Communication

Intelligent Wind Turbine Prediction System

Extra Form
Country	China
Year	2023
Award	WINNER
Client	Fourth Paradigm (Beijing) Data & Technology Co., Ltd.
Affiliation	Jianxin Technology Co., Ltd.
Designer	Yun Xu, Wei Wang, Shaoxiong Bai, Xinjie Li, Yanfen Zhang
English	Through the visualization of wind turbine models and data graphs, we can quickly pinpoint the fault in wind turbines and efficiently safeguard the operations and maintenance process. Through the “drag and drop” data and model settings, we can meet the custom settings of “same type and same module” and “multiple types and multiple modules” from various machine learning models to hone targeted operations and maintenance trouble-shooting capabilities.
Native	利用三维可视化建模、虚拟仿真、监控可视化等技术手段将物理现实空间在数字系统中智能展示，提供可解释、可溯源的机器学习能力，模拟用户物理世界中真实的交互操作逻辑。帮助相关企业故障预警平均提前20分钟，整体模型准确率达到81%。智能风机预测系统内置了第四范式 HyperCycle 机器学习底层技术，帮助用户拥有独立的数据回流和机器学习模型自学习的能力。本系统是在HyperCycle 机器学习方法论和智能运营决策的指导下，面向风力发电新能源行业的人工智能解决方案。智能风机预测系统做到了构建风机健康模型，分析风机健康影响因素，为风机故障定向检修、制定维护计划、经济运行等提供科学依据，帮助全球加快完成碳中和提供数据智能服务，推动社会发展。基于风力新能源场景下专有机器学习模型，具有独立的自学能力，同时和整体模型中心进行打通，满足客户能够针对“同型同模”和“多型多模”的不同需求完成快速建模和低门槛项目落地。
Positive Comments