No abstract

No abstract

No abstract

Highlight: A new optimization algorithm inspired by how duelist improve their skill in duel  In duelist algorithm, different treatment is given to each duelist based on the duel result  Duelist algorithm provided good result compared to the other optimization algorithms such as genetic algorithm, particle swarm optimization algorithm and imperialist competitive algorithm

No abstract

The development of green building has been growing in both design and quality. The development of green building was limited by the issue of expensive investment. Actually, green building can reduce the energy usage inside the building especially in utilization of cooling system. External load plays major role in reducing the usage of cooling system. External load is affected by type of wall sheathing, glass and roof. The proper selection of wall, type of glass and roof material are very important to reduce external load. Hence, the optimization of energy efficiency and conservation in green building design is required. Since this optimization consist of integer and non-linear equations, this problem falls into Mixed-Integer-Non-Linear-Programming (MINLP) that required global optimization technique such as stochastic optimization algorithms. In this paper the optimized variables i.e. type of glass and roof were chosen using Duelist, Killer-Whale and Rain-Water Algorithms to obtain the optimum energy and considering the minimal investment. The optimization results exhibited the single glass Planibel-G with the 3.2 mm thickness and glass wool insulation provided maximum ROI of 36.8486%, EUI reduction of 54 kWh/m2·year, CO2 emission reduction of 486.8971 tons/year and reduce investment of 4,078,905,465 IDR.

No abstract

A Neural Network Internal Model Control (NN-IMC) strategy is investigated, by establishing inverse and forward model based neural network (NN). Further for developing the model has been selected suitable adaptive filter. Two types of NN-based inverse model (i.e. with and without disturbance input) were accurately simulated. The results indicated that the neural networks are capable to establish forward and inverse model rapidly from the couple of input-output open loop data of single distillation column binary system with a good root mean square error (RMSE). The simulation results revealed that NN-IMC with appropriate learning rate -momentum is capable to pursue the set-point changes and to reject the disturbance changes without steady state error or oscillations. NN-IMC with inverse model which contains disturbance input (modified NN-IMC) offer better performance than without it (conventional NN-IMC).