CN103218673A - Method for predicating short-period output power of photovoltaic power generation based on BP (Back Propagation) neural network - Google Patents
Method for predicating short-period output power of photovoltaic power generation based on BP (Back Propagation) neural network Download PDFInfo
- Publication number
- CN103218673A CN103218673A CN2013101032880A CN201310103288A CN103218673A CN 103218673 A CN103218673 A CN 103218673A CN 2013101032880 A CN2013101032880 A CN 2013101032880A CN 201310103288 A CN201310103288 A CN 201310103288A CN 103218673 A CN103218673 A CN 103218673A
- Authority
- CN
- China
- Prior art keywords
- output
- neural network
- day
- data
- generated output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for predicating the short-period output power of photovoltaic power generation based on a BP (Back Propagation) neural network. According to the method, the BP neural network is adopted for predicating the output power of a photovoltaic power generation system, and the influence of weather factors on the output power of the photovoltaic power generation system is subjected to statistical analysis. The method comprises the following steps of firstly mapping weather types as day types used as the input data of the BP neural network, and utilizing power generation power during each time period of a prediction day as output data; then determining the quantity of hidden layer nodes through formula calculation and repeated cut-and-try operations according to the quantity of input and output units; performing normalization treatment on the input data, performing reverse normalization treatment on the output data, and training the BP neural network by utilizing the treated operating data; and finally predicating the power generation power of the predication day by utilizing a trained model, thereby obtaining the predication result. The data processing method and a prediction model can be used for effectively predicating the short-period output power in photovoltaic power generation under multiple weather types.
Description
Technical field
The present invention relates to photovoltaic generating system, relate in particular to a kind of photovoltaic generation short-term and go out force prediction method.
Background technology
Current, environmental pressures such as global fossil energy resources worsening shortages, climate change day by day increase.Sun power alleviating the world energy supplies anxiety, is optimized energy structure as a kind of cleaning, safety, reproducible green energy resource, and there is special advantages aspects such as protection environment.Solar electrical energy generation does not need to consume conventional energy resources, is the clean energy resource of green non-pollution, is paid attention to by countries in the world.As the principal mode that solar electrical energy generation utilizes, photovoltaic generation has obtained development at full speed in recent years.It is a kind of effective ways that make full use of sun power that large-scale photovoltaic generates electricity by way of merging two or more grid systems, and also is the main flow trend of photovoltaic generating system, and at present large-scale photovoltaic parallel in system is applied.
Because grid-connected exerting oneself has randomness, photovoltaic parallel in system is a uncontrollable power supply with respect to big electrical network, and its instability of exerting oneself is influential to the safe and stable operation of big electrical network.Therefore, the infiltration of large-scale photovoltaic generating inserts and must bring a series of influence to electrical network, photovoltaic generation is an intermittent energy source, be subjected to intensity of solar radiation, influences such as environment temperature, output power has uncertainty, after being connected to the grid, it make the short-term load forecasting accuracy of big electrical network reduce, photovoltaic is exerted oneself and is significantly changed the voltage that also must cause total system, frequency jitter, electric system exists frequency and voltage stable problem, increased traditional generating, the difficulty of control and operational plan is unfavorable for the conventional power supply of dispatching of power netwoks personnel placement and its coordinated scheduling.
So,, predict that accurately exerting oneself of photovoltaic generation has great importance in order to utilize photovoltaic generation safely and effectively.Can reduce the grid-connected negative effect that electric system is caused effectively on the one hand, improve the reliability and stability of Operation of Electric Systems, can reduce the spinning reserve capacity of electric system on the other hand, make full use of solar energy resources, obtain bigger economic benefit, social benefit and environmental benefit.Simultaneously, photovoltaic is exerted oneself and is predicted it is the needed basic data in design grid-connected photovoltaic power station.What photovoltaic generation was exerted oneself predicts the outcome, for on-position, access way and the method for operation of photo-voltaic power supply provides theoretical foundation, for the reasonable disposition of electric system Optimization Dispatching and energy storage device provides reliable foundation.
At present, the photovoltaic forecast method of exerting oneself mainly is divided into two kinds:
(1) directly prediction is meant and directly sets up the forecast model of exerting oneself by the historical data of output power, weather etc. is carried out statistical study, does not need external environment conditions such as meteorological temperature are predicted.
