-T_N-2;

ΔS_N=S_N-1-S_N-2;

ΔF_N=F_N-1-F_N-2;

ΔG_N=G_N-1-G_N-2;

N point is following time point, N-1 and N-2 is the time point occurred, and weigh the Changing Pattern of future time point with the Changing Pattern of the temperature of two time points above, humidity, wind speed, intensity of illumination here;

Wherein, Δ T_NFor the temperature change value of N number of collection point on working days, Δ S_NFor the humidity changing value of N number of collection point on working days, Δ F_NFor the wind speed changing value of N number of collection point on working days, Δ G_NFor the intensity of illumination changing value of N number of collection point on working days, N is N number of collection point on working days;

The real time meteorological data at the prediction weather data become more meticulous that weather data receiving trap described in described power prediction system acceptance sends and the photovoltaic generation realtime power that history generating server sends and corresponding photovoltaic electric station, power prediction system carries out process by parameter of described weather data receiving trap and the described history generating data information that sends of server obtain photovoltaic electric station predicted power by setting up predictive model, and the photovoltaic electric station predicted power that described power prediction system obtains inputs in described history generating server;

The temperature change value Δ T that the photovoltaic electric station predicted power that described power prediction system is obtained by described power prediction correcting system calculates in conjunction with the meteorological change system of described actual measurement_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, intensity of illumination changing value Δ G_N, and the history generating temperature correction coefficient Kt that counts of server_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, intensity of illumination adjusted coefficient K g_N, the power prediction value at photovoltaic electric station is revised, obtains the super short time test forecast value revision value at photovoltaic electric station, to improve the accuracy of the super short time test prediction in photovoltaic electric station; Its implementation is,

Step 5, in power prediction correcting system, revises photovoltaic electric station predicted power, and its method is:

P_X,N=P_y,N+ΔT_N·Kt_N+ΔS_N·Ks_N+ΔF_N·Kf_N+ΔG_N·Kg_N

Wherein, P_X,NFor the photovoltaic electric station predicted power modified value of N number of collection point on working days, P_y,NFor the photovoltaic electric station predicted power of N number of collection point on working days, P_y,NObtained by described power prediction system, Δ T_N、Kt_N、ΔS_N、Ks_N、ΔF_N、Kf_N、ΔG_N、Kg_NThe same;

The super short time test forecast value revision method solving certain month certain day type of above-mentioned photovoltaic electric station super short time test prediction unit, by its same method can be applied to certain month other 3 kinds day type super short time test forecast value revision method, also can be applied to 4 kinds of other month not super short time test forecast value revision methods of type on the same day with reason.

The useful effect of the present invention is: provide a kind of photovoltaic electric station super short time test prediction unit, this photovoltaic electric station super short time test prediction unit includes the real-time change information of each for meteorology key element (temperature, humidity, wind speed and intensity of illumination) in power prediction correcting system, the variable quantity of meteorological each key element provided in conjunction with the meteorological change system of actual measurement and the correction factor of meteorological each key element of history generating server offer realize the real-time correction of photovoltaic electric station super short time test prediction, it is possible to improve the accuracy of the super short time test prediction in photovoltaic electric station.

Accompanying drawing explanation

Fig. 1 is the structure iron of the super short time test prediction unit in photovoltaic electric station of the present invention.

Fig. 2 is the power prediction modification method figure of the super short time test device in photovoltaic electric station of the present invention.

Embodiment

Below in conjunction with accompanying drawing and embodiment, the present invention is described in further detail:

As shown in Figure 1 and Figure 2, the super short time test prediction unit in photovoltaic electric station, comprises weather data receiving trap, power station monitoring device, history generating server, the meteorological change system of actual measurement, power prediction system and power prediction correcting system;

Described weather data receiving trap receives the prediction weather data in photovoltaic electric station future as weather station, and the method NO emissions reduction that the prediction weather data of the big scale received is combined with power by adding up, obtain the prediction weather data become more meticulous;

Described power station monitoring device is used for measuring in real time real time meteorological data and the photovoltaic electric station realtime power at photovoltaic electric station, and it is sent to history generating server, data comprise temperature, humidity, wind speed, intensity of illumination and the photovoltaic electric station realtime power that described power station monitoring device is monitored, the real time meteorological data at described power station monitoring device monitoring photovoltaic electric station and the resolving power of photovoltaic electric station realtime power are 15 minutes;The data acquisition time of described power station monitoring device, from morning 5 every day to point in afternoon 7, when resolving power is 15 minutes, has 56 time collection points every day;

