基于随机森林的多站融合项目优先业务筛选机制

在“双循环”新发展格局和“电力数字化”战略中顺势发展“多站融合”项目。基于多站融合布局,提出了一个多站融合项目区域性优先业务筛选的技术方法,从经济发展与能源产业视角,依据各区域属性特征将中国31个省市划分为四类区域,并结合多站融合项目业务清单所梳理的三大类业务范畴,利用随机森林算法的大数据分析法,通过选择区域数据样本,形成了多站融合项目区域性优先业务筛选机制,进而依据优先业务匹配组合预测,实现了各区域优先发展业务的精准甄别和科学选择。实证分析验证了基于随机森林算法的多站融合项目区域性优先业务筛选模型的科学合理性以及有效可靠性,以期为多站融合推进提供有益参考。     

基于改进U-net和CNN的绝缘子自爆检测方法研究

针对绝缘子自爆故障人工检测效率低,成本高的问题,基于改进U-net和卷积神经网络（CNN）模型,提出一种可有效识别绝缘子自爆故障的双阶段目标检测算法。首先,在语义分割阶段使用改进U-net模型,通过翻倍提高图像分辨率的方法有效提高图像分割精度。其次,在图像分类阶段提出更适合所提问题且有效提高分类准确度的新型CNN模型。最后,使用无人机拍摄的绝缘子图片为实验数据进行实验。实验结果表明所提算法识别精度较高。     

A power fluctuation evaluation method of PV plants based on RankBoost ranking

Fluctuation evaluation is an important task in promoting the accommodation of photovoltaic (PV) power generation. This paper proposes an evaluation method to quantify the power fluctuation of PV plants. This consists of an index system and a ranking method based on the RankBoost algorithm. Eleven indices are devised and included in the index system to fully cover diverse fluctuation features. By handling missing and invalid data effectively, the ranking method fuses multiple indices automatically and provides a systematic and comprehensive comparison of power fluctuation. Simulation results based on power data from six PV plants indicate that the evaluation list obtained by the RankBoost ranking method is better represented and more comprehensive than that derived by the equal weight method.     

Impact of communication time delays on combined LFC and AVR of a multi-area hybrid system with IPFC-RFBs coordinated control strategy

In this paper, the impact of communication time delays (CTDs) on combined load frequency control (LFC) and automatic voltage regulation (AVR) of a multi-area system with hybrid generation units is addressed. Investigation reveals that CTDs have significant effect on system performance. A classical PID controller is employed as a secondary regulator and its parametric gains are optimized with a differential evolution - artificial electric field algorithm (DE-AEFA). The superior performance of the presented algorithm is established by comparing with various optimization algorithms reported in the literature. The investigation is further extended to integration of redox flow batteries (RFBs) and interline power flow controller (IPFC) with tie-lines. Analysis reveals that IPFC and RFBs coordinated control enhances system dynamic performance. Finally, the robustness of the proposed control methodology is validated by sensitivity analysis during wide variations of system parameters and load.     

A Web Server for Designing Molecular Switches Composed of Two Interacting RNAs

The programmability of RNA–RNA interactions through intermolecular base-pairing has been successfully exploited to design a variety of RNA devices that artificially regulate gene expression. An in silico design for interacting structured RNA sequences that satisfies multiple design criteria becomes a complex multi-objective problem. Although multi-objective optimization is a powerful technique that explores a vast solution space without empirical weights between design objectives, to date, no web service for multi-objective design of RNA switches that utilizes RNA–RNA interaction has been proposed. We developed a web server, which is based on a multi-objective design algorithm called MODENA, to design two interacting RNAs that form a complex in silico. By predicting the secondary structures with RactIP during the design process, we can design RNAs that form a joint secondary structure with an external pseudoknot. The energy barrier upon the complex formation is modeled by an interaction seed that is optimized in the design algorithm. We benchmarked the RNA switch design approaches (MODENA+RactIP and MODENA+RNAcofold) for the target structures based on natural RNA-RNA interactions. As a result, MODENA+RactIP showed high design performance for the benchmark datasets.     