(2) indirect predictions is meant weather condition or intensity of sunshine is predicted, sets up Changes in weather and intensity of sunshine model, and the relation by these parameters and photovoltaic generation amount calculates predicted value.
Exert oneself Study on Forecast starting early for photovoltaic abroad, mainly concentrate on short-term prediction and the ultrashort phase prediction of exerting oneself of exerting oneself.In short-term is exerted oneself Study on Forecast, obtain weather information in following 1-2 days according to weather forecast, utilize the historical data prediction to obtain surface solar radiation intensity, again exerting oneself according to photovoltaic array efficient conversion formula prediction photovoltaic generation.Wherein, the prediction of exerting oneself of ultrashort phase is a motion conditions according to cloud layer in the following several hrs of the variation prediction of satellite cloud picture, the cloud layer index is predicted, set up the mapping relations between cloud layer index and the earthbound solar energy radiation intensity, obtain the predicted value of solar radiation, the efficient conversion formula by photovoltaic array calculates exerting oneself of photovoltaic plant again.The prediction of exerting oneself of short-term and ultrashort phase is real-time, but the scheduling of electrical network needs the certain reaction time, therefore the adaptability to changes of electrical network is had higher requirement.
The photovoltaic generation forecasting research of China still is in the starting stage, and the research of exerting oneself at photovoltaic also is not very abundant.Domestic to photovoltaic exert oneself directly, the indirect predictions Study of model mainly contains based on the mathematical statistics Forecasting Methodology with based on the artificial intelligence Forecasting Methodology.Based on free serial method of mathematical statistics forecast method etc., photovoltaic plant is gone out force data regard a periodically variable in time random time sequence as, have advantages such as predetermined speed is fast, forecasting process is simple, extrapolation is good, but this method does not consider to influence the environmental factor of photovoltaic generation.Forecasting Methodology based on artificial intelligence has the neural network method, utilize association, study, the memory function of neural network photovoltaic to be exerted oneself predict, because photovoltaic plant output is affected by environment bigger, complex environment changes makes neural network produce error greatly when training and prediction, even lost efficacy, therefore when setting up forecast model, carried out the submodule division, avoid the not influence of type to predicting the outcome on the same day, the division of submodule needs the data of photovoltaic plant are handled in a large number before making and predicting, has limited the versatility of model.
Summary of the invention
Technical matters to be solved by this invention is at above-mentioned problems of the prior art, proposes a kind of photovoltaic generation short-term based on the BP neural network and goes out force prediction method, and this method can predict effectively that the photovoltaic short-term of different weather type exerts oneself.
The present invention is for solving the problems of the technologies described above by the following technical solutions:
A kind of photovoltaic generation short-term based on the BP neural network goes out force prediction method, comprises the steps:
Step 101 according to exerting oneself the time period in the historical generating of the photovoltaic plant data analysis one day, determines that prediction day needs exerting oneself the time period of prediction;
Step 102, statistics photovoltaic plant in one period every day in the generated output data of each time period of exerting oneself, with these first group of input data as the BP neural metwork training;
Step 103 is added up the generated output data under all kinds of weather patterns, calculates the different weather type at the generated output mean value of step 102 in described a period of time;
Step 104, generated output mean value according to the different weather type, calculating the multiplying power relation between the generated output mean value of different weather type, be a day type index with this multiplying power relationship map, then with day the type index import data as second group of BP neural metwork training;
Step 105 according to input, output unit number, is adjusted the hidden layer node number of determining the BP neural network through formula to calculating with trying to gather, and sets up BP neural metwork training model;
Step 106 before utilizing the BP neural metwork training, is carried out normalized to generated output data and a day type index;
Step 107 is utilized the data after step 106 normalization, and the BP neural network is trained;
Step 108, the BP neural network after the utilization training are predicted the generated output prediction of day, draw to predict the outcome.
Go out the further prioritization scheme of force prediction method as a kind of photovoltaic generation short-term based on the BP neural network of the present invention, described BP neural network comprises input layer, output layer and hidden layer; Wherein the input of hidden layer and output layer node is the weighted sum of last node layer output, and the incentive degree of each node is by its excitation function decision; Wherein:
Being input as of k node of output layer:
Wherein, o
jOutput for hidden layer; w
JkBe the connection weights between hidden layer and the output layer node; Q is the node number of hidden layer;
K node of output layer is output as:
o
k=f(n
k) （2）
Wherein f () is an excitation function,
In the following formula, θ
jExpression biasing or threshold values are worked as θ
jFor on the occasion of the time excitation function is moved to right along horizontal axis, otherwise opposite; θ
0Be used to regulate the shape of e function.