Described history generating server receives real time meteorological data and the photovoltaic electric station realtime power at the photovoltaic electric station that described power station monitoring device sends, history generating server is for the real time meteorological data of the history generated energy and the photovoltaic electric station corresponding with history generated energy that record and add up photovoltaic electric station, history generating server record and the data comprise photovoltaic electric station realtime power of statistics, temperature, humidity, wind speed and intensity of illumination, the resolving power of history generating server record and the above-mentioned data of statistics is 15 minutes; The prediction weather data that described weather data receiving trap receives is sent in history generating server;

Described history generating server to above-mentioned data according to 4 kinds day type classify, 4 kinds day type comprise A, B, C, D tetra-kinds, wherein A class day type is fine, fine with occasional clouds, cloudy with some sunny periods weather, B class day type is cloudy, the nether world is cloudy, cloudy, cloudy the moon, greasy weather gas, C class day type is rain with snow, light rain, sleet, drizzle or moderate rain, snow shower, slight snow, light to moderate snow shower, thundershower, thundershower is with hail weather, D class day type is extra torrential rain, moderate snow, heavy snow, sudden and violent snow, torrential rain is to extra torrential rain, moderate rain, heavy rain, heavy rain, torrential rain, moderate or heavy snow, heavy to torrential snow, sandstorm, moderate rain or heavy rain, heavy or torrential rain, thunderstorm is to Heavy Rain, on this basis, history generating server counts the temperature correction coefficient Kt of same history cycle one month, on the same day type_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_N; Its implementation is,

Step one, history generating server in, photovoltaic electric station historical data is monthly set up 4 kinds of day type set, 4 kinds day type as previously mentioned; Its specific implementation method is: taking the meteorological parameter of same cycle history and photovoltaic generation power as significant parameter foundation, monthly set up data acquisition, and monthly the meteorological change at the photovoltaic electric station of type and generated output have similarity on the same day;

Step 2, in history generating server, every data acquisition chooses the photovoltaic electric station historical data on 6 working dayss weekly, build together vertical monthly 24 groups working days data, the photovoltaic electric station historical data wherein often organizing working days at least comprises temperature T, humidity S, wind speed F, intensity of illumination G, photovoltaic electric station actual measurement power P_c, and the photovoltaic electric station predicted power P of power prediction system_y, the acquisition resolution of photovoltaic electric station historical data is 15 minutes;

Step 3, in history generating server, monthly calculates the temperature correction coefficient Kt of continuous two collection points of every data acquisition_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_N;

The photovoltaic electric station actual measurement power of N number of collection point and predicted power difference DELTA P_NFor:

${\mathrm{\ΔP}}_N={\mathrm{\ΔT}}_{N,M}\·{\mathrm{Kt}}_{N,M}+{\mathrm{\ΔS}}_{N_{,M}}\·{\mathrm{Ks}}_{N,M}+{\mathrm{\ΔF}}_{N,M}\·{\mathrm{Kf}}_{N,M}+{\mathrm{\ΔG}}_{N,M}\·{\mathrm{Kg}}_{N,M};$

Wherein, Δ T_NFor the temperature change value of N number of collection point on working days, Δ S_NFor the humidity changing value of N number of collection point on working days, Δ F_NFor the wind speed changing value of N number of collection point on working days, Δ G_NFor the intensity of illumination changing value of N number of collection point on working days; Δ P_N=P_c,N-P_y,N, wherein P_c,NFor the photovoltaic electric station actual measurement power of N number of collection point on working days, P_y,NFor the photovoltaic electric station predicted power of N number of collection point on working days;

For the history generating data at photovoltaic electric station, above-mentioned Δ P_NFormula there are 4 unknown numbers, i.e. Kt_N,M、Ks_N,M、Kf_N,M、Kg_N, it is necessary to 4 groups of data can obtain Kt_N,M、Ks_N,M、Kf_N,M、Kg_N, but in data acquisition correspondence often plant a day type exist 6 groups working days data, therefore can obtainPlant and separate, specifically, for temperature correction coefficient Kt_N, there is Kt_N,BIndividual solution, B is the number of 1 to 15, for humidity adjusted coefficient K s_N, there is Ks_N,BIndividual solution, B is the number of 1 to 15, for wind speed adjusted coefficient K f_N, there is Kf_N,BIndividual solution, B is the number of 1 to 15, for intensity of illumination adjusted coefficient K g_N, there is Kg_N,BIndividual solution, B is the number of 1 to 15;So,

