Method for Prediction for Nonlinear Seasonal Time Series

• US 20110160927A1
• Filed: 12/30/2009
• Published: 06/30/2011
• Est. Priority Date: 12/30/2009
• Status: Abandoned Application

• Current Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Electric Research Laboratories, Inc. (Mitsubishi Electric Corporation)
• Inventors: Jiang, Xiaoqian, Wilson, Kevin W.
• Application Number:
  US12/650,452
• Publication Number:
  US 20110160927A1
• Time in Patent Office: Days
• US Class Current: 700/291
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202

Abstract

A method predicts nonlinear seasonal time series data. More particularly, the method predicts short term power demand. The method uses an exponential smoothing technique and dynamic modeling of seasonal time series data with kernels. The kernels are used to predict future values of the time series from past values using nonlinear functions such as a least-squares radial basis function or support vector regression using a Gaussian kernel.

Claims

1. A method for predicting future values of a time series from past values of the time series, wherein L_t and I_t are values of an initial local mean level, and a seasonal index at a time t, and α

and δ

are smoothing parameters for updating the mean level and the seasonal index, respectively, and p is a duration of a seasonal pattern, comprising for each next local mean level L_t the steps of;

determining {circumflex over (X)}_t (1)=I_t-pF_t(L_t), wherein {circumflex over (X)}_t (1) is a prediction made at time t one time step in the future, and F_t is a nonlinear function for a vector L_t of local level means;

updating the local mean L_t and the seasonal index according to

L_t=α

(X_t/I<sub>t−

p</sub>)+(1−

α

)({circumflex over (X)}<sub>t−

1</sub> (1)/I<sub>t−

p</sub>)

I_t=δ

(X_t/L_t)+(1−

δ

)I<sub>t−

p</sub>; and

estimating a next F_t (L) from past values of L, wherein the determining, updating and estimating steps are performed in a processor.

2. The method of claim 1, wherein the time series is short term power demand.

3. The method of claim 1, wherein the time series have seasonal patterns and nonlinear relationships between the past values and the future values.

4. The method of claim 1, wherein the nonlinear function is a least-squares radial basis function.

5. The method of claim 1, wherein the nonlinear function is a support vector regression using a Gaussian kernel.