Fast network intrusion detection system using adaptive binning feature selection

Aiming at the problems of the low detection rate of traditional intrusion detection systems and the long training and detection time of intrusion detection systems based on deep learning,an adaptive binning feature selection algorithm using the information gain is proposed,which is combined with LightGBM to design a fast network intrusion detection system.First,the original data set is preprocessed to standardize the data;then the redundant features and noise in the original data are removed through the adaptive binning feature selection algorithm,and the original high-dimensional data are reduced to the low-dimensional data,thereby improving the accuracy of the system and reducing the training and detection time;finally,LightGBM is used for model training on the training set selected by the characteristics to train an intrusion detection system that can detect attack traffic.Through verification on the NSL-KDD data set,the proposed feature selection algorithm only takes 27.35 seconds in feature selection,which is 96.68% lower than that by the traditional algorithm.The designed intrusion detection system has an accuracy rate of 93.32% on the test set,and its training time is low.Compared with the existing network intrusion detection system,the accuracy rate of the proposed system is higher,and its model training speed is faster.     

Optimization algorithm for ultra-sparse arrays in regional energy focusing

The setting of ultra-sparse arrays is one of key factors for the performance of regional energy focusing.This paper focuses on one of its typical applications:Precision Electronic Warfare by UAV,and proposes an optimization algorithm for ultra-sparse arrays to maximize the performance of regional energy focusing.First,we establish the joint optimization model of the interference waveform and the positions of elements.Second,particle swarm optimization (PSO) is adopted to iterate the positions of elements.Then,the target value of the subproblem is solved for the fitness criterion,as the current array is known in each step of the iteration.Finally,we get the optimal solution of the interference waveform and the ultra-sparse array by this method.Numerical results indicate that the algorithm can optimize the array effectively,and improve the indicators of regional energy focusing obviously.Whether the localization errors of elements exist or not,the algorithm provides a better performance than existing algorithms which only design the interference waveform.     

“三全育人” 格局下高校辅导员队伍角色定位与转型论析

高校围绕立德树人根本使命,正在积极构建全员全过程全方位育人格局。在此形势下,作为高校思政工作一线骨干力量的辅导员队伍面临新的角色转型,找准自己在高校"大思政"中的新定位,要成为全员育人的骨干力量,做思想政治教育和管理工作的组织者与育人资源的协调者;要成为全过程育人的融通力量,做学生全面成长的指导者与持续发展的推动者;要成为全方位育人的串联力量,做"十大"育人体系落地的终端实施者与成效评估者。在三全育人的新格局下,辅导员队伍要与校内外育人队伍一起,协同攻关、形成合力,共同推进一体化育人机制。     

基于相对熵的KNN文本分类方法的研究

摘要:为提高处理文本相似度的效果,提出了一种基于相对熵度量文本差异的KNN算法.该算法首先对文本进行预处理(分字与删去停用字)和构建特征字字典; 然后计算训练集中所有文本特征字的概率,并组成训练集(特征字概率矩阵); 最后计算预测文本的特征字概率向量,并通过计算和统计K个预测文本与训练集文本间相对熵最小的文本类别个数后将数目最多的类别作为测试样本的类别.实验结果表明,该算法的分类效果不仅显著优于传统KNN、SVM、Decision Tree、朴素Bayes算法的分类效果,且在小样本数据情况下     

惩罚函数结合遗传算法的PID参数优化

针对传统优化算法对PID参数优化时,由于目标函数选择不当出现的超调量过大,从而引起系统的反应速度变慢和稳定性变差的问题,提出以传统遗传算法为基础,引入惩罚函数的方法,根据系统中出现的超调量来构造新的目标函数实现对PID的参数优化。实验表明,该方法可以有效减少系统中的超调量,加快优化以及运行速度的同时保证系统的稳定性,实现对PID参数的优化。