BESS High‑Level Requirements & Reference System Block Diagram

Reference document: ISGF Report on Energy Storage System (ESS) Roadmap for India (2019–2032)

1) High‑Level Requirements (HLR) with Traceability

Scope: “BESS” includes battery, PCS/inverter, controls/communications, protection, monitoring/O&M.

ID	Requirement (High level “shall”)	Trace to report heading
BESS-HLR-01	The BESS shall support peak shaving by charging during low/off‑peak periods and discharging during peak periods, with peaker‑like attributes (fast start/low response time, rapid output variation, efficient part‑load operation).	Section 2.5.2 — Peak Shaving
BESS-HLR-02	The BESS shall be capable of providing ancillary services for grid reliability/security, including active power support (load following), reactive power support, and black start (as applicable).	Section 2.5.3 — Ancillary Services
BESS-HLR-03	The BESS shall support frequency response / frequency support consistent with India’s grid operation context discussed in the roadmap.	Section 2.5.3 — Ancillary Services
BESS-HLR-04	The BESS shall enable deferral of T&D upgrades by reducing peak loading on distribution assets (feeders/DTs) where upgrades are otherwise required only for limited peak hours.	Section 2.5.4 — Transmission and Distribution Grid Upgrade Deferral
BESS-HLR-05	The BESS solution shall support stacking multiple services/value streams (e.g., peak shaving + voltage/reactive support + frequency support + time shifting) to improve project feasibility.	Tariffs section (roadmap discussion; after consolidated roadmap)
BESS-HLR-06	Where PV output exceeds local load and drives voltage rise, the BESS shall mitigate voltage rise by absorbing surplus energy and/or providing suitable control support.	Section 3 — Assessment of MV/LV Stabilization & Optimization for 40 GW RTPV: Technical Issues and Challenges
BESS-HLR-07	The BESS shall help keep MV/LV voltages within acceptable limits to avoid over/under‑voltage conditions and reduce nuisance inverter tripping.	Section 4 — Load Flow Studies on MV/LV Lines with RTPV
BESS-HLR-08	The BESS/PCS shall provide reactive power capability (or coordinate with smart inverter functions) to address PF concerns and provide reactive power compensation.	Section 3 — Technical Issues and Challenges
BESS-HLR-09	The BESS shall support smoothing/mitigation of rapid PV output changes (e.g., cloud effects / ramp rates) to reduce voltage fluctuations in weak networks.	Section 3 — Technical Issues and Challenges
BESS-HLR-10	The BESS/PCS integration shall not degrade power quality and shall support mitigation of harmonic‑related PQ issues.	Section 3.4 — Power Quality (PQ) and Harmonics
BESS-HLR-11	The BESS solution shall support PQ conditioning approaches (customer‑side or utility‑side), including compatibility with smart inverters and/or active power filter (APF/SAPF) concepts where applied.	Section 3.4 — Power Quality (PQ) and Harmonics
BESS-HLR-12	The BESS solution shall be deployable as part of a portfolio to increase VRE/RTPV hosting capacity on MV/LV feeders while avoiding adverse impacts.	Section 1.3.2 — Hosting Capacity of VRE on MV/LV Feeders
BESS-HLR-13	BESS design and operation shall respect technical criteria that constrain hosting capacity/interconnection: thermal ratings, short‑circuit/fault level, voltage regulation, PQ (flicker/harmonics), islanding considerations, and protection coordination impacts.	Section 1.3.2 — Hosting Capacity of VRE on MV/LV Feeders
BESS-HLR-14	The PCS/inverter shall provide smart‑inverter‑class capabilities, including digital architecture, bidirectional communications, robust software infrastructure, and performance monitoring.	Section 3.5 — Comparison of Regular and Smart Inverters (Autonomous and SCADA Controlled)
BESS-HLR-15	The BESS shall support fast messaging and granular data sharing with owners/utilities/stakeholders to support monitoring and operations.	Section 3.5 — Comparison of Regular and Smart Inverters…
BESS-HLR-16	The BESS shall support remote diagnostics/maintenance and remote parameter upgrades to reduce O&M burden.	Section 3.5 — Comparison of Regular and Smart Inverters…
BESS-HLR-17	The BESS monitoring/control stack shall provide API capabilities (or equivalent integration interfaces) to enable enterprise tools and portfolio/fleet management integration.	Section 3.5 — Comparison of Regular and Smart Inverters…
BESS-HLR-18	The PCS/inverter shall support advanced grid functions (as applicable), such as ramp rate control, power curtailment, fault ride‑through, and voltage support through VARs.	Section 3.5 — Advantages of using Smart Micro Inverters (SMI)
BESS-HLR-19	The BESS planning approach shall support determination of optimal storage size and technology selection for RTPV/VRE integration use cases.	Section 5.1 — Description and Overview (Energy Storage India Tool / ESIT)
BESS-HLR-20	The BESS solution (or associated planning tool/process) shall use network load data and optimize storage capacity, supporting analysis at feeder, DT, and customer levels.	Section 5.1 — Description and Overview (and location‑level analyses in Section 6)
BESS-HLR-21	The BESS evaluation shall account for monetizable and non‑monetizable benefits, including reliability‑related avoided losses.	Section 5.1 — Description and Overview
BESS-HLR-22	The BESS planning/evaluation approach shall work with partial/incomplete load data and support flexible time‑interval inputs.	Section 5.1 — Description and Overview
BESS-HLR-23	The BESS shall provide operational statistics for lifecycle planning (e.g., cycling, energy throughput, average SoC, replacement planning outputs).	Section 5.2 — Techno‑Commercial Evaluation of ESS Projects
BESS-HLR-24	The BESS shall support dispatch logic addressing heavily loaded feeders/DTs, including local peak shaving and reverse power flow absorption, as part of T&D deferral and multi‑use operation.	Section 5.3 — Consideration of Multiple Use‑Cases

