Home / Solutions / Industrial Networking & Datacom / Power Grid IP Backbone & Dispatch Network

Power Grid IP Backbone & Dispatch Network: One Network Carrying Dispatch, Protection and Video Without Ever Merging Their Planes

A grid dispatch network answers to a different clock than an office network: a relay-protection packet that arrives late is operationally the same as one that never arrived, and the security regulation that governs it will not accept a flat network as an answer. We design the IP backbone that carries dispatch automation, comprehensive data and video on one physically hard-sliced infrastructure — SRv6 with FlexE hard pipes keeping each plane isolated on shared fibre, dual-plane reliability so a single failure cannot take down both, and sub-50 ms protection switching where relay traffic needs it. This is the grid's own production and dispatch backbone — distinct from the utility's office campus network (see Energy & Utility Campus Network): that page covers the workplace, this one covers the wires that keep the grid itself coordinated. Sized honestly for a municipal grid, a provincial grid or a cross-provincial backbone.

Why a Grid Backbone Cannot Be Just Another WAN

Four realities we design around on every grid backbone project:

Protection and dispatch traffic cannot tolerate jitter or interruptionRelay protection and dispatch automation ride on a timing budget measured in milliseconds, not seconds. A packet that arrives late during a fault is operationally the same as one that never arrived — and unlike an office outage, the consequence is a grid event, not a help-desk ticket.

Dispatch, comprehensive data and video have to sit on one physical network, hard-isolatedA regional grid genuinely cannot afford three separate physical networks for dispatch automation, comprehensive/enterprise data and video — but it also cannot afford to let them share a network the way an office does. The requirement is one infrastructure, several hard-isolated planes, not a flat network with VLANs and good intentions.

The old backbone was sized for a grid that did not have this much new energy on itDistributed PV, wind and storage sites are joining the grid faster than most backbones were planned for, and each one adds a monitoring and control traffic load the original bandwidth budget never accounted for. A backbone sized for yesterday's grid runs out of headroom quietly, then all at once.

Substations and dispatch sites are spread across cities and provincesA dispatch network is not one building — it is every substation, every metering point and every control centre across a city, a province or several provinces, each needing reliable, low-latency reach back to dispatch. The interconnect problem grows with the map, not with the headcount.

Architecture: SRv6 + FlexE Hard Slicing, Dual-Plane Reliability, Sub-50 ms Protection

One backbone, several hard-isolated planes, and a protection-switching budget measured in milliseconds:

Architecture drawn by AtlasCommTech following carrier-grade design practice. Diagram labels are kept in English for engineering clarity.

Why us: our founder spent 13 years inside the Huawei partner ecosystem delivering carrier networks — the same SRv6 and hard-slicing design discipline a dispatch backbone needs, not adapted from an office template. Our own Zhongxin Shutong industrial switches are built for exactly this substation yard: rated for roughly −40 to +75 °C, DIN-rail mounted and hardened to IEC 61850-3 class immunity — a solid fit for substation access, while the backbone routing core stays open to whichever brand suits your regulator, your budget and your team.

Equipment Options

The solution is sized to your requirements and budget first — the same architecture can be delivered on several vendors' product lines. We help you choose by supply availability in your destination country, budget and your team's operating habits.

Huawei — enterprise campus, WAN and security linesMature ecosystem with a global service network.

ZTE Wantong — comparable datacom linesPrice-performance direction; supply runs smoother in some markets.

H3C — campus and data-center linesWidely deployed campus and data-center portfolio.

Zhongxin Shutong industrial switches — substation access, our own lineRated for roughly −40 to +75 °C, DIN-rail mounted, hardened to IEC 61850-3 class immunity — built for the substation yard, not the comms room. We answer for this access layer ourselves; the backbone routing core above it stays open to any brand that suits your country and your regulator.

What the Design Delivers

Six things a properly engineered dispatch backbone does that a generic WAN never will:

Dispatch, data and video in separate hard slicesSRv6 with FlexE hard slicing gives dispatch automation, comprehensive/enterprise data and video their own guaranteed-bandwidth pipe on the same physical fibre — not a priority queue on a shared one. A flood on the video plane cannot borrow bandwidth from the protection plane, because there is no path for it to borrow through.

Dual-plane reliability, not just a redundant linkThe dispatch plane and the integrated data plane are engineered as two logically independent planes over physically diverse paths, so a single fibre cut, card failure or misconfiguration on one plane cannot propagate into the other.

Sub-50 ms protection switchingWhere relay protection traffic rides the backbone, TI-LFA and SRv6 fast-reroute bring path switching down into sub-50 ms territory — the only budget that matches what a protection scheme actually needs.

Headroom sized for new energy, not just today's loadBandwidth planning accounts for the PV, wind and storage sites already in your interconnection queue, not only the substations already on the network — so the backbone still has room when the next solar farm energizes.

SDN-based centralized visibility across every siteA controller with BGP-LS topology collection and streaming telemetry gives dispatch a live map of the whole backbone, instead of per-site logs that only make sense after the fact.

Substation access built for the yardAccess equipment at the substation is specified to substation conditions — temperature range, EMC and vibration — not to the comms-room conditions a standard enterprise switch assumes.

