CN102184453A - Wind power combination predicting method based on fuzzy neural network and support vector machine - Google Patents
Wind power combination predicting method based on fuzzy neural network and support vector machine Download PDFInfo
- Publication number
- CN102184453A CN102184453A CN2011101256331A CN201110125633A CN102184453A CN 102184453 A CN102184453 A CN 102184453A CN 2011101256331 A CN2011101256331 A CN 2011101256331A CN 201110125633 A CN201110125633 A CN 201110125633A CN 102184453 A CN102184453 A CN 102184453A
- Authority
- CN
- China
- Prior art keywords
- model
- training
- prediction
- wind
- neural network
- 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
Abstract
The invention provides a wind-field power combination predicting method based on a fuzzy neural network and a support vector machine, which comprises the following steps of: acquiring and pre-processing data; setting up a fuzzy neural network model by using a normalized training sample set and a prediction sample set and predicting; setting up a support vector machine model and predicting; linearly combining the prediction results of the two algorithms to obtain a wind speed prediction value; and setting up a wind speed power expert table via historical data, and inquiring the expert table according to the predicted wind speed value so as to obtain a power prediction result. By the method provided by the invention, the short-term prediction of a wind speed sequence can be effectively realized, the power prediction precision is improved, and fewer computing resources are consumed.
Description
Technical field
The present invention relates to a kind of data modeling and Forecasting Methodology, relate in particular to a kind of wind power combination forecasting method based on fuzzy neural network and support vector machine based on artificial intelligence technology.
Technical background
Along with the fast development of wind-power electricity generation installed capacity, the ratio of wind-powered electricity generation in electrical network constantly increases.Because wind-powered electricity generation is a kind of intermittence, the undulatory property energy, large-scale wind-powered electricity generation inserts safety, the stable operation of electric system and guarantees that the quality of power supply brought severe challenge.If can make prediction more accurately to the wind speed and the generated output of wind field, then can effectively alleviate the influence of wind-powered electricity generation to whole electrical network.To help dispatching of power netwoks department in time to formulate reasonable Operation Mode and adjust operation plan exactly by wind power prediction, thereby guarantee reliable, high-quality, the economical operation of electric system.In general, according to the object difference of forecast model, Forecasting Methodology can be divided into based on the direct predicted method of power with based on the indirect predictions method of wind speed, and the present patent application belongs to the latter.
The domestic existing prognoses system time series methods that adopt based on the autoregression linear model more, because model itself is linear, precision of prediction is often not ideal enough according to this, and the single neural network method of existing employing needs more training sample usually, calculation consumption is excessive on the one hand, generalization ability preferably can't be guaranteed on the other hand, when sample information is insufficient, the better prediction precision can't be obtained again simultaneously.The present invention adopts fuzzy neural network and support vector machine to realize the combined prediction algorithm, fuzzy neural network possesses the advantage of powerful learning ability of neural network and fuzzy logic processes uncertain information simultaneously, and support vector machine processing small sample higher-dimension problem has a clear superiority in than traditional neural network, therefore, the binding energy of the two effectively overcomes existing ultrashort phase forecasting problem, can realize accurate ultrashort phase prediction with less historical sample.
Summary of the invention
The invention reside in the defective that overcomes prior art, a kind of intelligent data modeling algorithm that is used for the prediction of wind energy turbine set generated output is proposed, this method synthesis has utilized the advantage of fuzzy neural network and support vector machine, can in forecasting process, add the expertise experience, also can under the sufficient inadequately situation of blower fan historical data, realize the accurate short-term prediction of wind series, in the precision that improves power prediction, only consume less computational resource.
The object of the present invention is achieved like this:
A kind of wind power combination forecasting method based on fuzzy neural network and support vector machine comprises the steps:
(1) data obtains and pre-service.
Wind energy turbine set power prediction system utilizes data acquisition module to obtain data such as wind field wind speed, wind direction, temperature, humidity, atmospheric pressure and blower fan output power in the fixed time scope from the anemometer tower of wind field and wind field central monitoring system, obtains sample set thus and carries out normalized.Training sample set is input as:
V (t)=[X (t-m) ..., X (t-2), X (t-1), Vs (t-1), Vc (t-1), T (t-1), H (t-1), P (t-1)], wherein, X is a wind speed, and Vs is the wind speed sine, Vc is a wind speed cosine, and T is a temperature, and H is a humidity, P is an atmospheric pressure, and t is the moment to be predicted, and m is the length as the wind series of model input.
Training sample set is output as t wind speed X (t) constantly, and then [V (t), X (the t)] training sample that partners is right.By time range and the right number of time interval decision training sample.Forecast sample then is the V of current time.
The normalized formula is:
Wherein, d (t) is a raw data, and X (t) is the data after the normalization.