2) Reference System Block Diagram (with possible interfaces)

2.1 Integrated controls + power interface view

flowchart TB
    %% Utility Level
    subgraph UtilityLevel ["Utility / SLDC / DERMS"]
        UtilityNode["Utility / SLDC / DERMS<br/>[Utility control center / State Load Dispatch Centre / Distributed Energy Resource Management System]<br/>(dispatch, limits, events, telemetry)"]
    end

    %% Communication / OT Level
    Gateway["Secure OT Gateway/Firewall<br/>[OT edge security gateway + firewall]<br/>(router, VPN, RBAC, logs)"]
    UtilityNode -- "IEC 60870-5-104 / DNP3 / IEC 61850<br/>(VPN/TLS over WAN)" --> Gateway

    subgraph ControlLAN ["Ethernet (LAN, VLANs)<br/>[Local Area Network using Ethernet; VLANs for segmentation]"]
        direction TB
        SCADA["SCADA / RTU<br/>[Supervisory Control and Data Acquisition / Remote Terminal Unit]<br/>(alarms/events, protocol bridge)"]
        PlantCtrl["Plant Controller<br/>[BESS Plant Controller / Power Plant Controller (PPC)]<br/>(P/Q/V/f modes, ramp, SoC, FFR)"]
        EMS["EMS / Optimizer<br/>[Energy Management System with optimizer]<br/>(peak shaving, value stacking)"]
        Historian["Historian / Data Platform + APIs<br/>[Time-series historian + data platform with APIs]"]
    end

    Gateway --- SCADA
    Gateway --- PlantCtrl
    Gateway --- EMS
    PlantCtrl -. "REST/OPC UA/MQTT" .-> Historian
    EMS -.-> Historian

    %% Hardware / Inverter Level
    PCS_Ctrl["PCS Controller<br/>[Power Conversion System controller]<br/>(DSP/FPGA: PLL, current loops,<br/>Volt-VAR/Watt, Freq-Watt, FRT)"]
    PlantCtrl -- "Modbus TCP / IEC 61850 / OPC UA" --> PCS_Ctrl

    PCS["PCS / Bidirectional Inverter + LCL<br/>Filter + Contactor/Precharge<br/>(optional APF/STATCOM mode)<br/>[Bidirectional inverter with grid filter and switching/safety gear]"]
    PCS_Ctrl -- "Internal gate drives / fast I/O" --> PCS

    %% DC Side
    DC_Bus{"DC Bus<br/>[High-voltage DC link]<br/>(e.g., 600–1500 Vdc)"}
    PCS === DC_Bus

    OptionalDCDC["Optional DC/DC<br/>[Optional bidirectional DC-DC converter]<br/>(bi-dir buck/boost)"]
    DC_Bus === OptionalDCDC

    BMS_Master["BMS Master<br/>[Battery Management System master controller]<br/>(SoC/SoH, limits, balancing, trips, logs)"]
    BMS_Slaves["BMS Slaves (cell/module CMUs)<br/>[Cell/Module Monitoring Units]<br/>(V,T sensing, balancing control)"]

    BMS_Master -- "CAN / RS485 / Ethernet" --> PlantCtrl
    BMS_Master -- "Hardwired Interlocks" --> PCS
    BMS_Master -- "CAN / daisy-chain / RS485" --> BMS_Slaves

