Multi-objective placement and sizing of DGs in distribution networks ensuring transient stability using hybrid evolutionary algorithm

Review of distributed generation (DG) system planning and optimisation techniques: Comparison of numerical and mathematical modelling methods

“…Among these technical indicators, power loss [272], short circuit current capacity [273], voltage stability [274] and power system transient stability [275] are more relevant to DG distribution (i.e. topography of DG-based generators and their injection points to power grid), microgrid power management and detailed DG expansion planning [276].…”

“…For instance, Tabu Search is combined with Simulated Annealing (SA) to shorten its computational time but maintaining its strength for diversified and extensive solution search [314]; in another case, simplicity of MILP is complemented by excellent global solution exploration skill of SA [347]. In addition, incorporation of FischerBurneister (FB) algorithm is recommended to increase the convergence speed of Self-adapted Evolutionary Strategy (SAES) besides overcoming the infeasible solution generation from mutation step [265], whereas dispersed vector search of PSO and Particle Artificial Bee Colony (PABC) algorithms is helpful to avoid the premature convergence problem of Shuffled Frog Leaping Algorithm (SFLA) [275], Harmony Search (HS) Algorithm [272] and Big Bang -Big Crunch (BB-BC) Algorithm [348]. Similarly, Differential Evolution (DE) is used to compensate the population diversity decay in SFLA with respect to computation iteration number, in order to disperse its search vectors in all possible state spaces for avoiding sub-optimality trap [337].…”

“…More effective with fewer control parameters [333]; Fast convergence [334]; robust and generic [335] Fitness of solution is proportional to objective function [335]; Weak local search capability [336] Shuffled Frog Leaping Algorithm (SFLA) Simple structure, and low storage requirement [275] Prone to premature convergence; quick population diversity decay with increasing iteration number [337] Krill Herd Algorithm (KHA) Involvement of fixed parameters in computation, and able to achieve optimality within short time frame (i.e. few seconds) [338] Requirement of parametric tuning depending on problem structure; Local optimality may result due to inappropriate parameter inputs [338] Swine Influenza Model-Based Optimisation (SIMBO) Model Able to solve multi-modal problems with short computational time and high accuracy; simple execution and better exploration ability [339] Large iteration number and slow convergence for large-scale problem [339] Minimum total cost (TC) and operational cost (OC) Biomass generation turn-down ratio, battery storage operation, and inverter operation Solar PV and biomass generation capacities; Energy and power-related energy storage capacities Investigates the daily and weekly modes of stored energy dispatch scheduling, and DG design variation with respect to TC and OC; Evaluates the compatibility of energy storage system with daily and weekly operational modes Ho et al [26] Minimum total annual cost, and carbon emissions Energy balance, equipment constraints, trade-off constraint with power grid, and thermal energy interchange constraint Capacities of solar thermal, CHP, natural gas boiler, and thermal storage systems Applies to optimisation of CHP-based DG system integrated with district heating system; Total DG capacity increases with incorporation of thermal energy interchange mechanism; DG cost effectiveness is sensitive to energy cost uncertainties Wu et al [65] Minimum outsourced electricity Energy Energy balance, water mass balance, PHS water outlet flow rate constraint, water gate operation, and renewable energy penetration rate Sizes of wind turbine and diesel generators; Capacity of PHS system Applies to solar PV/hybrid wind/diesel generator/PHS DG system optimisation; Considers sequential multi-objective optimisation approach in lieu of fuzzy optimisation approach; Installation of additional wind turbine reduces the upper reservoir size, whereas decreased cycle efficiency of PHS increases the number of wind turbine for reliability DG operation Chen et al [282] Minimum outsourced electricity and NPC Energy balance, inverter and rectifier operation, and energy storage system power rating Capacities of solar PV, wind turbine, and biomass generation systems; Type and capacity of energy storage system; Power dispatch schedule Applies to a hypothetical eco-industrial park comprising of batch and continuous processes; Lead-acid battery system is chosen in economic perspective, but superconducting mag...…”

“…Optimal DG siting is based on voltage sensitivity and apparent power load; Specific DG operational schedule is considered; Optimised DG system with lagging power factor exhibits better technical and economic performances Gampa Muttaqi et al [283] Minimum operation and power reserve costs, and emission-related environmental cost Market balance, DG generation capacity, spinning reserve limit, power flow balance, demand reserve limit, and load shedding limit Load and reserve power ratings of multiple DG technologies Uncertainty in power dispatch scheduling associated with renewable variability is incorporated into cost function; Concurrent scheduling of electric vehicle (EV) and responsive load assists in peak shaving, load shifting, reserve power supply, operating cost and emission reduction, and system reliability improvement Rabiee et al [266] Stochastic (improved) PSO Hybrid PSO -SFLA Minimum real power loss, voltage deviation, and CCT index DG power limit, voltage limit, branch power flow constraint, and power flow balance Size and capacity of (multiple) microturbine DG PSO is coupled with SFLA to overcome its premature convergence problem; CCT index is introduced to quantify power system transient stability, and serves to minimise short-circuit fault in the system design; More security-based objective functions are anticipated Nayeripour et al [275] Hybrid In addition, as inspired by multi-modelling regression approach, development of a multi-algorithm optimisation framework for energy planning context could provide an alternative solution besides model hybridisation. This method shall embed different algorithms with varying computational time, exploration and exploitation skill levels.…”

Optimal Configuration of DG in Distribution System: An Overview

“…Evolutionary algorithm [26] x Efficient performance for finding the global minimum x Easy to find example from literature x Harder to simulate, premature convergence x Low precision factor PSO [27,28] x This technique makes it yield good results for number of DG and responds well for variation. Table 3.…”