Essay: Economic/Environmental Dispatch (EED)

I have proposed an Economic/Environmental Dispatch (EED) which satisfies the operational power constraint and also considering the pollutant emission and cost of fuel. EED addresses the issue of environment during economic dispatch. First the problem is formulated and Pareto Archive Multiobjective Particle Swarm Optimisation (PAMPSO) algorithm is proposed to deal with this problem. The main aim is to less the cost of operations and pollutants. PAMPSO uses the local search method to increase its search efficiency. The two main problems are EED and MEED which takes care of environmental issues and further it also deals with the operational cost and emission of pollutants in multiple areas. It also sees to the fuel cost optimization. The main objective is to less the fuel cost in power generation considering the overall demand. Optimization should follow some constraints like demand balance, tie line capacity limit and capacity limit.
Population based search algorithm like Particle Swarm Optimization (PSO) algorithm is used. In this algorithm each individual is taken as particle and its movement is observed in particular dimension. The velocity and position of each particle is taken and used to calculate the optimal solution. PAMPSO algorithm uses the basic idea of PSO algorithm with some other additional things like archive storage. Mainly archive stores all the non important solutions obtained during the optimization process. By using those stored position theoretical Pareto front is found. Like this EED and MEED is used to control the fuel cost and emission of pollutants.
The advantages of this paper are the aim of this paper is to store all the non dominated solutions found during the optimisation process, the best individuals in each generation are always preserved and the algorithm also helps us to get close to the Pareto optimal front. Improvements required in this paper are no biclustering method has been used, Cost is relatively high, and Transmission is higher as compared to other methods.
[2] L. J. Cheng, et al., explains the concept of clonal selection which is very important in artificial immune system concept. Principle used in clonal selection is it only recognises those cells which have the capability to see the antigens and can be augmented. Those cells are kept back in the immune system. Cells which cannot recognise the antigens are not selected. Self learning and adaptive nature is very high in this clonal selection. Clonal selection is used to obtain the diversity in immune system with biological constraints. The drawbacks of Clonal Selection Algorithm are overcome by the Improved Clonal Selection Algorithm. The problems need to be focused in algorithm are while antibody mutation will be happening to match antigen antibody gene crossover is not supported. Another problem is after multiple group evolution there will be huge number of antibodies which has close or identical affinity. Because of this situation there will be a lack of diversity in the antibody group.
In improved Clonal algorithm crossover is made possible which results in genetic recombination in higher level whereas mutation cannot provide so much combinational results. Multiple antibodies can be obtained by the process of crossover than mutation. By hyper mutation process antibody diversity is increased by cross-over of antibody. Size of antibody cloning is determined by affinity of antibody. If more affinity then more number of antibody clones are possible. For high affinity antibodies antibody concentration is calculated. Based on the concentration of antibodies antibody population is restrained or increased. High affinity antibodies are cloned to avoid premature convergence. Level of antibody is calculated and population scale of antibody is adjusted accordingly.
In this paper advantages are clonal selection algorithm has good self-learning and adaptive capacity, it is the key steps to analyze understand and describe the images, cross-over is made and clonal population is controlled based on the requirement and affinity of antibody is taken care before clonning. Improvements in this paper are computation is highly intensive, it has optimization problem and selection of antibodies from large number of cloning is difficult.
[3] Yong-Sheng Ding, et al., discusses about the optimization of target in wireless sensor network using multi-object concept. In wireless sensor network coverage of network is an important aspect. In network connection constraint target cover is taken as multi-objective vertex. Multi-objective immune co-evolutionary algorithm (MOICEA) is used to get the cover set of target. This algorithm uses the biological mechanisms like hypermutation, clonal selection, elimination based on immunity,