Question 1

What is a Virtual Power Plant (VPP)?

Accepted Answer

A Virtual Power Plant (VPP) is a network of distributed energy resources-such as batteries, solar panels, EV chargers, and smart thermostats-that are aggregated and controlled as a single entity to provide grid services. Unlike traditional power plants that generate electricity in one location, VPPs coordinate thousands of small, distributed assets to act like a conventional power plant: dispatchable, reliable, and capable of responding to grid signals in real time. VPPs participate in capacity markets, demand response programs, and wholesale energy markets, turning distributed assets into grid-grade infrastructure. Learn more about [how Molecule Systems enables grid-grade VPP performance](/platform) and explore the [VPP Maturity Model](/blog/what-is-vpp-maturity-and-why-it-matters).

Question 2

What does Molecule Systems do?

Accepted Answer

Molecule Systems is the Execution Layer for Virtual Power Plants, providing edge-based infrastructure that enables distributed energy resources to operate as grid-grade assets. We sit between OEM device software and energy markets, delivering deterministic, real-time control at the site level. Our MOS 350 edge operating system runs on gateways deployed at each site, ensuring batteries, inverters, and other DERs respond reliably to dispatch signals-even during internet outages. Molecule doesn't replace DERMS platforms or optimization algorithms; we execute their commands with millisecond-level precision. Our customers include battery OEMs like EG4 Electronics, fleet operators, and utilities who need VPPs that perform like power plants, not prototypes. Explore our [platform architecture](/platform) or learn about [DividendVPP for OEMs](/dividend-vpp).

Question 3

What is the Execution Layer?

Accepted Answer

The Execution Layer is infrastructure that sits between OEM device software and energy markets, providing deterministic, real-time control of distributed energy resources at the site level. It's the difference between knowing *what* to do (intelligence, forecasting, optimization) and actually *doing it* reliably (execution). OEM firmware is designed for energy management-charge/discharge scheduling, backup power transitions, time-of-use arbitrage. But grid services require millisecond-level dispatch precision, real-time telemetry, and autonomous operation during internet outages. The Execution Layer provides this. Molecule Systems' AERA (Adaptive Energy Response Architecture) architecture decouples control intelligence from execution, allowing optimization logic to come from anywhere-OEM cloud, third-party DERMS, utility dispatch systems-while ensuring local, reliable, real-time execution at every site. Read the full technical deep-dive: [Why VPPs Need More Than OEM Software](/blog/the-execution-layer-why-vpps-need-more-than-oem-software).

Question 4

What is AERA (Adaptive Energy Response Architecture)?

Accepted Answer

AERA (Adaptive Energy Response Architecture) is Molecule Systems' execution architecture that decouples control intelligence from deterministic asset operation. Under AERA, optimization logic-forecasting, dispatch algorithms, market bidding strategies-can come from anywhere: OEM cloud platforms, third-party aggregators, utility DERMS systems, or custom algorithms. But execution happens locally, at the edge, via MOS350 (Molecule's internal edge OS) running at each site. This separation ensures that even if cloud connectivity fails, dispatch logic continues executing. It also means optimization algorithms can be hot-swapped without downtime, device firmware updates, or site visits. AERA's principle: "Intelligence can come from anywhere. Execution has to be reliable." This architectural approach allows OEMs to keep building hardware, utilities to keep running optimization models, and Molecule to handle the deterministic execution that makes VPPs perform like power plants. Learn more about [AERA and edge-first architecture](/platform).

Question 5

How is Molecule different from a DERMS platform?

