Search

Accurate electricity price forecasting (EPF) is important for the purpose of bidding strategies and minimizing the risk for market participants in the competitive electricity market. Besides that, EPF becomes critically important for effective planning and efficient operation of a power system due to deregulation of electricity industry. However, accurate EPF is very challenging due to complex nonlinearity in the time series-based electricity prices. Hence, this work proposed two-fold contributions which are (1) effective time series preprocessing module to ensure feasible time-series data is fitted in the deep learning model, and (2) an improved long short-term memory (LSTM) model by incorporating linear scaled hyperbolic tangent (LiSHT) layer in the EPF. In this work, the time series pre-processing module adopted linear trend of the correlated features of electricity price series and the time series are tested by using Augmented Dickey Fuller (ADF) test method. In addition, the time series are transformed using boxcox transformation method in order to satisfy the stationarity property. Then, an improved LSTM prediction module is proposed to forecast electricity prices where LiSHT layer is adopted to optimize the parameters of the heterogeneous LSTM. This study is performed using the Australian electricity market price, load and renewable energy supply data. The experimental results obtained show that the proposed EPF framework performed better compared to previous techniques.

Various fault location methods have been developed in the past to identify the faulty phase, fault type, faulty section, and distance. However, this identification is commonly conducted in a separate manner. An effective fault location should be able to identify all of these at the same time. Therefore, in this work, a method using a support vector machine (SVM) to identify the fault type, faulty section, and distance considering the faulty phase is proposed. The proposed method uses voltage sag magnitude of the distribution system as the main feature for the SVM to identify faults. The fault type is classified using a directed acyclic graph SVM. The possible faulty sections are identified by estimating the fault resistance using support vector regression and matching the voltage sag data in the database with the actual voltage sag data. The most possible faulty sections are ranked using ranking analysis. The fault distance for the possible faulty sections is then identified using support vector regression analysis and its overfitting or underfitting issues are addressed by the proper selection of a regularization parameter. The feasibility of the proposed method was tested on an actual Malaysian distribution system. The results of faulty phase, fault type, faulty section, and fault distance are analyzed. The performance of the proposed method is compared with various other intelligent methods such as the artificial neural network, deep neural network, extreme learning machine, and kriging method. The test results indicate that the faulty phase and fault type yield 100 % accurate results. All the faulty sections are identified and the proposed method obtains reliable fault location.

This paper presents a load shedding scheme (LSS) based on system frequency and voltage stability (VS). In the proposed scheme, the amount of power imbalance is determined using a rate of change of frequency, and the load shedding priority is carried out based on a VS index. The load with the highest tendency for voltage collapse (based on the index value) is given first priority to be shed. Thus, by shedding the most sensitive load, the distribution system can be saved from power collapse and then the system can return to its nominal state after load shedding. The proposed LSS is validated through simulation using PSCAD/EMTDC software on a Malaysian distribution network that consists of two mini-hydro generators. The simulation results show that the proposed scheme manages to shed the optimal amount of load when compared to conventional and adaptive frequency LSS. Moreover, apart from frequency stability, the proposed scheme also shows a significant improvement in the voltage profile of the islanded system.

Under-voltage load shedding (UVLS) is an important technique to maintain the voltage stability and frequency of a power system network. UVLS has been applied widely in transmission systems to avoid system blackouts. However, with increasing penetration of distributed generation such as photovoltaic (PV) systems, the application of UVLS becomes important for islanded distribution systems. Under this condition, the network does not have a frequency reference as when it is connected to the grid. In this condition, when the load demand exceeds the PV capacity, UVLS is the only option to stabilize the system by shedding the load based on the changes of the voltage magnitude. In this work, a new UVLS scheme based on voltage stability indices is proposed. Four voltage stability indices are used as indicators for load shedding. Based on the stability indices, the loads that have the highest tendency of voltage collapse shall be the first ones to be shed. The proposed scheme is tested on a practical distribution network energized by a grid, a mini hydro generator, and a PV system. The test results on various scenarios prove that the proposed method is able to restore the system stability.

The appropriate output of distributed generation (DG) in a distribution network is important for maximizing the benefit of the DG installation in the network. Therefore, most researchers have concentrated on the optimization technique to compute the optimal DG value. In this paper, the comparative learning in global particle swarm optimization (CLGPSO) method is introduced. The implementation of individual cognitive and social acceleration coefficient values for each particle and a new fourth term in the velocity formula make the process of convergence faster. This new algorithm is tested on 6 standard mathematical test functions and a 33-bus distribution system. The performance of the CLGPSO is compared with the inertia weight particle swarm optimization (PSO) and evolutionary PSO methods. Since the CLGPSO requires fewer iterations, less computing time, and a lower standard deviation value, it can be concluded that the CLGPSO is the superior algorithm in solving small-dimension mathematical and simple power system problems.

This paper compares the performance of three population-based algorithms including particle swarm optimization (PSO), evolutionary programming (EP), and genetic algorithm (GA) to solve the multi-objective optimal power flow (OPF) problem. The unattractive characteristics of the cost-based OPF including loss, voltage profile, and emission justifies the necessity of multi-objective OPF study. This study presents the programming results of the nine essential single-objective and multi-objective functions of OPF problem. The considered objective functions include cost, active power loss, voltage stability index, and emission. The multi-objective optimizations include cost and active power loss, cost and voltage stability index, active power loss and voltage stability index, cost and emission, and finally cost, active power loss, and voltage stability index. To solve the multi-objective OPF problem, Pareto optimal method is used to form the Pareto optimal set. A fuzzy decision-based mechanism is applied to select the best comprised solution. In this work, to decrease the running time of load flow calculation, a new approach including combined Newton-Raphson and Fast-Decouple is conducted. The proposed methods are tested on IEEE 30-bus test system and the best method for each objective is determined based on the total cost and the convergence values of the considered objectives. The programming results indicate that based on the inter-related nature of the objective functions, a control system cannot be recommended based on individual optimizations and the secondary criteria should also be considered. Refereed/Peer-reviewed