• Experienced LG Energy Solution Vertech data scientist highlights the growing importance of data analysis in guaranteeing system safety and optimizing performance in the BESS market.
• LG Energy Solution Vertech pioneers data-driven strategies for sustainable energy solutions, ultimately optimizing energy systems for a sustainable future.

LG Energy Solution Vertech, the US energy storage division of LG Energy Solution, provides integrated and reliable Alternating Current (AC) energy storage systems and services that enable customers to optimize revenues while shaping a sustainable future.

Kiran Kumar, Director of Data Science and Data Analytics at LG Energy Solution Vertech, was recently featured on the Battery Intelligence Podcast hosted by Bradley Newton from Lawrence Harvey, an international recruitment business specializing in tech recruitment. Kumar leveraged this platform to share his insight into the Battery Energy Storage System (BESS) landscape and discuss the increasing importance of data products.

LG Energy Solution Vertech plays a crucial role in integrating advanced lithium-ion batteries manufactured by its parent company into complete AC storage systems with inverters, auxiliaries, and system-level controls and supervisory systems. Given the unique nature of each project, there are numerous factors to consider during the design process. Integration of domain knowledge and data ensures battery safety by monitoring parameters such as temperature and capacity degradation. Through on-site operations, the early detection of anomalies and optimization of energy usage across thousands of battery racks can be achieved. Considering the extensive scale of installation units, applying sophisticated algorithms to orchestrate operations becomes paramount for maximizing efficiency.

According to Kumar, the data science team at LG Energy Solution Vertech is tasked with analyzing a wide range of data from the storage system, which is collected through various means, including on-site battery pack installations. This collaborative effort involves not just data scientists but also professionals from various disciplines. With the market’s rapid growth and the volume of data increasing exponentially, the team is constantly working on new ways to add value and improve system performance to keep pace with this dynamic landscape.

Collecting data from nodes and ensuring its consistency and reliability is extremely challenging by itself and having to synchronize this data too adds an extra layer of complexity. Once data is cleaned and available, data scientists use the data to analyze the behavior of the system from various perspectives, develop an algorithm to identify the issue, and use the information to optimize the system performance. LG Energy Solution Vertech’s data science team utilizes advanced techniques such as Convolutional Neural Network (CNN)¹⁾ for complex pattern recognition, Logistic Regression²⁾ for binary classification problems, Monte Carlo Data Analysis³⁾ for probabilistic simulations, and Large Language Model (LLM)⁴⁾ for analyzing data with correlated onsite actions such as parts replacement, service notes, and more.

In the future, LG Energy Solution Vertech’s data science team plans to adopt Machine Learning Operations (MLOps⁵) more actively to automate specific data products training, validation and testing, building deployment packages, and evaluating its field performance on a periodic basis. The overall goal here is to leverage the MLOps infrastructure to improve the productivity of the team.

The Data Science team has been growing rapidly, expanding from 2 to 13 members in just one and a half years. When recruiting data scientists or software engineers, candidates with diverse backgrounds are encouraged to apply. While an energy systems background is a plus, it’s not a must as LG Energy Solution Vertech believes candidates can acquire domain-specific knowledge after they join. LG Energy Solution Vertech values data science or software engineering skills coupled with a strong academic and professional background, especially any practical experience applying data science/mathematical principles to real-world challenges.

In conclusion, Kumar underscored the significant growth and future prospects within the energy sector, especially in the realm of clean energy, highlighting the plethora of opportunities available at LG Energy Solution Vertech for individuals with problem-solving abilities. For those interested in delving deeper into LG Energy Solution Vertech and data science and analytics within this dynamic field, further insights can be found on LinkedIn.

1) A regularized feed-forward neural network that self-learns feature engineering via filters (or kernel) optimization.
2) Statistical model that models log-odds of an event as a linear combination of one or more independent variables.
3) A broad class of computational algorithms that rely on repeated random sampling to obtain numerical results.
4) A language model that achieves general-purpose language generation and other natural language processing tasks such as classification.
5) A core function of Machine Learning engineering that streamlines the process of taking machine learning models to production while maintaining and monitoring them.