Accepted Answer

Molecule Systems provides execution infrastructure (the *how*), while DERMS platforms provide optimization intelligence (the *what*). DERMS (Distributed Energy Resource Management Systems) forecast load, optimize dispatch schedules, bid into markets, and coordinate fleets at the portfolio level. They're the brain. Molecule's AERA architecture is the nervous system-it executes those commands reliably, in real time, at the site level. AERA is designed to work *with* any DERMS: utility-owned systems like Autogrid or Stem, third-party aggregator platforms, or custom optimization algorithms. We don't replace DERMS-we make DERMS work better by ensuring their dispatch signals are executed deterministically, with sub-second telemetry verification and autonomous edge-based control. Think of it this way: DERMS tells the VPP what to do. Molecule ensures it actually does it. Explore [AERA's fleet coordination capabilities](/platform) and see how Molecule complements DERMS platforms.

Question 6

What is the VPP Maturity Model?

Accepted Answer

The VPP Maturity Model, introduced by EnergyHub, categorizes VPP programs across five levels based on operational capability, from basic demand response (Level 1) to grid-grade, multi-program optimization with real-time telemetry and deterministic control (Level 5). Most VPPs today operate at Level 2 or 3, suffering from partial fleet response, 24-hour data delays, and ~50% underperformance vs. modeled capability. The gap between Level 3 and Level 5 isn't a forecasting problem or a market design problem-it's an execution infrastructure problem. VPPs at Level 4-5 have edge-based control, sub-second telemetry, and deterministic dispatch that works even during internet outages. Molecule Systems builds the execution layer that enables VPPs to progress from Level 3 (basic aggregation) to Level 5 (grid-grade infrastructure). The maturity gap creates a double economic penalty: underperformance vs. capacity commitment *and* grid operator discounting for future bids. Closing that gap is what Molecule does. Read the full analysis: [What is VPP Maturity and Why It Matters](/blog/what-is-vpp-maturity-and-why-it-matters).

Question 7

Can Molecule integrate with my existing battery inverters?

Accepted Answer

Yes. Molecule Systems is protocol-agnostic and integrates with 1000+ device makes and models via Modbus RTU/TCP, OCPP 1.6/2.0, JSON APIs, CAN bus, and proprietary OEM protocols. We've integrated with EG4, Tesla Powerwall, Enphase, SolarEdge, LG, BYD, RCT Power, and many others. Integration typically takes less than 30 minutes for standard protocols using drag-and-drop configuration in the Molecule dashboard-no custom code required. For devices with proprietary protocols or unique communication interfaces, we provide Lua scripting via our Optimization SDK, allowing custom integration logic to run at the edge. Our MOS 350 gateways normalize protocol diversity, so whether you're controlling a Modbus-connected battery or an API-based inverter, your optimization logic sees a unified abstraction. No firmware changes to your devices. No dependency on OEM cloud platforms. Just deterministic, real-time control via the execution layer. Explore our [developer portal](/developer) for integration guides and protocol documentation.

Question 8

What devices and protocols does Molecule support?

Accepted Answer

Molecule Systems supports 1000+ device makes and models across multiple protocols: **Modbus RTU and TCP** (most battery inverters and industrial controllers), **OCPP 1.6 and 2.0** (EV chargers), **JSON/REST APIs** (cloud-connected devices like Tesla Powerwall), **CAN bus** (automotive-grade battery systems), and **proprietary OEM protocols** (via Lua scripting in our Optimization SDK). Device brands include EG4, Tesla, Enphase, SolarEdge, LG Chem, BYD, RCT Power, Schneider Electric, and many others. Molecule's edge execution runtime normalizes protocol diversity-your optimization logic doesn't care whether it's controlling a Modbus battery, an OCPP charger, or an API-based inverter. The gateway handles translation. We also support whole-building device orchestration: batteries + solar + EV chargers + HVAC behind a single meter, all coordinated as a single resource. If your device speaks a standard protocol or has an accessible API, we can integrate it. Visit our [SDK documentation](/platform/sdk) or [contact us](/contact) for device-specific integration details.

Question 9

What is MOS 350 and how does it work?