The temperature correction coefficient Kt of the N collection point under one day type_NFor:

${\mathrm{Kt}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kt}}_{N,B}}{15};$

The humidity adjusted coefficient K s of the N collection point under one day type_NFor:

${\mathrm{Ks}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Ks}}_{N,B}}{15};$

The wind speed adjusted coefficient K f of the N collection point under one day type_NFor:

${\mathrm{Kf}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kf}}_{N,B}}{15};$

The intensity of illumination adjusted coefficient K g of the N collection point under one day type_NFor:

${\mathrm{Kg}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kg}}_{N,B}}{15};$

Said temperature adjusted coefficient K t_NFor rise in temperature 1 unit is to the influence value of photovoltaic electric station power, humidity adjusted coefficient K s_NFor humidity rises 1 unit to the influence value of photovoltaic electric station power, wind speed adjusted coefficient K f_NFor wind speed rises 1 unit to the influence value of photovoltaic electric station power, intensity of illumination adjusted coefficient K g_NFor intensity of illumination rises 1 unit to the influence value of photovoltaic electric station power, said temperature adjusted coefficient K t_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_NFor normalization coefficient;

The meteorological change system of described actual measurement receives the real time meteorological data at the photovoltaic electric station that described power station monitoring device sends, and then analyze the change information of real time meteorological data, the meteorological change system of actual measurement is the important evidence super short time test prediction in photovoltaic electric station corrected, the temperature T of the first two collection point that described power station monitoring device is collected by the meteorological change system of described actual measurement, humidity S, wind speed F, intensity of illumination G carry out data comparative analysis, calculate the temperature change value Δ T between the first two collection point_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, and intensity of illumination changing value Δ G_N; Its implementation is,

Step 4, in the meteorological change system of actual measurement, calculates the temperature change value Δ T between every workday continuous two collection points_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, intensity of illumination changing value Δ G_N;

Its implementation is:

ΔT_N=T_N-1-T_N-2;

ΔS_N=S_N-1-S_N-2;

ΔF_N=F_N-1-F_N-2;

ΔG_N=G_N-1-G_N-2;

N point is following time point, N-1 and N-2 is the time point occurred, and weigh the Changing Pattern of future time point with the Changing Pattern of the temperature of two time points above, humidity, wind speed, intensity of illumination here;

Wherein, Δ T_NFor the temperature change value of N number of collection point on working days, Δ S_NFor the humidity changing value of N number of collection point on working days, Δ F_NFor the wind speed changing value of N number of collection point on working days, Δ G_NFor the intensity of illumination changing value of N number of collection point on working days, N is N number of collection point on working days;

The real time meteorological data at the prediction weather data become more meticulous that weather data receiving trap described in described power prediction system acceptance sends and the photovoltaic generation realtime power that history generating server sends and corresponding photovoltaic electric station, power prediction system carries out process by parameter of described weather data receiving trap and the described history generating data information that sends of server obtain photovoltaic electric station predicted power by setting up predictive model, and the photovoltaic electric station predicted power that described power prediction system obtains inputs in described history generating server;

The temperature change value Δ T that the photovoltaic electric station predicted power that described power prediction system is obtained by described power prediction correcting system calculates in conjunction with the meteorological change system of described actual measurement_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, intensity of illumination changing value Δ G_N, and the history generating temperature correction coefficient Kt that counts of server_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, intensity of illumination adjusted coefficient K g_N, the power prediction value at photovoltaic electric station is revised, obtains the super short time test forecast value revision value at photovoltaic electric station, to improve the accuracy of the super short time test prediction in photovoltaic electric station; Its implementation is,

Step 5, in power prediction correcting system, revises photovoltaic electric station predicted power, and its method is:

P_X,N=P_y,N+ΔT_N·Kt_N+ΔS_N·Ks_N+ΔF_N·Kf_N+ΔG_N·Kg_N

Wherein, P_X,NFor the photovoltaic electric station predicted power modified value of N number of collection point on working days, P_y,NFor the photovoltaic electric station predicted power of N number of collection point on working days, P_y,NObtained by described power prediction system, Δ T_N、Kt_N、ΔS_N、Ks_N、ΔF_N、Kf_N、ΔG_N、Kg_NThe same.

