当前位置:首页 > 报告详情

将机器学习与冶金学见解相结合实现可持续的电弧炉性能.pdf

上传人: 可*** 编号:991960 2025-12-07 21页 1.28MB

1、AI-Driven Optimization Framework for EAF OperationsNarottam BeheraDr.Hany HamedEMSTEEL R&DPublic 2EMSTEEL GROUPEmirates steel process routeEmirates steel process routeRaw MaterialDirect ReductionSteel Making (EAF)Ladle Treatment Billet and beam blank)2 MTPA0.5 MTPA1.0 MTPA4.2 MTPA3.6 MTPAPublic 3EMS

2、TEEL GROUPEmsteel Decarbonization StrategyEmsteel Decarbonization StrategySteel Business performance vs global industry(2024)2.331.921.431.050.870.6700.511.522.5BF-BOF-WSA*Global -WSADRI-EAF-WSAES Emission-WSA(Without offsets)ES Emission-WSA(With offsets)ES Scope 1&2(GHGprotocol)*Net Zero-2050 carbo

3、n intensity in the sion compared to WSA averageSteel Business2030 CO2e:1.9Mt 2023 CO2e:2.2Mt 2019 CO2e:3.2Mt 32%Reduction achieved40%Reduction Target setLevers deployed in 2024Increase our clean and renewable electricity through“Energy Attribute Certificates”1st Steel maker worldwide to capture part

4、 of our CO2 emissions 1st of-a-kind MENA region green hydrogen to green steel pilot facilityRoad to 2030-Key InitiativesCarbon CaptureAlternative Fuel SwitchingUsed recycled Steel ScrapClean and renewable EnergyElectrificationEnergy EfficiencyDigital TransformationPublic 4EMSTEEL GROUPProblem Statem

5、entProblem StatementFirst Measured Temp.DRI Feed Rate T/min1500155016001650145017005101520253035404550Time(min)Temperature oC234516DRI Feeding(T/min)Energy LossTap TempTarget(1600-1620 o COptimumTemp.Profile234516Tap Temp.DRI CompositionCarbon:2-2.5%Fe(Metallic):82-85%DRI FeO%:5-8%Hot DRI temp:400-4

6、50 deg CEAF SpecificationsTap weight-150 TonTransformer capacity-130 MVACharge mix Hot DRI,Cold DRI and ScrapTap to Tap time 40-50 Min Four injectors(Natural gas,oxygen and carbon)M150 Ton+30 Ton Hot HeelOxygen and carbon injectionRate 3-5 t/min130 MVA250 t/hrNo continuous bath temperature measureme

word格式文档无特别注明外均可编辑修改,预览文件经过压缩,下载原文更清晰!
三个皮匠报告文库所有资源均是客户上传分享,仅供网友学习交流,未经上传用户书面授权,请勿作商用。
全文主要介绍了AI驱动的电弧炉(EAF)优化框架,旨在提高钢厂效率和减少碳排放。关键点如下: 1. **目标**:通过优化EAF操作,减少特定能量消耗、提高生产率和降低CO2排放。 2. **技术**:结合机器学习(Gaussian Process Regression, GPR)和遗传算法(GA)进行预测和优化。 3. **成果**: - 能量消耗减少10 kWh/吨。 - 生产率提高7吨/小时。 - CO2排放减少10公斤/吨钢。 4. **方法**: - 使用GPR预测特定能量、电弧时间和炉渣FeO%。 - GA优化操作参数,如氧气、碳、石灰和白云石添加量。 5. **未来工作**: - 扩展到动态时间序列建模。 - 开发EAF数字孪生。 - 实施多目标优化。 - 推进Industry 4.0集成。
"EAF优化框架揭秘" "AI助力钢铁业减碳新突破" "钢铁生产效率革命性提升!"
客服
商务合作
小程序
服务号
折叠