Accepted Answer

MOS 350 is Molecule Systems' edge operating system that runs on site-level gateways, providing deterministic, real-time control of distributed energy resources without cloud dependency. Think of it as aerospace-grade firmware for VPPs. MOS 350 handles three critical functions: (1) **Protocol translation** - it speaks Modbus, OCPP, API, CAN, and other protocols, normalizing communication with batteries, inverters, and chargers. (2) **Local execution** - dispatch logic runs at the edge, not in the cloud, enabling millisecond-level response times and autonomous operation during internet outages. (3) **Hot-swappable optimization** - algorithms can be uploaded, updated, or replaced via the cloud without downtime, device reboots, or site visits. MOS 350 is self-healing: if the gateway loses connectivity, it continues executing the last-known dispatch logic. When connectivity returns, telemetry syncs and cloud-based optimization resumes. It's built for grid-grade reliability: 99.9%+ uptime, sub-second telemetry, and deterministic control precision. Learn more about [MOS 350 and edge-first architecture](/platform/mos-350).

Question 10

How does edge-first architecture benefit VPP performance?

Accepted Answer

Edge-first architecture places control logic at the site level, eliminating cloud latency, reducing internet dependency, and enabling millisecond-level dispatch precision required for ancillary services like frequency regulation and fast frequency response. Cloud-based dispatch introduces hundreds of milliseconds to seconds of latency-acceptable for demand response, unacceptable for grid services that require sub-second response times. Edge-first also means autonomous operation: if internet connectivity fails during a grid event (when it's most likely to fail), the gateway continues executing dispatch logic locally. No dependency on cloud servers, API rate limits, or network reliability. Additionally, edge-first reduces data backhaul costs and privacy concerns-sensitive site data (customer usage patterns, backup power preferences) stays local unless explicitly shared. For VPP operators, edge-first means higher fleet response rates (90%+ vs. 50-60% for cloud-only systems), verified performance (sub-second telemetry for pay-for-performance settlement), and eligibility for high-value ancillary service markets. Read more: [The Execution Layer: Why VPPs Need More Than OEM Software](/blog/the-execution-layer-why-vpps-need-more-than-oem-software).

Question 11

What is hot-swappable optimization logic?

Accepted Answer

Hot-swappable optimization logic allows VPP algorithms to be updated or replaced without downtime, device firmware changes, or site visits. In traditional systems, changing dispatch logic requires OEM firmware updates (slow, risky, requires device reboots) or re-integrating with new cloud platforms (requires reconfiguration, recertification, customer disruption). With Molecule's AERA architecture, optimization logic is injected into MOS 350 gateways via the cloud and executed locally. If you want to switch from a simple time-of-use strategy to a sophisticated multi-program optimizer, you upload new logic-it takes effect immediately, no downtime. If a market program changes (new tariff structure, new ancillary service offering), you adapt the logic without touching device firmware. This is critical for VPPs that need to evolve as market conditions change. It's also why OEMs choose Molecule: they don't need to rebuild firmware every time a new grid program launches. They inject new logic into the execution layer, and it just works. Learn about [DividendVPP's multi-program optimization](/dividend-vpp) for a real-world example.

Question 12

What is real-time telemetry and why does it matter?

Accepted Answer

Real-time telemetry is sub-second data reporting from distributed energy resources to VPP operators, grid operators, and settlement systems. It matters because pay-for-performance programs (like California's ELRP or CAISO's regulation energy markets) require verified delivery-you don't get paid for promised capacity, you get paid for *measured* capacity delivered during dispatch events. Most VPPs today rely on utility meter data, which is delayed by 24 hours. If a dispatch event happens at 5 PM today, you don't know how your fleet performed until tomorrow. That's too late to adjust, too late to troubleshoot, and often too late to claim payment (settlement windows close). Real-time telemetry solves this. Molecule's MOS 350 gateways report battery state-of-charge, power output, availability status, and dispatch compliance every second (or faster, if needed). VPP operators see fleet performance live. Grid operators verify delivery in real time. Settlement systems get the data they need for pay-for-performance verification. Real-time telemetry is the difference between a VPP that *should* perform and a VPP that *proves* it performed. Explore [EMS for real-time site control](/platform/mosems).

