Drive Unit Monitoring With True-Positive Fault Discrimination
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing monitoring methods for drive devices, particularly internal combustion engines, rely heavily on large amounts of data and complex algorithms, leading to high hardware and software costs, and often result in unreliable false-positive detections, which can cause unnecessary alarms and maintenance, without adequately addressing individual drive system characteristics.
Innovation Solution
A monitoring method that qualifies fault detection results as either true-positive or false-positive, using a discrimination unit to send genuine results to the control connection for immediate action and artificial results to a feedback loop for algorithm improvement, reducing reliance on extensive data and customized models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large amounts of data and complex algorithms are used for monitoring, then measurement precision and reliability improve, but device complexity and costs increase
Solution Approach 1:
The monitoring system is segmented into multiple independent algorithms (e.g., knock detection, combustion analysis, anomaly detection) that operate separately but contribute to overall fault detection. This modular approach improves measurement precision through specialized algorithms while managing device complexity by dividing the monitoring task into manageable segments.
Solution Approach 2:
The monitoring system employs universal algorithms that can detect multiple types of faults using the same sensor data infrastructure. The control device performs diverse monitoring functions (knock detection, combustion quality assessment, anomaly identification) using a unified data acquisition system, thereby improving measurement precision across different fault types without proportionally increasing device complexity.
2Measurement precision
If complex monitoring algorithms are applied, then measurement precision improves, but loss of time increases due to data processing
Solution Approach 1:
The system performs preliminary processing of sensor data by continuously monitoring combustion parameters and pre-identifying potential anomalies before they develop into critical faults. The control device maintains running assessments of combustion quality and knock patterns, enabling rapid fault detection without requiring extensive post-processing time.
Solution Approach 2:
When anomalies are detected, the system rushes through the data processing pipeline by prioritizing critical fault detection algorithms and bypassing less relevant analysis steps. The control device can skip detailed analysis of normal operating conditions and focus computational resources on detecting and confirming fault conditions, thereby reducing overall processing time while maintaining measurement precision.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a monitoring method for the operation of a number of one or more drive units, in particular for the operation of an internal combustion engine, wherein a drive unit comprises a control unit and an operating parameter of the drive unit is monitored based on operating data acquired by the control unit, comprising the steps of: - monitoring the operating parameter of the drive unit with a monitor algorithm for the operating data, in particular a monitor algorithm of an observer module of the control unit, - applying a fault detection algorithm for the acquired operating data of a fault detection system to detect a fault concerning the operating parameter of the drive unit, - sending a fault indication from the fault detection system for the drive unit based on a result of the fault detection algorithm;- Receiving the fault indication and checking the result, in particular receiving and checking in a server facility of a control instance, - activating a control connection to the drive device with respect to the operating parameter and feeding back a monitoring value for the operating parameter to a feedback connection to the monitor algorithm. According to the invention, it is provided that the result is qualified as false positive (artificial) or true positive (genuine) after testing, and - the fault detection has a discrimination unit that - sends the result qualified as true positive (genuine), in particular only the result qualified as false positive (artificial), to the control connection to an actuator and - sends the result qualified as false positive (artificial), in particular only the result qualified as false positive (artificial), to the feedback connection to the observer module.