(2) utilize training sample set and forecast sample collection after the normalization to set up fuzzy neural network model and prediction:
(2.1) the input and output dimension number of samples of determining the number of training sample and model need to determine to consider requirements such as computing time of model training and prediction and accuracy of predicting.Model adopts the output of multidimensional input one-dimensional.
(2.2) determine model structure and parameter and cycle of training number adopt fuzzy subtractive clustering to determine the structure of fuzzy model, determine that by selecting different cluster radius the model structure of optimum is to guarantee less training error; Cycle of training number definite computing time of need considering model training, avoid over training to guarantee extensive performance simultaneously.
(2.3) model training.Determined after the model structure that adopt training sample set pair model training, target is to minimize training error, the Error Calculation formula is:
Wherein, V
MiBe i actual wind speed constantly, V
PiBe i prediction wind speed constantly, n is a number of samples;
(2.4) model prediction.Model training is input to model with the forecast sample collection after finishing, and obtains predicted value, then predicted value is added the forecast sample collection, re-enters model, and circulation obtains the multi-step prediction value according to this.
(2.5) predicted value aftertreatment.The anti-normalization of model predication value is reduced to actual value, judges whether to exist abnormity point simultaneously, if there is then in addition smoothing processing.Anti-normalized formula:
Y(t)＝u(t)*max(d(t))-min(d(t))+min(d(t)
Wherein, u (t) is the model output valve, and Y (t) is the data after the anti-normalization.
(3) set up supporting vector machine model and prediction:
(3.1) determine the number of training sample and the input and output dimension of model
Consider the advantage of support vector machine on small sample higher-dimension problem, the number of training sample can be less here, and the input dimension is desirable higher.Because scale is calculated in the training of support vector machine and training sample number exponentially changes, so the number of training sample is chosen the computing power that will consider particular hardware simultaneously.
(3.2) determine the super parameter value of model
There are three main super parameters in support vector machine: the insensitive loss factor, kernel function parameter and punishment parameter.Wherein, the selection of the insensitive loss factor and kernel function parameter is bigger to the model performance influence, therefore can adopt the crosscheck method to determine concrete parameter value.
(3.3) model training
Determined that training sample is right, selected radially basic kernel function for use, good super parameter value is determined in input, the beginning model training, and output support vector and Lagrange multiplier difference are finished in training.
(3.4) model prediction
Input forecast sample collection obtains the single step predicted value to the model that trains, and then predicted value is added former forecast sample collection and forms new forecast sample collection, re-enters model, and circulation obtains the multi-step prediction value according to this.
(3.5) predicted value aftertreatment
The model predication value of support vector machine is reduced to actual value, judges whether to exist abnormity point simultaneously, if there is then in addition smoothing processing.
(4) with two kinds of algorithm predicts results linear combination in addition, obtain the forecasting wind speed value.If historical predicated error is respectively e
1, e
2, now predicting the outcome is V
1, V
2, then predicting the outcome after the combination is respectively:
Be that historical predicated error is more little, this predicts that shared weight is then big more.
(5) set up wind speed power expert table by historical data, obtain the power prediction result thereby remove to inquire about expert's table according to the air speed value of prediction.
Owing to adopted above-mentioned technical solution, the inventive method is with the beneficial effect that existing method is compared:
One, the present invention has adopted fuzzy neural network (ANFIS) model, this model had both possessed the ability of carrying out reasoning and study in uncertain and out of true environment, fast convergence rate, good stability are arranged simultaneously, and the precision advantages of higher has innate advantage to handling the wind field environmental data;
Two, the present invention has adopted support vector machine (SVM) model simultaneously, and this model is the data digging method that grows up on the Statistical Learning Theory basis, obviously is better than traditional neural network on processing small sample higher-dimension problem.Therefore this model is suitable for the power prediction under the limited sample situation of wind field, simultaneously since its based on structural risk minimization, the generalization ability of model can be protected;
Three, based on the combined prediction algorithm of two kinds of models, can adjust the weight of two class models according to historical precision of prediction in real time, when obtaining the higher forecasting precision, strengthened the stability of prognoses system greatly.
Description of drawings
Fig. 1 prognoses system structural drawing;
Fig. 2 is the process flow diagram of fuzzy neural network modeling among the present invention;
Fig. 3 is the modeling process flow diagram of support vector machine among the present invention;
Fig. 4 is that the present invention gives weighted array with two kinds of algorithm predicts results, obtains the forecasting wind speed value, the combinational algorithm synoptic diagram;
Fig. 5 is a prognoses system program flow diagram of the present invention;
Fig. 6 is predicted power and real power curve map.
Embodiment
Below in conjunction with accompanying drawing the present invention is described in further detail.
