OPC · KEPWARE / KEPSERVEREX
2,461 mappings·1,027-field schema·ISA-95 categorized·deterministic
Kepware already gives you every tag in one place. What it does not give you is agreement about what the tags mean.
KEPServerEX solves transport: every device arrives over OPC UA in one namespace. It does not solve semantics. Channel1.Device1.S1Temp and Channel2.Device1.SPINDEL_AUSLASTUNG are neighbours in the same browse tree and still speak different languages, because the vendor underneath did.
OPC also carries quality out of band. A tag can return a perfectly well-formed number with quality Bad, and a client that reads only Value will average a stale or invalid reading into a KPI without ever knowing. Forge reads the wrapper: a Bad quality value is nulled with opc_quality_bad as the stated reason, and a Good one passes through unchanged.
Send the leaf browse name, the last segment of the path, and resolution is an exact corpus hit at full confidence, whichever vendor sits underneath the channel.
| Raw vendor tag | Canonical field | ISA-95 category | |
|---|---|---|---|
| S1Temp | → | spindle_temperature | equipment_condition |
| SPINDLE_SPEED_ACT | → | spindle_speed_rpm | equipment_performance |
| Betriebsstunden | → | operating_hours | equipment_state |
| AC_Power | → | inverter_output_kw | energy_consumption |
Every row above is produced by the shipping engine, not written by hand. This page fails to build if any of them stops resolving to the field shown.
Resolution is a lookup, not an inference. A tag is matched against a curated corpus of 2,461 confirmed vendor mappings; only if that misses does it fall to weaker layers: a normalized fold that strips unit suffixes and separators, then a signal classifier that reads structure and unit evidence. Each layer carries a lower confidence, and the layer that answered is named in the response. A mapping you disagree with is visible rather than buried.
Below a confidence threshold the engine abstains. An evidence gate refuses a match it cannot justify and reports the tag unresolved, by name, with the score and the reasons, because a wrong field at high confidence is far more expensive than an honest miss. Six read-time validators then check what survives: sentinel detection, physics bounds, stuck sensors, dropout, operating mode and correlation.
Units are converted to SI where the source unit is known, and every conversion is reported with its factor. A value that fails a physics bound is nulled with the bound it violated attached, not silently dropped.
It does not invent fields. A tag outside the corpus that the signal layer cannot justify comes back unresolved and counted, and those names feed the roadmap for the next pack rather than a plausible guess in your data. It is a translation layer, not a historian: it holds no time series and replaces no SCADA system. And the container is a sandbox: it ships the deterministic layers and the canonical schema, not the production embedding, research and corpus-learning layers.
No key, no account, no sales call. The sandbox runs locally and answers with the same canonical schema production does.
Two limits worth stating before you find them yourself. A raw node-ID string (ns=2;s=Channel1.Device1.SpindleSpeed) is not parsed today. And a fully-qualified dotted path is resolved only by the signal layer, which is weaker than an exact match: Channel1.Device1.S1Temp does resolve, Channel1.Device1.ACTFEED does not. Split the path and send the leaf, a one-line client-side transform, and every tag above resolves exactly. Native node-ID and browse-path parsing is on the roadmap.
Export a tag list and send it as JSON. See platform licensing for embedding the canonical data model in your own product, or pricing for usage-based API access.