Ultra-short-term load prediction method and system including error correction
A technology of load forecasting and error correction, applied in forecasting, data processing applications, instruments, etc., can solve the problems of failure to effectively combine the advantages of the two models, singleness, etc., and achieve the effect of improving forecasting accuracy and high forecasting accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0040] This embodiment discloses an ultra-short-term load forecasting method including error correction, including:
[0041] Step 1: Obtain user load historical data;
[0042] Collect the historical data of user load data and divide it into three parts: HW training set, used to obtain the Holt-Winter prediction model; error test set, used to obtain the training set of error prediction; combined test set, used to verify the method prediction accuracy. Considering the seasonality and periodicity of the load data, the data from January to June of the year is used as the HW training set, the data from July to September is used as the error test set, and the data from October to December is used as the combined test set, with a data interval of 15 minutes.
[0043] Step 2: Preprocessing the load history data;
[0044] Data stability: The unit root (augmented Dickey Fuller, ADF) was used to test the data stability to ensure that the Holt-Winter method was applicable.
[0045] Dat...
Embodiment 2
[0057] The purpose of this embodiment is to provide an ultra-short-term load forecasting system including error correction.
[0058] In order to achieve the above purpose, this embodiment provides an ultra-short-term load forecasting system including error correction, including:
[0059] The data acquisition module obtains user load historical data, and the user load historical data includes a training data set and a test set of a specified period;
[0060] Holt-Winter predictor building block, training Holt-Winter predictor based on training dataset;
[0061] Error predictor construction module, based on the Holt-Winter predictor, the user load of the specified period is predicted, and according to the predicted value and test set of the specified period, an error prediction training set is obtained; based on the error prediction training set training is based on Error predictors for extreme learning machines;
[0062] The load forecasting module obtains a combined forecast...
Embodiment 3
[0064] The purpose of this embodiment is to provide an electronic device.
[0065] An electronic device, comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, when the processor executes the program, the following steps are implemented, including:
[0066] Obtaining user load historical data, the user load historical data includes a training data set and a test set of a specified period;
[0067] Training the Holt-Winter predictor based on the training data set, and predicting the user load of the specified time period based on the Holt-Winter predictor;
[0068] Obtaining an error prediction training set according to the predicted value and the test set in the specified time period;
[0069] Based on the error prediction training set, the error predictor based on the extreme learning machine is trained;
[0070] Based on the Holt-Winter predictor and the error predictor, a combined forecasting model is obtained for load...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More