An approach to the identification of temperature in intelligent building based on feed forward neural network and genetic algorithm

2010 ◽  
Author(s):  
Zhang Zhen-Ya ◽  
Cheng Hong-Mei ◽  
Zhang Shu-Guang
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Feed Forward Neural Network ◽  
Intelligent Building ◽  
Feed Forward

An Optimization Model for the Identification of Temperature in Intelligent Building

2011 ◽  
Vol 4 (2) ◽  
pp. 61-69 ◽  
Author(s):  
ZhenYa Zhang ◽  
HongMei Cheng ◽  
ShuGuang Zhang
Keyword(s):  
Neural Network ◽  
Optimization Problem ◽  
Optimization Problems ◽  
Fitness Function ◽  
Temperature Data ◽  
Temperature Fields ◽  
Observation Data ◽  
Feed Forward Neural Network ◽  
Intelligent Building ◽  
Feed Forward

Methods for the reconstruction of temperature fields in an intelligent building with temperature data of discrete observation positions is a current topic of research. To reconstruct temperature field with observation data, it is necessary to model the identification of temperature in each observation position. In this paper, models for temperature identification in an intelligent building are formalized as optimization problems based on observation temperature data sequence. To solve the optimization problem, a feed forward neural network is used to formalize the identification structure, and connection matrixes of the neural network are the identification parameters. With the object function for the given optimization problem as the fitness function, the training of the feed forward neural network is driven by a genetic algorithm. The experiment for the precision and stability of the proposed method is designed with real temperature data from an intelligent building.

An Optimization Model for the Identification of Temperature in Intelligent Building

2013 ◽  
pp. 116-124
Author(s):  
ZhenYa Zhang ◽  
HongMei Cheng ◽  
ShuGuang Zhang
Keyword(s):  
Neural Network ◽  
Optimization Problem ◽  
Optimization Problems ◽  
Fitness Function ◽  
Temperature Data ◽  
Temperature Fields ◽  
Observation Data ◽  
Feed Forward Neural Network ◽  
Intelligent Building ◽  
Feed Forward

Methods for the reconstruction of temperature fields in an intelligent building with temperature data of discrete observation positions is a current topic of research. To reconstruct temperature field with observation data, it is necessary to model the identification of temperature in each observation position. In this paper, models for temperature identification in an intelligent building are formalized as optimization problems based on observation temperature data sequence. To solve the optimization problem, a feed forward neural network is used to formalize the identification structure, and connection matrixes of the neural network are the identification parameters. With the object function for the given optimization problem as the fitness function, the training of the feed forward neural network is driven by a genetic algorithm. The experiment for the precision and stability of the proposed method is designed with real temperature data from an intelligent building.

AN IMPROVED GENETIC-ALGORITHM-BASED NEURAL-TUNED NEURAL NETWORK

2008 ◽  
Vol 07 (04) ◽  
pp. 469-492
Author(s):  
F. H. F. LEUNG ◽  
S. H. LING ◽  
H. K. LAM
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Network Structure ◽  
Neuron Model ◽  
Search Space ◽  
Superior Performance ◽  
Activation Functions ◽  
Improved Genetic Algorithm ◽  
Feed Forward Neural Network ◽  
Feed Forward

This paper presents a neural-tuned neural network (NTNN), which is trained by an improved genetic algorithm (GA). The NTNN consists of a common neural network and a modified neural network (MNN). In the MNN, a neuron model with two activation functions is introduced. An improved GA is proposed to train the parameters of the proposed network. A set of improved genetic operations are presented, which show superior performance over the traditional GA. The proposed network structure can increase the search space of the network and offer better performance than the traditional feed-forward neural network. Two application examples are given to illustrate the merits of the proposed network and the improved GA.

Sign in / Sign up

Export Citation Format

Share Document