Power load prediction method and device, and terminal equipment

Abstract

The invention discloses a power load prediction method and device, and terminal equipment. The method comprises the steps of obtaining power load data; based on a Spark engine, inputting the power load data into a model based on a K-means algorithm, optimizing the model by adopting an optimizer to obtain a clustering model, and outputting clustered power load data to be predicted; dividing the to-be-predicted power load data into a training set and a prediction set; converting the training set and the prediction set into a first RDD data set and a second RDD data set; and inputting the first RDD data set into an XGboost model based on a Spark engine, optimizing the model by using an optimizer to obtain a load prediction model, and performing power load prediction on the second RDD data set. According to the embodiment of the invention, by automatically selecting the optimal parameters of the K-means-based model and the XGboost model, the time of model training is greatly shortened, and the precision of power load prediction is improved.

Description

technical field [0001] The present invention relates to the field of power data processing, in particular to a power load forecasting method, device and terminal equipment. Background technique [0002] As the smart grid and clean energy gradually become the development direction of the power industry, power load forecasting has attracted more and more attention from power workers. Accurate power load forecasting can provide an important basis for the dispatching strategy of the power system and the adjustment of the operation structure of the power grid, and effectively improve the stability of the power system operation. [0003] At present, the commonly used methods of power load forecasting can be divided into classical methods and artificial intelligence methods. The classical method is a method based on mathematical modeling, including time series analysis, regression analysis, etc. However, the classical model is a single linear model, and its prediction accuracy can...

AI Technical Summary

Problems solved by technology

The classical method is a method based on mathematical modeling, including time series analysis, regression analysis, etc. However, the classical model is a single linear model, and its prediction accuracy cannot meet the status quo. There are more than 32 factors affecting the current power load , simple regression analysis has been unable to fit the model well

Artificial intelligence methods such as: neural network, SVM, random forest, etc., have good performance in fitting multidimensional data. However, when using artificial intelligence methods to build models, engineers need to use their own experience to evaluate the performance of the model. Optimization, including tuning and selecting model parameters, and selecting the loss function and regularization items of the model, not only greatly increases the time of model training, but also cannot ensure that the optimized model is the optimal model, which affects power load forecasting Accuracy of results

Images