Short-term power load prediction method based on hybrid model
A technology of short-term power load and forecasting method, which is applied in the field of power system, can solve the problem of large data forecasting error and achieve the effect of improving accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0054] Such as figure 1 , figure 2 and image 3 As shown, a short-term power load forecasting method based on a hybrid model includes the following steps:
[0055] S1: Establish an autoregressive differential moving average model based on the original power load data to obtain a stationary sequence y t with predicted value
[0056] S2: through the stationary sequence y t with predicted value Get the residual sequence e t ;
[0057] S3: For the residual sequence e t Using time convolutional network model to model and get the result
[0058] S4: Linearly combine the prediction results of the autoregressive differential moving average model and the time convolutional network model to obtain the final prediction result;
[0059] S5: For the finally obtained prediction result, use the model in step S3 to calculate the performance index to evaluate the prediction effect.
[0060] In the above scheme, the hybrid model structure is a combination of the autoregressive d...
Embodiment 2
[0090] Such as Figure 4 , Figure 5 and Figure 6 shown, yes Figure 4 ARIMA modeling is performed on the real original power load data of a certain place shown; this data is the training sample in this embodiment, which is the load data of a certain place from May 1, 2016 to December 5, 2016, and the sampling frequency is 1 Once an hour, a total of 5256 hours of load data.
[0091] 1.1) Carry out a stationarity test on the original data to determine that the data is not stable.
[0092] 1.2) According to the variation trend of the original data, it is preliminarily determined that its differential order d=1. After the first order difference of the original data, we can get Figure 5 The differential sequence diagram shown. right Figure 5 The data shown are tested for stationarity and found that the sequence has been stationary.
[0093] 1.3) Calculate the average value, variance, autovariance function, autocorrelation function, partial autocorrelation function, etc....
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More