Question 13

How does DividendVPP generate revenue for battery owners?

Accepted Answer

DividendVPP generates revenue through multi-program value stacking: time-of-use arbitrage, demand response, resource adequacy capacity payments, wholesale market participation, clean peak optimization, and demand charge management (for commercial systems). Instead of enrolling batteries in a single program (like basic TOU arbitrage, which earns $300-$600/year), DividendVPP optimizes across six revenue streams simultaneously, delivering $1,600-$2,400/year for residential systems - up to a 2-4x improvement compared to typical OEM-led API aggregation with limited site control (VPP 1.0). Results depend on program mix, asset configuration, and deployment depth. Here's how it works: the system charges batteries during off-peak hours (cheap electricity), holds charge to avoid demand peaks (if applicable), discharges during on-peak TOU windows *and* high grid carbon intensity periods (maximizing TOU revenue + clean peak incentives), reserves capacity for demand response events (ELRP pays $2/kWh), and bids into wholesale markets during price spikes. All of this is coordinated in real time by edge-based execution (Molecule) + forecasting intelligence (Lightsmith Energy). Battery owners see monthly payouts. OEMs see recurring revenue per installed system. Learn more: [DividendVPP: Maximizing Revenue for Battery Storage](/blog/dividendvpp-maximizing-revenue-for-battery-storage).

Question 14

What is the DividendVPP value stack?

Accepted Answer

The DividendVPP value stack is the combination of six revenue streams optimized simultaneously for maximum earnings per battery system: **(1) Time-of-Use Arbitrage** - charge cheap, discharge expensive ($300-$600/year for residential). **(2) Demand Response** - participate in utility DR programs like California's ELRP ($2/kWh for verified load reduction, $200-$800/year). **(3) Resource Adequacy** - provide firm capacity during peak availability windows (monthly capacity payments, $400-$1,200/year). **(4) Wholesale Market Participation** - bid into real-time energy markets to capture price spikes ($300-$800/year, highly variable). **(5) Clean Peak Optimization** - align discharge with high grid carbon intensity to qualify for clean energy incentives ($100-$400/year). **(6) Demand Charge Management** - reduce peak power draw for commercial/industrial customers ($500-$5,000/year depending on rate structure). Total potential: $1,600-$4,000/year per residential battery (10-15 kWh). Standard aggregator models capture $400-$800/year because they only optimize one or two programs. DividendVPP captures the full stack. Explore the [EG4 partnership case study](/blog/eg4-electronics-partners-with-molecule-systems) for real-world results.

Question 15

Who developed DividendVPP?

Accepted Answer

DividendVPP was co-developed by Molecule Systems and Lightsmith Energy, combining execution infrastructure (Molecule) with forecasting and optimization intelligence (Lightsmith). Molecule provides the edge-based control layer-MOS 350 gateways, real-time telemetry, deterministic dispatch, protocol-agnostic integration-ensuring batteries respond reliably to dispatch signals. Lightsmith provides the value-stack algorithms-forecasting TOU windows, predicting demand response event likelihood, bidding into wholesale markets, optimizing across multiple programs simultaneously. Together, they deliver a fully-integrated, white-labeled VPP stack for battery OEMs. OEMs get recurring revenue per installed system, customers get up to 2-4x more earnings per battery. Results depend on program mix, asset configuration, and deployment depth. The partnership structure allows each company to focus on what they do best: Molecule handles execution, Lightsmith handles intelligence. For OEMs, it's plug-and-play: no firmware redesign, no new development, no disruption to installer networks. Just a white-labeled VPP that works. Learn more about [DividendVPP for OEMs](/dividend-vpp/oems) and [our partnership with Lightsmith](/about/partnerships).

Question 16

How much more revenue can DividendVPP generate?