Go out the further prioritization scheme of force prediction method as a kind of photovoltaic generation short-term based on the BP neural network of the present invention, described BP neural network adopts one deck hidden layer, and its hidden layer node number is calculated according to input and output node number and adjustment is gathered in examination.
Go out the further prioritization scheme of force prediction method as a kind of photovoltaic generation short-term of the present invention, also comprise step 109: will predict that the data behind day actual motion count training data, strengthen the predictive ability of BP neural network based on the BP neural network.
Go out the further prioritization scheme of force prediction method as a kind of photovoltaic generation short-term of the present invention based on the BP neural network, described day concrete acquiring method of type index of step 104 is as follows: find out the pairing weather pattern of minimum value in the pairing generated output mean value of the described different weather type of step 103, the day type index of this weather pattern is made as 1, with the pairing generated output mean value of all the other each weather patterns respectively divided by the minimum value in this generated output mean value, obtaining the multiplying power relation between the generated output mean value of different weather type respectively, be a day type index with this multiplying power relationship map.
Go out the further prioritization scheme of force prediction method as a kind of photovoltaic generation short-term of the present invention based on the BP neural network, the described normalized of step 106 is meant according to following formula carries out normalized to the input data, and output data is carried out anti-normalized:
Wherein:
Be the data after the normalized, x
iBe the output data after reality input data or the anti-normalized, described input data are meant generated output, day type index of respectively exert oneself interior every day of statistics phase time period; Described output data is meant the generated output of the time period of respectively exerting oneself of prediction day; x
MinAnd x
MaxBe respectively the minimum value and the maximal value of all types of data in the timing statistics section.
The present invention adopts above technical scheme compared with prior art, has following technique effect:
The present invention adopts based on neural network and has set up the general and efficient photovoltaic forecast model of exerting oneself, to day type set up corresponding weather index, this forecast model has considered to influence the weather conditions that photovoltaic plant is exerted oneself when prediction, adopt a forecast model can predict the different weather type generated output of each time period of next day, adopt the measured value comparative result that predicts the outcome with photovoltaic plant of this method to show that predicting the outcome of this method is comparatively accurate, by the test of actual light overhead utility service data, can predict effectively that the photovoltaic system short-term under the various weather patterns is exerted oneself.
Description of drawings
Fig. 1 is a BP neural network model structural representation of the present invention.
Embodiment
Below in conjunction with accompanying drawing technical scheme of the present invention is described in further detail:
As shown in Figure 1, the present invention is based on the BP neural network and realize the photovoltaic short-term forecasting of exerting oneself, the BP neural network model adopts structure shown in Figure 1.
Among Fig. 1, x
1X
nBe the node of input layer, the training sample data and the weather index of corresponding input, y
1Y
mBe the output layer node, the output result of corresponding prediction.w
IjBe the connection weights between input layer and the hidden layer node, w
JkBe the connection weights between hidden layer and the output layer node, the input of hidden layer and output layer node is the weighted sum of last node layer output, and the incentive degree of each node is by its excitation function decision.
Being input as of k node of output layer:
Wherein, o
jOutput for preceding one deck.
It is output as:
o
k=f(n
k) （2）
F () is an excitation function,
Wherein, θ
jExpression biasing or threshold values, positive θ
jExcitation function is moved to right along horizontal axis; θ
0Regulate the shape of e function.
The BP neural network adopts one deck hidden layer, and its hidden layer node number calculates and try to gather adjustment according to input and output node number.
Exert oneself and the weather conditions of photovoltaic plant have much relations, drawing the photovoltaic plant factor that has the greatest impact of exerting oneself by the analysis to the photovoltaic plant historical data is weather pattern, fine generated energy with the rainy day has a long way to go, and therefore mainly considers the influence of day type to exerting oneself when photovoltaic plant is exerted oneself forecast model setting up.The influence that day type is exerted oneself to photovoltaic generation is embodied directly on day generated output, so generated output has also embodied the influence that day type is exerted oneself to photovoltaic.
