Special issue on intelligent tools and techniques for signals, machines and automation

Srivastava, Smriti; Malik, Hasmat; Sharma, Rajneesh

doi:10.3233/jifs-169773

Cited by 66 publications

(5 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…After the model development, training and testing are performed to validate the model performance, and finally obtained results are compared to obtain the best model for future forecasting applications. For more detail regarding demonstration of step-step-wise procedure of implementation of feature extraction and selection, the reader may refer [25][26][27][28][29][30][31] and [25][26][27][32][33][34], respectively.…”

Section: Datasets and Methodologymentioning

confidence: 99%

“…At the same time, RNN has limited memory to store the operation of each stage and suffers from gradient problems [25][26][27], such as vanishing or exploding, etc. To resolve these problems, an advanced version of RNN was formulated, LSTM, in 1997 [32]. LSTM also works based on the sequential structure of 4 states.…”

Section: Deep Neural Network (Dnn)mentioning

confidence: 99%

“…), ( 5) train the DNN model (using function trainNetwork), ( 6) forecast future time steps (using the function of predictAndUpdateState), and ( 7) update the DNN network state with observed values. For more detail regarding LSTM implementation, the reader may refer to [25][26][27]32].…”

Section: Deep Neural Network (Dnn)mentioning

confidence: 99%

See 2 more Smart Citations

Machine Learning Approach for Short-Term Load Forecasting Using Deep Neural Network

Alotaibi

2022

Energies

View full text Add to dashboard Cite

Power system demand forecasting is a crucial task in the power system engineering field. This is due to the fact that most system planning and operation activities basically rely on proper forecasting models. Entire power infrastructures are built essentially to provide and serve the consumption of energy. Therefore, it is very necessary to construct robust and efficient predictive models in order to provide accurate load forecasting. In this paper, three techniques are utilized for short-term load forecasting. These techniques are deep neural network (DNN), multilayer perceptron-based artificial neural network (ANN), and decision tree-based prediction (DR). New predictive variables are included to enhance the overall forecasting and handle the difficulties caused by some categorical predictors. The comparison among these three techniques is executed based on coefficients of determination R2 and mean absolute error (MAE). Statistical tests are performed in order to verify the results and examine whether these models are statistically different or not. The results reveal that the DNN model outperformed the other models and was statistically different from them.

show abstract

Section: Datasets and Methodologymentioning

confidence: 99%

Section: Deep Neural Network (Dnn)mentioning

confidence: 99%

See 1 more Smart Citation

Machine Learning Approach for Short-Term Load Forecasting Using Deep Neural Network

Alotaibi

2022

Energies

View full text Add to dashboard Cite

show abstract

“…This control strategy used for the converter basically influences the impact of the control of power quality. A variety of works like delineating control strategies, which are the hysteresis control, bum control, model farsighted control, comparing essential control, relative full control, dull control, and nonlinear solid control are there [28][29][30][31][32][33][34] (Table 1).…”

Section: Modeling and Optimization Techniques Used For Ressmentioning

confidence: 99%

Energy Management System for Hybrid Energy System: Renewable Integration, Modeling and Optimization, Control Aspects and Conceptual Framework

Kumari

Sharma

Singh

et al. 2020

Metaheuristic and Evolutionary Computation: Algorithms and Applications

View full text Add to dashboard Cite

In developing Countries like India, the interest of energy has been expanding surprisingly because of the speculation of the agricultural, modern just as residence exercises. The rising an Earth-wide temperature boost marvels and an expansion in the consumption of petroleum derivatives has been the most crucial main thrust towards the thoughtfulness regarding misuse of Renewable Energy Sources. Alongside different favorable circumstances of these sources, there comes a heap of complexities connected to them because of their irregular and variable nature. Consequently, to maintain a strategic distance from these vulnerabilities it is important to give these assets appropriate planning and proper energy management. This work here a detailed study of different optimization techniques which can be applied to the renewable energy resources including the multi-agent solution as well as the Artificial Intelligence and Micro-grid controller which can offer a clear vision for the researchers in this field. Certain recommendations considering the challenges in renewable energy (RE) development are also been provided. In the proposed framework, the smart grid is optimized by the use of different optimization methods.

show abstract

“…It has many advantages over the stand alone system such as it is reliable, economic and environmental friendly. In literature various study reported by researchers such as photovoltaic (PV)-wind (W) [3][4][5][6][7][8][9][10][11][12][13][14], W-battery (B) [15][16][17], W-diesel (D) [18][19][20][21][22][23], PV-W-D-B [24][25][26][27][28][29], PV-D-B [30][31][32], PV-D [33][34][35], PV-W-B [36], hydro-PV-W-fuel cell [37] and PV-W-hydro [38][39][40][41][42]. The benefit of hybrid system is continuity of power to load is increases but to enhance maximum power remains an important issue which can be achieved by incorporating tracking in hybrid systems.…”

Section: Introductionmentioning

confidence: 99%

Novel Application of Grid Search Algorithm for Optimization of Photovoltaic-Wind-Diesel Hybrid Systems with and Without Tracking Systems for Power Generation

Yadav

2020

Metaheuristic and Evolutionary Computation: Algorithms and Applications

View full text Add to dashboard Cite

Power from hybrid renewable energy sources (RES) are increasing drastically in recent years but in mountainous regions it can play an important role where population is sparse and geographically isolated. This chapter aims to study first time the techno-economic feasibility analysis of photovoltaic (PV)-wind (W)-diesel (D), PV-D, PV-W, W-D, PV and W only with and without tracking systems for mountainous region of Sikkim in India. For this load demand of 88.9 kWh/day of hostel in National Institute of Technology Sikkim is used for analysis. The cost of Energy (COE) $/kWh for PV-D, PV-W-D, PV-W, W-D, PV and W are 0.284, 0.305, 0.334, 0.375, 0.377 and 0.687 respectively, showing PV-D combination has least COE and can be beneficial. PV-D combined with different tracking systems and it is found that horizontal tracking system is producing more power.

show abstract

Special issue on intelligent tools and techniques for signals, machines and automation

Cited by 66 publications

References 46 publications

Machine Learning Approach for Short-Term Load Forecasting Using Deep Neural Network

Machine Learning Approach for Short-Term Load Forecasting Using Deep Neural Network

Energy Management System for Hybrid Energy System: Renewable Integration, Modeling and Optimization, Control Aspects and Conceptual Framework

Novel Application of Grid Search Algorithm for Optimization of Photovoltaic-Wind-Diesel Hybrid Systems with and Without Tracking Systems for Power Generation

Contact Info

Product

Resources

About