{"title":"pymops:基于多代理模拟的电力调度优化软件包","authors":"Awol Seid Ebrie , Young Jin Kim","doi":"10.1016/j.simpa.2024.100616","DOIUrl":null,"url":null,"abstract":"<div><p>In response to the NP-hard power scheduling problem, the <span>pymops</span> package is developed as a robust solution. The package operates within a multi-agent simulation environment, where power-generating units are represented as reinforcement learning (RL) agents. The environment is designed to account for a comprehensive range of constraints. It also accommodates thermal valve point effects (VPEs) within cost and emissions functions. Moreover, in cases of constraint violations, the environment makes real-time contextual adjustments. Within the environment, the power scheduling problem is broken down into sequential Markov decision processes (MDPs), which serve as inputs for training a deep RL model aimed at solving the optimization problem.</p></div>","PeriodicalId":29771,"journal":{"name":"Software Impacts","volume":null,"pages":null},"PeriodicalIF":1.3000,"publicationDate":"2024-01-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S2665963824000046/pdfft?md5=651e2ac4e426bc6fbbf5275437df2a6b&pid=1-s2.0-S2665963824000046-main.pdf","citationCount":"0","resultStr":"{\"title\":\"pymops: A multi-agent simulation-based optimization package for power scheduling\",\"authors\":\"Awol Seid Ebrie , Young Jin Kim\",\"doi\":\"10.1016/j.simpa.2024.100616\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><p>In response to the NP-hard power scheduling problem, the <span>pymops</span> package is developed as a robust solution. The package operates within a multi-agent simulation environment, where power-generating units are represented as reinforcement learning (RL) agents. The environment is designed to account for a comprehensive range of constraints. It also accommodates thermal valve point effects (VPEs) within cost and emissions functions. Moreover, in cases of constraint violations, the environment makes real-time contextual adjustments. Within the environment, the power scheduling problem is broken down into sequential Markov decision processes (MDPs), which serve as inputs for training a deep RL model aimed at solving the optimization problem.</p></div>\",\"PeriodicalId\":29771,\"journal\":{\"name\":\"Software Impacts\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.3000,\"publicationDate\":\"2024-01-29\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.sciencedirect.com/science/article/pii/S2665963824000046/pdfft?md5=651e2ac4e426bc6fbbf5275437df2a6b&pid=1-s2.0-S2665963824000046-main.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Software Impacts\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S2665963824000046\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"COMPUTER SCIENCE, SOFTWARE ENGINEERING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Software Impacts","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2665963824000046","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, SOFTWARE ENGINEERING","Score":null,"Total":0}
pymops: A multi-agent simulation-based optimization package for power scheduling
In response to the NP-hard power scheduling problem, the pymops package is developed as a robust solution. The package operates within a multi-agent simulation environment, where power-generating units are represented as reinforcement learning (RL) agents. The environment is designed to account for a comprehensive range of constraints. It also accommodates thermal valve point effects (VPEs) within cost and emissions functions. Moreover, in cases of constraint violations, the environment makes real-time contextual adjustments. Within the environment, the power scheduling problem is broken down into sequential Markov decision processes (MDPs), which serve as inputs for training a deep RL model aimed at solving the optimization problem.