As shown in Figure 2, the power prediction modification method of the super short time test prediction unit in above-mentioned photovoltaic electric station, comprises above-mentioned steps one to step 5.

The super short time test forecast value revision method of above-mentioned certain month certain day type solving photovoltaic electric station super short time test prediction unit, by its same method can be applied to certain month other 3 kinds day type super short time test forecast value revision method, also can be applied to 4 kinds of other month not super short time test forecast value revision methods of type on the same day with reason.

In the present invention, photovoltaic electric station predicted power is obtained by power prediction system, and power prediction system can be set up predictive model by four kinds of power forecasting methods in background technology and obtain photovoltaic electric station predicted power. a kind of super short time test prediction unit in photovoltaic electric station of the present invention and Forecasting Methodology thereof introduce temperature variation, humidity variable quantity, wind speed variable quantity, and these meteorological parameters of intensity of illumination variable quantity are used for weighing real-time weather change, also namely meteorological changing factor constantly is taken into account, could effectively understand currently concrete temperature, humidity, wind speed, the changing conditions of intensity of illumination, accurately analyze current meteorological Changing Pattern, the power prediction result of next time point more accurately is corrected by current meteorological Changing Pattern, this power correction method can more accurately weigh certain day type under Changes in weather on the impact of photovoltaic power station power generation power, thus the photovoltaic electric station predicted power of original power prediction model is revised in real time, to improve the accuracy of the super short time test prediction in photovoltaic electric station.

Return to background technology to be carried, at burst weather such as cloudy, the heavy rains of happening suddenly type is fine day day, under burst cloudy weather, on certain time point, the intensity of illumination of meteorological parameter can change, under burst rainstorm weather, on certain time point, the humidity of meteorological parameter can change, therefore the meteorological changing conditions happened suddenly under can predicting certain day type, it is possible to the generated output at photovoltaic electric station is realized super accurately predicting in short-term.

The foregoing is only the better embodiment of the present invention, not in order to limit the present invention, all amendment, equivalent replacement and improvement etc. done within the spirit and principles in the present invention, all should be included within protection scope of the present invention.

Claims (2)

1. the super short time test prediction unit in photovoltaic electric station, it is characterised in that, comprise weather data receiving trap, power station monitoring device, history generating server, the meteorological change system of actual measurement, power prediction system and power prediction correcting system;

Described weather data receiving trap receives the prediction weather data in photovoltaic electric station future as weather station, and the method NO emissions reduction that the prediction weather data of the big scale received is combined with power by adding up, obtain the prediction weather data become more meticulous;

Described power station monitoring device is used for measuring in real time real time meteorological data and the photovoltaic electric station realtime power at photovoltaic electric station, and it is sent to history generating server, data comprise temperature, humidity, wind speed, intensity of illumination and the photovoltaic electric station realtime power that described power station monitoring device is monitored, the real time meteorological data at described power station monitoring device monitoring photovoltaic electric station and the resolving power of photovoltaic electric station realtime power are 15 minutes;

Described history generating server receives real time meteorological data and the photovoltaic electric station realtime power at the photovoltaic electric station that described power station monitoring device sends, history generating server is for the real time meteorological data of the history generated energy and the photovoltaic electric station corresponding with history generated energy that record and add up photovoltaic electric station, history generating server record and the data comprise photovoltaic electric station realtime power of statistics, temperature, humidity, wind speed and intensity of illumination, the resolving power of history generating server record and the above-mentioned data of statistics is 15 minutes;The prediction weather data that described weather data receiving trap receives is sent in history generating server;