According to statistical study to photovoltaic plant actual operating data under the different weather type, the generated output minimum of rain (snow) day, with its day the type index be made as 1, cloudy, cloudy and fine day type index is respectively the N of rain (snow) light type index
1, N
2And N
3Doubly, N
1, N
2And N
3Determine according to the per day generated output of statistics cloudy day, cloudy weather and fine day in the phase and the ratio of the per day generated output of rainy day.
The concrete implementation step of Forecasting Methodology is as follows:
The first step: the generated output data in the statistics photovoltaic plant time period, determine that prediction day needs the generated output of the time period of prediction;
Second step: the generated output of the time period of day output electric energy in the statistics photovoltaic plant time period, with its training input data as the BP neural network;
The 3rd goes on foot: add up interior weather pattern of this time period, it is mapped as a day type index, during the BP neural metwork training, the day type exponential sum of prediction day proxima luce (prox. luc) is predicted the day type index of day is also imported data as the training of neural network;
The 4th step: output data is the prediction generated output of day each time period on the same day;
The 5th step:, through formula to calculating with repeatedly try to gather, determine the hidden layer node number according to input, output unit number;
The 6th step: the input data are carried out normalized, output data is carried out anti-normalized;
The 7th step: adopt the service data after handling that the BP neural network is trained;
The 8th step: utilize the model after the training that the generated output of prediction day is predicted, draw and predict the outcome.
As an exert oneself specific embodiment of short-term forecasting method of the photovoltaic that the present invention is based on the BP neural network, comprise the steps:
(1) according to exerting oneself the time period in the historical generating of the photovoltaic plant data analysis one day, determines exerting oneself the time period of prediction;
(2) generated output of one period interior each time period of every day in period of statistics photovoltaic plant is with these input data of exerting oneself as the BP neural metwork training;
(3) (comprise the cloudy rain that changes according to fine, cloudy (comprise clear to cloudy, cloudy turn to fine), cloudy day, cloudy turn to overcast), rain (comprising snow) classifies weather pattern (day type), add up the generated output under all kinds of weather patterns, calculate in one section timing statistics not the generated output mean value of type on the same day;
(4) according to the average generated output of type on the same day not, calculating the multiplying power relation between the generated output, be a day type index with this multiplying power relationship map, and this index is the input of BP neural network;
(5) generated output and a day type index are carried out normalized according to formula (4) before training.
In the formula (4):
Be the data after the normalized, x
iBe the output data after reality input data or the anti-normalized, described input data are meant generated output, day type index of respectively exert oneself interior every day of statistics phase time period; Described output data is meant the generated output of the time period of respectively exerting oneself of prediction day; x
MinAnd x
MaxBe respectively the minimum value and the maximal value of all types of data in the timing statistics section.
(6) utilize normalization data that the BP neural network is trained;
(7) the BP neural network after the utilization training is predicted the prediction of exerting oneself of day;
(8) data behind the prediction day actual motion count training data, strengthen the predictive ability of network.
Claims (6)
1. the photovoltaic generation short-term based on the BP neural network goes out force prediction method, it is characterized in that, comprises the steps:
Step 101 according to exerting oneself the time period in the historical generating of the photovoltaic plant data analysis one day, determines that prediction day needs exerting oneself the time period of prediction;
Step 102, statistics photovoltaic plant in one period every day in the generated output data of each time period of exerting oneself, with these first group of input data as the BP neural metwork training;
Step 103 is added up the generated output data under all kinds of weather patterns, calculates the different weather type at the generated output mean value of step 102 in described a period of time;
Step 104, generated output mean value according to the different weather type, calculating the multiplying power relation between the generated output mean value of different weather type, be a day type index with this multiplying power relationship map, then with day the type index import data as second group of BP neural metwork training;
Step 105 according to input, output unit number, is adjusted the hidden layer node number of determining the BP neural network through formula to calculating with trying to gather, and sets up BP neural metwork training model;
Step 106 before utilizing the BP neural metwork training, is carried out normalized to generated output data and a day type index;
Step 107 is utilized the data after step 106 normalization, and the BP neural network is trained;
Step 108, the BP neural network after the utilization training are predicted the generated output prediction of day, draw to predict the outcome.
