Parimal Acharjee

Expert algorithm based on adaptive particle swarm optimization for power flow analysis

An Expert algorithm based on particle swarm optimization (PSO) technique with adaptive PSO parameters has been developed for power flow analysis under critical conditions and multiple power flow solutions. Depending on the objective functions of the current and best solutions in the present generation, unique and innovative formulas are designed for two sets of PSO parameters, inertia weight and learning factors. For faster, sure convergence and overcome the limitations of conventional methods, PSO parameters are so designed that they are highly adaptive. To the best of our knowledge, it is the first report of applying Adaptive PSO (APSO) to solve power flow problems. Multiple power flow solutions which are used for voltage stability analysis can be obtained if proposed method is used with local search as accelerating technique. The proposed algorithm proves its robustness providing reliable and better convergence under high R/X ratios and maximum loadability limits. The effectiveness and efficiency has been established showing results for standard and ill-conditioned test systems.

Multiple Solutions of Optimal PMU Placement Using Exponential Binary PSO Algorithm for Smart Grid Applications

For smart grid execution, one of the most important requirements is fast, precise, and efficient synchronized measurements, which are possible by phasor measurement unit (PMU). To achieve fully observable network with the least number of PMUs, optimal placement of PMU (OPP) is crucial. In trying to achieve OPP, priority may be given at critical buses, generator buses, or buses that are meant for future extension. Also, different applications will have to be kept in view while prioritizing PMU placement. Hence, OPP with multiple solutions (MSs) can offer better flexibility for different placement strategies as it can meet the best solution based on the requirements. To provide MSs, an effective exponential binary particle swarm optimization (EBPSO) algorithm is developed. In this algorithm, a nonlinear inertia-weight-coefficient is used to improve the searching capability. To incorporate previous position of particle, two innovative mathematical equations that can update particles position are formulated. For quick and reliable convergence, two useful filtration techniques that can facilitate MSs are applied. Single mutation operator is conditionally applied to avoid stagnation. The EBPSO algorithm is so developed that it can provide MSs for various practical contingencies, such as single PMU outage and single line outage for different systems.

Robust load flow based on local search

A heuristic approach to the load flow solutions is proposed in this paper. The method is based on a perturbation technique. Voltage and phase angle of a bus is perturbed around their current values and the effect of perturbation is judged on a local network formed around the node concerned. Perturbations generating improved solutions are accepted. The method is slow but robust and is applicable in those cases when the conventional approaches fail. It can be used in finding the loadability limit, generating P-V/Q-V curves and also in finding multiple power flow solutions.

Multiple low voltage power flow solutions using hybrid PSO and optimal multiplier method

In this paper a method for computing multiple power flow solutions of power system is proposed. The proposed method determines the multiple solutions in pairs. Particle Swarm Optimization (PSO) technique is used to find the starting values of the rectangular Newton-Raphson power flow. From these starting values, the rectangular Newton-Raphson load flow determines a low voltage solution close to the first one. Optimal multipliers are then used to determine a second low voltage solution. Test results of standard IEEE 14 bus system have been shown.

Multiple solutions of optimal PMU placement using exponential binary PSO algorithm

In modern power systems, phasor measurement unit (PMU) is the most advanced measurement device to measure precise, fast and reliable voltage and current phasor. Due to high installation cost and other practical limitations, optimal PMU placement (OPP) is needed. An exponential binary particle swarm optimization (EBPSO) algorithm is proposed to solve the OPP problem for a completely observable network. Various practical contingencies such as zero injection, single PMU outage are considered in the proposed algorithm along with the normal operating condition. Multiple solutions for OPP problem can improve the feasibility of the placement methodology in practical environment. Even though any bus is selected as candidate location but it may not be possible to install a PMU on that bus due to the lack of necessary infrastructure. On the contrary, few buses in practical systems which require close and precise monitoring should be directly observed by PMU. Placing some extra PMUs can solve this problem but economically it is not preferable. Hence, having alternative solutions can be very effective. To ensure multiple solutions and improve the performances, an adaptive exponentially decaying inertia weight coefficient is developed. A sigmoid function is introduced to update the position of the particles in binary form. Both inter connected (IEEE 14-bus and 30-bus) and radial (IEEE 39-bus) system are tested to check the feasibility and effectiveness of the algorithm.

