FUNCTION GENERATION DEVICE, HEAT SOURCE EQUIPMENT CONTROL DEVICE, AIR CONDITIONING SYSTEM, FUNCTION GENERATION METHOD, AND PROGRAM FOR FUNCTION GENERATION
PROBLEM TO BE SOLVED: To generate a model function capable of achieving local high accuracy and the stabilization of global characteristics.SOLUTION: A function generation device (100) includes: a storage part (101) for storing learning data related to the input and output of an object system; a div...
Saved in:
Main Authors | , , , , , |
---|---|
Format | Patent |
Language | English Japanese |
Published |
03.12.2015
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | PROBLEM TO BE SOLVED: To generate a model function capable of achieving local high accuracy and the stabilization of global characteristics.SOLUTION: A function generation device (100) includes: a storage part (101) for storing learning data related to the input and output of an object system; a division part (102) for dividing a plurality of learning data stored in the storage part for each of input spaces (H, L) divided into a plurality of pieces; a setting part (103) for setting a mathematical programming problem; and a function estimation part (104) for estimating a model function indicating the object system in accordance with the mathematical programming problem set by the setting part. The mathematical programming problem includes: a constraint condition related to an error between the learning data in each of the divided input spaces and the output value of a function defined for each of the divided input spaces; and a constraint condition related to a difference between the output values of the respective functions at a boundary between the divided input spaces.
【課題】局所的な高精度化と大域的な特性の安定化を両立したモデル関数を生成する。【解決手段】本発明に係る関数生成装置(100)は、対象システムの入力と出力とに関する学習データを記憶する記憶部(101)と、前記記憶部に記憶された複数の前記学習データを、複数に分割された入力空間(H、L)毎に分割する分割部(102)と、数理計画問題を設定する設定部(103)と、前記設定部によって設定された前記数理計画問題に従って、前記対象システムを表すモデル関数を推定する関数推定部(104)とを備え、前記数理計画問題は、分割された入力空間の夫々における前記学習データと前記分割された入力空間毎に定義した関数の出力値との誤差に関する制約条件と、夫々の前記分割された入力空間の境界における前記関数の夫々の出力値の差異に関する制約条件とを含むことを特徴とする。【選択図】図3 |
---|---|
Bibliography: | Application Number: JP20140096514 |