2. a kind of photovoltaic generation short-term based on the BP neural network according to claim 1 goes out force prediction method, it is characterized in that described BP neural network comprises input layer, output layer and hidden layer; Wherein the input of hidden layer and output layer node is the weighted sum of last node layer output, and the incentive degree of each node is by its excitation function decision; Wherein:
Being input as of k node of output layer:
Wherein, o
jOutput for hidden layer; w
JkBe the connection weights between hidden layer and the output layer node; Q is the node number of hidden layer;
K node of output layer is output as:
o
k=f(n
k) （2）
Wherein f () is an excitation function,
In the following formula, θ
jExpression biasing or threshold values are worked as θ
jFor on the occasion of the time excitation function is moved to right along horizontal axis, otherwise opposite; θ
0Be used to regulate the shape of e function.
3. a kind of photovoltaic generation short-term based on the BP neural network according to claim 1 and 2 goes out force prediction method, it is characterized in that, described BP neural network adopts one deck hidden layer, and its hidden layer node number calculates and try to gather adjustment according to input and output node number.
4. a kind of photovoltaic generation short-term based on the BP neural network according to claim 1 goes out force prediction method, it is characterized in that, also comprises step 109: will predict that the data behind day actual motion count training data, strengthen the predictive ability of BP neural network.
5. a kind of photovoltaic generation short-term based on the BP neural network according to claim 1 goes out force prediction method, it is characterized in that, described day concrete acquiring method of type index of step 104 is as follows: find out the pairing weather pattern of minimum value in the pairing generated output mean value of the described different weather type of step 103, the day type index of this weather pattern is made as 1, with the pairing generated output mean value of all the other each weather patterns respectively divided by the minimum value in this generated output mean value, obtaining the multiplying power relation between the generated output mean value of different weather type respectively, be a day type index with this multiplying power relationship map.
6. a kind of photovoltaic generation short-term based on the BP neural network according to claim 1 goes out force prediction method, it is characterized in that, the described normalized of step 106 is meant according to following formula carries out normalized to the input data, and output data is carried out anti-normalized:
Wherein:
Be the data after the normalized, x
iBe the output data after reality input data or the anti-normalized, described input data are meant statistics respectively exert oneself in the phase generated output, day type index of time period; Described output data is meant the generated output of the time period of respectively exerting oneself of prediction day; x
MinAnd x
MaxBe respectively the minimum value and the maximal value of all types of data in the timing statistics section.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101032880A CN103218673A (en) | 2013-03-27 | 2013-03-27 | Method for predicating short-period output power of photovoltaic power generation based on BP (Back Propagation) neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013101032880A CN103218673A (en) | 2013-03-27 | 2013-03-27 | Method for predicating short-period output power of photovoltaic power generation based on BP (Back Propagation) neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103218673A true CN103218673A (en) | 2013-07-24 |
Family
ID=48816438
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013101032880A Pending CN103218673A (en) | 2013-03-27 | 2013-03-27 | Method for predicating short-period output power of photovoltaic power generation based on BP (Back Propagation) neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103218673A (en) |
Cited By (25)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103489038A (en) * | 2013-09-09 | 2014-01-01 | 国家电网公司 | Photovoltaic ultra-short-term power prediction method based on LM-BP neural network |
| CN103996082A (en) * | 2014-06-03 | 2014-08-20 | 华北电力大学（保定） | Method for predicating solar radiation intensity based on double-random theory |
