The views and opinions expressed in the interviews published on Made in CA are those of the interviewees and do not reflect the official policy or position of Made in CA.

The information provided through these interviews is for informational purposes only and does not constitute an endorsement or recommendation of any products, services, or individuals featured. We strongly encourage readers to consult with appropriate professionals or authorities in the relevant fields for accurate information and advice.

Sharat Singh

Quadrical Ai helps improve solar generation by 2-4% and O&M efficiency by 15% with a Digital Twin AI-based Asset Management Platform. We are differentiated via our product-first approach built on proprietary Digital Twin Technology, which results in industry-leading performance benchmarking, and the most accurate combination of Prediction/Anomaly Detection AI productized for solar asset operations.

Tell us about yourself?

I was a Global VP at Adobe Global, with previous stints as CTO at MakeMyTrip, and MD at Scale-Ups, prior to a career in the US with startups and companies like Microsoft. I did B. Tech at IIT Delhi with an MBA from New York University. I’m passionate about making the benefits of AI and Big Data accessible to everyone, esp. to drive energy transition goals via Renewables. Many of our early customers were in the solar industry in India, where we have scaled up a development subsidiary to support our current global customer base.

If you could go back in time a year or two, what piece of advice would you give yourself?

Brace and plan proactively for industry shifts. An example would be Storage gaining momentum worldwide as an enabler of Renewables. That actually pushed us to reassess our product portfolio and start building Storage capabilities to complement the rest of Quadrical’s offerings.

What problem does your business solve?

Quadrical AI Platform Image
Quadrical AI Platform Image

Plummeting PPA & panel costs are forcing O&M and Asset Managers to figure out ways to improve both operational efficiency and yield. They are also plagued by the unrealistic yield expectations set by industry standards like PVSyst, lack of effective investments in technology (which is why we witness abundant use of manual tools like excel, stats-based tools etc.) and skyrocketing costs of drone/thermal imaging every 6-12 months.

To put into perspective, a 1MW plant means approx. 70 strings (with 50 panels each) over 5-6 acres – often with inadequate instrumentation or calibration quality, coupled with unclear expectations of weather, micro-weather, uneven degradation scenarios across the plant, plus shading, soiling, and scores of short-term, fixable issues. When multiplied by 1000s for plant & fleet size – the sheer scale of issues indicates that plant managers are ultimately able to do only bare minimum fixes on time, mostly tending to outage or high-threshold alerts.

Additionally, standard industry benchmarks like PVSyst are just not accurate enough to validate plant performance against. Challenges with traditional contracted energy, resulting from inaccuracies and noise, blind reliance on pyranometer data (which could be very unreliable due to frequent calibration requirement, limited number of devices covering large plant areas etc.), no consideration of efficiency and behaviour of individual SCBs at different stages of aging – means that the benchmark may work at an aggregate level (for entire plant for a longer period of time), but is completely ineffective to make actionable, measurable improvements at SCB or weekly granularity.

To solve the challenges posed by unrealistic yield expectations and razor-thin O&M margins, Quadrical uses its proprietary Digital Twin technology to create highly accurate and granular performance benchmark, personalized for any plant and each underlying device, using historical and real-time data, for acceptable DC, Inverter, AC performance. This leads to Condition-based Monitoring – with deviations from the benchmark being tracked, characterized, and quantified for weather, structural degradations, shading, soiling, new-fixable issues – ultimately converted into actionable, revenue prioritized work orders. We have worked with customers to replace their Preventative Maintenance schedules with this Prescriptive approach to help improve their performance by 2-4% and O&M efficiency by 15% over a period of 6 months of continuous performance audit and issue remediation. Additionally, the Digital Twin level accuracy has led customers to have strong confidence in their OEM claims and due diligence processes.

What is the inspiration behind your business?

Plummeting PPA & panel costs are forcing O&M and Asset Managers to figure out ways to improve both operational efficiency and yield. They are also plagued by the unrealistic yield expectations set by industry standards like PVSyst, lack of effective investments in technology (which is why we witness abundant use of manual tools like excel, stats-based tools etc.) and skyrocketing costs of drone/thermal imaging every 6-12 months.