    OptionalDCDC === BatteryStrings["Battery Racks/Strings<br/>[Physical battery assemblies]"]
    BMS_Slaves -.- BatteryStrings

    %% Power Path
    subgraph PowerPath ["POWER PATH / PCC & PROTECTION"]
        direction TB
        MV_Feeder["Utility MV Feeder<br/>[Medium-voltage distribution feeder]<br/>(11/22/33 kV typical)"]
        MV_Switchgear["MV Switchgear (VCB/Isolator)<br/>[Switchgear with Vacuum Circuit Breaker and isolator]<br/>+ Protection IED"]
        Transformer["MV/LV Transformer<br/>[Medium-voltage to low-voltage transformer]"]
        LV_Bus{"LV AC Bus<br/>[Low-voltage AC busbar]<br/>(415/690 Vac typical)"}
        RevenueMeter("Revenue Meter<br/>[Billing/settlement-grade energy meter]")
        PQMeter["PQ Meter / PMU<br/>[Power Quality meter / Phasor Measurement Unit]<br/>(V, I, f, THD, flicker)"]

        MV_Feeder ==> MV_Switchgear
        MV_Switchgear ==> Transformer
        Transformer ==> LV_Bus
        LV_Bus ==> PCS

        LV_Bus --- RevenueMeter
        LV_Bus --- PQMeter

        MV_Switchgear -. "Status/Trips (IEC 61850 GOOSE)" .-> SCADA
        RevenueMeter -. "Modbus/IEC 61850" .-> SCADA
    end

    %% Styling
    classDef hardware fill:#f9f,stroke:#333,stroke-width:2px;
    classDef control fill:#ccf,stroke:#333,stroke-width:2px;
    classDef grid fill:#ddf,stroke:#333,stroke-width:2px;

    class PCS,PCS_Ctrl,BMS_Master,BMS_Slaves,OptionalDCDC hardware;
    class SCADA,PlantCtrl,EMS,Gateway,Historian control;
    class MV_Feeder,MV_Switchgear,Transformer,LV_Bus,RevenueMeter,PQMeter grid;

2.2 Interface legend (typical)

Utility ↔ Gateway/RTU: IEC 60870‑5‑104 or DNP3 (WAN), sometimes IEC 61850 (substation context)
RTU/SCADA ↔ IEDs/meters: IEC 61850 (incl. GOOSE) and/or Modbus TCP/RTU
Plant Controller ↔ PCS: Modbus TCP, IEC 61850, OPC UA (vendor‑dependent)
Plant Controller/EMS ↔ Historian/IT: REST APIs, OPC UA, MQTT (plus file exports for settlement)
BMS Master ↔ PCS/Plant Controller: Ethernet/RS485/CAN (plus hardwired “permit‑to‑operate” interlocks)
Time sync: GPS → Grandmaster → PTP (IEEE 1588) or NTP across OT LAN
Safety: hardwired E‑Stop chain, fire alarm I/O, breaker/contactor status, door interlocks

3) Functionality of each sub-block

Legend: OT = Operational Technology, IT = Information Technology, PCC = Point of Common Coupling, PCS = Power Conversion System, BMS = Battery Management System, IED = Intelligent Electronic Device, PMU = Phasor Measurement Unit.

