Time series prediction techniques have been used in many real-world applications such as financial market prediction, electric utility load forecasting , weather and environmental state prediction, and reliability forecasting. The underlying system models and time series data generating processes are generally complex for these applications and the models for these systems are usually not known a priori. Accurate and unbiased estimation of the time series data produced by these systems cannot always be achieved using well known linear techniques, and thus the estimation process requires more advanced time series prediction algorithms. This paper provides a survey of time series prediction applications using a novel machine learning approach: support vector machines (SVM). The underlying motivation for using SVMs is the ability of this methodology to accurately forecast time series data when the underlying system processes are typically nonlinear, non-stationary and not defined a-priori. SVMs have also been proven to outperform other non-linear techniques including neural-network based non-linear prediction techniques such as multi-layer perceptrons.The ultimate goal is to provide the reader with insight into the applications using SVM for time series prediction, to give a brief tutorial on SVMs for time series prediction, to outline some of the advantages and challenges in using SVMs for time series prediction, and to provide a source for the reader to locate books, technical journals, and other online SVM research resources.

Time Series Prediction Using Support Vector Machines: A Survey

This paper presents a comprehensive review and assessment of the latest research and advancement of electric vehicles (EVs) interaction with smart grid portraying the future electric power system model. The concept goal of the smart grid along with the future deployment of the EVs puts forward various challenges in terms of electric grid infrastructure, communication and control. Following an intensive review on advanced smart metering and communication infrastructures, the strategy for integrating the EVs into the electric grid is presented. Various EV smart charging technologies are also extensively examined with the perspective of their potential, impacts and limitations under the vehicle-to-grid (V2G) phenomenon. Moreover, the high penetration of renewable energy sources (wind and photovoltaic solar) is soaring up into the power system. However, their intermittent power output poses different challenges on the planning, operation and control of the power system networks. On the other hand, the deployment of EVs in the energy market can compensate for the fluctuations of the electric grid. In this context, a literature review on the integration of the renewable energy and the latest feasible solution using EVs with the insight of the promising research gap to be covered up are investigated. Furthermore, the feasibility of the smart V2G system is thoroughly discussed. In this paper, the EVs interactions with the smart grid as the future energy system model are extensively discussed and research gap is revealed for the possible solutions.

Electric vehicles and smart grid interaction: A review on vehicle to grid and renewable energy sources integration

Electrifying transportation is a promising approach to alleviate the climate change issue. The adoption of electric vehicle into market has introduced significant impacts on various fields, especially the power grid. Various policies have been implemented to foster the electric vehicle deployment and the increasing trend of electric vehicle adoption in the recent years has been satisfying. The continual development of electric vehicle power train, battery and charger technologies have further improved the electric vehicle technologies for wider uptake. Despite the environmental and economical benefits, electric vehicles charging introduce negative impacts on the existing network operation. Appropriate charging management strategies can be implemented to cater for this issue. Furthermore, electric vehicle integration in the smart grid can bring many potential opportunities, especially from the perspective of vehicle-to-grid technology and as the solution for the renewable energy intermittency issue. This paper reviews the latest development in electric vehicle technologies, impacts of electric vehicle roll out and opportunities brought by electric vehicle deployment.

A review on the state-of-the-art technologies of electric vehicle, its impacts and prospects

A comprehensive survey on support vector machine classification: Applications, challenges and trends

Nowadays, deep learning is a current and a stimulating field of machine learning. Deep learning is the most effective, supervised, time and cost efficient machine learning approach. Deep learning is not a restricted learning approach, but it abides various procedures and topographies which can be applied to an immense speculum of complicated problems. The technique learns the illustrative and differential features in a very stratified way. Deep learning methods have made a significant breakthrough with appreciable performance in a wide variety of applications with useful security tools. It is considered to be the best choice for discovering complex architecture in high-dimensional data by employing back propagation algorithm. As deep learning has made significant advancements and tremendous performance in numerous applications, the widely used domains of deep learning are business, science and government which further includes adaptive testing, biological image classification, computer vision, cancer detection, natural language processing, object detection, face recognition, handwriting recognition, speech recognition, stock market analysis, smart city and many more. This paper focuses on the concepts of deep learning, its basic and advanced architectures, techniques, motivational aspects, characteristics and the limitations. The paper also presents the major differences between the deep learning, classical machine learning and conventional learning approaches and the major challenges ahead. The main intention of this paper is to explore and present chronologically, a comprehensive survey of the major applications of deep learning covering variety of areas, study of the techniques and architectures used and further the contribution of that respective application in the real world. Finally, the paper ends with the conclusion and future aspects.