| CN104104116A (en) * | 2014-07-01 | 2014-10-15 | 杭州电子科技大学 | Design method for photovoltaic microgrid supply-demand control system containing distributed energy sources |
| CN104346659A (en) * | 2014-11-26 | 2015-02-11 | 北京邮电大学 | Short-term power generation prediction method applied to high-concentration-ratio photovoltaic power generation system |
| CN104463356A (en) * | 2014-11-27 | 2015-03-25 | 国网浙江省电力公司嘉兴供电公司 | Photovoltaic power generation power prediction method based on multi-dimension information artificial neural network algorithm |
| CN104463349A (en) * | 2014-11-11 | 2015-03-25 | 河海大学 | Photovoltaic generated power prediction method based on multi-period comprehensive similar days |
| CN104573879A (en) * | 2015-01-30 | 2015-04-29 | 河海大学 | Photovoltaic power station output predicting method based on optimal similar day set |
| CN104915727A (en) * | 2015-05-22 | 2015-09-16 | 南京国电南自电网自动化有限公司 | Multi-dimensional isomorphic heterogeneous BP neural network optical power ultrashort-term prediction method |
| CN105373849A (en) * | 2015-10-22 | 2016-03-02 | 华北电力大学（保定） | Stepped temperature prediction method for photovoltaic battery assembly |
| CN105512760A (en) * | 2015-12-04 | 2016-04-20 | 北京国电通网络技术有限公司 | Neural network-based calculation method and calculation system for calculating power-generating capacity of photovoltaic station |
| CN105844343A (en) * | 2016-03-15 | 2016-08-10 | 武汉大学 | Water-optical daily joint scheduling method based on analysis of photovoltaic properties |
| CN106169771A (en) * | 2016-08-01 | 2016-11-30 | 河海大学常州校区 | A kind of combining inverter of measurable generated energy data |
| CN106602595A (en) * | 2016-11-28 | 2017-04-26 | 国网青海省电力公司 | Method for evaluating AC-side impedance balance index of grid-connected photovoltaic inverter |
| CN106712086A (en) * | 2017-01-17 | 2017-05-24 | 无锡协鑫分布式能源开发有限公司 | Microgrid optimization control mode |
| CN106779154A (en) * | 2016-11-22 | 2017-05-31 | 浙江工业大学 | Area distribution formula photovoltaic power generation output forecasting method based on satellite cloud picture |
| CN106961116A (en) * | 2017-01-17 | 2017-07-18 | 无锡协鑫分布式能源开发有限公司 | The integrated micro-grid system of demand sidelight storage |
| CN109615131A (en) * | 2018-12-07 | 2019-04-12 | 国能日新科技股份有限公司 | Photovoltaic power prediction technique and device |
| CN109657881A (en) * | 2019-01-14 | 2019-04-19 | 南京国电南自电网自动化有限公司 | A kind of neural network photovoltaic power generation prediction technique and system suitable for small sample |
| CN110601204A (en) * | 2019-10-14 | 2019-12-20 | 国网辽宁省电力有限公司盘锦供电公司 | Photovoltaic grid-connected system probability power flow analysis method based on random variable state time sequence simulation |
| CN111027816A (en) * | 2019-11-21 | 2020-04-17 | 浙江大学 | Photovoltaic power generation efficiency calculation method based on data envelope analysis |
| CN111489028A (en) * | 2020-04-09 | 2020-08-04 | 国网江苏省电力有限公司苏州供电分公司 | Thundercloud trajectory tracking-based photovoltaic power prediction method under lightning condition |
| CN113052389A (en) * | 2021-04-01 | 2021-06-29 | 中国电力科学研究院有限公司 | Distributed photovoltaic power station ultra-short-term power prediction method and system based on multiple tasks |
| CN113159466A (en) * | 2021-05-27 | 2021-07-23 | 沃太能源股份有限公司 | Short-time photovoltaic power generation prediction system and method |
| CN113449370A (en) * | 2021-07-12 | 2021-09-28 | 西安石油大学 | Masonry component uniaxial compressive strength calculation method based on BP artificial neural network |
| CN114362154A (en) * | 2021-12-28 | 2022-04-15 | 湖南安华源电力科技有限公司 | Photovoltaic power station operation and maintenance method and system with energy storage system based on coordination control |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930347A (en) * | 2012-10-15 | 2013-02-13 | 河海大学 | Method for forecasting short term load under demand response |
-
2013
- 2013-03-27 CN CN2013101032880A patent/CN103218673A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930347A (en) * | 2012-10-15 | 2013-02-13 | 河海大学 | Method for forecasting short term load under demand response |
Non-Patent Citations (3)
| Title |
|---|
| 丁明等: "基于改进BP神经网络的光伏发电系统输出功率短期预测模型", 《电力系统保护与控制》, vol. 40, no. 11, 1 June 2012 (2012-06-01) * |
| 王守相等: "基于灰色神经网络组合模型的光伏短期出力预测", 《电力系统自动化》, vol. 36, no. 19, 10 October 2012 (2012-10-10) * |
| 陈昌松等: "基于神经网络的光伏阵列发电预测模型的设计", 《电工技术学报》, vol. 24, no. 9, 30 September 2009 (2009-09-30), pages 153 - 158 * |
