Frequently Asked Questions

A forensic diagnostic report covering: true State-of-Health (SoH) calibrated against BMS firmware drift, lithium plating risk window (charge rate vs. temperature envelope), yield gap attribution (what is underperforming and why), forensic fault replay (reconstructing the timeline of any fault events), and a prioritised action list. Delivered as a structured PDF with engineering walkthrough included.

S$4,800 fixed fee, per site, per audit cycle. One site = one physical asset location (e.g., a single BESS system or a single rooftop solar installation). There are no variable pricing tiers — the S$4,800 covers the full diagnostic scope regardless of system size. Enterprise volume pricing is available for portfolio operators with 5+ sites.

Five business days from clean data receipt to report delivery. The clock starts once we confirm your submitted telemetry is complete and parseable. We will flag any data gaps on Day 1 so you can resolve them quickly.

Our physics-informed pipeline detects: sub-second clipping events the BMS does not log, lithium plating induced by unsuitable charge profiles, capacity fade from calendar ageing vs. cycle ageing (separated), thermal runaway precursor signatures, and yield shortfalls from tracker misalignment, soiling, or inverter clipping on the solar side. We process the signal layer — not just the BMS summary fields.

Yes. Verify covers both BESS and Solar PV (utility-scale and rooftop). For Solar PV, the audit includes irradiance-normalised yield gap analysis, module-level anomaly detection, and string imbalance attribution. For co-located BESS + Solar systems, we audit the interaction between the two assets in the same report.

A clean Verify report is a valuable deliverable in itself — it provides an independent, physics-grounded certificate of asset health that you can table to investors, insurers, or lenders. We deliver the report regardless of outcome, with full methodology disclosure.

Verify is a one-time forensic diagnostic — a snapshot. Horizon is a continuous edge-deployed safety and dispatch intelligence engine that runs on your site hardware. Horizon monitors your BESS in real-time using the same physics-informed signal models, flags degradation trajectories before they become faults, and exposes a dispatch API so your control system can act on SoH-aware cycling recommendations. Verify is the diagnosis. Horizon is the ongoing treatment.

Yes, and this is intentional. Verify produces the calibrated asset baseline — true SoH, degradation mode fingerprint, and fault history — that Horizon uses to initialise its real-time models. Deploying Horizon without a Verify baseline means the real-time engine starts from OEM-default assumptions rather than your asset's actual condition. The Verify → Horizon sequence is the recommended path.

Edge-first. Horizon runs on a dedicated compute node at your site (your hardware or Oxaide-supplied). All raw telemetry is processed locally — your signal data never leaves the site perimeter. Horizon exposes a local dispatch API that your BMS, SCADA, or EMS can poll. Optional encrypted summary telemetry (aggregated, not raw) can be pushed to a cloud dashboard if you want remote visibility, but this is opt-in.

A REST API with four primary endpoints: GET /soh (real-time SoH estimate with confidence interval), GET /dispatch (current recommended charge/discharge rate ceiling), GET /alerts (active degradation flags and fault precursors), and GET /forecast (remaining useful life projection at current cycling profile). Your EMS polls these endpoints to make SoH-aware dispatch decisions.

Yes. Horizon ingests from standard industrial protocols: Modbus TCP/RTU, DNP3, IEC 61850, OPC-UA, and direct historian pull (OSIsoft PI, Ignition, InfluxDB). Most BMS manufacturers output to at least one of these. For proprietary BMS formats, we provide a parser configuration layer — contact us with your BMS vendor and we will confirm compatibility before you commit to a pilot.

Four inputs: (1) BMS raw logs — cell-level voltage, temperature, current at the native sample rate (minimum 1-second resolution, 6–12 months of history preferred); (2) SCADA or historian export — site-level power, SOC, and fault event log; (3) OEM datasheet — capacity, chemistry, and thermal specifications; (4) Irradiance or GHI data if Solar PV is in scope. Secure upload instructions are provided at checkout. We accept CSV, TDMS, Parquet, and most historian export formats.

A 12-month BMS log for a 1 MWh BESS system at 1-second resolution is typically 2–8 GB compressed. For Solar PV yield gap analysis, the SCADA export is usually under 500 MB. We provide a secure upload link — no email attachments. Large datasets can also be transferred via SFTP or temporary object storage with pre-signed URLs.

Yes, within limits. Our pipeline is designed for real-world industrial data which is rarely clean. We handle timestamping gaps, sensor dropouts, firmware-version discontinuities in BMS logs, and resampling from irregular intervals. We will flag any gaps that materially affect specific analysis modules in the report. If a critical data gap prevents a module (e.g., no irradiance data for yield gap analysis), we will tell you before completing the audit, not after.

Your submitted telemetry is used exclusively to produce your forensic report — nothing else. We do not use your data to train or calibrate shared models. Data is encrypted in transit (TLS 1.3) and at rest (AES-256). All Verify processing occurs within Singapore-hosted infrastructure. Upon report delivery, you may request deletion of your submitted data at any time. We retain data for 90 days post-delivery by default, then purge.

No. Horizon is edge-first by design. The physics-informed inference engine runs locally on your site hardware. Raw telemetry — cell voltage, temperature, current — never leaves your network. If you opt into the cloud dashboard, only aggregated SoH summary metrics are encrypted and transmitted. You can disable this entirely and operate Horizon in fully air-gapped mode.

Yes. Rocketship Pte. Ltd. (trading as Oxaide) is registered with Singapore's PDPC and operates under PDPA. We are GDPR compliant for European clients. For industrial telemetry data specifically, it is typically classified as operational / non-personal data — however, we apply the same security and data handling standards regardless. A Data Processing Addendum (DPA) is available for clients who require it contractually.

Three steps: (1) Submit an enquiry via oxaide.com/verify — takes 2 minutes; (2) We confirm data requirements and send a secure upload link within 1 business day; (3) Submit your data and we begin the 5-day audit clock. Invoice is issued at step 2, payable via bank transfer or card. No upfront scoping calls required for standard Verify engagements.

Horizon pilots are scoped individually — system size, BMS protocol, and deployment environment all affect the setup. The recommended path is: run a Verify audit first (establishes the asset baseline), then book a 30-minute technical call with our engineering team to confirm Horizon compatibility with your BMS and SCADA stack. Horizon pilots are S$35,000+, priced per site.

Yes. For portfolio operators and O&M firms, we offer quarterly Verify re-audits at a bulk rate, and annual diagnostic subscription packages that combine scheduled audits with priority access to our engineering team. Contact us at sales@oxaide.com with your portfolio size for a custom quote.

Yes — this is a growing use case. Verify reports are increasingly used as independent technical evidence for: insurance underwriting (battery health at policy renewal), M&A / asset acquisition due diligence (validating OEM-reported SoH), and project finance (demonstrating performance to lenders). We can structure the report deliverable to meet specific mandate requirements — contact us to discuss.