Ship Fuel Switch Detection Using Distributed SO2 Sensor Units
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for enforcing Sulfur Dioxide (SO2) exhaust emissions regulations in the maritime industry are ineffective, as existing detection systems are expensive, limited in coverage, and lack reliable means to confirm fuel switching compliance within Sulfur Emission Control Areas (SECA) boundaries, leading to challenges in identifying non-compliant ships.
Innovation Solution
A distributed sensor system using neural networks and blockchain-secured data integrity, deployed on ships with self-contained, self-powered units that collect and process SO2, CO2, and other data to detect fuel switching patterns, providing corroborating evidence for compliance and reducing the burden on cloud resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If permanent ship-mounted sensors are deployed globally, then monitoring coverage and compliance detection capability are improved, but installation and maintenance costs become prohibitively expensive
Solution Approach 1:
The system divides the monitoring function into multiple independent, low-cost sensor units distributed across the ship's exhaust system. Each unit independently samples and processes data, eliminating the need for a single expensive centralized sensor system. This segmentation allows global deployment while maintaining reliable compliance detection through aggregated data from multiple units.
Solution Approach 2:
Instead of using one expensive laboratory-grade sensor, the system employs multiple copies of simpler, cheaper sensor units. These units replicate the basic sensing function but at a fraction of the cost, allowing widespread installation without prohibitive expense while collectively providing the same monitoring reliability as a single high-end system.
2Productivity
If cloud computing resources are used to process sensor data, then data processing capability is improved, but system cost and resource requirements scale proportionally with deployment size
Solution Approach 1:
The data processing function is segmented and distributed to each sensor unit through embedded neural networks. Each unit independently processes its own data locally, filtering and analyzing emissions patterns without requiring centralized cloud processing. This distributes the computational load across many simple units rather than concentrating it in expensive cloud infrastructure.
Solution Approach 2:
Each sensor unit is self-sufficient with onboard processing capabilities that allow it to independently analyze its own data and make local determinations about compliance. The units self-manage their data processing needs through embedded intelligence, reducing or eliminating the need for external cloud computing resources while maintaining high processing capability.
3Area of stationary object
If distributed sensor units are deployed on each ship, then monitoring coverage is improved, but data transmission and power requirements increase
Solution Approach 1:
The sensor units perform sampling and data processing in periodic cycles rather than continuously. Each unit activates sensors and processing functions at scheduled intervals, allowing the ship to operate with minimal power consumption between cycles. This periodic operation maintains monitoring coverage while dramatically reducing average power requirements compared to continuous operation.
Solution Approach 2:
The sensor units harvest energy from the ship's exhaust heat through thermoelectric generators, making them self-powered. This energy harvesting capability allows the units to operate independently without drawing significant power from the ship's main system, enabling widespread deployment without proportionally increasing the ship's power burden.
Data Source
AI summary
A system for the maritime shipping industry to aid enforcement of the Sulfur Dioxide (SO2) exhaust emissions regulations which uses neural networks and a novel sampling process to detect and record compliant operation of a ship regarding the fuel switching aspect of the regulation. The processing load of neural network training can be distributed over multiple identical self-contained, self-powered, self-communicating sensor units on each of the monitored ships.