Fig. 1 is a power prediction of the present invention system pie graph, and it has illustrated formation, effect and the implementation procedure of native system.System at first obtains wind field weather information data from the measurement module of wind field, obtain the power output data of blower fan from wind field central monitoring system (SCADA), by data processing module with in the data importing database, the prediction algorithm service routine extracts the historical sample data and forms training sample set forecast sample collection from database, be input to ANFIS and SVM model again, combinational algorithm draws final predicting the outcome, and will predict the outcome and store in the database server.At last, predicted data is shown to the user, and the power prediction result is reported dispatching of power netwoks mechanism by suitable communication interface by man-machine interface.
Fig. 2 is the process flow diagram of fuzzy neural network modeling among the present invention.Prognoses system is utilized data acquisition module to obtain data such as wind field wind speed, wind direction, temperature, humidity, atmospheric pressure and blower fan output power in the fixed time scope from the anemometer tower of wind field and wind field central monitoring system and is carried out normalized.Determine the number of training sample and the input and output dimension of model after utilizing normalization then, key elements such as the computing time of definite need consideration model training of number of samples and prediction and accuracy of predicting.Next be to adopt fuzzy subtractive clustering to determine the structure of fuzzy model, determine that by selecting different cluster radius optimum model structure is to guarantee less training error; Cycle of training number definite computing time of need considering model training, avoid over training to guarantee extensive performance simultaneously.Determined after the model structure that adopt training sample set pair model training, target is to minimize training error.Model training is input to model with the forecast sample collection after finishing, and obtains predicted value, then predicted value is added the forecast sample collection, re-enters model, and circulation obtains the multi-step prediction value according to this.Be the predicted value aftertreatment at last, model predication value is reduced to actual value, judge whether to exist abnormity point simultaneously, if there is then in addition smoothing processing.
Fig. 3 is a support vector machine modeling process flow diagram among the present invention.The normalized process is to determine the number of training sample and the input and output dimension of model with preceding identical afterwards.Consider the advantage of support vector machine on small sample higher-dimension problem, the number of training sample can be less here, and the input dimension is desirable higher.Because scale is calculated in the training of support vector machine and training sample number exponentially changes, so the number of training sample is chosen the computing power that will consider particular hardware simultaneously.Be to determine the super parameter value of model then.There are three main super parameters in support vector machine: the insensitive loss factor, kernel function parameter and punishment parameter.Wherein, the selection of the insensitive loss factor and kernel function parameter is bigger to the model performance influence, therefore can adopt the crosscheck method to determine concrete parameter value.Be the model training step then.Determined that training sample is right, selected radially basic kernel function for use, good super parameter value is determined in input, the beginning model training, and output support vector and Lagrange multiplier difference are finished in training.Import the forecast sample collection then to the model that trains, obtain the single step predicted value, then predicted value is added former forecast sample collection and form new forecast sample collection, re-enter model, circulation obtains the multi-step prediction value according to this.Model predication value with support vector machine is reduced to actual value at last, judges whether to exist abnormity point simultaneously, if there is then in addition smoothing processing.
Fig. 4 is a combinational algorithm synoptic diagram of the present invention.If historical predicated error is respectively e
1, e
2(general desirable preceding 5 prediction result statistics), now predicting the outcome is V
1, V
2, then predicting the outcome after the combination is respectively:
Be that historical predicated error is more little, this predicts that shared weight is then big more, and each prediction all can adjust according to historical prediction result, and real-time follow-up shows model preferably in the recent period.
After drawing the combined prediction result, can set up wind speed power expert table, obtain the power prediction result thereby remove to inquire about expert's table according to the air speed value of prediction by historical data.Fig. 5 has illustrated the real work process flow diagram of power prediction algorithm routine of the present invention, to predict 1 hour simultaneously and 4 hours generated outputs are example, Fig. 6 has provided the prediction effect figure of the inventive method, gets 40 sample points, and each sample point is the real 15 minutes mean value of power of sending out.Horizontal ordinate is the data number, and ordinate is the wind power size, and unit is MW.Three curves are arranged among the figure, and solid line is actual wind field generated output data, and dotted line is 1 one-hour rating predicted value, the asterisk line is 4 one-hour rating predicted values, 1 hour predicted root mean square error of gained is that 4.3%, 4 hour predicted root mean square error is 5.7%, and visible precision of prediction is good.Root-mean-square error adopts following formula to calculate:
Wherein, P
MiBe i real power constantly, P
PiBe i predicted power constantly, Cap is average start capacity, and n is a number of samples.The start capacity is 7.6MW in this example.
Below embodiment has been described in detail the present invention in conjunction with the accompanying drawings, and those skilled in the art can make the many variations example to the present invention according to the above description.Thereby some details among the embodiment should not constitute limitation of the invention, and the scope that the present invention will define with appended claims is as protection scope of the present invention.