Hybridization of cuckoo search algorithm and chemical reaction optimization for economic load dispatch problem

In this paper, a hybrid algorithm involving cuckoo search algorithm (CSA) and chemical reaction optimization (CRO) is proposed for solving the practical economic load dispatch (ELD) problem of power systems. The two important features of CSA i.e. Lévy Flight and Random Walk techniques are incorporated in the optimization process of CRO to produce new generation of solutions. Lévy Flight and Random Walk increase the exploitation capabilities of the next generation of solutions and on the other hand, the collision techniques of CRO increase the exploration of the solutions. The result thus obtained is better in terms of quality, convergence and computational time. This hybrid CSA-CRO algorithm can handle practical constraints such as prohibited operating zones, transmission losses, valve-point effects and multiple fuels for the ELD problem. The overall results indicate the superiority of the proposed algorithm over the existing algorithms.

Optimal Allocation of DG Using Exponentential PSO with Reduced Search Space

In the current deregulating environment, integration of Distributed Generation (DG) in the radial distribution network is one of the reliable and efficient options which can be used for reduction of power loss, improving the voltage profile of the system and stability. Optimal allocation of DG units is essential for improving the quality of supply and reliability of the network. Using voltage stability index, weak and healthy zone are determined. DG with same size is placed in weak and healthy zone separately. The voltage profile improvement, cost of energy saving and reduction of losses can be maximized by placing DG in weak zone. To reduce the computational time required for optimal allocation of DG, it is proposed to conduct its performance analysis only at the weak bus locations of the system. Therefore, the search space for optimal allocation of DG can be restricted only to the weak zone of the system. Taking account of operational constraints, a new objective function is formulated considering Voltage Profile Improvement Index (VPII) and Benefit to Cost ratio (BCR). An Exponential Particle Swarm optimization (EPSO) method is proposed for optimal placement and sizing of DG considering both full and reduced search space. The proposed algorithm is compared with other types of Particle swarm optimization techniques (PSO) such as Simple Particle Swarm Optimization (SPSO) and Adaptive Particle Swarm Optimization (APSO). The best performance in terms of computational efficiency and solution quality is achieved for the proposed EPSO method.

Investigation of the power scenario in India for the implementation of smart grid

In this paper, the current power scenario of India is elaborately described and the prospect, potential of energy for the developing countries is analyzed. The concept of Smart Grid (SG) and its necessity for the modern power systems are demonstrated. The effect of SG on social, economical and power sector is explained. The main problems to the implementation of SG in India and their remedies are discussed. The recent initiatives already taken by the Government of India (GoI) are sequentially described. Considering social, economical, political and environmental circumstances, steps to implement SG in India is evidently suggested.

Comparative Performance Analysis of DG and DSTATCOM Using Improved PSO Based on Success Rate for Deregulated Environment

A new and improved particle swarm optimization (PSO) technique with adaptive inertia weight (w) based on success rate is proposed to find the optimal allocation of distributed generation (DG) and distribution static compensator (DSTATCOM) considering security limits. For the optimal sizing and siting of the device, technical, economic, and social objectives are considered. Logical and innovative indexes namely voltage profile enhancement index, benefit cost ratio, and emission cost benefit index are formulated to judge the impact of the device on the system. The developed algorithm is executed for both full search space (all load bus locations) and reduced search space (only unhealthy zone/locations) to prove that the optimal allocation is obtained only at the unhealthy zone. For the IEEE 33 and IEEE 69 bus systems, the proposed algorithm is compared with other techniques such as differential evolution, real-coded genetic algorithm, and PSO based on randomized inertia weight, and PSO based on linearly decreasing inertia weight. The best performance in terms of computational efficiency and solution quality is achieved for the proposed method. Further, a comparative performance analysis is presented between DG and DSTATCOM based on the economic, technical, and social impacts.

Decoupled PSO based rugged power flow method for loadability limit identification

In this paper, PSO based robust load flow is proposed. In the proposed decoupled based PSO technique, the decoupling features among the power flow variables are taken into consideration. Very simple technique is applied to prompt the convergence. Keeping all inherent properties of evolutionary technique, improvement method is developed to attain better performances. To detect the stability margin, maximum loadability limit (MLL) is identified using the proposed method. The developed algorithm shows that it can provide satisfactory solutions under stressed situations when classical standard methods fail. To establish the effectiveness and efficiency, the proposed algorithm is compared with other methods.