Patents
Literature
Patsnap Copilot is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Patsnap Copilot

38 results about "Weather patterns" patented technology

Wind power plant short-term wind speed forecasting method based on weather patterns

The invention discloses a wind power plant short-term wind speed forecasting method based on weather patterns. The wind power plant short-term wind speed forecasting method based on the weather patterns mainly solves the problems that the situation that different changing rules of a wind speed time sequence are presented under different weather conditions is not taken into account by the prior wind speed forecasting method, and a regression forecasting model is not appropriate to build because a clustering result given by a fuzzy cluster is spherical. The wind power plant short-term wind speed forecasting method based on the weather patterns comprises the following steps that feature expression is conducted on each wind speed sample point; second, subspace cutting is conducted on a feature space using a GPCA algorithm, and each sample point is projected in a corresponding subspace; third, a support vector machine forecasting model is built for sample points in the subspaces; fourth, integrated forecasting is conducted according to the membership degrees of a current sample point for each subspace and a forecasting result is given. The wind power plant short-term wind speed forecasting method based on the weather patterns can be applied to the field of large-scale wind power resource application.
Owner:HARBIN INST OF TECH

Ultra short-term chaos prediction method for photovoltaic output power

The invention relates to an ultra short-term chaos prediction method for photovoltaic output power. The duration of the ultra short-term lasts from zero to four hours, and the method comprises the following steps: using the C-C method to obtain the optimal delay amount l and the optimal embedding dimension m for the time sequence of photovoltaic output power; reconstructing the phase space for the time sequence of the photovoltaic power; determining the predicted center phase space point Pk according to the phase space for the time sequence of the photovoltaic power; selecting the adjacent phase space point Pkj corresponding to the predicted center phase space point and calculating the weight Wj of the adjacent phase space point Pkj; according to the weight Wj of the adjacent phase space point Pkj, building a photovoltaic weight first-order local linear regression model; calculating the optimal linear fitting coefficient matrix; and calculating the predicted photovoltaic output power value according to the optimal linear fitting coefficient matrix. Compared with the prior art, the method of the invention does not have to obtain metrological data in advance and does not need to establish prediction models for different weather patterns. The model can be built simply. The prediction consumes a short time but higher prediction accuracy can be achieved.
Owner:SHANGHAI UNIVERSITY OF ELECTRIC POWER

Day-ahead optical power prediction method based on similar day clustering and meteorological factor weighting

The invention discloses a day-ahead optical power prediction method based on similar day clustering and meteorological factor weighting. The method comprises the steps of determining and obtaining a day meteorological factor feature vector of a prediction day, and employing a clustering method to screen out photovoltaic power station historical data similar to the prediction day from a historicaldata set according with the season type and weather mode of the prediction day, and making a similar day training sample set; establishing an optical power prediction model based on an Elman neural network; determining input and output factors of the model; calculating a Pearson correlation coefficient of each meteorological factor sequence and the solar photovoltaic power sequence in each similarday training sample, weighting an input factor of the training sample based on the Pearson correlation coefficient, and training an optical power prediction model; and obtaining meteorological prediction data of the prediction day, and predicting the predicted solar photovoltaic power by using the trained optical power prediction model. The method can predict the day-ahead power of the photovoltaic power station, and is suitable for the built photovoltaic power station.
Owner:TIANJIN UNIV

LED panel light for simulating weather conditions

The invention relates to an LED panel light for simulating weather conditions. The LED panel light for simulating weather conditions comprises a light source plate, a power module and a control module, wherein N light bead groups are arranged on the light source plate; each of the light bead groups comprises a plurality of cool-colour light beads and a plurality of warm-colour light beads; the power module supplies power to the control module and light bead modules, and the power module comprises N power circuits; each of the power circuits outputs two kinds of current signals, the N1 current signal flows to a cool-colour light bead bunch, and the N2 current signal flows to a warm-colour light bead bunch; and the control module comprises a weather condition obtaining unit, a colortemperature control unit, a weather display control unit and a control unit. Hollow-out patterns for showing weather conditions are arranged on a diffusion plate, and the weather condition of the current state of the moment can be showed by illuminating the light bead groups corresponding weather patterns, coupled with light color temperature adjustments and through intermittent lighting of two kinds of different color temperature light beads to simulate the current moment color temperature, and the atmosphere of a room in the current weather condition is further improved.
Owner:ZHEJIANG KAIYAO LIGHTING

Wind and light typical scene construction method containing meteorological data based on density peak value-FCM

The invention discloses a wind and light typical scene construction method containing meteorological data based on a density peak value-FCM, and the wind and light typical scene construction method comprises the following steps: S1, obtaining the data of historical numerical weather forecast and the historical year actual measurement output data of a wind power plant and a photovoltaic power station, and constructing an original data set matrix; S2, performing dimension reduction processing on the original data set matrix constructed in the S1, and constructing a weather feature matrix after dimension reduction; S3, according to a fuzzy C-means clustering algorithm optimized by a density peak value, carrying out clustering processing on the weather characteristic matrix obtained in the S2 after dimension reduction; and S4, classifying the historical sunshine output data of the wind power plant and the photovoltaic power station according to a weather mode identification result in the S3, and then fitting to obtain wind and light typical sunshine output scenes in different weather modes. According to the method, dimension reduction processing is carried out on various meteorological feature data, so that original data features can be reserved to the maximum extent, data redundancy can be avoided, and the calculation efficiency is improved.
Owner:国网青海省电力公司清洁能源发展研究院 +3

Trajectory prediction method and device for reciprocating sounding system

The invention provides a trajectory prediction method and device for a reciprocating sounding system. The method comprises the following steps: determining the radius of a balloon when the system ascends or floats according to thermodynamic properties of inflation gas and balloon skin in combination with high-temporal-spatial-resolution atmospheric environment parameters predicted by a numerical weather mode; according to the first drag coefficient, the effective cross sectional area of the balloon and the volume of the balloon, determining the vertical acceleration of the system under the static atmosphere condition; according to the vertical acceleration and the mode high space-time atmospheric environment three-dimensional velocity field, predicting trajectory of a system in an ascending stage and a floating stage; and predicting the descending section trajectory according to the second drag coefficient and the atmospheric environment parameters. Therefore, the high-temporal-spatial-resolution atmospheric environment parameters, the high-temporal-spatial-resolution atmospheric three-dimensional velocity field, the drag coefficient with higher precision and the thermodynamic property which are predicted by directly adopting the high-resolution numerical weather mode improve the precision of trajectory prediction.
Owner:NATIONAL METEOROLOGICAL CENTRE +1
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products