Claims (7)
1. the wind power combination forecasting method based on fuzzy neural network and support vector machine comprises the steps:
Step 1, data acquisition and pre-service:
Wind power forecasting system utilizes data acquisition module to obtain wind field wind speed, wind direction, temperature, humidity, atmospheric pressure and blower fan output power data in the fixed time scope from the anemometer tower of wind field and wind field monitoring system, Various types of data is provided with limits, out-of-limit abnormal data is corrected, obtained sample set thus and carry out normalized;
Step 2 utilizes training sample set and forecast sample collection after the normalization to set up fuzzy neural network model and prediction;
Step 3 is set up supporting vector machine model and prediction;
Step 4 with two kinds of algorithm predicts results linear combination in addition, obtains the forecasting wind speed value, that is:
If historical predicated error is respectively e
1, e
2, now predicting the outcome is V
1, V
2, then predicting the outcome after the combination is respectively:
Be that historical predicated error is more little, this predicts that shared weight is then big more;
Step 5 is set up wind speed power expert table by historical data, obtains the power prediction result thereby remove to inquire about expert's table according to the air speed value of prediction.
2. a kind of wind power combination forecasting method based on fuzzy neural network and support vector machine according to claim 1 is characterized in that:
Wind speed in the described fixed time scope in the step 1, wind direction, temperature, humidity and atmospheric pressure data are meant in chronological sequence one group of data of tactic constant duration, training sample set is input as:
V(t)＝[X(t-m)，...，X(t-2)，X(t-1)，Vs(t-1)，Vc(t-1)，T(t-1)，H(t-1)，P(t-1)]，
Wherein, X is a wind speed, and Vs is the wind speed sine, and Vc is a wind speed cosine, and T is a temperature, and H is a humidity, and P is an atmospheric pressure, and t is the moment to be predicted, and m is the length as the wind series of model input;
Training sample set is output as t wind speed X (t) constantly, and then [V (t), X (the t)] training sample that partners is right, by time range and the right number of time interval decision training sample.Forecast sample then is the V of current time;
Normalized formula in the step 1 is:
Wherein, d (t) is a raw data, and X (t) is the data after the normalization;
Anti-normalized formula:
Y(t)＝u(t)*max(d(t))-min(d(t))+min(d(t)
Wherein, u (t) is the model output valve, and Y (t) is the data after the anti-normalization.
3. a kind of wind power combination forecasting method based on fuzzy neural network and support vector machine according to claim 1 is characterized in that:
In the step 2 utilize the training sample set after the normalization and the forecast sample collection is built the ANFIS model and prediction comprises the following steps:
2.1 empirical method is determined the number that training sample is right and the input dimension of model;
Adopt fuzzy subtractive clustering to determine model structure 2.2 set suitable cluster radius, set number cycle of training simultaneously;
2.3 model training is promptly determined after the good model structure, adopts training sample set with blended learning method training fuzzy neural network parameter, target is to minimize training error, and the Error Calculation formula is:
Wherein, V
MiBe i actual wind speed constantly, V
PiBe i prediction wind speed constantly, n is a number of samples;
2.4 model prediction after promptly model training finishes, is input to model with the forecast sample collection, obtains predicted value, then predicted value is added the forecast sample collection, re-enters model, circulation obtains the multi-step prediction value according to this;
2.5 the predicted value aftertreatment is about to model predication value and is reduced to actual value, judge whether simultaneously to exist out-of-limit, if there is then in addition smoothing processing.
4. a kind of wind power combination forecasting method based on fuzzy neural network and support vector machine according to claim 1 is characterized in that:
Setting up the SVM model and predicting in the step 3 comprises the following steps:
3.1 determine the number of training sample and the input and output dimension of model;
3.2 select radially basic kernel function for use, adopt the grid search method to determine the super parameter value of model;
3.3 model training has determined that training sample is right, good super parameter value is determined in input, the beginning model training, and output support vector and Lagrange multiplier difference are finished in training;
3.4 model prediction, input forecast sample collection obtains the single step predicted value to the model that trains, and then predicted value is added former forecast sample collection and forms new forecast sample collection, re-enters model, and circulation obtains the multi-step prediction value according to this;
3.5 the predicted value aftertreatment is reduced to actual value with the model predication value of support vector machine, judges whether to exist abnormity point simultaneously, if there is then in addition smoothing processing.
5. a kind of wind power combination forecasting method based on fuzzy neural network and support vector machine according to claim 3 is characterized in that:
Step 2.1 is determined the number of training sample and the input and output dimension of model, be that computing time and the accuracy of predicting according to model training, prediction requires to determine number of samples, and model adopts multidimensional input one-dimensional to export.
6. a kind of wind power combination forecasting method based on fuzzy neural network and support vector machine according to claim 3 is characterized in that:
Step 2.2 determine model structure and parameter and cycle of training number, be to adopt fuzzy subtractive clustering to determine the structure of fuzzy model, determine that by selecting different cluster radius the model structure of optimum is to guarantee less training error; Cycle of training number definite computing time of need considering model training, avoid over training to guarantee extensive performance simultaneously.
