共查询到19条相似文献,搜索用时 343 毫秒
1.
2.
3.
4.
本文阐述了支持向量回归SVR模型的搭建方法,针对一栋虚拟的三维办公建筑,采用能耗模拟的方式获取空调系统能耗作为训练集,并通过模型参数的设置及6个特征数的选取等一系列步骤建立了SVR能源预测模型,最后通过测试集对该模型的性能进行了评价。结果表明,经数据样本训练后所获得模型的预测值与真实值之间的方均误差MSE为0.015,该SVR模型的预测精度达到合格要求,可以用来指导预测未来能耗,为建筑节能领域提供了方法与参考。 相似文献
5.
6.
介绍支持向量机的原理和支持向量回归模型,提出支持向量回归(SVR)模型的城市燃气短期负荷预测方法。探讨输入样本数据的选择和预处理方法、核函数和支持向量机参数的选择,结合某城市燃气日负荷数据进行燃气短期负荷预测。与BP神经网络预测方法相比,支持向量回归模型预测方法用于小样本情况下的燃气短期负荷预测精度略高。 相似文献
7.
8.
在供水管网水质评价体系中,节点水龄和余氯浓度是评价供水管网水质情况的两个重要指标。本文以某区供水管网为例进行分析,在建立水力模型的基础上,通过Water GEMS建模软件构建水龄分析模型和余氯浓度分析模型,对管网进行水力和水质的动态模拟。通过对管网不同点的水龄和余氯浓度模拟结果变化的分析,总结出了管网节点水龄突变型、周期性、线性递增的变化规律和管网整体余氯浓度呈衰减变化的规律;并归纳出了某一节点水龄与余氯的负相关关系,为评价管网水质情况及管网的科学化管理提供了科学的依据。 相似文献
9.
从管网模拟仿真的微观和宏观型入手,详细分析管网在正常及故障状态下的微观特性,在此基础上首次提出了管网故障局部停水后压力预测宏观数学模型,利用管网平差结果数据为宏观模型的拟合提供样本,采用“多元线性回归”方法进行了拟合、分析,并检验了宏观模型的显著性及拟合精确度。编制了一系列计算程度,对管网进行实例计算和分析,结果令人满意。 相似文献
10.
针对余氯量在供水系统内非线性变化的特性,建立了PSO-SVM与BP神经网络组合模型对管网末端余氯进行预测分析。该模型通过粒子群优化算法(PSO),对SVM的特性参数进行优化;采用BP神经网络对模型进行残差修正。通过对单一的BP模型和SVM模型、组合模型的预测精度进行分析。结果表明:组合模型预测比BP和SVM单一预测均方误差分别降低了62.30%、75.29%,平均相对误差降低了55.03%、54.27%。综上所述,该模型具有强大的非线性拟合能力,预测精度高,运行稳定性强,对供水企业控制余氯的投加量和设置二次加氯点有一定的指导作用。 相似文献
11.
将支持向量回归(SVR)算法引入岩土工程数值计算模型参数的辨识中可以充分发挥SVR算法的小样本、泛化性好和全局最优化的优点。但现阶段标准的SVR算法只能解决一维输出变量的回归问题,这就使其在反分析领域的应用受到限制。引入一种改进的SVR算法,这种算法通过将多维输出变量回归转化为多层标准一维输出变量回归来解决这个问题,并与十进制编码的遗传算法相结合,形成改进的GA-SVR算法,用遗传算法搜索最优的SVR模型参数以建立最优的待辨识参数与位移之间的非线性映射关系,然后用遗传算法进行待辨识参数的最优辨识。为对比这种改进GA-SVR算法的效果,将遗传算法与BP神经网络相结合,形成GA-BP算法且编制相应的计算程序。将这两种算法运用于同样的隧道工程三维弹塑性模型参数的智能辨识,数值算例表明改进的GA-SVR算法较GA-BP算法可以取得更高的辨识精度和更好的计算效率,可运用于类似岩土工程计算参数的辨识。 相似文献
12.
