05 Nov 2020 | 17:00SGT
Session 1.1: Welcome and Introductions
Session 1.2: Keynote: Big Data, IoT, Enterprise Data Management and the New Data Requirements of Drones and Robotic Inspection Devices – John McDonald, GE Grid Solutions
This talk begins by discussing key industry / societal trends and their impacts. To realize greater benefits from Intelligent Electronic Devices (IEDs) the two types of data within the IED, operational and non-operational, must be managed differently. The Internet of Things (IoT) and its benefits, and examples of new groups of analytics being developed, will be discussed. The importance of convergence of the Operations Technology and Information Technology groups within a utility is emphasized, and its enablement of enterprise data management. Lastly, the use of new sources of data - unmanned aerial vehicles (UAVs) and robotics – will be described, available data identified, use of the data shown, and data management principles to be implemented discussed.
Session 1.3: Digitization Strategy - Lessons learnt and results from development of an IoT and Data Analytics platform for Energy Forecasting and Predictive Maintenance applications – Sebastian Boehm and Stefan Hartleib, Stadtwerke Leipzig
In this talk, we will share lessons and results acquired over the past three years, concerning the development of an IoT- and Data science Platform. As a municipal energy provider, we are confronted with the trend of digitization and the resulting disruptive business models. Customer now ask for renewable energies and want to be self-sufficient or be an active part of the energy system leading to energy communities.
On the other hand, we must deal with the German energy transition to renewables and the increase in decentralized and small energy producers (like PV power). Renewable energy is very volatile und thus we need solutions which enables us to automatically monitor, analyze, forecast and control many thousand small energy producer and consumer in an optimal way. Use cases like forecasts, predictive maintenance or real-time energy markets and virtual powerplants will be discussed.
19 Nov 2020 | 17:00SGT
Session 2.1 Application of Artificial Neural Networks for Condition Monitoring of Wind Turbine Main Bearings – Niels Jessen, RWE Renewables
In a wind turbine, a failure of an important component, such as a main bearing, can lead to long-lasting downtimes and thus to a corresponding energy loss. In offshore wind energy, the problem is even more serious as maintenance work is not always possible due to adverse weather conditions and must be planned in advance. In order to save operational expenditure, wind farm operators are required to implement a maintenance strategy that enables them to predict a component’s failure as early as possible.
The RWE Renewables GmbH has developed an ANN based tool that predicts the temperature of undamaged main bearings based on a selection of SCADA signals. Anomalies are detected when the actual bearing temperature deviates from the predicted temperature. The tool was shown to be successful in detecting issues up to nine months before failure.
Session 2.2 Grid Fault Location Detection Using System Simulation and Machine Learning - Patrice Brunelle, Hydro Quebec; Graham Dudgeon, MathWorks
MathWorks and Hydro-Québec explore how both system simulation and machine learning can be used to develop algorithms that can detect the location of faults on electric grids using voltage sag measurements. System simulation is used to generate synthesized fault data that covers a broader operating envelope than measured data alone. The synthesized data is then used to train machine-learning classification algorithms. You’ll learn how the performance of classification algorithms may be used to provide further insight into the physical behavior of the system and any limitations associated with training data. You’ll also see how recommendations can be made from this insight to enhance system measurements and training data sets to improve overall classification accuracy.
Session 2.3 Signal Waveform Classification in Partial Discharge Applications for Underground Power Cables - Steffen Ziggler, IMCORP
Underground distribution cable system failures can be predicted! Predictive maintenance begins with understanding how cable system failures occur. Analyzing and interpreting results from partial discharge (PD) measurements taken in the field can be a complex task for humans. Machine learning algorithms and deep learning algorithms are used to automatically identify and categorize markers of defects contained in the PD measurements. These algorithms are used to categorize different defect types by risk of going to failures soon. Differentiating cables with “high to low risk defects” along with those that are “defect free” enables predictive maintenance. Examples of identified defects will be presented.
