首页> 外文会议>SME annual meeting and exhibit >PERFORMANCE OPTIMIZATION OF A COAL PREPARATION PLANTUSING GENETIC ALGORITHMS
【24h】

PERFORMANCE OPTIMIZATION OF A COAL PREPARATION PLANTUSING GENETIC ALGORITHMS

机译:煤炭制备规划遗传算法的性能优化。

获取原文

摘要

Modern coal processing plants utilize multiple cleaningcircuits to efficiently beneficiate different size coalfractions of run-of-mine coal. An older plant optimizationapproach is to maintain average product quality fromindividual cleaning circuits at the same level as the givenproduct specifications for the overall plant. However, manypast studies indicate that the equalization of product qualityapproach fails to produce the maximum plant-yield. Anewer approach, known as equalization of incrementalproduct quality, requires that the incremental productquality obtained from each circuit be maintained at thesame level to satisfy the overall plant product quality. Thisapproach ensures the maximization of plant yield whilesatisfying a single product quality constraint. However,while dealing with multiple product quality constraints, thisapproach by itself may not be sufficient to obtain thedesired maximum yield. For the simple reason that thedirtiest particle (or group of particles) with respect to onequality constraint may not be the same particle (or group ofparticles) with respect to another quality constraint, themass yield versus product quality relationship generated byequalizing different incremental product quality may not beexactly the same. Therefore, an additional search techniquehas to be utilized to determine the global-maximum valueof plant yield.Thus, the main objective of this study was to utilize anemerging optimization technique, known as geneticalgorithms (GA) to maximize plant yield while satisfyingmultiple product quality constraints. The maximum plantyield obtained from this approach was nearly same as themaximum yield obtained by the incremental product qualityapproach while satisfying one specific product qualityconstraint. The GA was applied on a coal preparation plantthat utilizes four circuit operations ? heavy medium bath,heavy medium cyclone, spiral and froth flotation. Theresults showed that using GA as an optimization processgives 2.56% higher yield that will result in additionalrevenue generation of $5,120,000 per annum than averageproduct quality approach.
机译:现代煤炭加工厂利用多次清洁 高效选矿的电路 煤的一部分。较旧的工厂优化 方法是保持产品的平均质量 与给定水平相同的单个清洁回路 整个工厂的产品规格。但是,很多 过去的研究表明,产品质量均等化 该方法无法产生最大的植物产量。一种 更新的方法,称为增量均衡 产品质量,要求增量产品 从每个电路获得的质量保持在 达到相同水平,以满足工厂整体产品质量。这 该方法可确保最大程度地提高植物产量,同时 满足单个产品质量约束。然而, 在处理多个产品质量限制时,这 本身的方法可能不足以获取 所需的最大产量。出于简单的原因, 相对于一个而言最脏的粒子(或一组粒子) 质量约束可能不是同一粒子(或 粒子)相对于另一个质量约束, 产生的批量产量与产品质量的关系 均衡不同的增量产品质量可能不会 一模一样。因此,另一种搜索技术 必须使用以确定全局最大值 植物产量。 因此,这项研究的主要目的是利用 新兴的优化技术,称为遗传 算法(GA),可在满足以下条件的同时最大化植物产量 多种产品质量限制。最大植物 通过这种方法获得的产量几乎与 通过提高产品质量获得最大的产量 同时满足一种特定产品质量的方法 约束。遗传算法应用于选煤厂 利用四个电路操作?重介质浴, 重介质旋风分离器,螺旋浮选器和泡沫浮选器。这 结果表明,使用遗传算法作为优化过程 使产量提高2.56%,这将导致额外的 每年平均创收$ 5,120,000 产品质量方针。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号