A Deep Learning‐to‐learning Based Control system for renewable microgrids

In this manuscript, a hybrid method for optimum energy management of micro grid (MG) connected system is proposed. The MG connected sources are Photovoltaic (PV) array, Wind Turbine (WT), Fuel Cell (FC), Micro Turbine (MT), Diesel Generator (DG), battery storage (BESS). The proposed technique is the consolidation of Artificial Neural Network (ANN), Artificial Bee Colony (ABC), and Firefly Algorithm (FA). The proposed method is utilized to maintain the power flow (PF) amid the energy sources and grid. Initially, the predicted load requirement based on the ANN technique predicts the load requirement and ABC optimizes the MG configuration. Finally, the FA technique helps to reduce the FC including operation with maintenance cost. The proposed method is executed in MATLAB/Simulink site and the performance is analyzed with different existing methods like online management, ABC, ABC-ABC, and ABC-Gravitational Search (GS). Here, the maximal power generation of photovoltaic is 5.9 kW, wind turbine is 5.9 kW, micro turbine is 4 kW, FC is 4 kW, DG is 6 kW, and battery is 0.5 kW. The total cost of ABC, ABC-ABC, ABC-GS, and the proposed technique is about 4.1$/h, 3.9$/h, 3.3$/h, and 2.7$/h. Optimum micro grid combination using proposed method, overall cost comparison using different methods, statistical evaluation of MG management, cost accuracy percentage (CAP) of proposed and existing techniques, efficiency under various number of trails are also analyzed in this work. The comparison results demonstrate that the proposed technique has minimal cost-efficient based on its load requirement.

Developing Hybrid Demand Response Technique for Energy Management in Microgrid Based on Pelican Optimization Algorithm

Prediction of PM2.5 concentration in urban agglomeration of China by hybrid network model

This paper proposes a multi-layered anomaly detection scheme to train feature extraction and to test anomaly prediction by using Convolutional Neural Networks (CNNs) layer, Bidirectional and Unidirectional Long Short-Term Memory (LSTM) Recurrent Neural Networks (RNNs), which is one of a novel deep architecture named stacked convolutional bidirectional LSTM network (SCB-LSTM). In the proposed model, the stacked CNNs perform feature extraction of vibration sensor signal patterns, and the result is used to feature learning with the stacked bidirectional LSTMs (SB-LSTMs). After this procedure, the stacked unidirectional LSTMs (SU-LSTMs) enhance the feature learning, and a regression layer finally predicts anomaly detections. The experimental results of bearing data not only show the accuracy of the proposed model in anomaly detection for rotating machinery diagnostics, but also suggest the better performance than other state-of-the-art algorithms such as a plain uni-LSTM or Bi-LSTM.

To satisfy the need to accurately monitor emotional stress, this paper explores the effectiveness of the attention mechanism based on the deep learning model CNN (Convolutional Neural Networks)-BiLSTM (Bi-directional Long Short-Term Memory) As different attention mechanisms can cause the framework to focus on different positions of the feature map, this discussion adds attention mechanisms to the CNN layer and the BiLSTM layer separately, and to both the CNN layer and BiLSTM layer simultaneously to generate different CNN–BiLSTM networks with attention mechanisms. ECG (electrocardiogram) data from 34 subjects were collected on the server platform created by the Institute of Psychology of the Chinese Academy of Science and the researches. It verifies that the average accuracy of CNN–BiLSTM is up to 0.865 without any attention mechanism, while the highest average accuracy of 0.868 is achieved using the CNN–attention–based BiLSTM.

Towards the integration of interconnected microgrids to deregulated electricity markets

As an important part of s mart grid, microgrid(M G) is a new form of s mart grid in the future. Microgrid( MG ) technology can effectively integrate the advantages of new energy and renewable energy generation and provide a novel way for large-scale applicat ions of new energy and renewable energy connecting to grid. This paper deals with the problem o f economic operation of microsources in the microgrid, such as micro -turbine(MT), fuel cell(FC), diesel generator(DG), photovoltaic cell(PV), wind turbine(WT), and battery storage. The proposed problem is formulated as a nonlinear constrained optimization problem. The paper takes into consideration the operation cost as well as the emission reduction of NO x , SO 2 , and CO2. So a mathematical optimal model is built to optimize operation of microgrid(M G) system, based on the characteristics of various microsources, the restraint of microgrid system and the predicting output of the next 24hours’ wind turbine and photovoltaic cell and load demand. An improved particle swarm algorith m is emp loyed to minimu m the co mprehensive benefit cost of microgrid operation including economic and environmental benefits which realizes the mu lti-objective optimization operation. Besides, this paper focuses on the effect of electricity price between microgrid and the main grid on system operation costs. The results demonstrate the efficiency of the proposed approach to satisfy the load and to reduce the operation cost and the emissions.

