Echo State Network based Noise Detection in Energy Internet Orienting Justice Blockchain Data

Energy Internet (EI) is a hot research topic which smartly combines the energy system and information and communication technologies to promote the efficiency and reduce the cost of energy generation, transmitting and utilizing, its related data need to be stored in justice blockchain for possible lawsuit. To ensure the integrate of EI data stored in justice blockchain, noise detection is a necessary function which can detect the amplitude of noise and further discover the cyber attacking. To solve this problem, many noise detecting algorithms are available, and the authors have firstly used three layers artificial neural network (ANN) to estimate the noise level quickly and efficiently. In this paper, based on previous research, the authors manage to use Echo State Network (ESN) to solve this problem, and get better simulation results in most scenes. As the simulation results show, the noise interference or cyber attacking behavior can be detected with probability 99% within 4 continuous samples. Applying this algorithm in justice blockchain will make the system more robust and secure, and its data handling cost can be largely reduced.