为有效控制供水管网中的三卤甲烷浓度,考察了以太湖和阳澄湖为双水源的管网水中三卤甲烷的变化趋势。采集S市分别经过A水厂(以太湖为水源)和B水厂(以阳澄湖为水源)处理后的供水管网水样,探讨三卤甲烷和余氯浓度随供水距离的变化规律,并进行了三卤甲烷-余氯和三卤甲烷-供水距离的线性回归分析。结果表明,在A、B水厂单独供水管网中,三卤甲烷随供水距离的增加而升高,余氯浓度随供水距离的增加而降低;在A、B水厂同时供水的对置供水管网中,三卤甲烷浓度由A水厂到B水厂呈现先升高后降低的变化规律,而余氯浓度则呈现先降低后增加的变化规律,对置供水管网中10 km处为水力交接点,此处余氯浓度最低,三卤甲烷浓度最高。经线性回归分析发现,三卤甲烷浓度与余氯浓度和供水距离之间存在显著的相关性,并且温度对模型的影响较大。 相似文献
13.
Conventional methods of calculating chlorine demand are dose dependent, making intercomparison of samples difficult, especially in cases where the samples contain substantially different concentrations of dissolved organic carbon (DOC), or other chlorine-consuming species. Using the method presented here, the values obtained for chlorine demand are normalised, allowing valid comparison of chlorine demand between samples, independent of the chlorine dose. Since the method is not dose dependent, samples with substantially differing water quality characteristics can be reliably compared. In our method, we dosed separate aliquots of a water sample with different chlorine concentrations, and periodically measured the residual chlorine concentrations in these subsamples. The chlorine decay data obtained in this way were then fitted to first-order exponential decay functions, corresponding to short-term demand (0-4h) and long-term demand (4-168 h). From the derived decay functions, the residual concentrations at a given time within the experimental time window were calculated and plotted against the corresponding initial chlorine concentrations, giving a linear relationship. From this linear function, it was then possible to determine the residual chlorine concentration for any initial concentration (i.e. dose). Thus, using this method, the initial chlorine dose required to give any residual chlorine concentration can be calculated for any time within the experimental time window, from a single set of experimental data. 相似文献
14.
在深入分析评价指标的基础上,引进一种改进的SVR算法,将这种改进SVR算法引入建设项目招标的评标工作,收集了大量评标实例用于改进的SVR网络训练,建立了建设项目评标的智能模型,并用一个实际工程招标的评标结果验证了该模型的可靠性,为建设项目评标提供了一种新的方法。 相似文献
15.
16.
介绍了用双硫腙分光光度法测定水中余氯的新方法,阐述了其测定试验过程,根据试验结果显示,其结论符合国家标准,表明该方法可用于自来水等样品中余氯含量的测定,结果比较准确。 相似文献
17.
Maintaining a chlorine residual is a major disinfection goal in many water distribution systems. A suitable general model of chlorine decay in the transported bulk water is an essential component for efficiently modelling chlorine concentration in distribution systems. The two-reactant model meets basic suitability criteria, including accurate prediction of chlorine residual over hundreds of hours, commencing with chlorine concentration 0-4 mg/L. This model was augmented with an equation that increases the decay coefficients with temperature according to Arrhenius theory. The augmented model was calibrated against decay-test data sets to obtain a single invariant set of parameters for each water. Model estimates of chlorine residuals over time closely matched decay-test data, over the usual operating ranges of initial chlorine concentration (1-4 mg/L) and temperature (3.5-28 °C). When the augmented model was fitted to partial data sets, it also predicted the data reserved for validation very well, suggesting that this model can accurately predict the combined effect of initial chlorine concentration and temperature on chlorine bulk decay in distribution systems, using a single set of invariant parameters for a given source water. 相似文献
18.
19.
The effect of chlorine on the analysis of three organic compounds (geosmin, 2-methylisoborneol (2-MIB) and methyl tert-butyl ether (MTBE)) in drinking water is elucidated. Three fibers for solid-phase microextraction (SPME) were employed for the extraction of the organic compounds from drinking water samples with and without free residual chlorine present. A gas chromatograph coupled with a mass spectrometer was used to analyze the compounds trapped by the fibers. The presence of chlorine substantially reduces the observed geosmin, 2-MIB, and MTBE concentrations. Depending on the analyte and chlorine concentrations, an experimental error of 10-70% may be observed due to the presence of free residual chlorine. The impact is larger for lower organic compound concentrations, and under higher residual chlorine conditions. To counteract the effect from residual chlorine, sodium thiosulfate was used to dechlorinate the water. After dechlorination the experimental error was less than 10%, suggesting that dechlorination is necessary when applying SPME for the extraction of organic compounds from chlorinated drinking water. 相似文献