Accepted Answer

DividendVPP delivers up to 2-4x more revenue per site compared to typical OEM-led API aggregation with limited site control and static dispatch (VPP 1.0). This uplift is driven by execution quality, asset staging, multi-DER coordination, and integrated forecasting and optimization provided by Lightsmith Energy. Results depend on program mix, asset configuration, and deployment depth. Residential battery systems (10-15 kWh) earn $1,600-$2,400/year with DividendVPP vs. $400-$800/year with single-program aggregators. Commercial systems (50-100 kWh) can earn $5,000-$15,000/year depending on demand charge rates and program participation. The difference comes from multi-program value stacking (six revenue streams optimized simultaneously) and execution infrastructure that ensures high fleet response rates (90%+ vs. 50-60% for cloud-only systems). Real-world example: EG4 Electronics partnered with Molecule Systems to deploy DividendVPP across 15,000-30,000 residential battery systems. At $1,600/year/system (conservative midpoint), that's $24M-$48M in total grid services revenue for EG4 customers over the program's life. For OEMs, DividendVPP creates recurring annual revenue per installed system-turning one-time hardware sales into subscription-like business models. Performance matters: pay-for-performance programs only pay for *verified* delivery, so edge-based execution + real-time telemetry = higher earnings. Explore the [EG4 case study](/blog/eg4-electronics-partners-with-molecule-systems) for details.

Question 17

What is the EG4 Electronics partnership?

Accepted Answer

EG4 Electronics, one of the fastest-growing residential battery manufacturers in North America, partnered with Molecule Systems to deploy DividendVPP across 15,000-30,000 battery systems over 18 months. This represents 75-150 MW of dispatchable capacity-equivalent to a small peaker plant, but distributed across thousands of residential sites for grid resilience. EG4 chose DividendVPP to access grid services revenue for their customers (earning $1,600-$2,400/year per system) and create recurring revenue for EG4 as an OEM. The partnership is white-labeled: EG4's brand, EG4's installer network, Molecule's execution infrastructure running behind the scenes. EG4's CEO James Showalter said: "Our mission has always been to make advanced energy systems accessible, reliable, and affordable. Partnering with Molecule gives us the software foundation to match the scale of our hardware ambitions." EG4's Grid Services Manager Clinton O'Neill emphasized execution over promises: "Trust is built by execution. Systems that perform reliably become utility-grade infrastructure, not ongoing pilots." Read the full announcement: [EG4 Electronics Partners with Molecule Systems](/blog/eg4-electronics-partners-with-molecule-systems).

Question 18

Who is Lightsmith Energy?

Accepted Answer

Lightsmith Energy is a grid services company specializing in forecasting and value-stack optimization for Virtual Power Plants. They co-developed DividendVPP with Molecule Systems, providing the intelligence layer-algorithms that forecast time-of-use windows, predict demand response event likelihood, optimize wholesale market bidding, and coordinate multi-program participation for maximum revenue. Lightsmith's team has deep expertise in ISO/utility market programs (CAISO, PJM, ERCOT), tariff structures, and pay-for-performance settlement systems. While Molecule handles execution infrastructure (edge control, real-time telemetry, deterministic dispatch), Lightsmith handles optimization intelligence (what to do, when to do it, how to maximize value). The partnership structure allows each company to focus on its core competency, delivering a fully-integrated VPP stack for OEMs without requiring them to build forecasting teams or execution infrastructure in-house. For more on the partnership, visit our [Partnerships page](/about/partnerships) or learn about [DividendVPP's architecture](/dividend-vpp).

Question 19

How do I become a Molecule partner?

Accepted Answer

Molecule Systems partners with battery OEMs, fleet operators, project developers, utilities, and Community Choice Aggregators (CCAs). Partner types include: **(1) OEM Partnerships** - white-labeled DividendVPP for residential/commercial battery manufacturers. You sell hardware, we provide VPP software, your customers earn grid services revenue, you earn recurring revenue per system. **(2) Fleet Operator Partnerships** - edge-based execution infrastructure for third-party-owned battery fleets (build-to-rent, C&I storage, community solar + storage). **(3) Developer Partnerships** - AERA and EMS integration for project developers deploying multi-site battery portfolios. **(4) Utility/CCA Partnerships** - DERMS-compatible execution layer for utility-owned or aggregated DER fleets. To explore partnership opportunities, [contact our team](/contact) with details about your fleet size, device types, target markets, and business model. We'll schedule a technical discovery call to assess integration feasibility, discuss revenue-sharing structures, and outline onboarding timelines. Typical onboarding takes 30-90 days depending on device protocols and program complexity.

