Dynamic Load Management, But with Extra Sauce - Artificial Intelligence (AI)

Regular DLM has its limitations, which revolve around the lack or shortage of data. Without real-time data, DLM is rigid. It can only work within the confines of its automated boundaries. For instance, while DLM can evenly distribute energy among EV loads without exceeding circuit limits, it cannot auto-adjust this action using arbitrary grid changes, low tariff signals, or EV driver behaviours.

The intersection of AI and DLM fixes this constraint with ease. So what does this look like?

• Load limit elasticity: The AI algorithm makes prescriptive decisions using the grid's response to demand. During peak periods, for instance, the AI algorithm takes data from the grid and enables the DLM technology to lower the load limit so that you can save cost and reduce strain on the grid. Conversely, the DLM technology raises the load limit when there’s low demand, allowing EVs to charge faster.
• Auto-distribution of energy based on chargers' capacities: EV chargers come with different capacities. The more powerful chargers sit on the top of the charging food chain. That is, they require more electrical power than others. The AI algorithm considers this data when coordinating the power flow among your chargers.
• Energy distribution using EV drivers' data: The typical DLM software distributes electricity evenly to EVs without paying attention to drivers' distinct details such as subscription plans, booking punctuality, etc. The AI algorithm learns these data, scales them on a priority list, and enables distribution using this list. For instance, a driver on a VIP tariff plan will take precedence over another driver on a guest tariff plan. Furthermore, through the data collated by the AI algorithm, priority is also given to the earliest charge bookers.
• Distribution based on vehicles' charging levels: Sometimes, an EV will continue to charge even with its State of Charge (SoC) over 80%, while other EVs plugged into the same circuit are yet to begin charging or are allocated less electricity. This is detrimental to both your finances and circuit loading. AI-enabled DLM carefully monitors each vehicle's SoC and distributes power accordingly, also maintaining the battery longevity by keeping the SOC in the “sweet spot”.
• EV profile-based distribution: EVs generally have differing charging behaviours. The AI algorithm runs through each EV's profile, checks its history, and runs a descriptive analysis to enable electricity allocation and speed based on each EV's profile.
• Load distribution using predictive analytics: One of the functions of an AI-enabled DLM is to forecast using past trends. Imagine, as a charge point operator, knowing the circuit limit you could need at a particular time in the future and then preparing for it. This is much better than operating a charging station without any foresight.

In summary, the benefits are enormous when your DLM software is enhanced with an AI algorithm that considers other data points mentioned above. Hive Power’s FLEXO is a cloud-to-cloud AI Vehicle-to-Grid (V2G) solution for EV Fleet Managers, OEMs, and Charge Point Operators that optimise electric vehicles' charging cycles to increase ROI and provide essential grid services. With Hive Power FLEXO controlling your DLM, you can

• Reduce the cost of your station’s energy bills
• Avoid unnecessary electrical system upgrade
• Improve your customers charging experience
• Expand your station’s charging infrastructure without worries.

DLM is only the beginning of the benefits of true smart charging. With the right solution partner, you can start extracting more business value from aggregating your EVs for ancillary services, optimising for renewable availability, and even V2X. The majority of automakers and charging providers are implementing this technology. Find out how you can save and earn by booking a call now.?