7. a kind of wind power combination forecasting method based on fuzzy neural network and support vector machine according to claim 4 is characterized in that:
Step 3.2 is determined the super parameter value of model, is to adopt the crosscheck method to determine concrete parameter value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101256331A CN102184453A (en) | 2011-05-16 | 2011-05-16 | Wind power combination predicting method based on fuzzy neural network and support vector machine |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011101256331A CN102184453A (en) | 2011-05-16 | 2011-05-16 | Wind power combination predicting method based on fuzzy neural network and support vector machine |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102184453A true CN102184453A (en) | 2011-09-14 |
Family
ID=44570625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011101256331A Pending CN102184453A (en) | 2011-05-16 | 2011-05-16 | Wind power combination predicting method based on fuzzy neural network and support vector machine |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102184453A (en) |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609766A (en) * | 2012-02-17 | 2012-07-25 | 中南大学 | Method for intelligently forecasting wind speed in wind power station |
| CN102682219A (en) * | 2012-05-17 | 2012-09-19 | 鲁东大学 | Method for forecasting short-term load of support vector machine |
| CN102750542A (en) * | 2012-05-29 | 2012-10-24 | 北京交通大学 | Support vector regression machine wind speed combination forecast method with interpolation being smoothed and optimized |
| CN103066594A (en) * | 2012-12-20 | 2013-04-24 | 河南省电力公司 | Power grid risk pre-controlling method based on province power grid topology and city power grid topology and county power grid topology |
| CN103091603A (en) * | 2013-01-14 | 2013-05-08 | 华北电力大学 | Breakdown intelligent classification and positioning method of electric transmission line |
| CN103117546A (en) * | 2013-02-28 | 2013-05-22 | 武汉大学 | Ultrashort-term slide prediction method for wind power |
| CN103218675A (en) * | 2013-05-06 | 2013-07-24 | 国家电网公司 | Short-term load prediction method based on clustering and sliding window |
| CN103516051A (en) * | 2013-10-11 | 2014-01-15 | 国家电网公司 | Centralized monitoring system platform for wind power plants |
| CN103530508A (en) * | 2013-09-30 | 2014-01-22 | 国家电网公司 | Method for establishing wind speed-power conversion probability model |
| CN104155423A (en) * | 2014-08-08 | 2014-11-19 | 天津大学 | Variable weight ing combination landscape water quality early warning method |
| CN104299044A (en) * | 2014-07-01 | 2015-01-21 | 沈阳工程学院 | Clustering-analysis-based wind power short-term prediction system and prediction method |
| CN104520861A (en) * | 2012-08-07 | 2015-04-15 | 韩国能源技术研究院 | Method for predicting wind power density |
| CN104573854A (en) * | 2014-12-23 | 2015-04-29 | 国家电网公司 | Iron steel electricity consumption forecasting method and device |
| CN104598984A (en) * | 2014-12-08 | 2015-05-06 | 北京邮电大学 | Fuzzy neural network based fault prediction method |
| CN104900061A (en) * | 2015-05-29 | 2015-09-09 | 内蒙古工业大学 | Road section travel time monitoring method and device |
| CN104915747A (en) * | 2015-02-03 | 2015-09-16 | 远景能源（江苏）有限公司 | Electricity generation performance evaluation method of generator set and equipment thereof |
| CN104933489A (en) * | 2015-06-29 | 2015-09-23 | 东北电力大学 | Wind power real-time high precision prediction method based on adaptive neuro-fuzzy inference system |
| CN104978605A (en) * | 2015-06-08 | 2015-10-14 | 上海电机学院 | Large-scale wind power prediction system and method based on deep learning network |
| CN105354763A (en) * | 2015-11-16 | 2016-02-24 | 华北电力科学研究院有限责任公司 | Method and device for measuring wind speed of incoming flow of wind turbine generator |
| CN105391083A (en) * | 2015-11-05 | 2016-03-09 | 河海大学 | Wind power range short-term prediction method based on variation mode decomposition and relevant vector machine |
| CN105930900A (en) * | 2016-05-09 | 2016-09-07 | 华北电力大学 | Method and system for predicting hybrid wind power generation |
| CN106326982A (en) * | 2016-08-16 | 2017-01-11 | 北京国能日新系统控制技术有限公司 | Neural network and pattern search optimization-based multi-meteorological wind speed fusion method |
| CN106351793A (en) * | 2015-07-17 | 2017-01-25 | 通用电气公司 | System and method for improved wind power generation |
| CN106383932A (en) * | 2016-09-05 | 2017-02-08 | 上海电机学院 | Wind power prediction method |
| CN106991477A (en) * | 2016-01-20 | 2017-07-28 | 南京艾溪信息科技有限公司 | A kind of artificial neural network compression-encoding device and method |