Item name (as in block diagram)	Short description (expanded / plain)	Detailed description (nature / functionality / characteristics)
Utility / SLDC / DERMS	Utility control center / State Load Dispatch Centre / Distributed Energy Resource Management System	External grid operator systems that send dispatch instructions (P, Q, modes, limits), receive telemetry/alarms/events, and may enforce grid codes. Typically operates over secure WAN links and requires strict data integrity, audit logging, and deterministic command handling.
dispatch, limits, events, telemetry	Control commands and monitoring data	Dispatch = power setpoints/schedules; limits = operating constraints (export/import caps, ramp limits, SoC limits, grid support modes); events = disturbances, trips, alarms; telemetry = continuous measurements (P, Q, V, I, f, SoC, status). Usually needs time-stamping and quality flags.
IEC 60870-5-104	IEC telecontrol protocol over TCP/IP	Common utility SCADA protocol for remote control/telemetry. Supports command/response and spontaneous events. Typically used on WAN links; requires secure transport (VPN/TLS), whitelisting, and rate limiting to prevent overload.
DNP3	Distributed Network Protocol v3	Utility-grade telemetry/control protocol widely used for SCADA. Supports event buffers, time-stamping, and robust comms. Often used over TCP/IP; security extensions exist but in practice commonly secured with VPN/firewalls.
IEC 61850	Substation automation communications standard	Standard for high-speed, structured data models and messaging in substations. Enables interoperable logical nodes and services; supports GOOSE for fast peer-to-peer trips/commands. Requires careful engineering of datasets, SCL files, and network design.
VPN/TLS over WAN	Virtual Private Network / Transport Layer Security on Wide Area Network	Secure communication method to protect data in transit between utility and site. VPN creates an encrypted tunnel; TLS secures sessions. Must manage keys/certificates, access control, and monitoring for cyber resilience.
Secure OT Gateway/Firewall	OT edge security gateway + firewall	The “front door” to the BESS OT network. Performs routing, segmentation (zones/conduits), firewalling, VPN termination, user access control (RBAC), logging, and sometimes protocol mediation. Critical for preventing lateral movement and restricting remote access.
router, VPN, RBAC, logs	Network services: routing, encryption, role-based access control, audit logs	Router forwards traffic; VPN encrypts; RBAC controls who can do what; logs provide traceability for compliance and incident response. In OT, these must be hardened, monitored, and configured with least privilege.
Ethernet (LAN, VLANs)	Local Area Network using Ethernet; VLANs for segmentation	Primary site communications fabric. VLANs separate traffic types (protection/controls/SCADA/vendor access) to reduce cyber risk and improve determinism. Requires managed industrial switches, QoS, and redundancy planning.
SCADA/RTU	Supervisory Control and Data Acquisition / Remote Terminal Unit	RTU/SCADA collects field I/O (status, analogs), manages alarms/events, and bridges protocols to the utility. Often the official interface point for “what the utility sees.” Must be reliable, deterministic, and support event buffering during comm loss.
alarms/events, protocol bridge	Alarm/event handling + protocol conversion	Alarm/event management includes prioritization, latching, acknowledgement, event time-stamps. Protocol bridge translates between site protocols (Modbus/61850/vendor) and utility protocols (104/DNP3).
Plant Controller	BESS Plant Controller / Power Plant Controller (PPC)	Real-time coordinator for the whole BESS. Converts external requests into PCS setpoints while enforcing constraints (SoC, thermal, grid limits). Implements operating modes: P/Q control, voltage support, frequency response logic orchestration, ramp control, black-start sequencing (if applicable). Typically runs on an industrial controller with redundant options.
(P/Q/V/f modes, ramp, SoC, FFR)	Active/reactive power, voltage/frequency control; ramp limiting; state-of-charge; fast frequency response	The controller selects control modes (e.g., constant P, Volt-VAR), manages ramp rates, ensures SoC stays within bounds, and triggers fast services (FFR) per grid requirements. Should include fallback strategies if comms fail.
EMS / Optimizer	Energy Management System with optimizer	Higher-level scheduler/optimizer that decides when to charge/discharge to meet business objectives (peak shaving, arbitrage, demand charge reduction) while respecting constraints. Runs on minutes-to-hours horizons and sends schedules/targets to Plant Controller. Typically integrates forecasts (load, solar, price) and supports reporting.
(peak shaving, value stacking)	Peak shaving and multiple-use value stacking	Peak shaving reduces peaks; value stacking coordinates multiple services without violating battery limits or grid restrictions. Requires prioritization logic (e.g., safety > grid code > contracted services > arbitrage).