Cited By (35)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103489038A (en) * | 2013-09-09 | 2014-01-01 | 国家电网公司 | Photovoltaic ultra-short-term power prediction method based on LM-BP neural network |
| CN103996082B (en) * | 2014-06-03 | 2017-06-20 | 华北电力大学（保定） | A kind of intensity of solar radiation Forecasting Methodology theoretical based on dual random |
| CN103996082A (en) * | 2014-06-03 | 2014-08-20 | 华北电力大学（保定） | Method for predicating solar radiation intensity based on double-random theory |
| CN104104116A (en) * | 2014-07-01 | 2014-10-15 | 杭州电子科技大学 | Design method for photovoltaic microgrid supply-demand control system containing distributed energy sources |
| CN104463349A (en) * | 2014-11-11 | 2015-03-25 | 河海大学 | Photovoltaic generated power prediction method based on multi-period comprehensive similar days |
| CN104346659A (en) * | 2014-11-26 | 2015-02-11 | 北京邮电大学 | Short-term power generation prediction method applied to high-concentration-ratio photovoltaic power generation system |
| CN104463356A (en) * | 2014-11-27 | 2015-03-25 | 国网浙江省电力公司嘉兴供电公司 | Photovoltaic power generation power prediction method based on multi-dimension information artificial neural network algorithm |
| CN104573879A (en) * | 2015-01-30 | 2015-04-29 | 河海大学 | Photovoltaic power station output predicting method based on optimal similar day set |
| CN104573879B (en) * | 2015-01-30 | 2017-07-25 | 河海大学 | Photovoltaic plant based on optimal similar day collection goes out force prediction method |
| CN104915727A (en) * | 2015-05-22 | 2015-09-16 | 南京国电南自电网自动化有限公司 | Multi-dimensional isomorphic heterogeneous BP neural network optical power ultrashort-term prediction method |
| CN104915727B (en) * | 2015-05-22 | 2019-04-16 | 南京国电南自电网自动化有限公司 | Various dimensions allomer BP neural network optical power ultra-short term prediction method |
| CN105373849A (en) * | 2015-10-22 | 2016-03-02 | 华北电力大学（保定） | Stepped temperature prediction method for photovoltaic battery assembly |
| CN105512760A (en) * | 2015-12-04 | 2016-04-20 | 北京国电通网络技术有限公司 | Neural network-based calculation method and calculation system for calculating power-generating capacity of photovoltaic station |
| CN105512760B (en) * | 2015-12-04 | 2019-08-20 | 北京中电普华信息技术有限公司 | The calculation method and system of built photovoltaic power station power generation amount neural network based |
| CN105844343A (en) * | 2016-03-15 | 2016-08-10 | 武汉大学 | Water-optical daily joint scheduling method based on analysis of photovoltaic properties |
| CN106169771A (en) * | 2016-08-01 | 2016-11-30 | 河海大学常州校区 | A kind of combining inverter of measurable generated energy data |
| CN106169771B (en) * | 2016-08-01 | 2018-12-25 | 河海大学常州校区 | A kind of gird-connected inverter of predictable generated energy data |
| CN106779154B (en) * | 2016-11-22 | 2020-12-01 | 浙江工业大学 | Regional distributed photovoltaic output prediction method based on satellite cloud picture |
| CN106779154A (en) * | 2016-11-22 | 2017-05-31 | 浙江工业大学 | Area distribution formula photovoltaic power generation output forecasting method based on satellite cloud picture |
| CN106602595A (en) * | 2016-11-28 | 2017-04-26 | 国网青海省电力公司 | Method for evaluating AC-side impedance balance index of grid-connected photovoltaic inverter |
| CN106602595B (en) * | 2016-11-28 | 2019-07-23 | 国网青海省电力公司 | A kind of grid-connected photovoltaic inverter exchange side impedance balance Index Assessment method |
| CN106961116A (en) * | 2017-01-17 | 2017-07-18 | 无锡协鑫分布式能源开发有限公司 | The integrated micro-grid system of demand sidelight storage |
| CN106712086A (en) * | 2017-01-17 | 2017-05-24 | 无锡协鑫分布式能源开发有限公司 | Microgrid optimization control mode |
| CN109615131A (en) * | 2018-12-07 | 2019-04-12 | 国能日新科技股份有限公司 | Photovoltaic power prediction technique and device |
| CN109657881A (en) * | 2019-01-14 | 2019-04-19 | 南京国电南自电网自动化有限公司 | A kind of neural network photovoltaic power generation prediction technique and system suitable for small sample |
| CN110601204A (en) * | 2019-10-14 | 2019-12-20 | 国网辽宁省电力有限公司盘锦供电公司 | Photovoltaic grid-connected system probability power flow analysis method based on random variable state time sequence simulation |
| CN110601204B (en) * | 2019-10-14 | 2024-02-02 | 国网辽宁省电力有限公司盘锦供电公司 | Random variable state time sequence simulation-based probability power flow analysis method for photovoltaic grid-connected system |