Three Sizes, One Design Logic

Tell us how many substations, how many dispatch sites and what your protection-service list looks like — the tier tells you the shape of the network:

Numbers we design around:

Protection-plane switching is a sub-50 ms budget — the protection-relay list sets the target, not a generic SLA

Dispatch and integrated planes are hard-isolated by design, verified before go-live — not separated by policy alone

Bandwidth is sized against the new-energy interconnection queue, not only against substations already live

Scale tier	Typical site	What the design includes
Municipal / prefecture grid	One city or prefecture bureau · a handful of substations · one dispatch centre	A backbone ring of core/aggregation routers linking the dispatch centre to each substation, SRv6 with FlexE hard slicing for dispatch and comprehensive-data planes, sub-50 ms protection switching on the segments carrying relay traffic, substation-hardened access equipment throughout.
Provincial grid	A province-wide grid · multiple prefecture dispatch centres · one provincial control centre	A provincial backbone interconnecting prefecture rings, SDN-based centralized management with BGP-LS topology collection, dedicated capacity planned for new-energy interconnection growth, and dispatch/integrated-plane separation enforced end-to-end rather than city by city.
Cross-provincial backbone	Backbone linking multiple provincial grids or a regional transmission operator	A high-capacity SRv6 backbone across provincial boundaries, coordinated protection-switching design across the full path, standardized zoning and naming so each province's segment is a copy of the same design, and a staged rollout aligned to each provincial dispatch centre's own maintenance windows.

Equipment Roles (Categories, Not Models)

The solution is built from these equipment categories — the brand is chosen with you at design stage. Exact models depend on your bandwidth, port counts, protection-service list and country — so we spec models after your requirements list, not before.

Role	What it does
Core / backbone routers	Carry the SRv6 backbone and the FlexE hard slices between dispatch centres and aggregation points — the layer sized against your bandwidth and new-energy growth plan.
Aggregation / PE routers	Collect substation and site traffic and hand it into the correct hard slice on the backbone — the layer where the dispatch/integrated-plane split is enforced.
Substation access routers/switches (hardened)	Live in the substation yard, rated for the temperature range, EMC and vibration a comms-room switch was never built for. Terminate relay protection, automation and video traffic locally.
Dispatch-network boundary security gateway	The single controlled crossing point where the dispatch/protection plane may exchange only explicit, named traffic with the integrated or enterprise side — everything else is refused by design, matching electric power monitoring system security requirements.
Clock / time-synchronization devices	Deliver the precision timing that relay protection and dispatch automation depend on — a requirement most enterprise WAN designs never have to think about.
SDN controller / network management platform	Centralized topology, configuration and fault visibility across the whole backbone — BGP-LS and telemetry feed a live map instead of per-site logs.

Send us your substation list, protection-service list, bandwidth forecast and new-energy interconnection queue — and the model list follows. That order keeps the design honest.

Design Notes & Honest Limits

Read this before you commit:

Power secondary systems are governed by electric power monitoring system security protection regulation — the dispatch network and the integrated/comprehensive network must be physically or hard-pipe isolated by rule, not by convenience. We check the regulation that applies to your grid before we draw the architecture, not after.
Relay protection has a millisecond-class latency and switching requirement. We ask for the actual protection-service list — which links carry which protection scheme — before equipment is selected, because that list is what sets the real target, not a generic SLA number.
New-energy interconnection growth is a forecast, not a fixed number. We size headroom against your interconnection queue and review it at each expansion, rather than promising an exact figure today that a busy solar season will outrun.
Licensing policy and product availability differ by brand and destination country. Our own industrial access line we can speak for directly; the core and aggregation layer we check and confirm for your country and your regulator at the design stage.
A single small dispatch site does not need the full multi-plane, multi-tier design. A hardened access ring and one boundary gateway serve a small site fine — we will tell you so if that is your case, rather than sizing up to sell more equipment.

FAQ

Why can't dispatch and comprehensive data just share VLANs on one network?

Because power secondary-system regulation in most jurisdictions requires the dispatch/protection network and the integrated/comprehensive network to be physically or hard-pipe isolated, not just logically separated. FlexE hard slicing gives each plane its own guaranteed pipe on shared fibre, which satisfies that requirement in a way a VLAN on a flat network does not — and it is also simply safer: a flood or misconfiguration on one plane has no path into the other.

What does sub-50 ms protection switching actually protect against?

It protects the relay protection scheme's own timing budget. When a fault occurs, protection relays exchange trip signals with a strict deadline; if the network path switch after a link failure takes longer than that budget, the protection scheme itself can fail to isolate the fault correctly. TI-LFA and SRv6 fast-reroute bring that switching time down to where it stops being the weak link — but the actual number your protection scheme needs has to come from your protection-service list, not from a datasheet.

How is this different from the Energy & Utility Campus Network page?

That page is the utility company's office and yard network — the campus where staff, inspection robots and enterprise IT live, separated from production by a dual-plane design. This page is the grid's own dispatch and production backbone — the network that carries relay protection, dispatch automation and inter-substation traffic. A utility typically needs both, and they are deliberately different designs for a reason: one protects a workplace, the other protects grid stability itself.

Can the same backbone really carry new-energy sites as they get added?

Yes, if the bandwidth and the hard-slice plan are sized against your interconnection queue rather than only today's substation list — that is precisely the planning step most backbones skip, and precisely why they run out of headroom the year a new solar or wind cluster energizes. We ask for the queue up front so the backbone still has room when it happens.

You make industrial switches — why are Huawei, ZTE & Wantong and H3C still on this page?

Because our line covers substation access, and a dispatch backbone is more than its access layer. The core and aggregation routing, the SDN controller and the boundary security gateway are layers where the right brand depends on your regulator's approved-vendor list, your existing operations skills and your country's supply situation — and on those, another brand is often the right call. We would rather deliver a backbone that passes your grid's compliance review than one where every box carries our name.

Send us your substation list and protection-service list

An engineer replies with a hard-sliced dispatch backbone design and the equipment-category list. Send us your requirements list — the model list follows.

WhatsApp an engineer →