REST/OPC UA/MQTT	Common IT/OT data interfaces: REST API / OPC Unified Architecture / MQTT	REST provides HTTP APIs for apps and reporting; OPC UA provides industrial-grade information modeling and secure client/server; MQTT is lightweight pub/sub for telemetry. Choice depends on enterprise integration and cybersecurity posture.
Historian / Data Platform + APIs	Time-series historian + data platform with APIs	Stores high-resolution telemetry, events, alarms, and calculated KPIs (throughput, cycles, PQ indices). Enables analytics, settlement evidence, and remote troubleshooting. Typically includes retention policies, time sync, and access controls.
Modbus TCP / IEC 61850 / OPC UA	OT protocols between controllers and devices	Used for Plant Controller ↔ PCS/meters/IEDs/RTU integration. Modbus is simple and common but less semantic; 61850 is structured and high-speed; OPC UA is secure and model-driven. Engineering must ensure consistent scaling, units, and update rates.
PCS Controller	Power Conversion System controller	Embedded real-time controller (DSP/FPGA/MCU) that runs the inverter control loops: PLL, current/voltage loops, modulation (PWM), protection, and grid-support functions. Must be deterministic (sub-millisecond), robust to disturbances, and tightly coupled to sensing and gate drives.
(DSP/FPGA: PLL, current loops, Volt-VAR/Watt, Freq-Watt, FRT, harmonic ctrl)	Digital Signal Processor / FPGA functions: phase lock, current control, grid functions	Implements grid-following synchronization (PLL), controls P/Q injection by regulating currents, provides Volt-VAR/Volt-Watt and Frequency-Watt droop, supports Fault Ride Through (FRT) where required, and may implement active harmonic compensation.
Internal gate drives / fast I/O	High-speed control signals and protections	Low-latency signals for IGBT/SiC gate drivers, hardware trips, DC-link monitoring, and interlocks. Typically isolated, noise-immune, and designed to fail-safe (trip on fault).
PCS / Bidirectional Inverter + LCL Filter + Contactor/Precharge (optional APF/STATCOM mode)	Bidirectional inverter with grid filter and switching/safety gear; optional power-quality modes	The “power interface” that converts DC↔AC. LCL filter reduces switching harmonics and meets grid THD limits. Contactor/precharge manages safe energization of DC-link capacitors. Optional APF/STATCOM mode enables dynamic reactive power support and harmonic compensation. Must meet efficiency, overload, and grid-code compliance.
DC Bus (e.g., 600–1500 Vdc)	High-voltage DC link	Main DC backbone between battery system and PCS (and optional DC/DC). Requires insulation monitoring, safe clearances, arc protection, and careful cable/busbar design for high currents and EMI. Voltage range depends on architecture and standards.
Optional DC/DC (bi-dir buck/boost)	Optional bidirectional DC-DC converter	Adds flexibility to decouple battery voltage variation from PCS DC-link requirements. Can improve efficiency and control range, enable wider SoC window, support multi-string architectures, and help with battery protection (current limiting). Adds cost/complexity and requires additional control/protection.
DC strings / contactors / fuses	Battery string electrical protection and isolation	Each battery string typically has fuses/breakers for fault protection, contactors for isolation, and sometimes precharge. Designed to clear DC faults safely and prevent cascade failures. Must coordinate with BMS and DC breaker strategy.
CAN / RS485 / Ethernet (vendor) + hardwired interlocks	BMS communications + safety interlocks	CAN/RS485/Ethernet carry SoC/SoH, limits, alarms, and commands. Hardwired interlocks (permit-to-operate, E-stop chain, fire trip, door switches) provide safety-critical control independent of software/comms—typically required for functional safety.
BMS Master	Battery Management System master controller	Supervises the complete battery system: aggregates cell data, computes SoC/SoH, enforces limits (charge/discharge current, voltage, temperature), controls contactors, and triggers trips. Provides event logs and communicates operating limits to PCS/Plant Controller. Central to safety and lifecycle performance.
(SoC/SoH, limits, balancing, trips, event logs)	State of Charge / State of Health; operating limits; balancing; protection trips; logs	SoC estimates available energy; SoH indicates degradation. Limits protect cells (C-rate, voltage, temperature). Balancing equalizes cell voltages. Trips isolate battery on unsafe conditions. Logs support root-cause analysis and warranty evidence.
BMS Slaves (cell/module CMUs)	Cell/Module Monitoring Units	Distributed measurement boards located near cells/modules measuring voltage and temperature, sometimes current. Provide balancing control at cell level and report to BMS Master. Must be accurate, noise-immune, and robust in harsh environments.
