QQ客服 官方微博 反馈留言

题名	面向家庭能源管理平台的用电策略研究
作者	牛丽肖
学位类别	硕士
答辩日期	2014-05-28
授予单位	中国科学院沈阳自动化研究所
导师	王正方
关键词	智能电网 需求响应 家庭能源管理 用电策略 电价预测
其他题名	Research on Home Energy Management Platform’Scheduling System
学位专业	计算机应用技术
中文摘要	随着现代信息技术在电网中的应用，智能电网以其经济、安全、高效等性能成为当前电力行业的研究热点，大量研究成果层出不穷，并成为未来电力系统的变革趋势。智能电网的一个重要特征就是强调用户信息和电网信息的双向流动，鼓励用户参与电网运行。这就为用户端需求响应的推广和实施提供了必要的技术基础。所谓用户端需求响应是指中断用户根据市场刺激信息，如电价信息或激励机制，主动改变自己的电力消费行为，调整需要运行的用电设备的使用时间，将其从用电高峰时刻转移到用电低谷时刻，从而提高电网的稳定性，减少发电端浪费，实现节能减排，同时降低了用户的电费支出。 文章首先就智能电网和需求响应的国内外发展现状进行了详细的调查研究，提出了家用电器调度模型及控制算法以辅助用户更好地参与需求响应。文章主要内包括以下几个方面： (1) 给出文章的研究背景和意义，指出课题是在能源短缺、浪费严重以及智能电网得到推广和实施的大前提下提出的；对用户端需求响应进行归类，需求响应主要包括基于激励型和基于价格型，并对这两种类型的需求响应进行了简要介绍，给出用户参与需求响应的过程，总结了用户参与需求响应需要解决的问题；归纳了国内外对需求响应和家庭能源管理系统方面的研究和进展，得出目前对家庭能源管理平台用电策略研究较少且没有形成商业化推广的结论；分析了家庭需求响应得到推广的决定因素；概况了常用的最优化求解算法；最后针对家庭能源管理平台系统进行了简要介绍。 (2) 调研了实时电价情况下电价预测的方法，以分析历史电价数据特点为前提，在前人提出的小波滤波理论、交叉验证理论、加权最小二乘支持向量机（Weighted LS-SVM, WLS-SVM）理论的基础上提出了组合小波多时段加权最小二乘支持向量机（Combination of Multi-period Wavelet Preprocessed Weighted LS-SVM, CMWPWLS-SVM）电价预测方法，并用此方法和小波-ARIMA模型、RBF神经网络及未经改进的WLS-SVM算法预测结果进行分析对比，结果验证了该方法的可行性； (3) 针对家用电器能源优化控制问题，根据家用电器的特征和使用情况，首先对家用电器的使用情况进行了合理的简化，规定了模型建立需满足的前提假设。针对不同家用电器的使用特征将家用电器进行分类，令不同类别的家用电器运行时必须满足的条件作为目标方程的约束条件，结合电价预测算法，建立了家用电器调度模型。除普通家用电器外，文章还针对储能设备进行分析，给出电器使用储能设备进行供电的经济前提。文章同时分别在有/无储能设备两种前提下给出对应的家用电器调度算法； (4) 文章给出了一个实际案例，案例选择了几种典型的家用电器，并对此案例进行仿真，仿真结果验证了家庭能源管理平台用电调度策略的有效性。
索取号	TM715/N47/2014
英文摘要	With the deployment of modern information technology in power grid, smart grid is becoming a new research hotspot of the power industry because of its economic, security, efficiency and other properties, a large number of research results on smart grid is emerging, and it is the future trend of the power system reform. One of the most important features of smart grid is the bidirectional flow of information between users and grid. Users are encouraged to participate in smart grid operation. The promotion of smart grid provides the must technology foundation of demand response. Demand response means the end users take the initiative to change their electricity consumption behavior or adjust the working time of electricity equipment to response the electricity price information or incentive mechanism. This moves electricity usage from peak time to low time. This transfer enhances the grid stability, reduces waste of generation side, achieves energy savings, while reduces the users’ electricity costs. According to the current researches of domestic and international development of smart grid and demand side response, this thesis proposed home appliances scheduling model and control algorithms to help family users participate the demand response better. This thesis mainly includes the following aspects: (1) Give the background and significance of this thesis, pointing out this research is proposed under these major premises: energy shortage, serious waste of the power industry as well as the implementation of smart grid; classify demand side response to two classes, incentive-based demand response and price-based demand response; give the procedure of users take part in demand response; summarize issues of users to participate in demand response need to be addressed; sum up the research results of consumer take part in demand response and home energy management system, then draw the conclusion that research on it is not very much, and there is no commercial usage of it; analysis the determinants to household demand response popularized; overview the most commonly used optimization algorithm; introduce the home energy management platform briefly; (2) Investigate the real-time electricity price forecasting methods, using the historic price data, based on previous theories such as wavelet filtering, cross-validation and weighted least squares support vector machine, this thesis proposed a new Combination of Multi-period Wavelet Preprocessed Weighted LS-SVM(CMWPWLS-SVM) model to predict future electricity price, then compared the forecasting result of the newly proposed method, wavelet-ARIMA model and RBF neural network, draw the conclusion that the newly proposed method is effective; (3) To solve the energy optimal problem for household appliances, provide some rational assumptions to simply the home appliance model. Classify home appliances to different groups based on their features, make these conditions of appliances in different groups be the constraints of the target equation. Combined with price forecasting algorithm, a home appliance scheduling model is built. In addition to common household appliances, energy storage equipment is also analyzed, giving the economic threshold while other appliances is charged by energy storage. Give home appliances scheduling algorithm under conditions whether contains energy storage condition or not; (4) An actual case was given to test the efficiency of this algorithm. Several typical household appliances was chosen, and the outcome of the simulated result show home appliances scheduling algorithm is effective.
语种	中文
产权排序	1
页码	63页
分类号	TM715
内容类型	学位论文
源URL	[http://ir.sia.ac.cn/handle/173321/14826]
专题	沈阳自动化研究所_工业控制网络与系统研究室
推荐引用方式 GB/T 7714	牛丽肖. 面向家庭能源管理平台的用电策略研究[D]. 中国科学院沈阳自动化研究所. 2014.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。