X-AMINOR
The current developments in the energy system pose new challenges to grid operators, as the increased integration of renewables and the introduction of dynamic loads into the power systems lead to a decoupling of consumption and production. Digitalization is one of the major tools for grid operators to tackle these challenges, as it allows them to implement flexible and advanced operation and planning strategies, which reduce costs while increasing service security.
Transformers constitute a core component in energy grids. Their availability and longevity can thus be seen as command variables in the context of service security. The outage of a transformer can lead to service loss in combination with high costs. Predictive maintenance is an important factor to increase the longevity of these essential infrastructure elements. At the start of its life, a transformer may fail due to mistakes made in the manufacturing process or its transportation and installation at the operational site. Towards the end of its life, the risk of failure increases due to its age and wear. Transformer fires, short circuits, mechanical faults, and overvoltage events are all destructive events that can occur during operation and cause significant damage to the transformer and its surrounding infrastructure. Unfortunately, destructive events can lead to a lack of data available for investigators during post-event forensics and failure analysis, which may impede the identification of the event’s cause and the further development of preventative measures, for instance fire protection.
Due to the long-life expectancy of transformers and the large number of legacy components in the energy grid, comprehensive retro-fitting of these components is not economically feasible and additionally poses a logistical challenge due to the amount of components and associated required engineering. To achieve this, data from existing monitoring systems will be combined with additional, new data modalities and advanced models to enable improved and dynamic monitoring of
transformers. In X-AMINOR, our solution aims at providing functionality for dynamic transformer monitoring and optimization of operation to reduce operation costs, preventing failures, and supporting intelligent grid planning strategies.
Goals
X-AMINOR aims at developing novel approaches for the exploitation of audio, video, thermal and other data towards achieving better lifecycle monitoring of transformers, without requiring a comprehensive retrofitting of these essential components of the power grid. The technical result of X-AMINOR is a mobile lifecycle-monitoring robotic and cloud backend solution to perform minimally invasive transformer monitoring to be deployed alongside standard existing transformer monitoring strategies. The robotic platform is equipped with a cross-/multisensor platform (audio, video, thermal), which can gather data similar to a traditional on-site inspection of the transformer. The system is initialized in the vicinity of a transformer and performs autonomous monitoring and assessment. Advanced data analytics are used to build data models which provide the basis for predictive maintenance and continuous product improvements. The project will demonstrate this functionality on a system level in the form of a demonstrator. X-AMINOR has been developed in the context of two application scenarios: final acceptance test after production and normal operation. A number of evaluation scenarios have been used for an in-depth quantitative and qualitative evaluation of the system and have allowed the assessment of the quality of the developed methods and the benefit of such a system in the context of automated condition monitoring.
Approach
The use of continuous monitoring and predictive maintenance is essential to prolong the life of transmission systems and reduce any unexpected outages. Standard monitoring solutions are integrated into a SCADA system to monitor the key performance indicators (KPIs) of power transformers. The type of data which can be gathered from a transformer depends on the transformer model. Generally, modern transformers have access to a greater range of information compared to older models (like winding hotspot sensors and dissolved gas analyzers). In contrast, in the X-AMINOR project, we have developed a robotic multi-sensor platform and cloud backend solution to perform minimally invasive transformer monitoring to be deployed alongside standard existing transformer monitoring strategies (Figure 1). The robotic platform provides transformer on-site data periodically, which is processed and made available to stakeholders by means of the cloud backend. Our system can, among other capabilities, support the lifecycle-monitoring of the transformer and provide data for post-event forensics in the case of destructive events.
Expected and Achieved Results
X-AMINOR is designed to collect and analyze visual and acoustic information in addition to already available operating parameters in order to enable a more precise assessment of the transformer’s condition. To guarantee scalability, edge computing has been implemented via GPU-enabled computing nodes. This allows nodes to preprocess data streams and perform first analyses, while the backend performs computation-heavy analytics as well as model training and development.
Project Details
Contact
No member found