This study proposes the integration of a hybrid Proof of Work/Proof of Stake (PoW/PoS) consensus mechanism with a Long Short-Term Memory (LSTM) model for anomaly detection in blockchain-based microgrids. The hybrid PoW/PoS model is designed to address common issues in blockchain systems, such as 51% attacks, double-spending, and high energy consumption, by optimizing energy usage and enhancing security. Simulation results show that the system can process transactions with an average throughput of 37.25 transactions per second (TPS), an average latency of 26.84 milliseconds per transaction (ms/tx), and extremely efficient energy consumption per transaction (0.01 kWh/tx). The LSTM model applied for anomaly detection achieves an 89.10% detection rate, a 0.00% false positive rate, and a 0.12 s recovery time, indicating the system's reliability in facing attacks. The hybrid PoW/PoS system demonstrates advantages in both energy efficiency and resilience to attacks compared to individual PoW and PoS systems. This research contributes to the development of safer, more efficient, and scalable blockchain-based microgrids by integrating Artificial Intelligence (AI) to strengthen the system against anomalies and threats.
This paper provides an analysis of recent advancements and potential applications of blockchain technology in microgrids. Microgrids are small-scale electrical networks that can operate autonomously or integrate with a central power grid, offering enhanced security, resilience, and efficiency in energy utilization through blockchain implementation. Blockchain's decentralization, transparency, and security capabilities present innovative solutions for optimizing microgrid operations. This study explores blockchain applications in microgrids, addressing fundamental principles, prior research, existing gaps, case studies, technical and regulatory challenges, and future optimization opportunities. The findings indicate that while blockchain adoption in microgrids is still in its early stages, it holds significant promise for transforming energy systems. With continuous technological advancements and supportive policy frameworks, blockchain can facilitate the transition towards a more sustainable and efficient energy future. Furthermore, the global proliferation of blockchain-based microgrids highlights the need for Indonesia, endowed with abundant renewable energy resources, to enact regulatory measures supporting this emerging technology
Penyakit Virus Corona 19 (COVID-19) merupakan penyakit menular yang disebabkan oleh virus corona yang baru-baru ini ditemukan. Saat ini COVID-19 menjadi fenomena permasalahan untuk kita semua namun hingga sekarang belum ada obat yang ditemukan ampuh dalam mengatasinya. Persoalan lain adalah pada proses pendeteksian orang terjangkit. Hasil pendeteksian Covid-19 menggunakan PCR Swap masih dianggap sangat lambat dan menggunakan Rapid Tes bahkan dianggap kurang meyakinkan dengan melihat beberapa kasus yang ada sebelumnya. Tujuan penelitian ini untuk pendeteksian orang terjangkit COVID-19 lebih cepat dengan tingkat akurasi yang tinggi menggunakan metode Artificial Intelligence yang lebih khusus menggunakan Deep Learning arsitektur Convolutional Neural Network (CNN). Metode penelitian yang digunakan adalah literature review, dimana artikel dikumpulkan dan diproses menggunakan aplikasi mendeley, kriteria artikel yang digunakan adalah yang diterbitkan tahun 2020 yang berkaitan dengan penanganan COVID-19 khususnya yang memanfaatkan Artificial Intelligence dalam pembahasannya. Dengan mengumpulkan dan membahas beberapa penelitian yang ada maka dapat dikatakan bahwa dengan menggunakan Artificial Intelligence sistem dapat mendeteksi terjangkitnya seseorang melalui analisa pola yang ada pada hasil CT Scan Paru dengan memanfaatkan tingkat akurasi data latih yang ada.