Data Thinning of Ship Engine Room Equipment Based on Improved Douglas-Peucker Algorithm

With the widespread application of ship engine room monitoring systems, vast amounts of engine room equipment data are stored in shore-based systems, leading to increased costs associated with data storage, querying, and processing. Although the traditional Douglas-Peucker (DP) algorithm effectively...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:International Conference on Transportation Information and Safety (Online) s. 1196 - 1203
Hlavní autoři: Dong, Shengli, Jiang, Chenji
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 16.07.2025
Témata:
Approximation algorithms
automated threshold selection
Costs
data compression
Data models
Douglas-Peucker algorithm
Engines
Heuristic algorithms
Load modeling
Manuals
Marine vehicles
Monitoring
Real-time systems
ship engine room equipment data
spatial similarity
ISSN:2832-899X
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:With the widespread application of ship engine room monitoring systems, vast amounts of engine room equipment data are stored in shore-based systems, leading to increased costs associated with data storage, querying, and processing. Although the traditional Douglas-Peucker (DP) algorithm effectively retains key information from the original data, its reliance on manual threshold selection introduces inefficiency, high subjectivity, and poor adaptability. To address this issue, this paper designs an improved Douglas-Peucker algorithm based on spatial similarity for threshold setting. On the basis of the Douglas-Peucker algorithm, an improved formula is proposed to convert absolute thresholds into relative thresholds. The compression quality of the simplified data is evaluated based on data offset and simplification quantity, and the spatial similarity between the data before and after simplification under different thresholds is calculated. A mapping model between relative thresholds and spatial similarity for multiple equipment data sources is established. Based on this model, the corresponding threshold can be calculated through spatial similarity. Experimental results show that the proposed method achieves a compression ratio of approximately 1% for engine room equipment data (e.g., main engine load), with the error between the target and actual spatial similarity being less than 0.4%. The designed algorithm can effectively obtain the corresponding threshold and efficiently compress engine room equipment information, which helps alleviate the operational pressure of engine room monitoring systems and provides technical reference for the construction of intelligent ship engine room systems.
ISSN:2832-899X
DOI:10.1109/ICTIS68762.2025.11214982