multi objective particle swarm optimization

The widespread use of residential air conditioning (RAC) systems in modern power systems has resulted in an increase in the phenomenon of fault-induced delayed voltage recovery (FIDVR). This phenomenon leads to short-term voltage instability and sometimes even voltage collapse. To address this issue, parallel FACTS devices such as SVC and STATCOM can be used. In this paper, a data-driven hybrid approach based on volt-ampere reactive (VAR) placement is proposed to reduce FIDVR events. This approach uses a new and efficient index for voltage evaluation after faults and determines the optimal location and size of VAR resources considering economic and technical constraints. A multi-layer perceptron (MLP) neural network is used to solve the multi-dimensional mapping problem considering reactive power injections into buses. Then, a multi-objective optimization is proposed to identify the optimal size of VAR resources to address short-term voltage instability and prevent FIDVR events using intelligent optimization methods. First, optimization is performed for the single-objective function with predefined weights by the PSO algorithm, and then the results are compared with the artificial bee colony (ABC) algorithm, ant colony optimization for continuous domains (ACOR), and differential evolution (DE) algorithms. Additionally, this paper focuses on identifying a Pareto front of non-dominated solutions using multi-objective particle swarm optimization (MOPSO). The proposed approach is tested on the 39-bus IEEE system considering a time-varying dynamic model for residential air conditioning loads. The results show that this approach is highly effective in solving reactive power optimization problems and reducing FIDVR effects.

Control of residual chlorine concentration within a desirable range throughout water distribution systems can cause the destruction of potentially harmful pathogens without chlorine adverse health effects & its toxic by-products. Hence, optimal scheduling of booster chlorination stations in the WDSs to ensure healthy water supply with the lowest dose of chlorine consumption is vital. The aim of the present study is to develop a multi-objective optimization model in order to minimize the mass injection rate as well as the probability of chlorine violation in the WDSs, which has been implemented in the MATLAB-EPANET platform. Multi-objective krill herd and multi-objective particle swarm optimization algorithms have been utilized as optimizers to obtain the desired Pareto front in the real-scale Brushy Plains network. The resulted Pareto fronts showed that in most of their solutions, as long as the mass injection rate increased, the probability of chlorine violation decreased. In this study, the solution with the less PCV in each Pareto was selected as the optimal solution to assure the healthy water supply. Though the MOPSO resulted Pareto showed more solution diversity, MKH optimal solution has a better MIR function than MOPSO optimal solution with the same amount of PCV. Analyzing the residual chlorine concentration profiles of the monitoring period corresponding to the MKH optimal solution showed that the chlorine concentration of the most nodes of Brushy Plains network exist in the desirable range of 0.2 to 0.8 mg/L and the residual chlorine of 100% of nodes exist in the range of 0.2 to 1.6 mg/L. Also, the MKH results are superior to those of the previous studies in terms of the total mass injection rate. Generally, in addition to economic advantages, minimizing chlorine injection rate and the probability of chlorine violation simultaneously in the water distribution systems reduces the adverse health effects of the disinfectant by-products.

The bridges as important links are connected all road networks and railways That a huge investment is needed to build them. In addition, bridges are considered as important points of vital arteries. If the bridges due to old age, exhaustion, overload, weather conditions, natural disasters and etc. destroyed, repair works are much more costly than the budget needed for the maintenance of them. Available budgets for reconstruction and maintenance, usually is not enough to maintain a constant state throughout the lifetime of the bridges. Due to the large number of bridges and high maintenance costs, it is not possible to repair and pay for maintenance of bridges simultaneously. In this study, the life cycle cost is calculated based on maintenance period of railway using multi-objective particle swarm optimization algorithm. Sensitivity analysis, furthermore, is done on deterioration rate and reduction of deterioration rate and effect of these parameters on system performance are analyzed.

Laser shock welding process has recently attracted the attention of many researchers. Similar to explosive and magnetic welding, this process may also be used for impact welding using solid state welding principle. Impact welding is based on the influence of high-velocity collision of two base metals and generation of metallurgical atomic bonding in the solid phase at the contact area at ambient temperature. This process is also used to clad a sheet metal with a thin layer of other metal, named flyer plate. The base metal is serrated with certain angle and depth and the flyer plate is moved rapidly to collide with the base plate to generate bonding. The energy required to move the flyer plate, is produced by plasma pressure created by laser impact on the surface of the flyer plate. The main advantage of this connection is its capability to attach two dissimilar metals in order to enhance physical, chemical, or mechanical properties on one side of a cheaper metal.Same as other welding methods, it is very important to forecast and optimize the weld quality obtained by this process. Hence, finite element method using ABAQUS software, was employed to simulate the laser welding or cladding process in this research and verified by experimental data. Impact speed, serration angle and depth are the main affecting parameters on weld quality. Therefore, multi-objective particle swarm optimization (MOPSO) algorithm for a certain thickness of the flyer plate was utilized to maximize the welded area of two plates and minimize the cost of machining the base plate for making serration using the data generated by finite element analysis, linked to MATLAB for optimization of these objectives. The optimization results indicate an increase in joined area at the connection point as well as reduced number of grooves which leads to decrease in manufacturing cost.