AI Anomaly Detection for EV Charging Networks

Stop Revenue Loss. Protect Your Network. Build User Trust.

About our AI Anomaly Detection Module

EV charging networks are growing fast—but so are the threats. From cloned RFID cards to ghost sessions and hardware faults, hidden anomalies are silently draining revenue and damaging trust. Most systems don’t catch them.

Solidstudio’s AI Anomaly Detection Module was built specifically for Charge Point Operators and eMobility Service Providers to solve this.

Business Impact

Financial Loss Prevention

Financial Loss Prevention

Cloned cards. Double sessions. Implausible travel speeds.

Our module detects fraud at machine speed—before it appears in your ledger. Each detector hunts a different threat, like overlapping sessions at distant stations or hacked payment sequences.

Enhanced User Trust & Security

Enhanced User Trust & Security

Your drivers deserve reliability.

When a user sees “phantom” sessions or unexpected charges, trust erodes fast. By monitoring every session for anomalies and catching issues early, we help you deliver peace of mind and a reputation for safety.

Data-Driven Insights

Data-Driven Insights

Every anomaly contains a signal.

Statistical and predictive detectors spot behavioral shifts that help you:

  • Identify underperforming stations
  • Anticipate peak loads
  • Redesign infrastructure with precision
Operational Efficiency

Operational Efficiency

Manual reviews are slow and error-prone.

Our system automates session monitoring, highlights real threats, and minimizes false alarms—so your team can fix issues, not hunt for them.

Operational Efficiency

Operational Efficiency

The threats evolve. So should your tools.

Our detector-based architecture is modular:

  • Plug it into your existing CPMS
  • Extend with custom models
  • Prioritize the fraud risks that matter most today—it’s built to scale with you, not to hold you back.
Smart, Targeted Detection
flash

Smart, Targeted Detection — How It Works?

The system incorporates multiple specialized Machine Learning detection mechanisms.

Geographical Detector

Stops card cloning and impossible travel.

Built on precise Haversine-distance calculations, this detector compares session coordinates and timestamps to verify that movement between charging events is physically possible.

  • check Flags overlapping sessions from the same RFID at distant locations
  • check Calculates implied travel speed between events
  • check Triggers alerts when the data suggests “teleporting” behavior or bilocation

Statistical Anomaly Detector

Learns what’s normal and spots what’s not.

Using unsupervised models like Z-score and IQR, this detector benchmarks each charging session against typical behaviors for that station, user, and tariff.

  • check Tracks metrics like energy draw, session duration, and start time
  • check Identifies subtle irregularities that don’t follow the bell curve
  • check Suppresses false positives from seasonality or edge-case usage

Predictive Analysis Layer

Looks ahead to stop fraud and faults before they surface.

This engine treats each token’s activity as time-series data forecasting what “should” happen based on historical patterns.

  • check Detects sudden usage spikes, long inactivity periods, or strange sequences
  • check Highlights signs of coordinated fraud, token farming, or hardware degradation
  • check Allows proactive maintenance i.e. swapping a €10 part before it causes a €3,000 failure

What is the role of software within the EV charging ecosystem?

Our CEO, Paweł, shared some valuable insights about the place that software holds within the eMobility environment in one of our latest publications. The ebook breaks down the cases of what are the involved entities, why should they use separate digital tools, what are the challenges of creating EV charging software and many more.

Dekra Certificate

ISO 270001 Certificate

Believing that our continued dynamic growth requires standardization and a systematic approach, including ensuring information security, we have decided to implement, maintain, and improve an Information Security Management System in accordance with the international standard PN-EN ISO/IEC 27001:2017.

Intercharge CHECK Software Certificate

Intercharge CHECK Software Certificate

In pursuit of the highest quality behind our products and their matching with industry standards, we have successfully passed Hubject’s certification. The certification process conducted between Hubject GmbH and Solidstudio sp. z o.o., affirmed the seamless interoperability of our EV charging digital solutions with the Hubject platform.

OCPP Certification Program

OCPP Certification Program

Recognizing the importance of interoperability and adherence to industry standards, we have successfully obtained the official OCPP Certification issued by the Open Charge Alliance. This certification confirms that our implementation is fully compliant with the OCPP 1.6/2.0.1 specification, ensuring reliable communication and seamless integration within the EV charging ecosystem.

We’re here to talk!
Whether you have a complete vision in mind or just exploring options, drop us a message.
location

VISIT US

Karmelicka 27/101, 31-131 Cracow

We need your name and email address to get back to you. Check our Privacy policy to learn more about how we process your personal data and what your rights are.