| CN107239852A (en) * | 2017-05-05 | 2017-10-10 | 南京邮电大学 | A kind of electric quantity consumption Forecasting Methodology based on deep learning |
| CN107480474A (en) * | 2017-08-01 | 2017-12-15 | 山东师范大学 | Grader modeling evaluation method of calibration and system based on gut flora abundance |
| CN107766929A (en) * | 2017-05-05 | 2018-03-06 | 平安科技（深圳）有限公司 | model analysis method and device |
| CN108734359A (en) * | 2018-06-08 | 2018-11-02 | 上海电机学院 | A kind of wind power prediction data preprocessing method |
| CN108846206A (en) * | 2018-06-15 | 2018-11-20 | 中广核研究院有限公司 | A kind of method and apparatus based on a variety of restructing algorithms reconstruct reactor Temporal And Spatial Distribution Model |
| CN109063936A (en) * | 2018-10-12 | 2018-12-21 | 南京千智电气科技有限公司 | A kind of method and device for establishing wind power prediction model |
| CN109376426A (en) * | 2018-10-23 | 2019-02-22 | 株洲中车时代电气股份有限公司 | A kind of wind-electricity integration power dispatching method and device |
| CN109449929A (en) * | 2018-11-22 | 2019-03-08 | 南方电网科学研究院有限责任公司 | Distributed generation resource influences prediction and evaluation method and product to distribution network feeder utilization rate |
| CN109726867A (en) * | 2018-12-27 | 2019-05-07 | 北京恒泰实达科技股份有限公司 | A kind of high-resolution electric power method of meteorological forecast based on multi-mode set |
| CN110276468A (en) * | 2018-03-14 | 2019-09-24 | 北京金风科创风电设备有限公司 | Method and device for predicting generated power of wind generating set |
| CN110672312A (en) * | 2019-10-12 | 2020-01-10 | 北京工业大学 | Method for predicting bolt residual clamping force based on BP neural network |
| US10635741B2 (en) | 2016-11-23 | 2020-04-28 | Industrial Technology Research Institute | Method and system for analyzing process factors affecting trend of continuous process |
| CN112418476A (en) * | 2019-08-23 | 2021-02-26 | 武汉剑心科技有限公司 | Ultra-short-term power load prediction method |
| CN113297704A (en) * | 2021-06-17 | 2021-08-24 | 上海交通大学 | Harmonic reducer power real-time prediction method and system based on hybrid deep neural network |
-
2011
- 2011-05-16 CN CN2011101256331A patent/CN102184453A/en active Pending
Cited By (60)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609766A (en) * | 2012-02-17 | 2012-07-25 | 中南大学 | Method for intelligently forecasting wind speed in wind power station |
| CN102682219A (en) * | 2012-05-17 | 2012-09-19 | 鲁东大学 | Method for forecasting short-term load of support vector machine |
| CN102682219B (en) * | 2012-05-17 | 2016-05-25 | 鲁东大学 | A kind of SVMs short-term load forecasting method |
| CN102750542A (en) * | 2012-05-29 | 2012-10-24 | 北京交通大学 | Support vector regression machine wind speed combination forecast method with interpolation being smoothed and optimized |
| CN104520861B (en) * | 2012-08-07 | 2018-01-09 | 韩国能源技术研究院 | Wind power concentration Forecasting Methodology |
| CN104520861A (en) * | 2012-08-07 | 2015-04-15 | 韩国能源技术研究院 | Method for predicting wind power density |
| CN103066594A (en) * | 2012-12-20 | 2013-04-24 | 河南省电力公司 | Power grid risk pre-controlling method based on province power grid topology and city power grid topology and county power grid topology |
| CN103091603A (en) * | 2013-01-14 | 2013-05-08 | 华北电力大学 | Breakdown intelligent classification and positioning method of electric transmission line |
| CN103091603B (en) * | 2013-01-14 | 2015-01-07 | 华北电力大学 | Breakdown intelligent classification and positioning method of electric transmission line |
| CN103117546A (en) * | 2013-02-28 | 2013-05-22 | 武汉大学 | Ultrashort-term slide prediction method for wind power |
| CN103117546B (en) * | 2013-02-28 | 2016-03-16 | 武汉大学 | A kind of Ultrashort-term slide prediction method for wind power |
| CN103218675A (en) * | 2013-05-06 | 2013-07-24 | 国家电网公司 | Short-term load prediction method based on clustering and sliding window |
| CN103218675B (en) * | 2013-05-06 | 2016-06-01 | 国家电网公司 | A kind of based on the short-term load forecasting method of cluster and moving window |
| CN103530508A (en) * | 2013-09-30 | 2014-01-22 | 国家电网公司 | Method for establishing wind speed-power conversion probability model |
| CN103530508B (en) * | 2013-09-30 | 2017-01-11 | 国家电网公司 | Method for establishing wind speed-power conversion probability model |
| CN103516051A (en) * | 2013-10-11 | 2014-01-15 | 国家电网公司 | Centralized monitoring system platform for wind power plants |
| CN103516051B (en) * | 2013-10-11 | 2016-02-24 | 国家电网公司 | Wind energy turbine set centralized monitoring system platform |
