Performance Analysis of an Optimized ANN Model to Predict the Stability of Smart Grid

Abstract

The stability of the power grid is concernment due to the high demand and supply to smart cities, homes, factories, and so on. Different machine learning (ML) and deep learning (DL) models can be used to tackle the problem of stability prediction for the energy grid. This study elaborates on the necessity of IoT technology to make energy grid networks smart. Different prediction models, namely, logistic regression, naïve Bayes, decision tree, support vector machine, random forest, XGBoost, k-nearest neighbor, and optimized artificial neural network (ANN), have been applied on openly available smart energy grid datasets to predict their stability. The present article uses metrics such as accuracy, precision, recall, f1-score, and ROC curve to compare different predictive models. Data augmentation and feature scaling have been applied to the dataset to get better results. The augmented dataset provides better results as compared with the normal dataset. This study concludes that the deep learning predictive model ANN optimized with Adam optimizer provides better results than other predictive models. The ANN model provides 97.27% accuracy, 96.79% precision, 95.67% recall, and 96.22% F1 score.

Machine Learning Approaches To Predict The Stability of Smart Grid

“…Moldovan and Salomie [18] A machine learning methodology was proposed to predict smart grid stability using features extraction, selection, and classification Raz et al [19] A regression tree-based approach had been employed to predict the power system stability margin and detect impending system Zhu et al [20] A HDLM had been developed to efficiently both quantitative and qualitative online transient stability prediction Alazab et al [1] A MLSTM model had been designed to predict the stability of the smart grid network. Gorza lczany et al [21] Genetically optimized FRBC had been used for the transparent and accurate prediction of DSGC stability Vanfretti and Arava [22] DT analyzed thousands of operating conditions and predicted voltage stability of power system Breviglieri et al [23] They optimized DL models to solve fixed inputs and equality issues in DSGC system Bashir et al [24] Comparative analysis of several state-of-the-art machine learning classifiers were carried out for the prediction of smart grid stability Zhang et al [25] The smart grid data was investigated with six classifiers to analyze the stability data Chahal et al [4] This work elaborates on the requirement of IoT and how proposed ANN predicts the stability of smart grid more accurately. Mishra et al [26] An optimized memetic algorithm-based ELM model is proposed to predict the stability of the smart grid.…”

“…The demand for renewable energy and the distribution of electricity in grid topology become more and more decentralized. The power generation and consumption can happen from any terminals which are called Prosumers [4]. Recently, this process does not happen in a central node anymore.…”

An improved ELM-WOA–based fault diagnosis for electric power

“…In their results, the decision tree and neural network have a good effect in the fault classification and location of electric wires (Ayushi et al, 2022). Moreover, some studies have used the neural network coupled with wavelet transforms to detect faults-specifically, some signals of the layer are extracted through wavelet transforms to judge whether there is a fault and then the neural network or regression decision tree is used to judge what the fault is (Ayushi et al, 2022;Singhal et al, 2022;Xin, 2022). Generally, the collected data are trained and located by simulating the fault type and fault location of the wire (Chen et al, 2022b;Singhal et al, 2022).…”

“…Chen et al (2022a) used the CNN-LSTM model to solve the problem of the slow transmission rate of high-frequency information in smart grid and improve the efficiency of information transmission (Xin, 2022). Because the distance of the transmission line is relatively long, the probability of failure of the transmission line is increased (Ayushi et al, 2022;Xin, 2022;Yuvaraja et al, 2022). Some scholars use neural networks to detect whether there is a current that is directly grounded (Xin, 2022).…”

“…Some scholars use neural networks to detect whether there is a current that is directly grounded (Xin, 2022). In their results, the decision tree and neural network have a good effect in the fault classification and location of electric wires (Ayushi et al, 2022). Moreover, some studies have used the neural network coupled with wavelet transforms to detect faults-specifically, some signals of the layer are extracted through wavelet transforms to judge whether there is a fault and then the neural network or regression decision tree is used to judge what the fault is (Ayushi et al, 2022;Singhal et al, 2022;Xin, 2022).…”