A Survey of Deep Learning and Its Applications: A New Paradigm to Machine Learning

The integration of a massive number of small-scale wind turbines and plug-in electric vehicles (PEVs) brought about urgent technical challenge to power distribution network operators (DNOs) in terms of secure power supply and energy dispatching optimization. In this paper, we exploited three coordinated wind-PEV energy dispatching approaches in the vehicle-to-grid (V2G) context, i.e., valley searching, interruptible and variable-rate energy dispatching, aiming to promote the user demand response through optimizing the utilization efficiency of wind power generation as well as meeting the dynamic power demands. This issue is addressed in a stochastic framework considering the uncertainties of wind power generation as well as the statistical PEV driving patterns. The performances of the proposed solutions are assessed through a comparative study through numerical simulation experiments covering sufficient system scenarios by the use of scenario generation and reduction techniques. The result demonstrates that the energy dispatch based on the latter two approaches can achieve better matching between power generation and demands as well as PEV user satisfaction. In addition, the suggested approaches can be adopted by DNOs in practice with minimal deployment hurdles to promote the energy supplies within microgrid with wind power sources and PEVs.

Coordinated Energy Dispatching in Microgrid With Wind Power Generation and Plug-in Electric Vehicles

For optimal microgrid (MG) operation, one challenge is the supply of cooling and electricity energy is a coupled co-optimization issue when considering the combined cooling, heating and power (CCHP) units and ice-storage air-conditioners. Another challenge is the inherent randomness of renewable energy within the MG should be accommodated by MG itself. In Part I of this two-part paper, the partial load performance of CCHPs and the performance of ice-storage air-conditioners are modeled, and the cooling and electricity coordinated MG day-ahead scheduling and real-time dispatching models are established. In day-ahead scheduling model, the uncertainty of wind and solar power is represented by multi-scenarios and the objective is to achieve the minimal expected MG operation cost. In real-time dispatching model, the different time-scale dispatch schemes are respectively applied for cooling and electricity to smooth out the fluctuations of renewable energy supply and to follow the variations of cooling and electricity demands by the fine dispatching of the components within MG such that the impact of MG to the connected main grid is minimal. The proposed MG multi time-scale cooling and electricity coordinated schedule achieves an integrated optimization for multi energy-type supply, and makes the MG be controllable as seen from the main grid.

A Multi Time-Scale and Multi Energy-Type Coordinated Microgrid Scheduling Solution—Part I: Model and Methodology

Power system faults are significant problems in power transmission and distribution. Methods based on relay protection actions and electrical component actions have been put forward in recent years. However, they have deficiencies dealing with power system fault. In this paper, a method for data-based line trip fault prediction in power systems using long short-term memory (LSTM) networks and support vector machine (SVM) is proposed. The temporal features of multisourced data are captured with LSTM networks, which perform well in extracting the features of time series for a long-time span. The strong learning and mining ability of LSTM networks is suitable for a large quantity of time series in power transmission and distribution. SVM, with a strong generalization ability and robustness, is introduced for classification to get the final prediction results. Considering the overfitting problem in fault prediction, layer of dropout and batch normalization are added into the network. The complete network architecture is shown in this paper in detail. The parameters are adjusted to fit the specific situation of the actual power system. The data for experiments are obtained from the Wanjiang substation in the China Southern Power Grid. The real experiments prove the proposed method’s improvements compared with current data mining methods. Concrete analyses of results are elaborated in this paper. A discussion of practical applications is presented to demonstrate the feasibility in real scenarios.

Data-Based Line Trip Fault Prediction in Power Systems Using LSTM Networks and SVM

In part II of this two-part paper, the improved particle swarm optimization (IPSO) algorithm for solving the microgrid (MG) day-ahead cooling and electricity coordinated scheduling is proposed. Significant improvements are made in comparison with the conventional PSO algorithm from two aspects. First, the mandatory correction is implemented to ensure the complex coupled constraints among the components of a particle are met after the particle's position is updated, which could enhance the algorithm performance when solving the problem including complex constraints. Second, it is assumed that a solution denoted by a particle occupies a neighboring area, the size of which decreases from a certain value to nearly zero as the iteration step increases to its limitation, which helps to avoid the pre-maturity of algorithm. For an MG composed of the combined cooling, heating and power (CCHP) units, PV panels, wind turbines, and storage batteries, a range of case studies under different MG operating modes are carried out through simulations. The simulation results demonstrate the proposed multi time-scale, multi energy-type coordinated MG scheduling solution can achieve the co-optimization of multi energy-type supply to meet customer's cooling and electricity demands, and make the MG be controllable as seen from the connected main grid.

Zhejing Bao

Papers

Coordinated Energy Dispatching in Microgrid With Wind Power Generation and Plug-in Electric Vehicles

A Multi Time-Scale and Multi Energy-Type Coordinated Microgrid Scheduling Solution—Part I: Model and Methodology

Data-Based Line Trip Fault Prediction in Power Systems Using LSTM Networks and SVM

A Multi Time-Scale and Multi Energy-Type Coordinated Microgrid Scheduling Solution—Part II: Optimization Algorithm and Case Studies

Analysis of cascading failure in electric grid based on power flow entropy