Short-Term Hydroscheduling with Discrepant Objectives Using Multi-Step Progressive Optimality Algorithm1

Cheng, Chuntian, Jianjian Shen, Xinyu Wu, and Kwok-wing Chau, 2012. Short-Term Hydroscheduling with Discrepant Objectives Using Multi-step Progressive Optimality Algorithm. Journal of the American Water Resources Association (JAWRA) 48(3): 464-479. DOI: 10.1111/j.1752-1688.2011.00628.x Abstract: With increase in the number and total capacity of hydropower plants in power systems, optimality algorithms with a single objective are not suitable for optimizing the operation of complex hydropower systems to meet complex demands. Hydropower plants should prioritize discrepant objectives, such as peak regulation and maximizing generation during solving of optimal operation problems of hydropower systems. In this article, we present a multi-step progressive optimality algorithm (MSPOA) for the short-term hydroscheduling (STHS) problem to improve the quality of optimal solutions and enhance the convergence speed of progressive optimality algorithm (POA). In MSPOA, the original problem is first decomposed into a sequence of problems with the longer time steps. Next, the problem with the longest time step is solved, and the optimal solution is used as the initial solution for the problem with the second longest time step. This process proceeds until the original problem with the shortest time step is solved. The proposed discrepant-objective method and solution technique are tested for two types of hydroelectric systems. The results show that MSPOA can give better solutions and cost less time than POA due to enlarging feasible range of decision variables and reducing the number of computational stages. Discrepant objectives among hydropower plants can express the operation characteristics of complex hydropower systems more accurately than unique objective or multiple objectives.     

建立洪水灾情等级模型的实用方案

为检验已订的洪水灾情等级标准的合理性,提出了一种新的洪水灾情模型－逻辑斯谛曲线（ＬＯＧ）模型,它的灾级是连续的实数值。根据ＬＯＧ模型参数值可以分析各灾情指标值对灾级的影响程度,从而检验原订灾级标准的合理性;并给出了基于实码遗传算法的ＬＯＧ建模的实施方案。实例研究说明了这套方案是实用和通用的,在其它灾情等级评估中也具有广泛的应用价值。     

A Monitoring Network Design Procedure for Three-Dimensional (3D) Groundwater Contaminant Source Identification

Finding the location and concentration of contaminant sources is an important step in groundwater remediation and management. This discovery typically requires the solution of an inverse problem. This inverse problem can be formulated as an optimization problem where the objective function is the sum of the square of the errors between the observed and predicted values of contaminant concentration at the observation wells. Studies show that the source identification accuracy is dependent on the observation locations (i.e., network geometry) and frequency of sampling; thus, finding a set of optimal monitoring well locations is very important for characterizing the source. The objective of this study is to propose a sensitivity-based method for optimal placement of monitoring wells by incorporating two uncertainties: the source location and hydraulic conductivity. An optimality metric called D-optimality in combination with a distance metric, which tends to make monitoring locations as far apart from each other as possible, is developed for finding optimal monitoring well locations for source identification. To address uncertainty in hydraulic conductivity, an integration method of multiple well designs is proposed based on multiple hydraulic conductivity realizations. Genetic algorithm is used as a search technique for this discrete combinatorial optimization problem. This procedure was applied to a hypothetical problem based on the well-known Borden Site data in Canada. The results show that the criterion-based selection proposed in this paper provides improved source identification performance when compared to uniformly distributed placement of wells.     

赤潮藻类图像自动识别的研究

为了分析海水中所含有害藻的种类和数量,做到赤潮藻类生物发展的早期监测、预报,开发了一赤潮藻类图像计算机自动识别系统.运用图像处理技术提取藻类图像形态、纹理特征等,运用遗传算法进行特征选择.在此基础上用神经网络建立分类器,对藻类图像进行分类识别.结果表明,该系统能有效提高学习能力和分类性能,对引发赤潮的3种主要藻类达到了很好的分类识别,分析结果与人工计数识别结果相差较少.     

