AI賦能的MoE油庫高效并行風險預測系統(tǒng)設(shè)計與實現(xiàn)

Design and Implementation of an AI-empowered Efficient Parallel Risk Prediction System for MoE oil Depots

Environment and Resource / 2024,6(4)：121-128 / 2025-03-07 36 40

查看全文 下載文檔

• 作者： 張韶田1      何俊霖1      王一婷1      黃李元鈞2      李滿弈1     
• 單位：

  1. 重慶科技大學石油與天然氣工程學院，重慶；
  2. 重慶科技大學電子智能材料與器件研究中心，重慶

• 關(guān)鍵詞： 人工智能；混合專家模型（MoE）；油庫安全管理；風險預測
• Artificial intelligence; Hybrid Expert Model (MoE); Oil depot safety management; Risk prediction; Efficient parallelism
• 摘要： 隨著人工智能技術(shù)的飛速發(fā)展，其在工業(yè)領(lǐng)域的應用日益廣泛。油庫作為能源儲存和轉(zhuǎn)運的重要設(shè)施，其安全管理直接關(guān)系到國家能源安全和環(huán)境保護。本文旨在探討如何利用混合專家模型（MoE）與人工智能技術(shù)，設(shè)計并實現(xiàn)一個高效并行的風險預測系統(tǒng)，以提升油庫的安全管理水平。該系統(tǒng)通過實時數(shù)據(jù)采集、智能分析與預測，能夠提前識別并預警潛在風險，為油庫的應急響應和安全管理提供有力支持。
• With the rapid development of artificial intelligence technology, its application in the industrial field is becoming increasingly widespread. As an important facility for energy storage and transportation, the safety management of oil depots is directly related to national energy security and environmental protection. This article aims to explore how to use a hybrid expert model (MoE) and artificial intelligence technology to design and implement an efficient parallel risk prediction system to improve the safety management level of oil depots. The system can identify and warn potential risks in advance through real-time data collection, intelligent analysis, and prediction, providing strong support for emergency response and safety management of oil depots.
• DOI： https://doi.org/10.35534/er.0604011
• 引用： 張韶田，何俊霖，王一婷，等．AI賦能的MoE油庫高效并行風險預測系統(tǒng)設(shè)計與實現(xiàn)［J］．環(huán)境與資源，2024，6（4）：121-128．