Described history generating server to above-mentioned data according to 4 kinds day type classify, 4 kinds day type comprise A, B, C, D tetra-kinds, wherein A class day type is fine, fine with occasional clouds, cloudy with some sunny periods weather, B class day type is cloudy, the nether world is cloudy, cloudy, cloudy the moon, greasy weather gas, C class day type is rain with snow, light rain, sleet, drizzle or moderate rain, snow shower, slight snow, light to moderate snow shower, thundershower, thundershower is with hail weather, D class day type is extra torrential rain, moderate snow, heavy snow, sudden and violent snow, torrential rain is to extra torrential rain, moderate rain, heavy rain, heavy rain, torrential rain, moderate or heavy snow, heavy to torrential snow, sandstorm, moderate rain or heavy rain, heavy or torrential rain, thunderstorm is to Heavy Rain, on this basis, history generating server counts the temperature correction coefficient Kt of same history cycle one month, on the same day type_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_N; Its implementation is,

Step one, history generating server in, photovoltaic electric station historical data is monthly set up 4 kinds of day type set, 4 kinds day type as previously mentioned; Its specific implementation method is: taking the meteorological parameter of same cycle history and photovoltaic generation power as significant parameter foundation, monthly set up data acquisition, and monthly the meteorological change at the photovoltaic electric station of type and generated output have similarity on the same day;

Step 2, in history generating server, every data acquisition chooses the photovoltaic electric station historical data on 6 working dayss weekly, build together vertical monthly 24 groups working days data, the photovoltaic electric station historical data wherein often organizing working days at least comprises temperature T, humidity S, wind speed F, intensity of illumination G, photovoltaic electric station actual measurement power P_c, and the photovoltaic electric station predicted power P of power prediction system_y, the acquisition resolution of photovoltaic electric station historical data is 15 minutes;

Step 3, in history generating server, monthly calculates the temperature correction coefficient Kt of continuous two collection points of every data acquisition_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_N;

The photovoltaic electric station actual measurement power of N number of collection point and predicted power difference DELTA P_NFor:

${\mathrm{\ΔP}}_N={\mathrm{\ΔT}}_{N,M}\·{\mathrm{Kt}}_{N,M}+{\mathrm{\ΔS}}_{N_{,M}}\·{\mathrm{Ks}}_{N,M}+{\mathrm{\ΔF}}_{N,M}\·{\mathrm{Kf}}_{N,M}+{\mathrm{\ΔG}}_{N,M}\·{\mathrm{Kg}}_{N,M};$

Wherein, Δ T_NFor the temperature change value of N number of collection point on working days, Δ S_NFor the humidity changing value of N number of collection point on working days, Δ F_NFor the wind speed changing value of N number of collection point on working days, Δ G_NFor the intensity of illumination changing value of N number of collection point on working days; Δ P_N=P_c,N-P_y,N, wherein P_c,NFor the photovoltaic electric station actual measurement power of N number of collection point on working days, P_y,NFor the photovoltaic electric station predicted power of N number of collection point on working days;

For the history generating data at photovoltaic electric station, above-mentioned Δ P_NFormula there are 4 unknown numbers, i.e. Kt_N,M、Ks_N,M、Kf_N,M、Kg_N, it is necessary to 4 groups of data can obtain Kt_N,M、Ks_N,M、Kf_N,M、Kg_N, but in data acquisition correspondence often plant a day type exist 6 groups working days data, therefore can obtainPlant and separate, specifically, for temperature correction coefficient Kt_N, there is Kt_N,BIndividual solution, B is the number of 1 to 15, for humidity adjusted coefficient K s_N, there is Ks_N,BIndividual solution, B is the number of 1 to 15, for wind speed adjusted coefficient K f_N, there is Kf_N,BIndividual solution, B is the number of 1 to 15, for intensity of illumination adjusted coefficient K g_N, there is Kg_N,BIndividual solution, B is the number of 1 to 15; So,

The temperature correction coefficient Kt of the N collection point under one day type_NFor:

${\mathrm{Kt}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kt}}_{N,B}}{15};$

The humidity adjusted coefficient K s of the N collection point under one day type_NFor:

${\mathrm{Ks}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Ks}}_{N,B}}{15};$

The wind speed adjusted coefficient K f of the N collection point under one day type_NFor:

${\mathrm{Kf}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kf}}_{N,B}}{15};$

The intensity of illumination adjusted coefficient K g of the N collection point under one day type_NFor:

${\mathrm{Kg}}_N=\frac{\underset{B=1}{\overset{15}{\Σ}}{\mathrm{Kg}}_{N,B}}{15};$