Question 20

How do I install and commission a Molecule gateway?

Accepted Answer

Installing and commissioning a Molecule gateway typically takes less than 30 minutes. Here's the process: **(1) Mount the gateway** - install the gateway enclosure near the battery or inverter (indoor or outdoor-rated enclosures available). **(2) Connect to devices** - wire the gateway to your battery/inverter via Modbus RTU (RS-485), Modbus TCP (Ethernet), or connect via API/OCPP (for cloud-connected devices). **(3) Provision via cloud** - log into the Molecule dashboard, add the gateway by serial number, assign it to a site, and configure device protocols using drag-and-drop. No custom code required for standard protocols. **(4) Run commissioning test** - the dashboard sends a test dispatch signal (e.g., "charge at 2 kW for 5 minutes"), verifies the device responds correctly, and checks telemetry accuracy. Once the test passes, the gateway is live and ready for dispatch. For devices with proprietary protocols, you can upload custom Lua scripts via our Optimization SDK. Installation guides, wiring diagrams, and protocol documentation are available in our [developer portal](/developer). Need hands-on support? [Contact us](/contact) to schedule installer training or on-site commissioning assistance.

Question 21

How long does integration take?

Accepted Answer

Integration typically takes less than 30 minutes for standard protocols like Modbus RTU/TCP or OCPP. Drag-and-drop configuration via the Molecule dashboard requires no custom code-you select the device type (battery, inverter, charger), specify the protocol (Modbus, API, OCPP), map data points (state-of-charge, power setpoint, availability), and run a test dispatch. For devices with proprietary protocols or unique communication interfaces, integration may take 1-3 days using Lua scripting in our Optimization SDK. Our SDK provides pre-built libraries for common operations (read Modbus registers, parse JSON APIs, send OCPP commands), so even custom integrations don't require starting from scratch. For OEM partnerships (like EG4), we often pre-integrate your device models into our device library, so installers can commission gateways without any custom configuration. Timeline example: EG4 partnership onboarding took ~60 days from contract signing to first live system, including device integration, installer training, and program enrollment. Explore our [technical documentation](/developer) or [schedule a demo](/contact) to see integration in action.

Question 22

Is there a developer portal?

Accepted Answer

Yes. Molecule Systems provides a comprehensive [developer portal](/developer) with technical documentation, integration guides, API references, and sandbox environments for testing. Key resources include: **(1) Lua Scripting Guides** - write custom logic for device integration, dispatch algorithms, or telemetry processing using our Optimization SDK. **(2) Modbus Integration Docs** - register maps, function codes, RTU vs. TCP configuration, and wiring diagrams for common inverter brands. **(3) API Reference** - RESTful API for fleet management, dispatch control, telemetry retrieval, and gateway provisioning (OAuth 2.0 authenticated). **(4) OCPP Documentation** - OCPP 1.6 and 2.0 message specs for EV charger integration. **(5) Sandbox Environments** - test dispatch logic, simulate grid events, and validate telemetry flows without live hardware. **(6) Troubleshooting Guides** - common integration issues, error codes, and debugging tips. The developer portal is public (no login required for docs), with API keys issued for authenticated sandbox access. Need custom integration support? Our team offers white-glove developer onboarding for OEM and fleet operator partners. Visit [developer.moleculesystems.com](/developer) to get started.

Frequently Asked Questions

General

Technical

DividendVPP

Partnerships

Getting Started

Explore Related Pages

Still Have Questions?