(V,T sensing, balancing control)	Voltage & temperature sensing with balancing	High-channel-count sensing with calibration and isolation. Balancing may be passive (resistor bleed) or active (energy transfer). Critical to prevent cell drift, extend life, and avoid unsafe over-voltage/over-temp conditions.
Utility MV Feeder (11/22/33 kV typical)	Medium-voltage distribution feeder	Grid connection point for the BESS plant. MV level reduces currents and losses, fits utility infrastructure. Requires compliance with grid protection, voltage regulation, and fault level constraints.
MV power	Medium-voltage power flow	Physical AC power flow at MV between BESS and grid. Impacts protection settings, thermal limits, and voltage profile on feeder.
MV Switchgear (VCB/Isolator)	Switchgear with Vacuum Circuit Breaker and isolator	Provides safe switching, isolation, and fault interruption on MV. VCB clears faults; isolator provides visible isolation for maintenance. Often includes earthing switch, surge arresters, and interlocking. Must meet short-circuit ratings and utility requirements.
+ Protection IED (O/C, E/F, UV/OV, UF/OF, RoCoF, Sync-check, directional as needed)	Protection relay functions: Overcurrent / Earth fault / Under-Over voltage / Under-Over frequency / Rate-of-Change-of-Frequency / Synchronism check	Protection IED detects faults and abnormal conditions and trips breakers to protect grid and plant. Directional elements may be needed for reverse power/fault direction. Sync-check ensures safe breaker closing (especially for black start or reclose). Settings coordination with utility is critical.
Status/Trips: hardwired + IEC 61850 GOOSE (optional)	Trip/status signaling via hardwired contacts and/or GOOSE messaging	Hardwired contacts provide deterministic fail-safe trip paths. IEC 61850 GOOSE provides very fast messaging (milliseconds) for interlocking/tripping across devices. Both can be used: GOOSE for speed + hardwired for ultimate safety.
MV/LV Transformer	Medium-voltage to low-voltage transformer	Steps MV down to LV for PCS connection (or steps up depending on configuration). Provides galvanic isolation, impedance for fault limitation, and voltage matching. May have OLTC (on-load tap changer) in some designs, though BESS often uses PCS Q/V control instead.
LV AC (415/690 Vac typical)	Low-voltage AC level	LV side where PCS connects. 415V or 690V common for industrial power electronics. Drives current magnitudes and busbar/switchgear sizing.
LV AC Bus	Low-voltage AC busbar	Common coupling point for PCS output, meters, and LV switchgear. Must handle rated current, harmonics, transient conditions, and provide adequate protection/segregation.
PQ/Revenue Metering -> Modbus/IEC 61850 -> RTU/SCADA	Metering data path via industrial protocols	Defines how metering measurements are transported to RTU/SCADA for reporting/settlement. Needs correct scaling, time sync, data quality flags, and event capture for disputes.
Revenue Meter	Billing/settlement-grade energy meter	Utility-accepted meter for import/export kWh/kVArh and demand. Must meet accuracy class and certification requirements. Provides tamper evidence and secure registers used for commercial settlement. Often installed per utility metering scheme.
PQ Meter / PMU (optional)	Power Quality meter / Phasor Measurement Unit	PQ meter records RMS V/I, THD, harmonics spectrum, flicker, sags/swells, and events. PMU adds synchronized phasors and high-speed dynamics (where deployed). Used to prove compliance, diagnose issues, and support advanced controls.
(V, I, f, THD, flicker, event recorder)	Voltage, current, frequency, total harmonic distortion, flicker, event recording	Captures steady-state and transient PQ performance. Event recorder time-stamps disturbances and helps correlate with trips or grid issues. Often requires accurate time sync (NTP/PTP/GPS).
PCS (shown above) → DC Bus → Battery Racks/Strings → Thermal Mgmt + Fire System	End-to-end energy flow and supporting safety systems	Indicates the complete energy chain: PCS interfaces to battery via DC bus; battery racks/strings store energy; thermal management maintains safe temperatures; fire system detects/mitigates thermal runaway events. All must be integrated with interlocks so faults force safe shutdown.
Battery Racks/Strings	Physical battery assemblies	Modular battery building blocks (cells → modules → racks → strings). Characteristics include nominal voltage, capacity (Ah), max current (C-rate), isolation, serviceability, and safety features. Architecture impacts redundancy, maintenance, and fault containment.
Thermal Mgmt + Fire System	Thermal management + fire detection/suppression	Thermal management (HVAC or liquid cooling) maintains cell temperatures within tight bands for safety and life. Fire system detects off-gassing/smoke/temperature excursions and triggers alarms, ventilation control, suppression, and system shutdown. Must integrate with BMS and site safety regulations.