To address the issue of frequency regulation in AC‐islanded microgrids (MGs), this paper introduces a novel approach for adjusting primary controller gains for AC‐islanded MGs. The proposed approach uses the linear matrix inequality (LMI) for polytopic linear parameter varying (P‐LPV) modelling based on the H ∞ control theory and principal component analysis (PCA) algorithm. The objective is to regulate the MG frequency while considering the nonlinearity and uncertainty of the system. To achieve optimal control gains, the method considers all the system uncertainties in a P‐LPV modelling of the system. The PCA algorithm reduces the scheduling parameter region, and the optimal control gains are computed by solving the relevant LMIs defined on the obtained P‐LPV model based on H∞ performance and stability achievement. The primary control gains are optimised to minimise the errors between the optimal and actual control signals. Importantly, the suggested method preserves the order and structure of the primary control, making it applicable to implement on digital hardware devices. In addition, the MG is simulated in MATLAB/Simulink, and the simulation results demonstrate the authenticity, effectiveness, and efficiency of the proposed process for MG frequency regulation in the presence of uncertainties, disturbances, nonlinearity, and dynamic changes in MGs.

Smart microgrids, as the foundations of the future smart grid, combine distinct Internet of Things (IoT) designs and technologies for applications that are designed to create, regulate, monitor, and protect the microgrid (MG), particularly as the IoT develops and evolves on a daily basis. A smart MG is a small grid that may operate individually or in tandem with the electric grid, and it is ideal for institutional, commercial, and industrial consumers, as well as urban and rural societies. A MG can operate in two methods (stand-alone and grid-connected), with the ability to transition between modes due to local grid faults, planned maintenance, expansions, deficits and failures in the host system, and other factors. Energy storage is the process of storing and converting energy that can be used for a variety of purposes, including voltage and frequency management, power backup, and cost optimization. IoT is designed to deliver solutions for optimal energy management, security protocols, control methods, and applications in the MG, with numerous distributed energy resources (DER) and interconnected loads. The use of IoT architecture for MG operations and controls is discussed in this research. With the use of power grid equipment and IoT-enabled technology, MGs are enabling local networks to give additional services on top of the essential supply of electricity to local networks that operate simultaneously or independently from the regional grid. Additionally, this review shows how hybrid AC/DC MGs are advantageous compared to AC and DC MGs. The state-of-the-art optimization techniques and trends in hybrid MG research are included in this work.

Cyber-physical system (CPS) security has become more important in the age of Industry 4.0 because of the quick integration of automation and the Internet of Things. The goal of this project is to create a strong intrusion detection and control system that can recognize and lessen security risks in CPS settings. The suggested approach makes use of deep learning (DL) and reinforcement learning (RL) techniques. To guarantee data consistency, pre-processing procedures such as mean-based imputation and min–max scaling come after data collection. ADASYN data augmentation is used to address class imbalance, while entropy analysis and statistical techniques are used to extract key features. The intrusion detection phase uses a combination of deep convolutional neural networks (DCNN) and bidirectional long short-term memory (BI-LSTM) networks to capture both spatial and temporal relationships in the data, while a hybrid feature selection technique improves the model’s performance. A deep Q-network (DQN) handles attack mitigation and uses reinforcement learning to adjust to new threats. Detecting attack patterns with high sensitivity (0.984), specificity (0.983), and accuracy (0.991626) for dataset 1, the accuracy of dataset 2 is 0.985 for 70% of training and 0.988 for 80% of training, and the Proposed-DBID-Net architecture enhances CPS security in Industry 4.0. The evaluation phase emphasizes how crucial feature selection is to maximize the model’s accuracy. In conclusion, this study offers a thorough and flexible method for protecting CPS in Industry 4.0, guaranteeing accuracy and scalability across changing cyber threats.