| CN104299044A (en) * | 2014-07-01 | 2015-01-21 | 沈阳工程学院 | Clustering-analysis-based wind power short-term prediction system and prediction method |
| CN104155423A (en) * | 2014-08-08 | 2014-11-19 | 天津大学 | Variable weight ing combination landscape water quality early warning method |
| CN104598984A (en) * | 2014-12-08 | 2015-05-06 | 北京邮电大学 | Fuzzy neural network based fault prediction method |
| CN104598984B (en) * | 2014-12-08 | 2018-03-02 | 北京邮电大学 | A kind of failure prediction method based on fuzzy neural network |
| CN104573854A (en) * | 2014-12-23 | 2015-04-29 | 国家电网公司 | Iron steel electricity consumption forecasting method and device |
| CN104915747B (en) * | 2015-02-03 | 2019-02-01 | 远景能源（江苏）有限公司 | A kind of the power generation performance appraisal procedure and equipment of generating set |
| CN104915747A (en) * | 2015-02-03 | 2015-09-16 | 远景能源（江苏）有限公司 | Electricity generation performance evaluation method of generator set and equipment thereof |
| CN104900061A (en) * | 2015-05-29 | 2015-09-09 | 内蒙古工业大学 | Road section travel time monitoring method and device |
| CN104978605A (en) * | 2015-06-08 | 2015-10-14 | 上海电机学院 | Large-scale wind power prediction system and method based on deep learning network |
| CN104933489A (en) * | 2015-06-29 | 2015-09-23 | 东北电力大学 | Wind power real-time high precision prediction method based on adaptive neuro-fuzzy inference system |
| CN106351793A (en) * | 2015-07-17 | 2017-01-25 | 通用电气公司 | System and method for improved wind power generation |
| US10443577B2 (en) | 2015-07-17 | 2019-10-15 | General Electric Company | Systems and methods for improved wind power generation |
| CN106351793B (en) * | 2015-07-17 | 2020-05-12 | 通用电气公司 | System and method for improving wind power generation |
| CN105391083A (en) * | 2015-11-05 | 2016-03-09 | 河海大学 | Wind power range short-term prediction method based on variation mode decomposition and relevant vector machine |
| CN105391083B (en) * | 2015-11-05 | 2018-05-25 | 河海大学 | Wind power interval short term prediction method based on variation mode decomposition and Method Using Relevance Vector Machine |
| CN105354763A (en) * | 2015-11-16 | 2016-02-24 | 华北电力科学研究院有限责任公司 | Method and device for measuring wind speed of incoming flow of wind turbine generator |
| CN106991477B (en) * | 2016-01-20 | 2020-08-14 | 中科寒武纪科技股份有限公司 | Artificial neural network compression coding device and method |
| CN106991477A (en) * | 2016-01-20 | 2017-07-28 | 南京艾溪信息科技有限公司 | A kind of artificial neural network compression-encoding device and method |
| CN105930900B (en) * | 2016-05-09 | 2018-02-27 | 华北电力大学 | The Forecasting Methodology and system of a kind of hybrid wind power generation |
| CN105930900A (en) * | 2016-05-09 | 2016-09-07 | 华北电力大学 | Method and system for predicting hybrid wind power generation |
| CN106326982A (en) * | 2016-08-16 | 2017-01-11 | 北京国能日新系统控制技术有限公司 | Neural network and pattern search optimization-based multi-meteorological wind speed fusion method |
| CN106326982B (en) * | 2016-08-16 | 2019-02-15 | 国能日新科技股份有限公司 | A kind of more meteorological wind speed fusion methods optimized based on neural network and pattern search |
| CN106383932A (en) * | 2016-09-05 | 2017-02-08 | 上海电机学院 | Wind power prediction method |
| US10635741B2 (en) | 2016-11-23 | 2020-04-28 | Industrial Technology Research Institute | Method and system for analyzing process factors affecting trend of continuous process |
| CN107766929B (en) * | 2017-05-05 | 2019-05-24 | 平安科技（深圳）有限公司 | Model analysis method and device |
| US11507963B2 (en) | 2017-05-05 | 2022-11-22 | Ping An Technology (Shenzhen) Co., Ltd. | Method and device of analysis based on model, and computer readable storage medium |
| CN107239852A (en) * | 2017-05-05 | 2017-10-10 | 南京邮电大学 | A kind of electric quantity consumption Forecasting Methodology based on deep learning |
| CN107766929A (en) * | 2017-05-05 | 2018-03-06 | 平安科技（深圳）有限公司 | model analysis method and device |
| CN107480474A (en) * | 2017-08-01 | 2017-12-15 | 山东师范大学 | Grader modeling evaluation method of calibration and system based on gut flora abundance |
| CN107480474B (en) * | 2017-08-01 | 2019-03-26 | 山东师范大学 | Classifier modeling evaluation method of calibration and system based on intestinal flora abundance |
| CN110276468A (en) * | 2018-03-14 | 2019-09-24 | 北京金风科创风电设备有限公司 | Method and device for predicting generated power of wind generating set |
| CN108734359B (en) * | 2018-06-08 | 2021-10-01 | 上海电机学院 | Wind power prediction data preprocessing method |