BESS — EMS vs Plant Controller vs PCS Controller - Unambiguous Boundary Spec

This document defines a clear functional boundary between: - EMS (Energy Management System / Optimizer), - Plant Controller (a.k.a. PPC: Power Plant Controller / BESS Plant Controller), and - PCS Controller (Power Conversion System controller inside the inverter/PCS).

It also provides example use cases (6) showing the flow and responsibility split.

1) Boundary definition: who owns what

1.1 Time‑scale ownership (hard rule)

PCS Controller (inner control): 0.1 ms → 50 ms (fast loops), up to ~200 ms for some grid support responses
Owns waveform-level control, instantaneous current regulation, converter self-protection.
Plant Controller / PPC (outer real-time supervisor): 200 ms → 5 s
Owns real-time setpoint generation, service arbitration, plant constraints, multi‑PCS coordination, PCC compliance.
EMS (scheduler/optimizer): 5 min → day‑ahead / season‑ahead
Owns economic optimization, schedules, reserve policy, service priorities and commercial constraints.

Non‑negotiable: EMS must not issue sub‑second commands to PCS. PCS must not decide market/economic intent.

1.2 Authority and override hierarchy (hard rule)

1) Protection / hardwired safety (trip/lockout)
2) BMS limits (battery safety)
3) PCS controller hardware limits (semiconductors, DC link, thermal)
4) Plant Controller (grid compliance + plant coordination)
5) EMS (economics/scheduling)

1.3 Command surface (what each layer is allowed to write)

EMS is allowed to write (Intent + Policy)

Writes: - Schedule / dispatch intent: P_schedule(t) or E_schedule(t) for next horizon (e.g., 24h) - SoC policy: SoC_min, SoC_max, SoC_target(t) - Reserve policy: SoC_reserve_up/down, headroom/footroom for FFR/voltage support - Commercial constraints: P_export_cap, P_import_cap, daily throughput cap (optional), contract service enable flags - Priority policy: ordered service priorities (e.g., Safety > Grid code > FFR > Voltage support > Peak shaving > Arbitrage)

EMS must NOT write: - P_set/Q_set every second or sub‑second - fast droop/curve parameters directly to PCS (except by referencing pre-approved profiles via Plant) - converter parameters, PWM settings, inner loop gains

Plant Controller is allowed to write (Real-time setpoints + mode selection)

Writes: - Instantaneous setpoints to PCS: P_ref, Q_ref or V_ref (if V/Q mode) - Mode selection: MODE (e.g., P/Q control, Volt‑VAR enabled, Freq‑Watt enabled, STATCOM mode) - Ramp limits (plant level): dP/dt_limit, dQ/dt_limit - Parameter set selection: profile_id (select pre‑approved droop/Volt‑VAR/FRT profiles) - Sharing commands: per‑PCS allocation factors or per‑unit P_ref_i, Q_ref_i

Plant Controller must NOT write: - PCS inner-loop gains, PWM parameters, or anything that bypasses PCS safety/protection - battery contactor direct commands (except via defined enable/permit interface; BMS owns contactors)

PCS Controller is allowed to write (converter actuation & self-protection)

Writes/Executes: - PWM/gate drive to meet commanded P_ref/Q_ref safely - Grid support algorithm execution using provided/approved profiles: - Volt‑VAR, Volt‑Watt, Freq‑Watt droop, ramp limiting, harmonic damping/compensation - Self‑protection trips/derates: overcurrent, DC link faults, thermal limits - Capability reporting: available P_max/Q_max, derate flags, fault codes

PCS must NOT write: - plant-level objectives (export cap at PCC, service priority decisions across multiple services) - EMS schedules or economic objectives

1.4 Avoiding functional duplication (the “same-looking feature” rule)

Even if both PCS and Plant “do Volt‑VAR”: - PCS owns: control law implementation + fast actuation - Plant owns: enable/disable, which curve/profile, priority vs other services, PCC-level compliance, multi‑PCS coordination - EMS owns: policy constraints (reserve/energy), whether service is contracted/allowed, schedule intent

2) Minimal interface contract (unambiguous)

2.1 EMS → Plant Controller (every 5–15 min + on changes)

P_schedule(t) (array for horizon) or E_schedule(t)
SoC_min/max, SoC_target(t)
P_export_cap, P_import_cap
reserve_up/down (headroom/footroom policy)
service_priority_list
service_enable_flags (FFR on/off, voltage support allowed, etc.)

2.2 Plant Controller → PCS (every 100 ms–2 s)

MODE (P/Q, V/Q, STATCOM, etc.)
P_ref, Q_ref (or V_ref)
ramp_limit
profile_id (select pre‑approved droop/Volt‑VAR/FRT profiles)
enable_flags (Volt‑VAR active, Freq‑Watt active, etc.)

2.3 PCS → Plant Controller (every 100 ms–2 s)

Measurements: P,Q,V,I,f (local or pass-through)
Capability: P_avail, Q_avail, derate flags
Status: running/standby/faulted, alarm codes, trip reasons

3) Use cases (flows + responsibilities)

Use Case 1 — Peak shaving (normal day, schedule‑driven)

Goal: reduce facility/feeder peak demand while staying within SoC limits.

Flow 1. EMS computes day-ahead/rolling schedule: discharge 18:00–22:00, maintain SoC_min=30%, keep reserve_down=10% for voltage events. 2. Plant Controller converts schedule to real-time P_ref considering SoC, available PCS power (derates), export cap, and ramp limits. 3. PCS tracks P_ref using fast current control and enforces converter limits.

Responsibilities - EMS: decides when and how much energy to shift; defines SoC limits + reserves. - Plant: decides instantaneous output and ramps; manages deviations and curtailment reasons. - PCS: executes waveform control and instantaneous limit protection.

Boundary statement - EMS never commands P_ref every second; PCS never decides the schedule window.

