20230136352. METHOD AND SYSTEM FOR PREDICTING A DAY-AHEAD WIND POWER OF WIND FARMS simplified abstract (Economic and Technological Research Institute of State Grid Liaoning Electric Power Co., Ltd.)

METHOD AND SYSTEM FOR PREDICTING A DAY-AHEAD WIND POWER OF WIND FARMS

Organization Name

Economic and Technological Research Institute of State Grid Liaoning Electric Power Co., Ltd.

Inventor(s)

XIAO Pan of Shenyang (CN)

MINGLI Zhang of Shenyang (CN)

LIN Zhao of Shenyang (CN)

NA Zhang of Shenyang (CN)

ZHUORAN Song of Shenyang (CN)

NANTIAN Huang of Jilin (CN)

JING Gao of Shenyang (CN)

XUMING Lv of Shenyang (CN)

HUA Li of Shenyang (CN)

MENGZENG Cheng of Shenyang (CN)

XING Ji of Shenyang (CN)

WENYING Shang of Shenyang (CN)

YIXIN Hou of Shenyang (CN)

SUO Yang of Shenyang (CN)

BO Yang of Shenyang (CN)

YUTONG Liu of Shenyang (CN)

LINKUN Man of Shenyang (CN)

XILIN Xu of Shenyang (CN)

HAIFENG Yang of Shenyang (CN)

FANGYUAN Yang of Shenyang (CN)

KAI Liu of Shenyang (CN)

JINQI Li of Shenyang (CN)

ZONGYUAN Wang of Shenyang (CN)

METHOD AND SYSTEM FOR PREDICTING A DAY-AHEAD WIND POWER OF WIND FARMS - A simplified explanation of the abstract

This abstract first appeared for US patent application 20230136352 titled 'METHOD AND SYSTEM FOR PREDICTING A DAY-AHEAD WIND POWER OF WIND FARMS

Simplified Explanation

The abstract describes a method for predicting the day-ahead wind power of wind farms. Here is a simplified explanation of the abstract:

• The method involves constructing a raw data set based on the correlation between the to-be-predicted daily wind power, numerical weather forecast meteorological features, and historical daily wind power.
• The raw data set is then clustered using k-means clustering, resulting in a data set with cluster labels.
• Massive labeled scenes are generated using robust auxiliary classifier generative adversarial networks based on the cluster labels.
• The cluster label category of the to-be-predicted day is determined based on known historical daily wind power and numerical weather forecast meteorological features.
• Multiple scenes with high similarity to the to-be-predicted daily wind power are screened out based on the cluster label category.
• Prediction results of the to-be-predicted daily wind power are obtained at multiple set times using the average value, upper limit value, and lower limit value of the to-be-predicted daily wind power.

Potential applications of this technology:

• Renewable energy management: The method can be used to predict the day-ahead wind power of wind farms, helping in the efficient management and utilization of renewable energy resources.
• Power grid planning: Accurate wind power predictions can assist in planning the power grid and optimizing the integration of wind energy into the existing infrastructure.

Problems solved by this technology:

• Uncertainty in wind power generation: By utilizing historical wind power data and numerical weather forecast meteorological features, this method addresses the challenge of predicting wind power generation accurately.
• Resource optimization: The method helps in optimizing the utilization of wind energy resources by providing reliable day-ahead predictions.

Benefits of this technology:

• Improved energy management: Accurate wind power predictions enable better planning and management of renewable energy resources, leading to more efficient energy generation and utilization.
• Cost savings: By optimizing the utilization of wind energy resources, this method can help reduce costs associated with energy generation and grid management.
• Enhanced grid stability: Reliable wind power predictions contribute to the stability and reliability of the power grid by facilitating better integration of intermittent renewable energy sources.

Original Abstract Submitted

a method for predicting a day-ahead wind power of wind farms, comprising: constructing a raw data set based on a correlation between the to-be-predicted daily wind power, the numerical weather forecast meteorological feature and a historical daily wind power; obtaining a clustered data set and performing k-means clustering, obtaining a raw data set with cluster labels, and generating massive labeled scenes based on robust auxiliary classifier generative adversarial networks; determining the cluster label category of the to-be-predicted day based on the known historical daily wind power and numerical weather forecast meteorological feature, and screening out multiple scenes with high similarity to the to-be-predicted daily wind power based on the cluster label category; and obtaining the prediction results of the to-be-predicted daily wind power at a plurality of set times based on an average value, an upper limit value and a lower limit value of the to-be-predicted daily wind power.