CPS Anomaly Classification and Diagnostics Under Sensor Failure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for anomaly monitoring in cyber-physical systems (CPS) are not well-equipped to handle the complex task of anomaly diagnostics, classification, filtering, and forecasting, particularly in scenarios where monitoring and measuring instruments (MMI) may fail or be unreliable.
Innovation Solution
The development of automated systems and methods for diagnosing and monitoring anomalies in CPS based on telemetry data, which involves classification of anomalies, diagnostics within each class, and subsequent monitoring to identify anomalies of each class.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If external monitoring systems are deployed, then anomaly detection capability is improved, but system cost and complexity increase
Solution Approach 1:
The automated anomaly diagnostics system serves multiple functions: it monitors anomalies across different classes, performs classification, conducts diagnostics, and provides forecasting. This multi-functional approach consolidates what would otherwise require multiple separate systems into a single unified platform, improving anomaly detection capability while avoiding the complexity multiplication that would result from deploying multiple specialized external monitoring systems.
Solution Approach 2:
The system segments anomalies into different classes based on their characteristics and performs specialized diagnostics for each class. This segmentation allows the system to handle diverse anomaly types efficiently through targeted analysis rather than applying a single complex monitoring approach to all anomalies, thereby improving detection capability while managing system complexity through modular organization.
2Productivity
If automated anomaly diagnostics system is implemented, then diagnostic efficiency is improved, but computational resources required increase
Solution Approach 1:
The system performs preliminary classification of anomalies into different classes before conducting detailed diagnostics. This preliminary action organizes the computational workload efficiently by grouping similar anomalies together, allowing subsequent diagnostics to be more targeted and less computationally intensive than analyzing each anomaly in isolation, thereby improving diagnostic efficiency while managing computational resource consumption.
Solution Approach 2:
The system applies different diagnostic approaches and characteristics tailored to each anomaly class rather than using a uniform diagnostic process for all anomalies. This local quality approach optimizes computational resources by applying the appropriate level and type of analysis specific to each class, improving overall diagnostic efficiency while avoiding the waste of applying excessive computational resources to all anomalies uniformly.
3Measurement precision
If classification of anomalies is performed, then anomaly monitoring precision is improved, but processing time increases
Solution Approach 1:
The system performs preliminary classification of anomalies into distinct classes based on their characteristics before conducting detailed monitoring and diagnostics. This preliminary action establishes an organized framework that enables more efficient subsequent processing, as anomalies within each class can be monitored using class-specific parameters and thresholds, improving monitoring precision without requiring excessive processing time for each individual anomaly.
Solution Approach 2:
The system applies different monitoring characteristics and evaluation criteria tailored to each anomaly class. This local quality approach allows the system to use optimized monitoring parameters specific to each class, improving the precision of anomaly detection and monitoring while reducing the overall processing time compared to applying a single complex monitoring protocol to all anomaly types.
Data Source
Figure 1a
Figure 1b
Figure 1c
AI summary
A method for diagnostics and monitoring of anomalies in a cyber-physical system (CPS) includes obtaining information related to anomalies identified in the CPS. The obtained information includes at least one value of one or more CPS variables. One or more classifying features of the identified anomalies in the CPS are generated based on the obtained information. Classification of the identified anomalies in the CPS into two or more anomaly classes is performed based on the generated classifying features. Each of the two or more anomaly classes is associated with one or more anomaly characteristics. Diagnostics of anomalies are performed in each of the two or more anomaly classes by calculating values of the anomaly characteristics associated with each of the two or more anomaly classes. Anomalies of each of the two or more anomaly classes are monitored based on the calculated values of the anomaly characteristics associated with each of the two or more anomaly classes.