Use Case 2 — Voltage rise mitigation (RTPV surplus midday)

Goal: keep PCC voltage within band during PV surplus.

Flow 1. PCC voltage rises above threshold (measured at PCC/PCS). 2. Plant Controller switches to Voltage Support Mode (if allowed by EMS policy), selects profile_id = Volt‑Watt + Volt‑VAR, and ensures SoC headroom. 3. PCS executes Volt‑Watt/Volt‑VAR response quickly and absorbs P / injects/absorbs Q. 4. Plant Controller monitors PCC objective and may temporarily override schedule. 5. EMS re‑optimizes later to recover SoC deviation.

Responsibilities - EMS: sets policy (“voltage support allowed”, reserve headroom). - Plant: enables the correct profile, arbitrates vs peak shaving/arbitrage, manages headroom. - PCS: executes curves and maintains stability.

Use Case 3 — Frequency dip (Fast Frequency Response / droop)

Goal: inject/absorb power within seconds/sub‑seconds to stabilize frequency.

Flow 1. Frequency dips below trigger. 2. PCS responds immediately using enabled Freq‑Watt droop/FFR profile. 3. Plant Controller ensures response stays within plant caps and SoC reserve; manages recovery to schedule. 4. EMS adjusts future schedule to restore SoC and account for energy imbalance.

Responsibilities - EMS: ensures reserve exists (SoC policy). - Plant: arms/disarms FFR service, sets profile_id, tracks reserve usage and recovery. - PCS: immediate response + converter protection.

Use Case 4 — Export cap enforcement (utility constraint)

Goal: never exceed an export limit at PCC (contract/technical constraint), regardless of schedule.

Flow 1. EMS schedule requests discharge, but PV surplus already pushes export near cap. 2. Plant Controller enforces P_export_cap in real time by clamping/adjusting P_ref. If needed and feasible, it commands absorption (charge) instead. 3. PCS follows P_ref and handles ramps safely. 4. EMS receives deviation report and re‑optimizes later.

Responsibilities - EMS: declares cap + schedule intent. - Plant: guarantees cap compliance; selects best action (reduce discharge, charge, or Q support). - PCS: executes commanded power.

Boundary statement - Export cap is a PCC-level objective → owned by Plant, not PCS.

Use Case 5 — Black start / islanded energization sequence (if applicable)

Goal: energize a dead bus and restore load safely.

Flow 1. Utility authorizes black start; site is de‑energized. 2. Plant Controller runs sequence: checks BMS readiness/SoC, sets PCS mode to grid‑forming/island (if supported), picks up load in steps, manages stability. 3. PCS Controller performs voltage/frequency forming (fast loops) and current limiting. 4. Plant coordinates synchronization and reconnection when grid returns (with protection sync‑check permissive).

Responsibilities - EMS: only policy constraints (minimum SoC) and logging; not in critical loop. - Plant: owns sequencing, interlocks, reconnection orchestration. - PCS: owns grid‑forming waveform control and fast stability.

Use Case 6 — Harmonics / power quality event (THD rise or resonance risk)

Goal: keep harmonic distortion within limits and avoid instability.

Flow 1. PQ monitoring indicates THD/event trend (or utility complaint). 2. Plant Controller selects a PQ-safe operating profile and may derate P/Q delivery or enable APF/STATCOM mode (if available). 3. PCS executes active damping/compensation (within capability) and maintains stable currents. 4. EMS re‑optimizes if capability is reduced.

Responsibilities - EMS: accepts reduced capability; reschedules economics. - Plant: profile selection, derate policy, and service arbitration. - PCS: harmonic mitigation execution + converter protection.

4) Unambiguous “Do / Don’t” summary

EMS — Do / Don’t

Do - Optimize and send schedules, SoC policies, reserves, priorities, contractual caps.
Don’t - Send fast real‑time P/Q setpoints; tune converter control.

Plant Controller — Do / Don’t

Do - Enforce PCC objectives, caps, grid compliance, service arbitration; coordinate multiple PCS.
Don’t - Implement PWM/current loops; bypass PCS/BMS protections.

PCS Controller — Do / Don’t

Do - Execute fast control and grid support algorithms; protect converter; report capability.
Don’t - Decide service priorities, economic intent, or plant-level PCC compliance strategy across assets.

5) Optional acceptance tests (quick sanity checks)

If EMS link is lost: Plant continues safe operation using last intent + fallback policy (no PCS inner-loop changes).
If a PCS derates: Plant redistributes setpoints across remaining PCS to meet PCC caps where possible.
If a frequency event occurs: PCS responds immediately only if Plant has armed the profile and reserve exists per EMS policy.