Said temperature adjusted coefficient K t_NFor rise in temperature 1 unit is to the influence value of photovoltaic electric station power, humidity adjusted coefficient K s_NFor humidity rises 1 unit to the influence value of photovoltaic electric station power, wind speed adjusted coefficient K f_NFor wind speed rises 1 unit to the influence value of photovoltaic electric station power, intensity of illumination adjusted coefficient K g_NFor intensity of illumination rises 1 unit to the influence value of photovoltaic electric station power, said temperature adjusted coefficient K t_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, and intensity of illumination adjusted coefficient K g_NFor normalization coefficient;

The meteorological change system of described actual measurement receives the real time meteorological data at the photovoltaic electric station that described power station monitoring device sends, and then analyze the change information of real time meteorological data, the meteorological change system of actual measurement is the important evidence super short time test prediction in photovoltaic electric station corrected, the temperature T of the first two collection point that described power station monitoring device is collected by the meteorological change system of described actual measurement, humidity S, wind speed F, intensity of illumination G carry out data comparative analysis, calculate the temperature change value Δ TN between the first two collection point, humidity changing value Δ S_N, wind speed changing value Δ F_N, and intensity of illumination changing value Δ G_N; Its implementation is,

Step 4, in the meteorological change system of actual measurement, calculates the temperature change value Δ T between every workday continuous two collection points_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, intensity of illumination changing value Δ G_N;

Its implementation is:

ΔT_N=T_N-1-T_N-2;

ΔS_N=S_N-1-S_N-2;

ΔF_N=F_N-1-F_N-2;

ΔG_N=G_N-1-G_N-2;

N point is following time point, N-1 and N-2 is the time point occurred, and weigh the Changing Pattern of future time point with the Changing Pattern of the temperature of two time points above, humidity, wind speed, intensity of illumination here;

Wherein, Δ T_NFor the temperature change value of N number of collection point on working days, Δ S_NFor the humidity changing value of N number of collection point on working days, Δ F_NFor the wind speed changing value of N number of collection point on working days, Δ G_NFor the intensity of illumination changing value of N number of collection point on working days, N is N number of collection point on working days;

The real time meteorological data at the prediction weather data become more meticulous that weather data receiving trap described in described power prediction system acceptance sends and the photovoltaic generation realtime power that history generating server sends and corresponding photovoltaic electric station, power prediction system carries out process by parameter of described weather data receiving trap and the described history generating data information that sends of server obtain photovoltaic electric station predicted power by setting up predictive model, and the photovoltaic electric station predicted power that described power prediction system obtains inputs in described history generating server;

The temperature change value Δ T that the photovoltaic electric station predicted power that described power prediction system is obtained by described power prediction correcting system calculates in conjunction with the meteorological change system of described actual measurement_N, humidity changing value Δ S_N, wind speed changing value Δ F_N, intensity of illumination changing value Δ G_N, and the history generating temperature correction coefficient Kt that counts of server_N, humidity adjusted coefficient K s_N, wind speed adjusted coefficient K f_N, intensity of illumination adjusted coefficient K g_N, the power prediction value at photovoltaic electric station is revised, obtains the super short time test forecast value revision value at photovoltaic electric station, to improve the accuracy of the super short time test prediction in photovoltaic electric station; Its implementation is,

Step 5, in power prediction correcting system, revises photovoltaic electric station predicted power, and its method is:

P_X,N=P_y,N+ΔT_N·Kt_N+ΔS_N·Ks_N+ΔF_N·Kf_N+ΔG_N·Kg_N

Wherein, P_X,NFor the photovoltaic electric station predicted power modified value of N number of collection point on working days, P_y,NFor the photovoltaic electric station predicted power of N number of collection point on working days, P_y,NObtained by described power prediction system, Δ T_N、Kt_N、ΔS_N、Ks_N、ΔF_N、Kf_N、ΔG_N、Kg_NThe same;

The super short time test forecast value revision method solving certain month certain day type of above-mentioned photovoltaic electric station super short time test prediction unit, by its same method can be applied to certain month other 3 kinds day type super short time test forecast value revision method, also can be applied to 4 kinds of other month not super short time test forecast value revision methods of type on the same day with reason.

2. the super short time test prediction unit in photovoltaic electric station as claimed in claim 1, it is characterised in that, the data acquisition time of described power station monitoring device, from morning 5 every day to point in afternoon 7, when resolving power is 15 minutes, has 56 time collection points every day.