| CN111027816B (en) * | 2019-11-21 | 2022-07-26 | 浙江大学 | Photovoltaic power generation efficiency calculation method based on data envelope analysis |
| CN111027816A (en) * | 2019-11-21 | 2020-04-17 | 浙江大学 | Photovoltaic power generation efficiency calculation method based on data envelope analysis |
| CN111489028A (en) * | 2020-04-09 | 2020-08-04 | 国网江苏省电力有限公司苏州供电分公司 | Thundercloud trajectory tracking-based photovoltaic power prediction method under lightning condition |
| CN113052389A (en) * | 2021-04-01 | 2021-06-29 | 中国电力科学研究院有限公司 | Distributed photovoltaic power station ultra-short-term power prediction method and system based on multiple tasks |
| CN113159466A (en) * | 2021-05-27 | 2021-07-23 | 沃太能源股份有限公司 | Short-time photovoltaic power generation prediction system and method |
| CN113449370A (en) * | 2021-07-12 | 2021-09-28 | 西安石油大学 | Masonry component uniaxial compressive strength calculation method based on BP artificial neural network |
| CN114362154A (en) * | 2021-12-28 | 2022-04-15 | 湖南安华源电力科技有限公司 | Photovoltaic power station operation and maintenance method and system with energy storage system based on coordination control |
| CN114362154B (en) * | 2021-12-28 | 2023-06-27 | 湖南安华源电力科技有限公司 | Photovoltaic power station operation and maintenance method and system with energy storage system based on coordinated control |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103218673A (en) | Method for predicating short-period output power of photovoltaic power generation based on BP (Back Propagation) neural network | |
| Dong et al. | A novel convolutional neural network framework based solar irradiance prediction method | |
| CN102930358B (en) | A kind of neural net prediction method of photovoltaic power station power generation power | |
| CN102694391B (en) | Day-ahead optimal scheduling method for wind-solar storage integrated power generation system | |
| CN103390200A (en) | Photovoltaic power station electricity generation output power forecasting method based on similar days | |
| CN102419394B (en) | Wind/solar power prediction method with variable prediction resolution | |
| CN104361406B (en) | A kind of photovoltaic plant can utilize solar power generation amount Forecasting Methodology | |
| CN103400204B (en) | Solar energy power generating amount Forecasting Methodology based on SVM Markov combined method | |
| CN101793907A (en) | Short-term wind speed forecasting method of wind farm | |
| CN106532764A (en) | Electric vehicle charging load regulation and control method for locally consuming photovoltaic power generation | |
| CN108364084A (en) | A kind of BP neural network photovoltaic power prediction technique based on Theory of Information Fusion | |
| CN104538992A (en) | Coordinating optimal dispatching method for large water electricity, small water electricity and wind electricity | |
| CN112862215B (en) | Micro-grid energy demand prediction method, system and equipment | |
| CN105678415A (en) | Method for predicting net load of distributed power supply power distribution network | |
| CN103916071B (en) | A kind of uniform output intelligent control system of wind light mutual complementing power generation and method | |
| CN103577893A (en) | Energy-saving optimization method for new energy and thermal power bidirectionally supplying power for high energy carrying loads | |
| CN110707754B (en) | Optimization method for water-wind-light power supply capacity configuration in micro-grid | |
| Vlasov et al. | Predictive control algorithm for a variable load hybrid power system on the basis of power output forecast | |
| CN109345129A (en) | A kind of photo-voltaic power generation station health state evaluation method and system | |
| CN113536581A (en) | Energy storage system multi-state reliability modeling method considering operation strategy | |
| CN105528517B (en) | Based on neural network, wavelet decomposition predicting power of photovoltaic plant method, system | |
| CN104239979A (en) | Ultra-short-term forecasting method of wind power plant generated power | |
| CN108416132B (en) | Automatic variable-step-size simulation acceleration method for distributed photovoltaic cluster | |
| CN106529716A (en) | Method for power dispatching by using mini hydropower classification short-period power prediction | |
| CN116885840A (en) | Distributed new energy online monitoring method and system based on real-time data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130724 |