scholarly journals DyAt Nets: Dynamic Attention Networks for State Forecasting in Cyber-Physical Systems

2019 ◽  
Author(s):  
Nikhil Muralidhar ◽  
Sathappan Muthiah ◽  
Naren Ramakrishnan
Keyword(s):  
Time Series ◽  
Performance Improvement ◽  
Load Forecasting ◽  
Cyber Physical Systems ◽  
Electric Load ◽  
Attention Networks ◽  
Physical Systems ◽  
Dynamic Attention ◽  
A Performance

Multivariate time series forecasting is an important task in state forecasting for cyber-physical systems (CPS). State forecasting in CPS is imperative for optimal planning of system energy utility and understanding normal operational characteristics of the system thus enabling anomaly detection. Forecasting models can also be used to identify sub-optimal or worn out components and are thereby useful for overall system monitoring. Most existing work only performs single step forecasting but in CPS it is imperative to forecast the next sequence of system states (i.e curve forecasting). In this paper, we propose DyAt (Dynamic Attention) networks, a novel deep learning sequence to sequence (Seq2Seq) model with a novel hierarchical attention mechanism for long-term time series state forecasting. We evaluate our method on several CPS state forecasting and electric load forecasting tasks and find that our proposed DyAt models yield a performance improvement of at least 13.69% for the CPS state forecasting task and a performance improvement of at least 18.83% for the electric load forecasting task over other state-of-the-art forecasting baselines. We perform rigorous experimentation with several variants of the DyAt model and demonstrate that the DyAt models indeed learn better representations over the entire course of the long term forecast as compared to their counterparts with or without traditional attention mechanisms. All data and source code has been made available online.

scholarly journals A Dependable Time Series Analytic Framework for Cyber-Physical Systems of IoT-based Smart Grid

2019 ◽  
Vol 3 (1) ◽  
pp. 1-18 ◽  
Author(s):  
Chang Wang ◽  
Yongxin Zhu ◽  
Weiwei Shi ◽  
Victor Chang ◽  
P. Vijayakumar ◽  
...  
Keyword(s):  
Time Series ◽  
Smart Grid ◽  
Cyber Physical Systems ◽  
Analytic Framework ◽  
Physical Systems

scholarly journals Accelerating privacy-preserving momentum federated learning for industrial cyber-physical systems

2021 ◽  
Author(s):  
Linlin Zhang ◽  
Zehui Zhang ◽  
Cong Guan
Keyword(s):  
Convergence Rate ◽  
Performance Improvement ◽  
Gradient Descent ◽  
Homomorphic Encryption ◽  
Cyber Physical Systems ◽  
Privacy Preserving ◽  
Fully Homomorphic Encryption ◽  
Physical Systems ◽  
Cloud Server ◽  
And Performance

AbstractFederated learning (FL) is a distributed learning approach, which allows the distributed computing nodes to collaboratively develop a global model while keeping their data locally. However, the issues of privacy-preserving and performance improvement hinder the applications of the FL in the industrial cyber-physical systems (ICPSs). In this work, we propose a privacy-preserving momentum FL approach, named PMFL, which uses the momentum term to accelerate the model convergence rate during the training process. Furthermore, a fully homomorphic encryption scheme CKKS is adopted to encrypt the gradient parameters of the industrial agents’ models for preserving their local privacy information. In particular, the cloud server calculates the global encrypted momentum term by utilizing the encrypted gradients based on the momentum gradient descent optimization algorithm (MGD). The performance of the proposed PMFL is evaluated on two common deep learning datasets, i.e., MNIST and Fashion-MNIST. Theoretical analysis and experiment results confirm that the proposed approach can improve the convergence rate while preserving the privacy information of the industrial agents.

