Hydraulic Event Detection for Predictive Maintenance Diagnostics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current hydraulic system maintenance practices are inefficient due to the lack of domain expertise and resources, leading to reactive maintenance and accelerated system wear, resulting in costly downtime and reduced production output.
Innovation Solution
A sensor system and platform for signal analysis that monitors hydraulic equipment at subcomponent and global levels, using a single set of sensors to provide real-time monitoring, forecasting, and troubleshooting capabilities, enabling predictive maintenance and extended equipment lifespan.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If distributed sensors and custom algorithms are used for full system monitoring, then measurement precision and reliability improve, but device complexity and cost increase
Solution Approach 1:
The patent applies universality by using a single set of sensors to monitor multiple hydraulic subcomponents (pump, motor, valves, actuators) through signal disaggregation. Instead of deploying distributed sensors for each component, the system processes signals from one location to diagnose the entire hydraulic system, reducing device complexity while maintaining diagnostic capability across all subcomponents
Solution Approach 2:
The patent applies segmentation by dividing the hydraulic system into subcomponent-level diagnoses (pump, motor, valves, actuators) through signal disaggregation. The system segments the overall hydraulic signal into component-specific signatures, enabling precise diagnosis of individual subcomponents without requiring physical segmentation of sensors across the system
2Productivity
If reactive maintenance is practiced without proper monitoring tools, then device complexity remains low, but productivity and reliability deteriorate due to unexpected downtime
Solution Approach 1:
The patent applies preliminary action by implementing continuous monitoring and predictive analytics to identify potential failures before they occur. The system analyzes hydraulic signals in real-time to forecast equipment degradation and schedule maintenance proactively, preventing unexpected downtime and maintaining both productivity and reliability
Solution Approach 2:
The patent applies feedback by continuously monitoring hydraulic system signals and using machine learning models to compare actual performance against expected patterns. The system provides real-time feedback on equipment health status and predicts future failures, enabling operators to take corrective actions before productivity or reliability deteriorates
3Ease of repair
If technicians replace components reactively without diagnostic tools, then ease of repair improves, but loss of time and productivity worsen due to inability to resolve root causes
Solution Approach 1:
The patent applies preliminary action by providing diagnostic information before repair is needed. The system identifies and flags potential issues with specific subcomponents through signal analysis, allowing technicians to prepare appropriate repair strategies in advance and resolve root causes rather than simply replacing components reactively, thereby reducing downtime
Solution Approach 2:
The patent applies intermediary by introducing an AI-based diagnostic system that mediates between raw hydraulic signals and technician decision-making. The system processes complex signal patterns and provides interpretable diagnostic information, enabling technicians to make informed repair decisions without requiring deep domain expertise, thus improving ease of repair while reducing downtime through accurate root cause identification
Data Source
AI summary
A system includes sensors for monitoring signals, and a processing system executes one or more methods for identification of system events, from the signals, corresponding to state changes and performance of the system and/or its subcomponents. Event identification is performed with classification and/or other machine learning algorithms, with generation of novel training data sets. The sensor(s) can also be used to determine power consumption information about the system and/or its subcomponents. The system processes event-associated outputs for execution of actions for improving system performance, along with other downstream applications.