面向能效的炼钢厂数据流实时预测方法实验数据集

本数据集源于面向炼钢厂能效管理的专项研究，研究在工业物联网技术支撑下完成，聚焦高能耗钢铁行业的能效实时监控与未来趋势预测，旨在破解行业能效管理碎片化、数据采集高延迟、预测模型低精度的核心难题。据集核心内容包括基于MQTT + OPC DA/UA 协议的炼钢厂边端设备能耗采集体系，以及基于改进粒子群优化的多核正则化极限学习机（IPSO-MKELM）能效预测模型。数据集包含1个文件夹，名称为面向能效的炼钢厂数据流实时预测方法实验数据集，容量31.7MB，内含1份专利、1个炼钢厂每日能效预测方法实验数据文件、全流程能源仿真系统和数据流实时预测系统与方法说明文档以及数据说明文件，同时包含2个子文件夹分别为预测模型代码和训练结果。

This dataset is derived from a targeted study on energy efficiency management for steel mills, which was carried out with the support of Industrial Internet of Things (IIoT) technologies. It focuses on real-time energy efficiency monitoring and future trend prediction for the energy-intensive iron and steel industry, aiming to address the core challenges of fragmented energy efficiency management, high latency in data collection, and low prediction accuracy prevalent in the sector.

The core contents of the dataset include the energy consumption acquisition system for edge devices in steel mills based on the MQTT + OPC DA/UA protocols, as well as the energy efficiency prediction model of multi-kernel regularized Extreme Learning Machine (MKELM) optimized by improved Particle Swarm Optimization (IPSO), abbreviated as IPSO-MKELM.

The dataset contains one top-level folder named "Experimental Dataset for Real-Time Data Stream Prediction Methods for Energy Efficiency-Oriented Steel Mills", with a total size of 31.7 MB. Inside this folder, there is one patent document, one experimental data file for daily energy efficiency prediction methods in steel mills, documentation for the full-process energy simulation system, real-time data stream prediction system and related methods, as well as a data description document. In addition, two subfolders are included, namely "Prediction Model Code" and "Training Results".

搜集汇总

背景与挑战

背景概述

该数据集源自炼钢厂能效管理研究，包含基于MQTT + OPC DA/UA协议的能耗采集体系以及改进的IPSO-MKELM能效预测模型。数据集提供了实验数据、系统文档和预测代码，用于支持能效实时监控与预测。