OPTICS Reachability Diagrams for Real-Time Sensor Anomaly Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing AI-based anomaly detection methods for technical systems fail to efficiently model multiple normal states and classify newly acquired sensor data points as normal or abnormal in real-time, particularly when time dependency is not considered, leading to high computational effort and inefficiency.

Innovation Solution

A density-based clustering function, specifically the OPTICS method, is adjusted to process new sensor data points by determining a predefined initial sequence using a reachability diagram, reducing computational effort and improving reliability through flexible parameter adjustment and rapid classification.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If clustering methods are used to model multiple normal states, then the ability to identify different operating modes is improved, but the computational effort required to classify each new data point increases significantly

Engineering Contradiction:
Improveability to model multiple normal statesVSAvoidcomputational effort
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent pre-calculates and stores the reachability diagram based on training data before actual monitoring begins. This preliminary clustering structure captures the relationships between different normal operating states, so that during real-time monitoring, new data points can be classified by simply querying the pre-built structure rather than performing full clustering computations, thus resolving the contradiction between modeling multiple states and computational complexity

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent segments the data processing into two distinct phases: an offline training phase where the reachability diagram is constructed from training data, and an online monitoring phase where new data points are classified using the pre-built structure. This segmentation allows complex computations to be performed only once during training, while runtime classification becomes computationally efficient

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If the entire clustering model is recalculated for each new data point, then classification accuracy is maintained, but the processing time increases and real-time detection becomes difficult

Engineering Contradiction:
Improveclassification accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The reachability diagram is constructed in advance during an offline training phase using training data that represents normal operating states. This pre-computed structure encodes the clustering information needed for accurate classification, allowing new data points to be classified rapidly during online monitoring without requiring recalculation of the entire clustering model, thus maintaining accuracy while reducing processing time

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system dynamically adapts to new data points by inserting them into the pre-built reachability diagram at appropriate positions based on their characteristics, rather than static recalculation. This dynamic insertion approach maintains classification accuracy for evolving operating conditions while avoiding the computational overhead of complete model recalculation

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4165484B1Monitoring device and method for detection of abnormalities
Publication Date: 2025.10.29 SIEMENS AG
  • EP4165484B1 patent drawingFigure 1
  • EP4165484B1 patent drawingFigure 2~3
  • EP4165484B1 patent drawingFigure 4~5

AI summary

Monitoring device for detecting anomalies on the basis of sensor data (102, z) of a technical system (100) which are continuously determined over time, comprising an input interface (201) which is designed to receive a new sensor data point (102, z) of the technical system (100), an analysis unit (202) which is designed to determine a minimum reachability distance (minRD) for the new sensor data point (102, z) with respect to a selection of adjacent training data points (x1, x2, x3) and to determine a minimum position (xi) in the sequence of training data points for the new sensor data point (102, z) in a reachability diagram (312) which was created by means of a density-based clustering function using predetermined training data points as input values, to insert the new sensor data point (102, z) in the reachability diagram after the minimum position (xi) and before the training data point having a greater reachability distance than the determined minimum reachability distance (minRD), to assign the new sensor data point (102, z) to a particular cluster on the basis of its position in the supplemented reachability diagram and to assign a status as normal or abnormal on the basis of the minimum reachability distance (minRD), and an output interface (203) which is designed to output the assignment determined for the new sensor data point (102, z) as the result of the anomaly detection.