The majority of failures in grid systems is related to aging and progressive deterioration of grid components (e.g. junctions, cable connectors, insulators). One of the main causes are “partial discharges”, localized dielectric breakdowns of a small portion of a solid or fluid electrical insulation under high voltage stress, which does not bridge the space between two conductors. Partial discharges can occur in a gaseous, liquid or solid insulating medium. It often starts within gas voids, such as voids in solid epoxy insulation or bubbles in transformer oil. Protracted partial discharge phenomena can erode solid insulation and eventually lead to breakdown of insulation. This is the reason why it would be most useful to detect them, in order to enforce preventive maintenance on the interested devices. This is hard to do, since even field inspectors often cannot detect them due to their nature (see attached images for examples of deteriorated devices). Partial discharges can be sensed in many other ways, through sound sensors or current distortion analysis for instance. The difficult part, and the focus of this challenge, is correlating the data collected by these sensors and the physical deterioration (i.e. the expected remaining lifetime) of the component under analysis.
A list of benefits that such innovative solution would bring includes but it is not limited to:
- Reduction of outages due to devices breakdown (improvement of QoS, avoiding penalties);
- Reduction of maintenance costs (maintenance only when needed).
Nowadays field inspectors set preventive maintenance measures on devices, but this is not sufficient. Ocular inspections hardly produce results, since a partial discharge is often not evident from the outside (as evident from the description below).
This Challenge’s objective is to find cost-efficient solutions that allow the detection of partial discharges in order to enforce preventive maintenance on the devices of interest.
Enel is looking for start-up or small and medium enterprises (SMEs) to partner with and develop an algorithm capable of using the information coming from the specific sensors (whose typology can be identified by Solvers themselves: sound, current, etc) to evaluate the remaining useful lifetime of network components in order to efficiently plan their maintenance.
In general, the solution needs to satisfy the following Solution Requirements:
- To be an algorithm capable of correlating the partial discharge data (i.e. frequency, intensity, in relationship with the type of monitored network device and historical data), collected by the chosen sensor system, to failure probability and estimated lifetime (must have);
- To make use of data collected by a sensor system of choice (e.g. sound, current, …), that should be scalable and cost-effective due to the great amount of components to be monitored (nice to have);
- Make use of a sensor system that is both scalable and cost-effective, due to the great amount of components to be monitored (must have);
- Possess an algorithm capable of correlating the partial discharge data collected by the sensor system (frequency, intensity, in relationship with the type of monitored device and historical data) to failure probability and estimated lifetime (must have).
- Any proposed solution should have at least TRL 4 (Technology Readiness Level), which means: design, development and lab testing of components / processes. Results provide evidence that performance targets may be based on projected or modeled systems.
[innovation], [technology], [artificial intelligences], [algorithm], [IoT], [predictive maintenance], [sensors], [safety], [maintenance efficiency]
The submitted proposals should include:
- Video - it is appreciated if the Solvers will submit a video that summarises the proposal and allows Enel to do a quick initial screening of proposals
- Collaboration Proposal including:
- A description of the proposed algorithm with an explanation of how the Solver proposes to address all the Solution Requirement. The Solver can withhold proprietary information, if necessary, but should provide convincing evidence for ENEL to appreciate the merits of the approach and be comfortable that the solution can effectively work;
- A brief discussion of capabilities and relevant prior experience that are relevant for the development of the solution and success of the post-Challenge collaboration;
- The Solver should explain what they can provide and what might be required of the Seeker. For example: “I can provide with the expertise, but I would need the Seeker to facilitate access to data”;
- A brief overview of the proposed path forward along with a plan for validation (e.g. deliverables, timelines, milestones, and cost estimates);
- The Solver should indicate the TRL (Technology Readiness Level) of the solution.
- General Information about the Solver including:
- The key contact person for this Challenge (including phone number and email address).
- Organization/Company name and address (including website, if available)
(NOTE: For most Challenges, Solvers are not allowed to include personal contact information; however, for an eRFP Challenge, it is required.)