26 Nov 2020 | 17:00SGT
Session 3.1: A MATLAB-based architecture for testing and deploying real time algorithms for Wide Area Monitoring Protection and Control of the Italian power system – Giorgio Giannuzzi and Cosimo Pisani, Terna
Keep the security high is one of the basic requirements for a Transmission System Operator (TSO). In order to do this, the acquisition of more and more pervasive information for predicting power systems dynamic evolution, in each operating condition, and for identifying the more appropriate and effective countermeasures which guarantee secure and stable operating conditions is essential. The major concerns in doing this are represented by the TSO inability to predict, with high accuracy, the system behavior but also the technological limits of the traditional Supervisory Control and Data Acquisition (SCADA) systems. A pragmatic solution to the highlighted issues is represented by the development of the Wide Area Measurement Systems (WAMS). Terna has developed a MATLAB-based architecture for testing and deploying real time algorithms for Wide Area Monitoring Protection and Control of the Italian power system. Thanks to this architecture power system engineers can, for instance, directly code real time algorithms in MATLAB, test them in a testing environment and once consolidated directly port on the production environment.
Session 3.2: Frequency Control Ancillary Services Delivery Verification to Manage the System Frequency and Ensure System Operations –LingXiao Situ, Hydro Tasmania
In the Australia NEM (National Electricity Market), as one of the marketized ancillary services - FCAS (Frequency Control Ancillary Services) is used by AEMO (Australia Electricity Market Operator) to manage the system frequency and ensure the system is operated in a safe, secure, and reliable manner. Based on the control objectives, the FCAS services can be classified into two categories, Regulation and Contingency.
After a contingency event, it is a mandatory process to evaluate the machine power delivery against its dispatch for the purpose of compliance. In the FCAS R6 case, it is required that the machine output will be recorded by 50ms high speed data, then by excluding the synchronous machine inertial response, the machine FCAS R6 is evaluated based on the net power injection from the governor primary frequency response in the first 6 seconds. However, in reality the evaluation can be rather challenging due to the number of the machines in our portfolio. Further, the issue associated with the data quality also created a lot of difficulty for us to interpret the unexpected delivery and interfere appropriate engineering decision making.
In this presentation, we will focus on the contingency FCAS. Using FCAS R6 (raise response within 6 seconds) to illustrate the FCAS delivery evaluation procedures and challenges, as well as our MATLAB applications to automate this process and gain the evaluation efficiency and system insights.
Session 3.3: Online Identification of the Source of Power Systems Oscillations using PMU – Slava Maslennikov, ISO-NE
Sustained forced and poorly damped natural oscillations represent the threat to power systems as potentially causing instability, uncontrolled cascading outages and undesirable mechanical vibration in equipment. The vast majority of sustained oscillations are forced oscillations caused by the failure of equipment, control systems and atypical operating conditions. Identification of the source of oscillation, which is typically a generator, is a key actionable information for the efficient mitigation requiring the elimination of actual physical forced signal or disconnection it from the network. ISO New England (MA, USA) has developed a MATLAB-based Oscillation Source Locating (OSL) application, which uses Phasor Measurement Units (PMU) data for online identification of the source of oscillations since 2017. The presentation discusses the OSL and online oscillation management at ISO New England.
Session 3.4: A Machine-learning Approach for Dynamic Security Margin Assessment of Power Grids with High Penetration of Renewable Energy – Jin Tan, National Renewable Energy Laboratory
Driven by the increasing of variable and uncertain renewable energy generation, power systems operation and planning are facing challenges to identify critical scenarios for contingency analysis, therefore it is hard to ensure the secure operation of the future grid. This presentation provides a machine-learning based approach to assess the dynamic security margin based on the steady-state dispatch results. A unified framework is proposed to assess the security margin of power system transient stability, frequency stability, and small signals stability simultaneously by using improved neural network and decision tree algorithm. All the machine-learning algorithms are developed on MATLAB platform. The Test results based on the 18-bus system and National Renewable Energy Laboratory (NREL) reduced the 240-bus Western Electricity Coordinating Council (WECC) system verify the effectiveness of the proposed machine-learning approach.>>Register Now