Q: How did Raycatch come to life?
With the drastic reductions in module prices, solar power generation became highly competitive and an attractive source for electricity production, leading to drastically lower margins. This also means that the benefits of performance optimization of existing assets are that much heighteneds a result with a more enhanced ROI. However, what did not change in this post-solar evolution era was the way asset owners, managers and operators handle their daily job. They relied on traditional processes to make their decisions. We decided to take on the challenge of revolutionizing that step as well and with that Raycatch was born by taking out the human factor and the guesswork to implement an automatic navigation for decision-making purposes.
Q: You mentioned the target groups benefiting, can you elaborate on the tangible benefits?
Owners can squeeze an additional of up to 8% on their assets. Asset portfolio management becomes easier and asset managers can improve their vendor relationship management. Operators can save money by being able to carry out specific tasks and avoid pre-scheduled maintenance work leading to a more efficient use of their time. Technicians will also be better directed and equipped with the right components.
Q: What role do AI and machine learning play in realizing these targets?
AI and machine learning are part of the game and at the heart of our solution. They allow for the generation of intelligence without being reliant and having physical eyes on the data. The intersection between AI, ML and solar is radically changing the way we operate as an industry whether it’s through better production forecasting or enabling software to examine large datasets to detect anomalies and make precise decisions based on continuous learning. Data analysis by its nature, is concurring and as such, its the best fit to ongoing operation and optimization driven, it enables to continuously and effortlessly diagnose your assets and reduces the cost of sporadic onsite testing.
Q: How is the training data obtained?
The data should be clean. The problems with the available data is its poor quality, the gaps in the data. To ensure the quality of the data, we’re working with O&Ms and collecting measurements and decisions of technicians, which we feed into our engine to further enhance our classification. We are currently focusing on Eastern Europe, Japan and Taiwan, inspecting over 2 GW of solar assets.
Q: How do you define an ROI-focused approach?
We have gathered insights from our clients, and created a checklist and inventory of the prices for various maintenance actions. Secondly, our system is able to calculate the losses separated according to the source of the problem allowing for more optimum categorizations of recommendations. These recommendations are engineered for specific returns on investment, meaning each action has its own unique ROI field. We are pushing for AI to better extend owners’ and operators’ reach, do a better job and handle more assets during lower margins periods whilst maintaining the same operation costs.
Q: What’s the vision for Raycatch described in one sentence?
Become the leaders of the next leap in global solar adoption.