A Study of Load Demand Forecasting Models in Electricity using Artificial Neural Networks and Fuzzy Logic Model
Abstract:Since load time series are very changeable, demand forecasting of the short-term load is challenging based on hourly, daily, weekly, and monthly load forecast demand. As a result, the Turkish Electricity Transmission Company (TEA) load forecasting is proposed in this paper using artificial neural networks (ANN) and fuzzy logic (FL). Load forecasting enables utilities to purchase and generate electricity, load shift, and build infrastructure. A load forecast was classified into three sorts (hourly, weekly and m… Show more
“…ANN commences computations to imitate the learning procedures of the human brain [19]. ANN is one of the CI techniques that can be employed as a tool to achieve the optimal solution.…”
Load shedding is generally regarded as the final option to evade voltage collapse and blackout following major overloads. The traditional method of load shedding curtails random loads regardless of their importance until the system's voltage is improved. Shedding random loads without considering their priority will lead to power interruption in vital infrastructures. Hence, to improve the existing power system protection scheme, development of a more effective and efficient load shedding method is necessary. In this paper, an optimal under voltage load shedding (UVLS) method is proposed for optimum prediction of amount of load shed and the best location for load curtailment. Moreover, the proposed method is designed to maintain the vital loads in the system during the load shedding process. In this work, the stability index (SI) and feed-forward backpropagation neural network (FFBPNN) were adopted to avoid voltage collapse and blackout by mitigating voltage instability following overloads in distribution system. The performance of the proposed method to several overload scenarios is investigated. Case studies performed on the IEEE 33-bus system exposed significant robustness and performance of the recommended technique. Compared to other approaches, the proposed approach is efficient in counteracting under-shedding occurrence, enhancing the voltage profile, and improving the stability of the system, whilst maintaining vital loads in the system during load shedding.
“…ANN commences computations to imitate the learning procedures of the human brain [19]. ANN is one of the CI techniques that can be employed as a tool to achieve the optimal solution.…”
Load shedding is generally regarded as the final option to evade voltage collapse and blackout following major overloads. The traditional method of load shedding curtails random loads regardless of their importance until the system's voltage is improved. Shedding random loads without considering their priority will lead to power interruption in vital infrastructures. Hence, to improve the existing power system protection scheme, development of a more effective and efficient load shedding method is necessary. In this paper, an optimal under voltage load shedding (UVLS) method is proposed for optimum prediction of amount of load shed and the best location for load curtailment. Moreover, the proposed method is designed to maintain the vital loads in the system during the load shedding process. In this work, the stability index (SI) and feed-forward backpropagation neural network (FFBPNN) were adopted to avoid voltage collapse and blackout by mitigating voltage instability following overloads in distribution system. The performance of the proposed method to several overload scenarios is investigated. Case studies performed on the IEEE 33-bus system exposed significant robustness and performance of the recommended technique. Compared to other approaches, the proposed approach is efficient in counteracting under-shedding occurrence, enhancing the voltage profile, and improving the stability of the system, whilst maintaining vital loads in the system during load shedding.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.