Sensor Placement for Hazardous Substance Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Industrial plants with hazardous chemicals face challenges in effectively detecting chemical leaks before they affect surrounding communities, as existing sensor placement methods are inefficient in identifying optimal sensor locations and coverage areas.
Innovation Solution
A system and method that calculate the optimal placement of sensors around industrial plants by identifying hazard points, simulating chemical releases, and determining sensor locations based on toxic levels of concern, wind direction, and dispersion models to ensure early detection and maximum coverage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensors are placed to ensure maximum coverage of hazard points, then detection reliability is improved, but the number of sensors required increases
Solution Approach 1:
The patent applies local quality by placing sensors at specific locations on the fenceline where they can detect releases from multiple hazard points. Instead of uniform distribution, sensors are strategically positioned based on local geometric relationships between hazard points and fenceline segments, allowing each sensor to serve multiple purposes and reducing total sensor count while maintaining coverage reliability.
Solution Approach 2:
The patent implements universality by designing a sensor placement system where each sensor can detect releases from multiple hazard points simultaneously. The optimization algorithm identifies sensor locations that provide multi-point coverage, making each sensor serve multiple functions and reducing the overall number of sensors needed while maintaining comprehensive detection capability.
2Loss of time
If sensor placement is optimized for early detection, then response time is improved, but computational complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the fenceline into discrete segments and associating each segment with specific hazard points. This segmentation allows the optimization algorithm to process sensor placement decisions in manageable units, reducing computational complexity while still achieving early detection capability by ensuring each segment has appropriate sensor coverage for its associated hazards.
Solution Approach 2:
The patent implements preliminary action by pre-calculating optimal sensor locations based on the geometric relationships between hazard points and fenceline segments before actual deployment. This advance optimization reduces the computational burden during operation, as the sensor positions are predetermined based on early detection requirements, allowing rapid response without real-time computational complexity.
3Adaptability or versatility
If sensors are placed to cover all wind directions, then detection coverage is improved, but sensor density increases
Solution Approach 1:
The patent applies asymmetry by recognizing that not all wind directions require equal sensor coverage. The optimization algorithm identifies dominant wind directions and places sensors asymmetrically to maximize coverage in those directions, rather than distributing sensors uniformly in all directions. This asymmetric placement reduces sensor density while maintaining adaptability for the most critical wind scenarios.
Data Source
AI summary
A system and method for placement of sensors for sensing hazardous substances released from a plurality of hazard points. A processor identifies a location of a hazard point, a fenceline of the plant-site, and a toxic level of concern (LOC) for the hazardous substance. The processor calculates a minimum amount of the hazardous substance (Q) for which a concentration at a centerline of a plume carrying the hazardous substance reaches the toxic LOC at the fenceline, and simulates a release of the hazardous substance in the calculated amount Q from the hazard point. The processor further calculates a pair of sensor locations where the concentration of the plume is equal to the minimum detectable concentration level of sensor based on the simulated release. The pair of sensor locations is then output by the processor.