Electricity Load Forecasting Using Support Vector Regression with Memetic Algorithms

Table 1

Performance metrics and their formulas in regression problems.

Metrics	Formula
MAPE	1MAPE=𝑁^𝑁𝑖∑=1𝑦𝑡+𝑖−𝑦𝑡+𝑖𝑦𝑡+𝑖×100
MASE	1MASE=𝑁^𝑁𝑖∑=1𝑦𝑡+𝑖−𝑦𝑡+𝑖(1/(𝑡−1))∑^𝑡𝑗=2𝑦𝑗−𝑦𝑗−1
DS	∑DS=^𝑁𝑖=2𝑑𝑖𝑁−1×100; 𝑑𝑖=1,if𝑦𝑡+𝑖−𝑦𝑡+𝑖−1𝑦𝑡+𝑖−𝑦𝑡+𝑖−1≥00,otherwise

 𝑁 is the number of forecasting periods, 𝑦_𝑡+𝑖 is the actual value at period 𝑡+𝑖, 𝑦_𝑡+𝑖 is the forecasting value at period 𝑡+𝑖, and 𝑦 is the mean of all values. In this study, the day-ahead (24 hours) short-term load is forecasted recursively, so the number of forecasting periods N equals 24.