To put into perspective, a 1MW plant means approx. 70 strings (with 50 panels each) over 5-6 acres – often with inadequate instrumentation or calibration quality, coupled with unclear expectations of weather, micro-weather, uneven degradation scenarios across the plant, plus shading, soiling, and scores of short-term, fixable issues. When multiplied by 1000s for plant & fleet size – the sheer scale of issues indicates that plant managers are ultimately able to do only bare minimum fixes on time, mostly tending to outage or high-threshold alerts.

Additionally, standard industry benchmarks like PVSyst are just not accurate enough to validate plant performance against. Challenges with traditional contracted energy, resulting from inaccuracies and noise, blind reliance on pyranometer data (which could be very unreliable due to frequent calibration requirement, limited number of devices covering large plant areas etc.), no consideration of efficiency and behaviour of individual SCBs at different stages of aging – means that the benchmark may work at an aggregate level (for entire plant for a longer period of time), but is completely ineffective to make actionable, measurable improvements at SCB or weekly granularity.

To solve the challenges posed by unrealistic yield expectations and razor-thin O&M margins, Quadrical uses its proprietary Digital Twin technology to create highly accurate and granular performance benchmark, personalized for any plant and each underlying device, using historical and real-time data, for acceptable DC, Inverter, AC performance. This leads to Condition-based Monitoring – with deviations from the benchmark being tracked, characterized, and quantified for weather, structural degradations, shading, soiling, new-fixable issues – ultimately converted into actionable, revenue prioritized work orders. We have worked with customers to replace their Preventative Maintenance schedules with this Prescriptive approach to help improve their performance by 2-4% and O&M efficiency by 15% over a period of 6 months of continuous performance audit and issue remediation. Additionally, the Digital Twin level accuracy has led customers to have strong confidence in their OEM claims and due diligence processes.

What is your magic sauce?

Our secret sauce is the Digital Twin Technology at the core of Quadrical Asset Management Platform.

Using both historical and real-time data, our team builds a personalized, digital replica of a plant by creating and adding up digital twins of each node. This often means at least 500 mini Twins for a 50 MW plant. This bottom-up addition of yield expectations for each string in real time produces the most accurate estimate (the Digital Twin benchmark) of the plant’s generation capacity at any time, and under any field condition, with unprecedented granularity and precision.

Next, we leverage multiple special purpose Digital Twins for identifying and quantifying shading, soiling, insulation, long-term degradation, anomalies, due diligence etc. We then optimize customers’ asset performance continually with true Condition-based Maintenance Guidance as our Digital Twins learn and adapt to their plants’ inherent characteristics over time. This is how we help customers improve generation by 2 –4% and O&M efficiency by 15% over a period of 6 months and beyond.

What is the plan for the next 5 years? What do you want to achieve?

By 2050, renewables will represent 70% of grid-connected power generation. $30 Trillion of investments will be needed to achieve a more stable grid. Renewable generators need new competencies to manage variability, distributed assets and to become better at predicting and balancing load. Our vision is to be the Data, AI, IoT, Image platform for all renewables, beginning with solar. We are building Digital Twin based products for Storage as well, to ensure better efficiency of Solar + Storage. As storage capacity grows worldwide, we’d like to ensure that we’re the company major global IPP’s look to for optimizing their solar and storage capacity.

What is the biggest challenge you’ve faced so far?

One of the biggest challenges would be customer acquisition in the new geographies we plan to enter. We are solving it by building specialized products targeted to addressing high-impact issues that our customers face.

How can people get involved?

If you are a solar asset owner/financer/O&M company/solar researcher/AI specialist/renewables enthusiast, feel free to reach us at sales@quadrical.ai and visit our website- www.quadrical.ai for more information. Our team is always happy to chat/have a virtual coffee anytime.

Quadrical AI LinkedIn – https://www.linkedin.com/company/quadrical

LinkedIn – Mr. Sharat Singh (CEO) – https://www.linkedin.com/in/sharatsingh/