|
|||||
|
|
| 光伏充气膜温室自跟踪发电系统发电量预测 | |
| 引用本文： | 徐小力,刘秋爽,见浪護.光伏充气膜温室自跟踪发电系统发电量预测[J].农业机械学报,2012,43(Z1):305-310,299. |
| 作者姓名： | 徐小力 刘秋爽 见浪護 |
| 作者单位： | 1. 北京信息科技大学现代测控技术教育部重点实验室,北京,100192 2. 北京七星华创电子股份有限公司集成电路工艺设备研发中心,北京,101312 3. 冈山大学工学部自然科学院,冈山700-8530 |
| 基金项目： | 国家自然科学基金资助项目（50975020）;国家科技重大专项资助项目（2009ZX04014—101）;北京市引进国外技术重点资助项目（B201101010） |
| 摘 要： | 针对光伏充气膜温室自跟踪发电系统提出了一种加入天气预报信息的自适应变异粒子群神经网络的发电量预测算法.首先结合历史发电量数据和气象数据分析了影响光伏充气膜温室自跟踪发电系统发电量的主要因素,建立了加入天气预报的神经网络预测模型,并针对传统神经网络预测模型中基于梯度下降的BP算法收敛慢、易陷入局部最优、训练难收敛等问题,通过自适应变异粒子群算法改进了神经网络.该算法通过将变异环节引入粒子群优化算法,进行隔代进化找到局部最优解.实验结果表明所采用的自适应变异粒子群的神经网络预测算法的全局收敛性能得到了显著提高,能有效避免粒子群优化算法中的早熟收敛问题. |
| 关 键 词： | 光伏充气膜温室 自跟踪发电系统 发电量预测 自适应变异粒子群神经网络算法 |
Generating Capacity Prediction of Automatic Tracking Power Generation System on Inflatable Membrane Greenhouse Attached Photovoltaic |
|
| Xu Xiaoli,Liu Qiushuang and Mamoru Minami.Generating Capacity Prediction of Automatic Tracking Power Generation System on Inflatable Membrane Greenhouse Attached Photovoltaic[J].Transactions of the Chinese Society of Agricultural Machinery,2012,43(Z1):305-310,299. | |
| Authors: | Xu Xiaoli Liu Qiushuang and Mamoru Minami |
| Institution: | Beijing Information Science and Technology University;Beijing Sevenstar Electronics Co., Ltd.;Okayama University |
| Abstract: | A method which can forecast generating capacity of automatic tracking power system on inflatable membrane greenhouse attached photovoltaic was proposed based on the self-adaptive variation particle swarm neural network by adding with weather information. Firstly, through combining historical data of electricity production and meteorological data, the main factors of the impact on generating capacity of power generation system on inflatable membrane greenhouse attached photovoltaic was analyzed. Then, the neural network forecasting model was established by combining the weather forecast. The self-adaptive variation particle swarm algorithm was introduced to improve the training effect by tackling the problems of slowly converging, easily falling into local optimum, and difficultly converging existed in traditional neural network forecasting model based on gradient-descent BP algorithm. The neural network was optimized with adaptable mutation particle swarm optimization(AMPSO) algorithm. The mutation was put into particle swarm optimization(PSO) algorithm to find local optimal value. Experimental results showed that the entire convergence performance was significantly improved by adopting AMPSO and the premature convergence problem can be effectively avoided in PSO. |
| Keywords: | Inflatable membrane greenhouse attached photovoltaic Automatic tracking photovoltaic power generation system Prediction of generating capacity Self-adaptive variation particle swarm neural network algorithm |
| 本文献已被 万方数据 等数据库收录！ | |
| 点击此处可从《农业机械学报》浏览原始摘要信息 | |
| 点击此处可从《农业机械学报》下载免费的PDF全文 | |