Load Forecasting using Time Series Techniques

2021 ◽  
pp. 11343-11357
Author(s):  
Shahida Khatoon, Ibraheem, Priti, Mohammad Shahid
Keyword(s):  
Time Series ◽  
Linear Models ◽  
Linear Trend ◽  
Moving Average ◽  
Load Forecasting ◽  
Electric Load ◽  
Time Data ◽  
Efficient Operation ◽  
Use Of Time ◽  
Distribution Company

Load Forecasting is of great significance for effective and efficient operation of power system. Use of time series is of much importance in load forecasting. In this study, effectiveness of different time series techniques is identified to gathered valuable information. The objective is to predict electric load efficiently and effectively. This paper analyses the prediction accuracy of variety of time series method in modeling Electric load forecasts. The study examines the time series forecasting methods applied to estimate future electric load, specifically, Moving Average (MA), Linear Trend, the Exponential and Parabolic Trend. A comparison of different forecasting techniques of Time Series is demonstrated on real time data. The data utilized for forecast is made available through a distribution company of India. The traditional linear models and hybrid models along with ANN are developed. These models are appraised for the forecasting capability.

scholarly journals Effective 5G Wireless Downlink Scheduling and Resource Allocation in Cyber-Physical Systems

Technologies ◽  
2018 ◽  
Vol 6 (4) ◽  
pp. 105 ◽  
Author(s):  
Ankur Vora ◽  
Kyoung-Don Kang
Keyword(s):  
Resource Allocation ◽  
Dynamic Programming Algorithm ◽  
Cyber Physical Systems ◽  
Programming Algorithm ◽  
Cross Layer ◽  
Communication Performance ◽  
Physical Systems ◽  
Downlink Scheduling ◽  
Machine Type Communication

In emerging Cyber-Physical Systems (CPS), the demand for higher communication performance and enhanced wireless connectivity is increasing fast. To address the issue, in our recent work, we proposed a dynamic programming algorithm with polynomial time complexity for effective cross-layer downlink Scheduling and Resource Allocation (SRA) considering the channel and queue state, while supporting fairness. In this paper, we extend the SRA algorithm to consider 5G use-cases, namely enhanced Machine Type Communication (eMTC), Ultra-Reliable Low Latency Communication (URLLC) and enhanced Mobile BroadBand (eMBB). In a simulation study, we evaluate the performance of our SRA algorithm in comparison to an advanced greedy cross-layer algorithm for eMTC, URLLC and LTE (long-term evolution). For eMTC and URLLC, our SRA method outperforms the greedy approach by up to 17.24%, 18.1%, 2.5% and 1.5% in terms of average goodput, correlation impact, goodput fairness and delay fairness, respectively. In the case of LTE, our approach outperforms the greedy method by 60%, 2.6% and 1.6% in terms of goodput, goodput fairness and delay fairness compared with tested baseline.

Electric Load Movement Forecasting Based on the DFT Interpolation with Periodic Extension

2015 ◽  
Vol 24 (08) ◽  
pp. 1550123
Author(s):  
Zong-Chang Yang
Keyword(s):  
Time Series ◽  
Load Forecasting ◽  
Arma Model ◽  
Ar Model ◽  
Electric Load ◽  
Ann Model ◽  
Method Performance ◽  
Bp Artificial Neural Network ◽  
Load Movement ◽  
Artificial Neural Network Ann

Electric load forecasting is increasingly important for the industry. This study addresses the load forecasting based on the discrete Fourier transform (DFT) interpolation. As the most common analysis method in the frequency domain, the conventional Fourier analysis cannot be directly applied to prediction. From the perspective of time-series analysis, electric load movement influenced by various factors is also a time-series, which is usually subject to cyclical variations. Then with periodic extension for the load movement, a forecasting approach based on the DFT interpolation is proposed for predicting its movement. The proposed DFT interpolation prediction model is applied to experiments of forecasting the daily EUNITE load movement and annual load movement of State Grid Corporation in China. The experimental results and analysis show potentiality of the proposed method. Performance comparisons indicate that the proposed DFT interpolation model performs better than the three commonly used interpolation algorithms as well as the classical autoregressive (AR) model, the ARMA model, and the BP-artificial neural network (ANN) model on the same forecasting tasks.

Sign in / Sign up

Export Citation Format

Share Document