| CN108734359A (en) * | 2018-06-08 | 2018-11-02 | 上海电机学院 | A kind of wind power prediction data preprocessing method |
| CN108846206A (en) * | 2018-06-15 | 2018-11-20 | 中广核研究院有限公司 | A kind of method and apparatus based on a variety of restructing algorithms reconstruct reactor Temporal And Spatial Distribution Model |
| CN109063936A (en) * | 2018-10-12 | 2018-12-21 | 南京千智电气科技有限公司 | A kind of method and device for establishing wind power prediction model |
| CN109376426B (en) * | 2018-10-23 | 2023-04-07 | 株洲中车时代电气股份有限公司 | Wind power grid-connected power scheduling method and device |
| CN109376426A (en) * | 2018-10-23 | 2019-02-22 | 株洲中车时代电气股份有限公司 | A kind of wind-electricity integration power dispatching method and device |
| CN109449929A (en) * | 2018-11-22 | 2019-03-08 | 南方电网科学研究院有限责任公司 | Distributed generation resource influences prediction and evaluation method and product to distribution network feeder utilization rate |
| CN109726867A (en) * | 2018-12-27 | 2019-05-07 | 北京恒泰实达科技股份有限公司 | A kind of high-resolution electric power method of meteorological forecast based on multi-mode set |
| CN109726867B (en) * | 2018-12-27 | 2020-07-28 | 北京恒泰实达科技股份有限公司 | High-resolution electric power weather forecasting method based on multi-mode set |
| CN112418476A (en) * | 2019-08-23 | 2021-02-26 | 武汉剑心科技有限公司 | Ultra-short-term power load prediction method |
| CN110672312A (en) * | 2019-10-12 | 2020-01-10 | 北京工业大学 | Method for predicting bolt residual clamping force based on BP neural network |
| CN113297704A (en) * | 2021-06-17 | 2021-08-24 | 上海交通大学 | Harmonic reducer power real-time prediction method and system based on hybrid deep neural network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102184453A (en) | Wind power combination predicting method based on fuzzy neural network and support vector machine | |
| Yang et al. | Day-ahead wind power forecasting based on the clustering of equivalent power curves | |
| US10443577B2 (en) | Systems and methods for improved wind power generation | |
| Shi et al. | Hybrid forecasting model for very-short term wind power forecasting based on grey relational analysis and wind speed distribution features | |
| CN102663513B (en) | Utilize the wind power combined prediction modeling method of grey relational grade analysis | |
| Wan et al. | Direct quantile regression for nonparametric probabilistic forecasting of wind power generation | |
| CN107591844B (en) | Active power distribution network robust reconstruction method considering node injection power uncertainty | |
| CN103390116B (en) | Use the photovoltaic power station power generation power forecasting method of stepping mode | |
| CN107425520B (en) | Active power distribution network three-phase interval state estimation method containing node injection power uncertainty | |
| CN107358060B (en) | The method that a kind of pair of wind power prediction burst error is estimated | |
| WO2023004838A1 (en) | Wind power output interval prediction method | |
| CN102102626B (en) | Method for forecasting short-term power in wind power station | |
| CN102269124B (en) | Ultra-short term wind power station generated power forecasting system | |
| CN104978608B (en) | A kind of wind electric powder prediction device and prediction technique | |
| CN106505631B (en) | Intelligent wind power wind power prediction system | |
| CN104899665A (en) | Wind power short-term prediction method | |
| CN105046374A (en) | Power interval predication method based on nucleus limit learning machine model | |
| CN106875033A (en) | A kind of wind-powered electricity generation cluster power forecasting method based on dynamic self-adapting | |
| CN102938562B (en) | Prediction method of total wind electricity power in area | |
| CN105069521A (en) | Photovoltaic power plant output power prediction method based on weighted FCM clustering algorithm | |
| CN102509026A (en) | Comprehensive short-term output power forecasting model for wind farm based on maximum information entropy theory | |
| CN110298494A (en) | A kind of wind power forecasting method based on Segment Clustering and Combinatorial Optimization | |
| CN106485603A (en) | A kind of short-term wind speed forecasting method being applied to wind-power electricity generation | |
| CN112215392A (en) | Method for predicting power generation capacity of wind power medium and long term region based on equipment state and environmental factors | |
| Zhang et al. | Joint forecasting of regional wind and solar power based on attention neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110914 |