Energy management of cooperative microgrids: A distributed optimization approach

Emotion recognition (ER) is crucial for understanding human behaviours, social interactions, and psychological well-being. Electroencephalography (EEG) has emerged as a promising tool for capturing the neural correlates of emotions. This work is a systematic review of articles in ER using EEG signals. A total of 120 articles from 1041 articles were selected based on PRISMA guidelines using defined inclusion and exclusion criteria, published between 2018 and 2024. This article aims to provide an in-depth understanding of the current landscape of ER from EEG signals utilizing deep learning (DL). This review offers valuable guidance for researchers and practitioners seeking more refined and reliable emotion classification systems. To explore the effectiveness of DL models in EEG-based ER, several potential DL models, such as convolutional neural network, long short-term memory (LSTM), gated recurrent unit (GRU), hybrid bidirectional LSTM (BiLSTM), bidirectional GRU, and advanced DL models such as convolutional recurrent neural network and EEG-Conformer models are applied to two popular datasets, SEED and GAMEEMO, respectively, to depict the full process of ER. Additionally, the performance of DL models is also compared with the performance of basic machine learning (ML) models such as SVM, k-nearest neighbors, logistic regression, and boosting algorithms such as AdaBoost, XGBoost and LightGBM. Through extensive experiments and performance evaluations, the performance of different models when applied to the datasets mentioned above is compared. The accuracy, precision, recall, and F1-scores are analysed to determine the most effective model for EEG-based ER. The findings of this study demonstrate that the performance of hybrid DL models is more efficacious than that of ML models. The best-performing model (BiLSTM) classified the emotions, with an accuracy of 90.54% when applied to the GAMEEMO dataset. This research contributes to the growing body of literature on ER and provides insights into the feasibility of using EEG signals to understand emotional states, and presents a structured roadmap for future exploration. The findings can aid in the development of more accurate and reliable ER systems, which can have wide-ranging applications in psychology, social sciences, and human-computer interactions.

The COVID-19 pandemic has triggered a global crisis in health and economic sectors, causing profound impacts on sea transport and trade. This research paper investigates the ramifications of the pandemic on maritime shipping prices, and, hence, their subsequent influence on the stocks of shipping companies. This global upheaval disrupted international trade significantly, resulting in an unprecedented demand surge for shipping services and a substantial spike in freight rates. This study is intended to propose a predictive method based on Multivariate Multi-step convolutional neural network - Bidirectional Long Short-Term Memory (Multivariate Multi-step CNNBiLSTM) networks in order to forecast the prices of three of the most prominent stocks of big organizations operating in maritime transport. The proposed method is composed of Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM), where the research utilizes CNN to help for feature extraction from the inputted data, alongside BiLSTM that forecasts the closing stock price for the upcoming five days, utilizing the extracted feature data. Hence, stock price prediction can be realized by applying a novel optimization strategy, which was founded on the Multivariate Multi-step CNN-BiLSTM model and utilizing the Adam optimizer. Prediction accuracy can be assessed by incorporating four metrics into the system: Root Mean Square Error, Mean Absolute Error, Mean Absolute Percentage Error, and Median Absolute Percentage Error. Experimental findings demonstrate that Multivariate Multi-step CNN-BiLSTM yields the most dependable stock price forecasts with the highest accuracy. The proposed prediction method, correctly applied, can yield economic benefits at the macr and micro levels; the prediction accuracy can help policy makers make better future outlook estimates in relation to inflation, gross domestic product (GDP), and unemployment levels that might be impacted by the volatile, uncontrolled, or unexpected fluctuations of stock prices of some leading economic sectors that are closely connected to global shipping and supply chain operations; thus, leading to serious impacts at the microeconomic level in relation to costs, supply and demand, and behavior of individual consumers and companies.

A novel deep learning-based modelling strategy from image of particles to mechanical properties for granular materials with CNN and BiLSTM

Social networks are popular for advertising, idea sharing, and opinion formation. Due to COVID-19, coronavirus information disseminated on social media affects people’s lives directly. Individuals sometimes managed it well, but it often hampered daily activities. As a result, analyzing people’s feelings is important. Sentiment analysis identifies opinions or sentiments from text. In this paper, we present an effective model that leverages the benefits of Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) to categorize Arabic tweets using a stacked ensemble learning model. First, the tweets are represented as vectors using a word embedding model, then the text feature is extracted by CNN, and finally the context information of the text is acquired by BiLSTM. Aravec, FastText, and ArWordVec are employed separately to assess the impact of the word embedding on the our model. We also compare the proposed method to various deep learning models: CNN, LSTM, and BiLSTM. Experiments are performed on three different Arabic datasets related to COVID-19 and vaccines. Empirical findings show that the proposed model outperformed the other models’ results by achieving F-measures of 76.76%, 87.%, and 80.5% on the SenWave, AraCOVID19-SSD, and ArCovidVac datasets, respectively.