基于随机森林算法的滁州市空气质量预报研究

利用2017—2019年滁州市6种空气污染物质量浓度和AQI资料、气象要素及ECMWF数值预报作为输入参数,构建基于随机森林算法的污染物质量浓度和AQI预报模型,其中AQI预报按季节划分为4个模型。结果表明：6种污染物中O3的预报效果最好,相关系数为0.84,PM2.5、PM10和NO2相关系数分别为0.76、0.72和0.72,SO2和CO预报效果略差;基于分季节模型AQI的24 h预报结果TS评分为0.77,空报率（FAR）和漏报率（PO）分别为15%和12%,相较于非季节模型预报效果更好;春季和秋季的TS评分分别为0.88和0.86,相较于冬季和夏季预报效果更好。     

基于迭代退火算法的高压变频器功率单元频繁故障诊断方法研究

现有的变压器故障诊断方法较为复杂且计算冗余度较高,在高压变频器的功率单元频繁发生故障时难以高效地检测故障。为此,提出基于迭代退火算法的高压变频器功率单元频繁故障诊断方法。采用小波包分解方法提取高压变频器功率单元的电压信号特征熵,将该特征熵输入到支持向量机模型。使用迭代退火算法优化支持向量机的训练参数,并输出诊断结果。研究结果表明：该方法提取的高压变频器单元故障的平均冗余度最低至3.2%,平均诊断时间为15.1 ms,可实现高压变频器功率单元频繁故障的高效诊断。     

引信步进应力加速试验贮存寿命预测研究

目的 针对某机电引信加速寿命试验数据,采用传统统计分析方法存在计算量大、寿命预测精度难以保证的问题,开展与智能算法相结合的引信贮存寿命预测研究。方法 针对步进应力加速寿命试验数据,采用贝叶斯理论的环境因子法,对各级应力下的贮存时间进行折合计算。利用进化策略对粒子群算法进行改进,进而对所建立的BP神经网络预测模型的全局参数进行调整和优化,突破传统方法的局限。将折合后的试验时间、样本量、应力水平作为网络输入,失效数作为输出,来预测引信贮存寿命。结果 利用训练好的 BP神经网络预测引信在正常应力水平下的失效数,计算其贮存可靠度。在迭代402次后,模型找到最优解,且预测误差在1%以内。结论 步进应力加速寿命试验与智能算法相结合的方法计算过程简单,预测精度较高,可有效提高引信贮存寿命的预测精度。     

基于随机搜索优化XGBoost的瓦斯涌出量预测模型*

为防治瓦斯灾害,解决井下瓦斯涌出量在预测过程中因影响因素繁多带来的精度较低问题,提出1种基于套索(Lasso)回归与随机搜索优化极限梯度提升(XGBoost)的模型进行瓦斯涌出量预测。以沈阳某煤矿综采面瓦斯涌出量历史数据为例,综合考虑影响瓦斯涌出量的影响因素。首先利用Lasso回归提取对瓦斯涌出量有重要影响的特征数据,作为预测输入;采用随机搜索算法对XGBoost模型4种主要参数进行寻优,选取最优参建立预测模型获得预测指标并分析比较其他模型。研究结果表明:Lasso回归筛选出的影响因素结合随机搜索获得的最优参数组合优化XGBoost比其他模型预测精度更高,平均相对误差为1.53%,均方根误差为0.140 3 m3/min,希尔不等系数为0.013 2,研究结果可为现场瓦斯管理提供参考依据。     

用投影寻踪分类模型进行环境监测优化布点

为处理各单项环境监测指标优选结果的不相容问题,提出用投影寻踪分类模型(PPC-RAGA)进行环境监测优化布点的新方法.利用该模型可把监测点多维环境指标值综合成一维投影值,投影值越大表示该监测点的环境综合质量越差,根据投影值的大小就可对环境监测点样本集进行合理分类.采用实码加速遗传算法进行PPC-RAGA建模,简化了投影寻踪技术的实现过程,克服了目前投影寻踪技术计算复杂、编程实现困难的缺点,为投影寻踪技术在各种系统工程中的广泛应用提供了新的工具.应用实例表明,直接由样本数据驱动的PPC-RAGA模型用于环境监测优化布点,简便可行,计算结果稳健,适用性和可操作性强,在非线性、非正态高维数据分类中具有广泛的应用价值.     

反演同化技术在激光雷达中的应用研究

为增强模式预报准确性,提高激光雷达数据在化学模式中的应用水平,建立激光雷达同化系统,将激光雷达数据引入模式,利用WRF-Chem模拟2016年12月6日～10日一次污染过程,将激光雷达监测的气溶胶光学深度(AOD)反演成近地面PM2.5浓度,通过三维变分同化技术订正模式原始场,调整模式预报结果.实验对比发现,反演同化后...