Find quiet failures
Surface readings that remain plausible but stop responding credibly as charging conditions change.
Sensor Confidence
A valid reading is only the beginning.
PlugStream does not simply read sensors. It checks whether their readings remain credible by looking at behaviour, timing, operating context, and independent evidence.
Native implementation
Implemented locally
Deterministic evaluation runs on the charger, close to the signals and operating state that give each reading meaning.
Compatible charge points
Delivered through current PlugStream firmware and maintained through over-the-air updates.
Evidence output
Domain states, evidence-backed findings and diagnostic telemetry are available to PlugStream Cloud for support and estate insight.
What changes for you
More useful evidence before the next investigation step
Sensor Confidence turns relationships inside the charger into practical diagnostic context for drivers, installers, support teams and estate operators.
Explain the evidence
Record what supported a finding, what conflicted with it and what evidence was missing or stale.
Improve remote diagnosis
Give PlugStream support and estate operators more useful context before deciding the next investigation step.
Validity and credibility
In range does not always mean credible
A conventional check can tell you that a sensor value falls within an acceptable range. It cannot always tell you whether that value still reflects what is happening.
Imagine a temperature sensor that continues to report 25 °C while the charger moves through changing operating conditions. The number may look ordinary. Its lack of response is the evidence that deserves attention.
The same principle applies beyond temperature. A current reading can remain at zero. An output-live indication can remain active. Protection feedback can remain unchanged. Each value may look plausible in isolation while telling an increasingly implausible story.
Sensor Confidence starts with the value, then looks at the behaviour around it.
How it works
Credibility comes from context
Sensor Confidence is designed to build an evidence trail in stages. Each stage asks a different question.
Step 1
Is the signal electrically plausible?
The first check considers whether the signal is present, readable, and within an expected electrical or numerical range.
Step 2
Does it behave credibly over time?
A changing system should not always produce an unchanging reading. Timing, persistence, rate of change, and recovery can all add useful context.
Step 3
Did it respond to a relevant stimulus?
Charging state, current flow, contactor commands, tests, and changing thermal conditions can create an expected response. Sensor Confidence is designed to assess whether the available evidence follows that relationship.
Step 4
Does independent evidence support it?
No single reading should have to tell the whole story when another signal can corroborate or challenge it. Cross-domain comparison turns isolated values into an explainable system view.
The result is not simply “valid” or “invalid”. It is a record of what the charger observed and how strongly the available evidence supports the reading.
Five domains
Five critical relationships
Sensor Confidence applies behaviour and corroboration across five domains to add meaning to an otherwise ordinary reading.
External temperature
Is the reading electrically plausible, and does it behave credibly as operating conditions change? Where independent evidence is available, Sensor Confidence can use it to test whether the temperature story makes sense.
Output current and CT measurement
Does measured output current change consistently with the charger’s operating state and other available electrical evidence? A plausible current reading can still deserve scrutiny when its behaviour does not match the wider system.
CP and current consistency
Does measured current make sense alongside the control-pilot state and the current made available to the vehicle? The relationship matters more than simple equality, because vehicle demand and operating state also shape the expected result.
Contactor and output feedback
Do commanded and reported states agree with independent evidence at the output? Comparing intent with electrical feedback can reveal a relationship that one signal alone cannot.
RCD and protection evidence
Is available protection feedback present, timely, and consistent with the expected state or test sequence? Sensor Confidence does not replace or change authoritative RCD protection.
Evidence architecture
From a reading to credible evidence
A sensor value is more useful when the charger can explain why it should be trusted. Sensor Confidence is designed to assess each reading in context, then record what supports it, what conflicts with it, and what remains unknown.
Sensor and subsystem signals
Readings and state information from the charger.
Electrical and range validation
Is the signal present, readable, and within an electrically plausible range?
Behaviour, timing, and stimulus
Does it change when operating conditions suggest that it should?
Independent and cross-domain corroboration
Do other available signals support or challenge the same conclusion?
Sensor Confidence
Deterministic local evaluation of credibility evidence.
Domain states, findings, and diagnostic telemetry
An explainable record for diagnostics and operational insight.
Sensor and subsystem signals first pass through electrical and range checks. Sensor Confidence then evaluates behaviour, timing, expected response to operating stimulus, and any independent supporting or conflicting evidence. It records domain states, findings, and diagnostic telemetry while established charger protection functions remain separate and authoritative.
FAQ
Questions about Sensor Confidence
What is Sensor Confidence?
Sensor Confidence is a deterministic evidence and plausibility layer for Sensor Confidence Ready PlugStream chargers. It evaluates whether available sensor and subsystem readings behave credibly in context and records the evidence behind its findings.
Why is an in-range reading not always credible?
Range checks look at the value itself. A reading can stay within range while becoming stuck, stale, mistimed, or inconsistent with another signal. Behaviour and corroboration provide the missing context.
What does Sensor Confidence assess?
It assesses behaviour over time, response to operating stimulus, timing, and the support or conflict provided by other available signals. The result is an explainable evidence record rather than an isolated reading.
Does Sensor Confidence replace charger protections?
No. Established thermal, RCD, contactor, PEN, current-limit, charging, and related protection functions remain authoritative. Sensor Confidence adds evidence about reading credibility without replacing those protections.
Which domains does it cover?
Sensor Confidence covers external temperature, output current and CT measurement, CP/current consistency, contactor/output feedback, and RCD/protection evidence.
Does it rely on cloud connectivity?
Sensor Confidence evaluates defined relationships locally on the charger. PlugStream Cloud can receive diagnostic evidence for fleet-level monitoring and insight, but cloud monitoring does not replace local protection.
How is Sensor Confidence related to PlugStream Cloud and Sentinel?
PlugStream Sensor Confidence provides the native device-level evidence story. PlugStream Sentinel uses PlugStream Cloud to extend available evidence into wider estate monitoring and operational insight. Sentinel and PlugStream Cloud do not replace local electrical protection.
From device evidence to fleet insight
Carry a clearer evidence story into PlugStream Cloud
PlugStream Sensor Confidence provides the native device-level evidence story. PlugStream Sentinel is the managed service that uses AmpNexus Sentinel through PlugStream Cloud to help operators understand evidence quality, trends, and operational concerns across supported charging estates.
PlugStream Cloud and Sentinel fleet insight do not replace local electrical protection.