Fluid Machine Anomaly Cause Estimation from External Sensor Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fluid machines like compressors face challenges in estimating the cause of anomalies due to the difficulty in installing sensors inside the machine, leading to incomplete measurement data and increased costs from high sampling frequency monitoring.
Innovation Solution
A cause-of-anomaly estimation device that includes a measurement value input unit, a performance calculation unit, an anomaly determination unit, a cause-of-anomaly estimation unit, and a results output unit, which uses pre-stored data in a cause-of-anomaly database to estimate the cause of anomalies based on measured values and performance calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensors are installed inside the fluid machine to obtain direct measurement data, then measurement precision is improved, but device complexity and installation difficulty increase due to the sealed casing structure
Solution Approach 1:
The patent uses external sensors as intermediaries to measure parameters (temperature, pressure, flow rate) that reflect the internal state of the fluid machine without requiring direct installation inside the sealed casing. These external measurements serve as mediators to infer internal conditions, resolving the contradiction between measurement precision and installation complexity.
Solution Approach 2:
The system creates a virtual model or copy of the internal state by measuring external parameters and calculating performance indicators that replicate the information obtainable from direct internal sensors, avoiding the need for physical installation inside the machine while maintaining diagnostic capability.
2Measurement precision
If high sampling frequency monitoring is implemented to capture detailed operational data, then anomaly detection precision is improved, but energy consumption and data processing burden increase
Solution Approach 1:
The system applies anomaly determination algorithms that focus on detecting specific anomaly patterns rather than continuously analyzing all data points at high frequency. By targeting only relevant anomalies and using calculated performance indicators, the system achieves sufficient detection precision without the energy cost of continuous high-frequency monitoring of all parameters.
Solution Approach 2:
The system transforms raw measurement data into calculated performance indicators that capture the essential operational state. This parameter transformation allows for lower sampling frequencies while maintaining anomaly detection capability, as the derived indicators consolidate information that would otherwise require high-frequency raw data monitoring.
3Loss of information
If multiple sensors are installed to monitor internal state comprehensively, then information completeness is improved, but cost and system complexity increase
Solution Approach 1:
The system uses a small set of external sensors that measure multiple parameters (temperature, pressure, flow rate) which collectively provide comprehensive information about the internal state. These multi-functional measurements replace what would require multiple specialized internal sensors, reducing system complexity while maintaining information completeness.
Solution Approach 2:
The system infers internal state information by measuring parameters from external dimensions (outside the casing) rather than requiring direct internal measurements. This dimensional shift allows comprehensive monitoring through external parameter combinations that correlate with internal conditions, avoiding the complexity of internal sensor installation.
Data Source
AI summary
In order to obtain a cause-of-anomaly estimation device for a fluid machine that enables simple and convenient, and highly precise estimation of a cause of an anomaly of the fluid machine, the present invention includes: a measurement value input unit to which a measurement value obtained from a sensor installed in the fluid machine is input; a performance calculation unit that calculates the performance of the fluid machine from the measurement value obtained by the measurement value input unit; an anomaly determination unit that determines the presence or absence of an anomaly in the measurement value obtained by the measurement value input unit and in the performance of the fluid machine as calculated by the performance calculation unit; a cause-of-anomaly estimation unit that estimates a cause of an anomaly of the fluid machine on the basis of a determination result regarding the presence or absence of an anomaly in the performance of the fluid machine as determined by the anomaly determination unit, and a cause of an anomaly of the fluid machine pre-stored in a cause-of-anomaly database; and a results output unit that outputs a cause-of-anomaly estimation result regarding the fluid machine as estimated by the cause-of-anomaly estimation unit.