Davide Astolfi, Silvia Iuliano, Antony Vasile, Alessandro Canali, Marco Pasetti, Francesco Castellani, Alfredo Vaccaro
A Comprehensive Methodology Based on SCADA Data Analysis for Diagnosing Static Errors Affecting Wind Turbine Performance

One of the most important root causes of non-optimal efficiency in the conversion of wind kinetic energy into electricity is represented by static errors affecting wind turbines. These can simply involve sensors (like the anemometers) or might deal with the set up of the rotor (as when the three blades do not have the same pitch reference). Based on such premise, in this work, a comprehensive framework is proposed for diagnosing static errors affecting wind turbine performance, based on SCADA data analysis. The three types of considered static errors are: static yaw error; absolute or relative blade pitch misalignment; anemometer bias. It is desirable to exploit as much as possible the SCADA-collected data for diagnosing such types of static errors. Yet, given the mediation of the control system, the SCADA data might even be misleading and, for example, might indicate a correct alignment of the rotor to the wind flow, while it is not correct. Hence, the objective of the work is identifying phenomena observable through SCADA data which can be related to such static errors. Acknowledged that some manifestations can be similar for the various errors, a work flow is set up for distinguishing between them. The selected phenomena are: under-performance, which is associated to all the errors; increased vibrations; shift in the nacelle anemometer measurements. Real-world test cases, namely wind farms operated by the ENGIE Italia company, are discussed. A combination of single wind turbine and fleet-wide methods is proposed. The latter involve dividing the fleet in potentially anomalous and normal wind turbines and modeling quantities of interest (like the power) as a function of the corresponding quantities of the healthy wind turbines. The collected case studies indicate that the proposed method is effective and allows distinguishing between the various types of errors related to the anemometer, the yaw and the pitch systems.

Théo Chacou Bertoldi, Viktoriya Mostova, Silvia Iuliano, Alfredo Vaccaro
On-Line Frequency Forecasting using Convolutional Neural Networks

In modern power systems the replacement of large synchronous generators with distributed inverter-based resources is lowering the power system inertia, making electrical grids more vulnerable to dynamic perturbations. For this purpose the European Network of Transmission System Operators for Electricity promoted the enhancement of the grid operation tools with specific functions for on-line estimation of the power system inertia, which are extremely useful in detecting critical thresholds and triggering proper mitigation strategies. For this purpose, the development of reliable frequency predictive models represents an essential requirement for enabling system inertia estimation. To try and address this issue, this paper explores the role of Convolutional Neural Networks (CNN) for on-line frequency estimation from grid measurements. The main idea is to model the grid frequency deviations by a CNN-based identification technique, which allows inferring the main parameters ruling the power system dynamics. Detailed simulation results obtained on several case studies demonstrate that the CNN model is able to detect data patterns, and discover hidden relationships maintaining low estimation errors even for multi-step ahead frequency predictions, thus offering a valuable tool for inertia estimation in modern power systems, especially with the increasing share of renewable energy.

Blair Hall
Communities of practice in metrology

Digital transformation in metrology will entail digitalisation of work practices in national metrology institutes (NMIs). This article argues that communities of practice (CoPs)—informal social structures—play a key role in how NMIs operate, making their dynamics a critical consideration in managing digitalisation. NMI work is empirical in nature, and the knowledge required to perform measurements at the highest level is embedded in practice-based forms within CoPs. However, digitalisation necessitates the development of abstract representations of processes and information. Moreover, scientific knowledge is usually communicated in a formal language using abstract concepts. So, digitalisation of NMI work products may be challenged by the distinction between empirical and abstract modes of knowledge capture. This article explores the conceptual tension between abstract and empirical modes of thought in metrology and the challenge of aligning them in the context of digital transformation. It draws on published anecdotal accounts and the author’s professional experience.

Diego Coppa, Blair Hall, Anjali Sharma
Metrology and Architectural Description: a case study of the Hybrid Comparison system

An architectural approach has been explored as a means to describe a system for managing a metrological process known as a hybrid comparison. This top-down approach, based on the international standard for architectural description—ISO/IEC/IEEE 42010—focuses on stakeholder concerns and contrasts with the more common bottom-up development of tools, services, and file formats typically seen in the metrology community. By clearly separating stakeholder concerns, the approach also reveals structural relationships with other external metrological systems. This exploratory study suggests that the perspective provided by architectural descriptions supports the coordinated development of digital metrological systems.

Marcel van Dijk, Walter Knulst, Devrim Nalbantoglu, Gertjan Kok
Metrological Validation of a Digital Model for a CMM including Digital Bias Correction

Virtual experiments and digital twins are virtual representations of simulation processes and can aid in crucial tasks, such as interpreting measurement data or evaluating uncertainties. Therefore, virtual experiments and digital twins are essential components in the digital transformation in metrology. However, it is important that the digital model is properly validated, to make sure its output is reliable. This research shows how to validate a virtual coordinate measurement machine (CMM) by means of reference measurements. The output of the digital model is compared to reference measurements of a ring gauge and its consistency is evaluated. The results show that the output of the digital model for the diameter of the ring gauge is consistent with the reference data. The roundness measurements seem to contain an inherent bias, introduced by the CMM. After correcting for this bias, the output of the digital model for the roundness of the ring gauge is also consistent with the reference data.